Acr Avaya PDF

Verint ContactStore for
Communication Manager
Release 7.8
Planning, Installation and Administration
Guide
Issue 5
May, 2008
© 2003-2008 Verint Systems, Inc. availability of the linked pages.
All Rights Reserved.
Trademarks
Notice
Avaya and Communication Manager are trademarks of Avaya Inc.
While reasonable efforts were made to ensure that the information in this
eQuality® is a registered trademark, and Quality is a trademark of
document was complete and accurate at the time of printing, Verint
Systems, Inc. and Avaya Inc. can assume no liability for any errors. Verint Systems, Inc.
Changes and corrections to the information in this document may be
All other trademarks identified by the ® or ™ are registered trademarks
incorporated in future releases.
or trademarks of their respective owners.
Preventing toll fraud
Third-party software license agreements
"Toll fraud" is the unauthorized use of your telecommunications system
by an unauthorized party (for example, anyone who is not a corporate This computer program is protected by U.S. and international copyright
employee, agent, subcontractor, or person working on your company's laws, patent laws, and other intellectual property laws and treaties.
behalf). Be aware that there may be a risk of toll fraud associated with Unauthorized use, duplication, publication and distribution of all or any
your system and that, if toll fraud occurs, it can result in substantial portion of this computer program are expressly prohibited and will be
additional charges for your telecommunications services. prosecuted to the maximum extent provided by law. Your rights in this
computer program are limited to the license rights granted under the
Avaya fraud intervention license agreement executed by you in hardcopy form (or if none, by
If you suspect that you are being victimized by toll fraud and you need acceptance of the clickwrap terms included with this computer program).
technical assistance or support, call Technical Service Center Toll Fraud If needed, please contact your vendor for an additional copy of those
Intervention Hotline at +1-800-643-2353 for the United States and terms. All other rights, title and interest are expressly restricted and
Canada. retained by Verint Systems, Inc. and its licensors.
Providing telecommunications security The following open source applications are included with this computer
Telecommunications security (of voice, data, and video communications) program ("Open Source"), with ownership in those applications as
is the prevention of any type of intrusion to (that is, either unauthorized or indicated below.
malicious access to or use of) your company's telecommunications
Ant, Axis, log4j and Tomcat are each copyright © 2000 - 2003 The
equipment by some party. Your company's "telecommunications
Apache Software Foundation. All rights reserved. Licensed under the
equipment" includes both this Avaya product and any other voice/data/
Apache License, Version 2.0, which can be obtained at http://
video equipment that could be accessed via this Avaya product (that is,
www.apache.org/licenses/LICENSE-2.0.
"networked equipment"). An "outside party" is anyone who is not a
corporate employee, agent, subcontractor, or person working on your PostgreSQL is portions copyright © 1996 - 2005, The PostgreSQL Global
company's behalf. Whereas, a "malicious party" is anyone (including Development Group, and portions copyright © 1994, The Regents of the
someone who may be otherwise authorized) who accesses your University of California. All rights reserved. A copy of the PostgreSQL
telecommunications equipment with either malicious or mischievous license can be obtained at http://www.postgresql.org/about/licence.
intent. Such intrusions may be either to/through synchronous
(time-multiplexed and/or circuit-based) or asynchronous (character-, jcifs is copyright © 2004 The JCIFS Project. All rights reserved. Licensed
message-, or packet-based) equipment or interfaces for reasons of: under the terms of the GNU Lesser General Public License as published
• Use (of capabilities special to the accessed equipment) by the Free Software Foundation, which can be obtained at http://
• Theft (such as, of intellectual property, financial assets, or www.gnu.org/licenses/licenses.html.
toll-facility access)
JoeSNMP is copyright © 2002 - 2003 Blast Internet Services, Inc, and
• Eavesdropping (privacy invasions to humans)
copyright © 1999 - 2001 Oculan Corp. Copyrights for modified and
• Mischief (troubling, but apparently innocuous, tampering)
included code are included in the individual source files. All rights
• Harm (such as harmful tampering, data loss or alteration,
reserved. Licensed under the terms of the GNU Lesser General Public
regardless of motive or intent)
License as published by the Free Software Foundation, which can be
Be aware that there may be a risk of unauthorized intrusions associated obtained at http://www.gnu.org/licenses/licenses.html.
with your system and/or its networked equipment. Also realize that, if
A license in each Open Source software application is provided to you in
such an intrusion should occur, it could result in a variety of losses to your
accordance with the specific license terms specified above. EXCEPT
company (including, but not limited to, human and data privacy,
WITH REGARD TO ANY WARRANTIES OR OTHER RIGHTS AND
intellectual property, material assets, financial resources, labor costs, and
OBLIGATIONS EXPRESSLY PROVIDED DIRECTLY TO YOU FROM
legal costs).
VERINT, ALL OPEN SOURCE SOFTWARE IS PROVIDED "AS IS'' AND
Responsibility for your company's telecommunications security ANY EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
The final responsibility for securing both this system and its networked
AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN
equipment rests with you, an Avaya customer's system administrator,
NO EVENT SHALL THE OWNERS OF THE OPEN SOURCE
your telecommunications peers, and your managers. Base the fulfillment
SOFTWARE OR ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT,
of your responsibility on acquired knowledge and resources from a
INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
variety of sources, including, but not limited to:
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
• Installation documents
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
• System administration documents
USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
• Security documents
HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
• Hardware-/software-based security tools
WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
• Shared information between you and your peers
NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
• Telecommunications security experts
USE OF THE OPEN SOURCE SOFTWARE, EVEN IF ADVISED OF
To prevent intrusions to your telecommunications equipment, you and THE POSSIBILITY OF SUCH DAMAGE.
your peers should carefully program and configure your Avaya-provided
telecommunications systems and their interfaces. Your Avaya-provided Avaya support
software applications, as well as their underlying hardware/software Avaya provides a telephone number for you to use to report problems or
platforms and interfaces. Any other equipment networked to your Avaya to ask questions about your contact center. The support telephone
products. number is 1-800-242-2121 in the United States. For additional support
telephone numbers, see the Avaya Web site:
Warranty
http://www.avaya.com
Avaya Inc. provides a limited warranty on this product. Refer to your
sales agreement to establish the terms of the limited warranty. In Select Support, and then select Escalation Lists. This Web site includes
addition, Avaya's standard warranty language, as well as information telephone numbers for escalation within the United States. For escalation
regarding support for this product, while under warranty, is available telephone numbers outside the United States, select Global Escalation
through the following Web site: List.
http://www.avaya.com/support
Link disclaimer
Avaya Inc. is not responsible for the contents or reliability of any linked
Web sites and does not necessarily endorse the products, services, or
information described or offered within them. We cannot guarantee that
these links will work all of the time and we have no control over the
Verint ContactStore for Communication Manager
Release 7.8
Planning, Installation and Administration Guide
About This Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Intended audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
.
Summary of information included in this guide . . . . . . . . . . . . . . . . . . . . . 12
Conventions used in this guide . . . . . . .. . . . . . . . . . . . . . . . . . . . . . 13
Additional references . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . 14
Introduction . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . 16
What’s New in Version 7.8 . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . 17
Server components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Verint ContactStore for Communication Manager Server . . . . . . . . . . . . . . . . 18
Optional Server Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
End-User tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Quality Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Bulk Search and Replay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Administration Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Recording Functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Sampled Recording for Quality Assessment . . . . . . . . . . . . . . . . . . . . . . 25
Bulk Recording Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Ad-hoc or Occasional Recording Modes . . . . . . . . . . . . . . . . . . . . . . . . 27
Replay Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Live Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Recording Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Service Observe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Single-step Conference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Conference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Beep Tone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
International support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Liability . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Audio Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
G.711 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Issue 5 May, 2008 3

Contents
G.729A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Storage Requirements . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 43
Storage at Each Recorder . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 43
Quality Monitoring. . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 44
Central Database Storage . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 45
Archive Call Storage . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 45
Backup Storage . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 45
Server Platform . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 46
Sizing . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 46
DVD+RW Drive . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 49
Network Issues. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 51
Load . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 51
Ports Used . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 51
Network Address Translation Routing . . . . . . . . . . . . . . . . . . . . . . . . . 51
Recorder Licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Server Licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Port Licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Other Licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Timed Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Avaya system prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Communication Manager . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Gateway Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
AE Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Expansion Interface Boards (TN570) . . . . . . . . . . . . . . . . . . . . . . . . . . 56
C-LAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
VoIP Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Multi-Connect Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
DMCC (IP_API_A) Licenses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
TSAPI Simultaneous User Licenses . . . . . . . . . . . . . . . . . . . . . . . . . . 60
VoIP Network Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Bulk Recording System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Sampled Quality Monitoring (only) System . . . . . . . . . . . . . . . . . . . . . . . 62
Bulk Recording + Quality Monitoring System . . . . . . . . . . . . . . . . . . . . . . 63
Large Bulk Recording Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Integrating with other systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Supplementary Tagging of Bulk Recordings . . . . . . . . . . . . . . . . . . . . . . 67
Explicit External Control of Bulk Recording . . . . . . . . . . . . . . . . . . . . . . . 67
Enhanced Quality Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Avaya System Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Communication Manager Configuration . . . . . . . . . . . . . . . . . . . . . . . . 71
AE Server Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
TSAPI Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Test Phonesets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Order in which to Install Applications . . . . . . . . . . . . . . . . . . . . . . . . . . 80
4 Witness ContactStore for Communication Manager Planning, Installation and Administration Guide
Contents
Platform Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Operating System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
DVD Drive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Network Connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Installing Verint ContactStore for Communication Manager . . . . . . . . . . . . . . . . 86
Installing Verint eQuality Balance (V5) . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Installing Verint eQuality Balance (V7) . . . . . . . . . . . . . . . . . . . . . . . . . . 89
If using selective quality recording . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
If using "ContactStore Plus" style quality recording . . . . . . . . . . . . . . . . . . . 89
Installing Viewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Installing Viewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Create an MMC Console . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Configuring Viewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Configuring Verint ContactStore for Communication Manager . . . . . . . . . . . . . . 93
Installing Archive . . .
. . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . 94
Overview . . . . . . . . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . 96
Accessing the System . . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . 97
URL . . . . . . . . . . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . 97
Initial User Account .
. . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . 97
Key Points . . . . .. . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . 98
Licensing . . . . . . . . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . 99
Terminology. . . . . . . .. .. . . .. . . . . . . . . . . . . . . . . . . . . . . . 99
Obtaining a License Activation Key . . .. . . . . . . . . . . . . . . . . . . . . . . . 100
Adding additional licenses . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . 101
Reinstalling on the same PC . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . 101
Reinstalling the Recorder on a new PC . . . . . . . . . . . . . . . . . . . . . . . . . 101
Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Securing the System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Windows Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
System Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
System Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Port Allocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Common Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Assigning Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
On Demand Recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Meeting Recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Station Bulk Recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Station Executive Recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Conferenced Recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Quality Server Ports (for Selective Quality Recording) . . . . . . . . . . . . . . . . . . 130
Unify or Externally Controlled Ports . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Phone Replay Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Live Monitor Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Issue 5 May, 2008 5

Contents
Search and Replay . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 136

Search and Replay Access Rights . .. . . . . . . . . . . . . . . . . . . . . . . . . 136
ActiveX Control Download . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 138
Modify Default Behavior . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 139
Backup/Restore . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 141
Application . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 141
Backing up the Database . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 141
Restoring data to a new PostgreSQL database . . . . . . . . . . . . . . . . . . . . . 142
Backing up Voice Recordings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Distributing User Instructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Those Using Recording or Live Monitor . . . . . . . . . . . . . . . . . . . . . . . . 145
Those entitled to replay calls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Configuring Avaya Support Remote Access . . . . . . . . . . . . . . . . . . . . . . . 147
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Status Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Alarms and Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
System Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Port States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Peak Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Audit Trail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Local Archive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
DVD+RW media control and care . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Loading a disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Changing disks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Labeling a disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Preventative Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Daily . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
Weekly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
Monthly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Every Six Months . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Use of SSL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Windows Domain Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Blocking Replay from the Recorder . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
Changing Passwords . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Linux User Accounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Postgres Database Owner. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Recorder Account on eWare/Viewer SQL Server . . . . . . . . . . . . . . . . . . . . 168
Encrypted File Storage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
PCI Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
Properties File . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
Standby Server. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
Central Replay Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Installation . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Configuration . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
User Defined Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Storing UDFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Contents
Search and Display UDFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

UDF Support in Viewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Viewer Compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Customizing Search and Replay . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
Default Configuration Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
Custom Layout Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
Internationalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
Usage Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Enabling the Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Accessing through URL: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Accessing the Usage report in a log file . . . . . . . . . . . . . . . . . . . . . . . . 192
.
Selective Record Barring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Configuration . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Limitations . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
Recording files . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
WAV files . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
XML files . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
Internal Database . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Recording details . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Configuration details . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
Recorder Interfaces . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
HTTP Interfaces Offered.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
eQuality Balance V5 or V7 in Selective Recording Mode . . . . . . . . . . . . . . . . 201
eQuality Balance V7 in "ContactStore Plus" mode. . . . . . . . . . . . . . . . . . . . 201
Other Recorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
External Control Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
Database Upload Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
Hints and Tips . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Where to Look for Clues . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Determining Current Version .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Application Logs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Tomcat Logs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Remote Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Specific Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
System Administration page problems . . . . . . . . . . . . . . . . . . . . . . . . . 208
Connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Search and Replay problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
Recording Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
Live Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Alarms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
Alarms Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
When to Use External Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
Issue 5 May, 2008 7

Contents
This Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234

Port Allocations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
Master + Slave Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
Java API Toolkit . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 237
TCP/IP Protocol Overview . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 238
Connection Method . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 238
General Protocol Specification . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 238
Channel Identification . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 239
XML Tagging . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Basic Call Tagging . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Fallback Mode . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . 241
Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
Third-party CTI Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
Additional Call Tagging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
TCP/IP Message Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Redundant SAN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
Duplicated recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
Standby Recorder Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
Prerequisites for high availability .. . . . . . . . . . . . . . . . . . . . . . . . . . . 254
Standby recorder licensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
Automated Configuration Shadowing . . . . . . . . . . . . . . . . . . . . . . . . . . 256
Manual Configuration of Standby . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
Known limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Distributed Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Supported failure modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Standby recorders and Unify/External Control . . . . . . . . . . . . . . . . . . . . . 260
Supported Master/Standby Topologies . . . . . . . . . . . . . . . . . . . . . . . . . 260
Mode of operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Standby configuration (automatic) . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Standby configuration (manual) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Power-On . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Standby mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Failure Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Disk Space Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Active mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
Return to Standby mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
Switchover Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
Restoring the Primary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Comparison with hardware switch-over units . . . . . . . . . . . . . . . . . . . . . . 280
Standby Recorder Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Configuration Differences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Upgrade Process for Primary/Standby . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
Modes of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
Principles of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
High Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Contents
Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Programmer’s Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
Enable(in string station, in SeqCouple tags) . . . . . . . . . . . . . . . . . . . . . . 290
Disable(in string station) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Start(in string station) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Stop(in string station) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Tag(in string station, in string inum, SeqCouple tags) . . . . . . . . . . . . . . . . . . 291
Assign() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Principles of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Choosing the dialler version . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Changing the log level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Controlling the service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Automatic restart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Disks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
NICS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
DVD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Soundcard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Kickstart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Installing Unlimited Strength Encryption . . . . . . . . . . . . . . . . . . . . . . . . . 302
Installing a Signed SSL Certificate . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Selecting a Certificate Authority (CA) . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Backing up the Keystore file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Creating the new Certificate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Generating a Certificate Signing Request . . . . . . . . . . . . . . . . . . . . . . . . 305
Importing the CA's certificates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305
Backing up the keystore file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
Changing Tomcat Port Numbers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
Encrypting Properties File entries . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
Configuring Viewer and Archive to use HTTPS . . . . . . . . . . . . . . . . . . . . . . 309
Importing the public certificates into Viewer and Archive . . . . . . . . . . . . . . . . . 309
Configure Viewer to use https and a secure port . . . . . . . . . . . . . . . . . . . . 310
Configure Archive to use https . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Configuring Archive for a Key Management Server . . . . . . . . . . . . . . . . . . . . 313
Configuring Archive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
Signing and Delivering the certificate . . . . . . . . . . . . . . . . . . . . . . . . . . 313
Installing the Certificate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
Issue 5 May, 2008 9

Contents
About This Guide
The Verint ContactStore for Communication Manager Planning, Installation and

Administration Guide provides details of the Verint ContactStore for Communication
Manager system, as well as recommended and required components.
Issue 5 May, 2008 11

About This Guide
Intended audience
This guide is designed for use by:
● I.T. managers
● Installers
● System Administrators
Summary of information included in this guide

The following table provides information about this guide.
Chapter Title Description
Chapter 1: This chapter provides an overview of the design options

System Overview for Verint ContactStore for Communication Manager.
Chapter 2: This chapter gives details of the prerequisites for Verint

Planning and Prerequisites ContactStore for Communication Manager.
Chapter 3: This chapter gives details of the steps to install Verint

Installation ContactStore for Communication Manager.
Chapter 4: This chapter gives details of the steps to configure Verint

Configuration ContactStore for Communication Manager.
Chapter 5: This chapter provides detais of regular maintenance

Operation, Administration required by Verint ContactStore for Communication
and Maintenance Manager.
Chapter 6: This chapter sets out the available and recommended

System Security steps you can take to secure the system.
Chapter 7: This chapter details the rarely used advanced settings of

Advanced Configuration Verint ContactStore for Communication Manager.
Appendix A: This appendix provides technical details about Verint

Technical Reference ContactStore for Communication Manager.
Glossary The glossary defines the terms you need to understand

Verint ContactStore for Communication Manager.
Conventions used in this guide
Conventions used in this guide

The following table shows how user input, output and instructions are highlighted in this
guide, as well as special notations that you will see as you use this guide.
To show... This style is used For example...
Information shown on Fixed width You should see the prompt below:
screen login:
Characters that you Fixed width, bold Enter the following command:
should type exactly as mount /mnt/cdrom
shown
Characters that you Fixed width, bold Browse to the new server by entering
should replace with italic http://servername:8080
appropriate information
Menu selections, Sans Serif, Bold Click on the Install button.
buttons and tabs
Helpful hints that can Tip: Tip:
improve the efficiency or If no part-time licenses are available,
effectiveness of your a full time license may be used
work instead.
Important details that we Note: Note:
want to make sure that Media Encryption may or may not
you do not overlook show up on this form.
Advice that can help you
avoid undesirable ! Important: ! Important:
results
If the network does not meet the three
conditions listed, there will be no
media resources.

About This Guide
To show... This style is used For example...
Situations that can result

in: ! CAUTION: ! CAUTION:
● Harm to software
Perform this procedure only after
● Loss of data
normal business hours. This
● An interruption in
procedure restarts all links on the
service interface, and can cause a temporary
loss of service.
Situations that can result
in harm to hardware or ! WARNING: ! WARNING:
equipment
Make sure that the disks are the
Update you require. RedHat and other
vendors still sometimes supply
Update 0 disks.
Additional references
The following guides contain additional information you may find helpful.
● Avaya Communication Manager Call Recording: A Design Approach for Device Media
and Call Control (DMCC, previously called CMAPI) (Compas ID 128862)
● Verint ContactStore for Communication Manager User Guide
● Viewer for Communication Manager Installation Guide
● Viewer for Communication Manager Quick Reference Guide
● Contact Archive for Communication Manager Installation Guide
● Contact Archive for Communication Manager System Administration Guide
● Unify for Communication Manager Installation Guide
● Witness Enterprise Deployment Guides Impact 360, Full-time Recorder, Enterprise
Security Administration Guide Release 7.8
● Avaya Communication Manager Guide to ACD Contact Centers
● Administrator’s Guide for Avaya Communication Manager
● Administration for Network Connectivity for Avaya Communication Manager
Note:
Note: Avaya Communication Manager documentation is available through the
Avaya online support Web site, http://www.avaya.com.
Chapter 1: System Overview
This chapter provides an overview of the design options for a Verint ContactStore for
Communication Manager system.

System Overview
Introduction
Verint ContactStore for Communication Manager provides an extremely efficient and
scalable, software only, voice recording platform, running on standard PC Hardware. It
uses Avaya's Device, Media and Call Control (DMCC) features to provide a wide range of
recording modes with all the benefits of VoIP-based recording but without the limitations of
passive tap IP recording systems.
This new approach to recording offers the following benefits:
● The recorder can record potentially any call on the switch. Traditional trunk and
extension modes cannot record internal and tandem calls respectively.
● There is no cabling to maintain as new trunks or extensions are added to the switch.
● Uses standard PC servers with no proprietary cards.
This chapter describes the major components of the Verint ContactStore for
Communication Manager system. It describes the Verint ContactStore for Communication
Manager (which is a mandatory component) and the optional components
Introduction
What’s New in Version 7.8

Version 7.8:
● Provides a range of enhanced security options as described in Chapter 6
● Supports Redhat Enterprise Linux (RHEL) Versions 4 and 5. (Existing installations on
RHEL 3 can be upgraded to CSCM 7.8 but this version is no longer supported for new
installations).
● Supports Avaya Communication Manager Version 4
● Supports numbering plans up to 16 digits
● Supports No-talk Service Observe - which does not require a time-slot on your port
network.
● Allows you to update call tagging details after recordings have completed.

System Overview
Server components
The Verint ContactStore for Communication Manager system can be installed as a single
server solution providing recording and replay of calls. Large systems can be built from
several independent recorders or in a master/slave topology. The former approach works
well for station-side recording while the latter provides many of the benefits of traditional
trunk-side approaches - without most of the disadvantages. Internal calls can still be
recorded, for example.
You can extend the scope of the system by adding additional optional server applications
in order to:
● perform quality monitoring and screen recording
● provide centralized shared search and replay
● provide centralized archive facilities
These optional server applications require their own physical server. All of the other
applications in the suite require Windows servers.
Note:
Note: This Guide is the detailed reference for the Verint ContactStore for
Communication Manager (CSCM) only. This component can record voice
calls over IP using Avaya DMCC softphones. It can also act as the Master -
controlling other recorders that provide additional capacity.
Each of the other components has a corresponding guide (as detailed in Additional
references on page 14), which you should refer to for more detail.
Verint ContactStore for Communication Manager Server

In most small to medium-sized (up to a few hundred channels) Bulk Recording systems,
one of these applications provides the entire recording and replay system (as shown in
Introduction on page 16).
In larger systems (where a single physical server is not powerful enough), you should
install multiple instances of this application on different physical servers, each providing a
subset of the system's overall functionality. The Verint ContactStore for Communication
Manager can:
● connect to the Avaya Communication Manager Server via Application Enablement
Services (AES) and control all voice recordings
● record and store telephone calls via IP
● archive the recordings it makes to a local DVD+RW drive
Server components
● provide basic search and replay services to users connecting via their browser or via
their telephone
● provide voice recording services to the Verint eQuality Balance (V5 or V7) application
● control other Verint ContactStore for Communication Managers - known as Slaves
● be controlled by another Verint ContactStore for Communication Manager - i.e. act as a
Slave
● act as a warm Standby to a Verint ContactStore for Communication Manager Master
● act as a dedicated, centralized replay server, holding details of recordings made by
other Verint ContactStore for Communication Managers
Optional Server Applications

Verint eQuality Balance Application
This application provides the features needed to automate your agent quality assessment
process, including:
● Design of evaluation forms
● Various call selection methods
● Screen content recording and storage
● Performance recording and reporting
See Verint eQuality Balance Administration, Evaluations, Graphic and Reporting, Server
Infrastructure, Troubleshooting and Installation Guides for full details of this component
and Installing Verint eQuality Balance (V5) on page 88 for how to integrate it into the
overall system.
Viewer
Although the Verint ContactStore for Communication Manager application includes search
and replay facilities within it, you will need to install this alternate replay server in the
following cases:
● If you wish to deploy centralized Archive
● If you wish to combine recordings with those made by other Verint recorder types
In such cases, you may install Viewer on an additional server and configure each of your
recorders to copy the details of its recordings to the database on this server. The database
holds detailed information on all recordings including where the audio content is stored (for
example, either on individual recorders or archived offline). The following figure shows a

System Overview
typical system, using Viewer to provide access to recordings from two independent
recorders.
See Viewer Installation Guide for full details of this component and Installing Viewer on
page 90 for how to integrate this application into the overall system.
Archive
Each recorder can archive the recordings that it makes itself via a single, locally connected
DVD+RW drive. If you need more sophisticated archival - from multiple recorders to a
common destination, such as a Network Attached Storage (NAS) device or of selected
Server components
recordings only, you can deploy the Archive application on a separate server as shown
below.
See Archive Installation Guide for full details of this component and Installing Archive on
page 94 for how to integrate this application into the overall system. Note that you must
first deploy Viewer in order to use Archive.
External Control
In addition to the data provided by the AES links, you may wish to control and/or tag your
recordings with details from other CTI feeds or application interfaces. These may include
third party systems and/or your own in-house applications. The recorder supports a wide
range of systems and allows them to be connected to your recording system. These
include:
● Avaya Interaction Center (AIC)
● Avaya Proactive Contact (formerly Predictive Dialing System, PDS)
Additional Verint Systems applications

As an integral part of Verint Systems' range of recording and analysis products, the
following applications can be layered on top of the recording system:

System Overview
● eQuality Now (agent training tools)

● eQuality Vision (advanced recording visualization)
End-User tools
End-User tools
To access the recordings held in the system, users have a variety of options.
Quality Monitoring
Supervisors and quality assessors use a Windows application (V5) or browser (V7) to
evaluate recordings - using both voice and screen content where available. They can
assess and review calls with their staff as well as analyze and report on the results of
assessments.
Audio is replayed via the desktop PC's soundcard or via the telephone.
Live monitoring of audio and screen content is also supported.
Verint eQuality Balance Evaluations Guide provides full details of the Quality Monitoring
application.
Bulk Search and Replay

Integral Search and Replay
The Verint ContactStore for Communication Manager includes a search and replay
application within it. This replay mechanism is a very simple and intuitive browser-based
interface, requiring the user to access it via Internet Explorer Version 5.0 or higher.
The Search and Replay application is hosted on a web server running on the recorder
itself. It uses a local database of recordings to allow users to search for bulk and ad hoc
recordings by:
● Call start date/time
● The name(s) and number(s) where provided of any party on the call (including ANI
and/or DID where provided by the switch)
● Agent ID and name
● Call duration
● Universal Call Identifier (UCID)
● Service Name and Number (also known as Vector Directory Number, VDN)
● User defined fields supplied by external controllers

System Overview
! Important:
Important: Verint ContactStore for Communication Manager does not retain those calls
recorded for Verint eQuality Balance; these use that product's own replay
mechanism.
Access restrictions determine which calls individual users are able to replay. Each
recording is assigned one or more "owners" at recording time (see Search and Replay
Access Rights on page 136 for further details).
The user can play and view details of any call that matches their search criteria and access
rights. When a call is played, a graphical representation of the audio level of the call, the
audio wave form, is displayed. The audio wave form shows silence and tones, so the user
can click beyond irrelevant sections and pinpoint parts of the call that are of interest. See
the accompanying User Guide for further details of this application.
Central Search and Replay

Where multiple Verint ContactStore for Communication Managers are deployed, an
additional server can be nominated as a Central Replay Server. This server is not used for
recording but can provide telephony replay ports.
The other recorders upload details of the recordings they have made into this server's
database allowing users to search for and replay recordings made on any recorder without
having to know which one recorded a particular call.
External Search and Replay

The recorder also supports replay from other applications including eQuality Balance (V5
and V7), Contact Viewer and Contact Vision.
Administration Tools
As the suite is designed specifically for Avaya systems, much of the complexity associated
with generic recording systems has been removed resulting in a system that is easy to
configure and maintain. With the exception of the Quality Monitoring (V5) system (which is
a Windows application), the recorders are administered via a web interface. The detailed
use of this interface is the subject of later Sections in this guide.
Recording Functionality
The Verint ContactStore for Communication Manager system records telephone calls
made on Avaya Communication Manager release 3.0 and above - using analog, digital or
IP-based stations. The recorder hosts a bank of IP softphones - each of which provides a
recording or replay port. These record calls by conferencing into a live call using either the
Service Observe or single-step conferencing features of the switch.
This manual only considers audio recording load. For Screen Content recording, please
refer to the Verint eQuality Balance (V5 or V7) Server Infrastructure Guide.
The first task in designing any recording system is to define what is to be recorded. This in
turn is often driven by the reason for wanting the recordings. As the recorder's ports can
also be used for replay and live monitor, it is important to consider these together before
specifying a system. This section introduces the various ways in which recorders' ports can
be used but you should refer to Port Allocation on page 114 for detailed functionality and
limitations of each mode.
Sampled Recording for Quality Assessment

If the end goal is to record only a sample of calls in order to assess the quality of your
interactions with customers, Verint eQuality Balance (V5 or V7) can be used alongside the
Verint ContactStore for Communication Manager. See this product's Administration,
Evaluations, Graphic and Reporting, Server Infrastructure, Troubleshooting and
Installation Guides to determine how you will use the Quality Monitoring system to control
recording.
Quality Recording Channels (eQuality Balance V5 only)

From the above manuals, you must determine the number of concurrent audio recording
"channels" that will be required on the Verint ContactStore for Communication Manager to
support your Quality Monitoring requirements.
Quality Replay Channels

Note that you will also need to consider how the Quality Application is to replay calls for
supervisors to review and coach their staff. If you choose to replay via their telephones,
then you must provide sufficient ports for this. This is in addition to the recording channels.
The overall port count required for selective quality recording is therefore the sum of
recording and replay requirements.

System Overview
Licensing
The Verint ContactStore for Communication Manager must be provided with a license key
that includes the required number of Quality Channels (the sum of recording and replay
ports assigned to the Quality Monitoring application).
Bulk Recording Modes

If you need to record all of the calls taken by specific stations, agents, skill groups or Vector
Directory Numbers (VDNs) then you require Bulk Recording - of which there are three
variants.
Station Bulk Recording

This recording mode can be used to record all calls occurring at specific stations.
Advanced configuration options let you filter the calls by VDN or Skill Group rather than
have to record every call on the nominated stations.
This mode uses the switch’s service observe feature and therefore requires a dedicated
port per station that is recorded.
Refer to Station Bulk Recording on page 121 for a full description of this mode and how to
configure it.
Conferenced Recording
Rather than dedicating ports to specific stations, a pool of ports can be used in conjunction
with an Avaya CT (also known as TSAPI) link to record calls on specific stations, agents,
skill groups or VDNs.
In this mode, the recorder uses single-step conferencing to connect into the calls to be
recorded.
Refer to Conferenced Recording on page 125 for a full description of this mode and how to
configure it.
Externally Controlled Recording

More complex recording requirements may be met by customized or specialist applications
that interface to other CTI feeds or customers' own applications. Such an application can
control ports on the Verint ContactStore for Communication Manager, allowing it to record
exactly what and when it requires. The application may also "tag" the recordings with
additional details such as customer number or account number.
Refer to Unify or Externally Controlled Ports on page 131 for a full description of this mode
and how to configure it.
Which to use?
In essence, Station Bulk recording is simpler and more economic to deploy than
Conferenced mode. Against this, however, it is a little less flexible. Externally controlled
recording is normally only considered when complex recording rules or additional CTI
control is needed.
It is imperative that you review the detailed functionality, limitations and caveats in Station
Bulk Recording on page 121 and Conferenced Recording on page 125.
Bulk plus Quality

Where Quality assessment is to be performed as well as bulk recording, Verint
ContactStore for Communication Manager can be used alongside Verint eQuality Balance
(V7.6). The recorder advises the quality system of all recordings that it makes and these
can be used for quality assessment.
Licensing
A "full-time recording channel license" is required for each recording port assigned to these
modes. In Station Bulk, this equates directly to one such channel license per station to be
recorded. In conferenced mode, a channel license is required per port assigned to this
mode.
Ad-hoc or Occasional Recording Modes

Three recording modes are provided for those requiring occasional recording:
On Demand Recording
This mode lets users of any phone on your system dial into or conference in a recording
port as and when they want to start recording a call.
One or more "pools" of ports on the recorder can be assigned to this recording mode and
accessed via Hunt Group numbers so that callers automatically reach an available port.
The recorder automatically answers the incoming call on its port and starts recording.
Refer to On Demand Recording on page 119 for a full description of this mode and how to
configure it.
Meeting Recording
A novel use for recording is in taking a detailed log of a meeting, either as an audio record
for those attending, or as a way to include non-attendees later. You can use any meeting

System Overview
room or office with a telephone that has a speakerphone or, ideally, conference phone
capabilities to record the meeting.
! Important:
Important: The audio recorded with Meeting Recording is the same as someone dialing
in would hear it on the phone used to record it. Place the phone so it picks
up the speech of all participants. They should speak loudly and clearly.
Experiment with this recording mode before relying on it to provide full and
complete records of your meetings. Verint cannot be held responsible for the
failure to pick up all of the audio intelligibly. Use this recording mode as an
aid to note taking, not a replacement for it.
One or more "pools" of ports on the recorder can be assigned to this recording mode and
accessed via Hunt Group numbers so that callers automatically reach an available port.
The user follows the spoken instructions to start the recording and specify which user(s)
can access it.
Refer to Meeting Recording on page 120 for a full description of this mode and how to
configure it.
Station Executive Recording

This option lets users of specific phones choose which of their calls are recorded. Unlike
the previous two modes, this method records the call from the beginning. Unless the user
chooses to keep the recording it is deleted as the call ends (or shortly afterwards according
to how the system is configured).
This recording mode is particularly suited to users who occasionally need to record a call,
and who might only recognize the need to record after the call is in progress or after it has
ended. The classic example of this is in "bomb-threat" or "abusive call" recording. In such
cases, it is only once the nature of the call becomes apparent that the agent realizes that
he should record the call. The advantage of this recording mode over other recording
techniques is that the entire call is being recorded all the time. The whole call can be
retained even if the nature of the call and the need to retain it is not recognized until the
end of the call or, if the delayed deletion option is enabled, after the call ends.
Refer to Station Executive Recording on page 123 for a full description of this mode and
how to configure it.
Scope
Ports assigned to On Demand or Meeting recording
● can be assigned to one or more hunt groups making them easily shared across one or
more user populations
● can be used not only by stations on the switch but also from outside the switch if they
are made accessible via a DID number.
Station Executive recording ports on the other hand are dedicated to specific stations and
can only be used to record those stations that can be service observed.
Licensing
As On Demand and Meeting recording ports are shared by all users, they typically record
many calls per day and hence a "full-time recording channel license" is required for each
port assigned to these modes.
Although Station Executive recording requires a dedicated port on the recorder for each
station that uses it, Verint recognizes that typically only a few recordings are made per
station per day. Hence the value of such a recording port is less - and a "part-time
recording channel license" is used.
Tip:
Tip: If no part-time licenses are available, a full time license may be used
instead.
Replay Options
Soundcard Replay
Many users choose to replay recordings via their browser and the soundcard on their PC.
This does not use any ports on the recorder and does not require any additional licensing.
See the eQuality Balance (V5) manuals for details of how to use this replay mechanism in
conjunction with this application.
Telephone Replay
However, the recorder also supports replay via the user's telephone and this does use a
port on the recorder.
eQuality Balance (V5) replay ports form part of the overall pool of ports assigned to Quality
Recording.
All other replay applications make use of telephony Replay ports on a Verint ContactStore
for Communication Manager. Ports configured in this way are compatible with Verint
"AudioServer" ports.
Refer to Phone Replay Ports on page 133 for a full description of this mode and how to
configure it.

System Overview
Licensing
Replay channels that form part of the eQuality Balance (V5) port pool require a Quality
channel license.
Telephony replay ports used by all other applications require a replay channel license.
Live Monitoring
Via eQuality Balance (V5)
Users of eQuality Balance (V5) can listen in to calls in real-time using their telephone. This
uses one of the replay ports within the Quality pool. See the eQuality Balance (V5)
manuals for details of this live monitoring mechanism.
Live Monitor Ports

On Communication Manager 3.x, a side-effect of using service observe to record calls (as
used by Station Bulk and Station Executive recording modes) is that this switch feature is
no longer accessible to supervisors who wish to use it. To mitigate this, one or more ports
on a Verint ContactStore for Communication Manager can be assigned for Live Monitor
use. Supervisors can dial into these ports and dial the number of the station they wish to
observe. The audio being recorded is relayed to them giving similar functionality to service
observe.
With Communication Manager 4.0 and above, two service observe sessions can be in
progress on the same call and it is therefore unlikely that you will need to use Live Monitor
ports.
Refer to Live Monitor Ports on page 133 for a full description of this mode and how to
configure it.
Licensing
Live monitoring via eQuality Balance (V5) uses a replay port within the Quality pool and is
therefore controlled by the overall Quality Channels license count.
Live Monitor ports on the recorder are licensed separately and a Live Monitor channel
license is required for each one.
Recording Methods
Recording Methods
The recorder uses three different methods to record calls. The table below shows which
mode is used by each mode.
Recording Mode Uses
On Demand Recording
Conference
Meeting Recording
Service Observe
Quality Monitoring for eQuality Balance V5
Externally Controlled (Service Observe mode)
Externally Controlled (Single-step conference Single-step Conference
mode) where the external controller instructs the
recorder to establish a single-step conference.
Externally Controlled (Single-step conference Depends how the external
mode) where the external controller is responsible controller establishes the call.
for connecting the recorder port into the call to be May be conference or
recorded. single-step conference.
The recording mechanisms differ in several respects as described below. Refer to Port
Allocation on page 114 onwards for details of the pros and cons of these modes as these
must be considered alongside the issues listed here.
Note:
Note: Regardless of which recording method is used, each recorder port that joins
a call to record it counts as an additional party on that call. Hence your
normal limit of 6 parties on a call includes all of the recordings being made.
Every recording made reduces the number of "real" participants you can
have on the call.

System Overview
Service Observe
Station Bulk, Station Executive and Quality recording modes use the Communication
Manager's Service Observe feature. This has the following characteristics.
Limited Number of Service Observers per Station

In Avaya Communication Manager 3.x, only one other station can service observe a
station at any one time. Therefore:
● a supervisor cannot use Service Observe to monitor stations or calls that are being
recorded. They will either fail to observe the call or will block the recorder from recording
it. To listen in to calls (in real time) on stations that are being recorded, users must dial
into Live Monitor ports on the recorder or use eQuality Balance (V5)’s Live Monitor
feature. Live Monitor ports can be included in any license. If you feel that Live Monitor
ports may be useful, request them when ordering the system.
● Quality Monitoring uses service observe. If a station is being recorded in another mode
using service observe, then Quality Monitoring can only work if the quality and other port
recording it are on the same recorder. If you need to perform bulk and quality recording,
use a single recorder or, if the load is too much for a single server, you must use a
single-step conference mode for the bulk recording.
● In Avaya Communication Manager 4.x, up to two stations can service observe a given
station. It is therefore rare that Live Monitor ports are needed in this case. However, it is
still a good idea to record a station on a single recorder where possible - due to the per
call limitations described below.
Limited Number of Service Observers per Call

In Avaya Communication Manager 3.x, if more than one party on a call is being service
observed, then only one of the observers will actually receive the audio from the call. The
other will be completely unaware that a call is in progress. Therefore:
● A call between a Station Bulk recorded station and a Station Executive recorded station
can only be visible to one or other of the two recording ports. Which one depends on
which station made the call. If the station executive port initiated the call, it will be
deleted unless the user deliberately retained it.
● An attempt to Quality Monitor a station on an internal call will fail if the other station on
the call is the one receiving the audio. If you need to quality monitor agents who solely
or predominantly take internal calls and are using bulk recording as well, you should use
a single-step conference mode in preference.
● If a user service observes a VDN, they will stop any call they hear from being recorded.
Use Class of Restriction (COR) settings to prohibit users from observing VDNs that
include calls that you are trying to bulk record.
Recording Methods
● In Avaya Communication Manager 4.x, each call can be service observed up to twice.
Normal, two-way internal calls can therefore be recorded twice and it is only with three
(or more) way calls that the above limitations apply. It is still important to avoid service
observing VDNs related to calls that you need to record.
Assign Ports to Stations not Agents/VDNs

Although Communication Manager can service observe VDNs and Agents, the recorder
only supports service observe of station numbers for the following reasons:
● Service observing a VDN only results in one call at a time on that VDN being observed.
This is not appropriate for a bulk recording mode.
● Service observing an agent id results in failure if that agent is not logged in. The
recorder cannot distinguish between failure due to this reason and other failure modes.
It would therefore raise alarms continuously.
Call segmentation
As the recorder port sees the same activity as a real user would do when service observing
a station, audio ceases to flow when the station being observed places a call on hold.
Placing a call on hold (automatically as part of a transfer/conference setup or deliberately)
therefore stops recording. A new recording is started when the call is retrieved from hold.
The segments of the call are stored as separate recordings (.wav files) and can be
searched and played as two calls.
! Important:
Important: If the call is retrieved from hold by another station, either by means of a
transfer or a "park" operation, the subsequent segment of the call is only
recorded if the other station is also being recorded.
The recorder's Search and Replay mechanism treats these segments as two separate
calls. However, if you find one call segment, you can easily find the other segments by
clicking on the link in the UCID column. Viewer, a separate server replay application, on
the other hand, "merges" these call records into one and shows the hold period as silence.
If an external controller is used, it may segment calls at points it chooses.
Bulk Recording requires dedicated port

Any attempt to service observe a station will - on perhaps one in a thousand attempts - fail.
While this is not a significant issue for a sampling quality recording system, it is
unacceptable for bulk recording. This is why the recorder must dedicate a port to each
phone that it bulk records via service observe. By doing so, it can establish service observe
as the recorder starts - and retry on the rare occasion that this fails - and then hold the
service observe permanently. This provides reliable recording.

System Overview
Personal Station Access (PSA)

PSA is not supported for the same reasons that agents cannot be targeted. As the address
is not always present, the recorder cannot maintain a service observe on it at all times as is
required for bulk recording.
Internal Calls via a VDN do not show who answered - CM 3.1 and earlier
If an agent handling a call dials a VDN to transfer, conference or simply consult on it, the
details of the party that answers the call are not visible to a station service observing that
call. The call will be tagged with the VDN number dialled but not the agent or station that
answered the call. This is corrected in CM3.1 load 632. If using an earlier version than this,
consider updating with Patch #11855.
Transfer via Coverage

Calls transferred to a group with coverage (e.g. Coverage Answer Group) and not picked
up by the original destination address cannot be observed. To overcome this limitation, use
a phantom extension and transfer the call to that rather than directly to a group with
coverage.
Bridged Lines
Where bridged line appearances are used on stations being recorded, calls made on the
secondary line appearance will be recorded but the details about the call will be limited to
the identification of the station being recorded. Only calls made on the primary line
appearance will include full call details.
Exclusion
If any party on the call invokes the Exclusion feature, the call cannot be recorded. (Check
with Avaya for changes to this from CM 3.1 onwards).
Beep Tone
If you do not need to inject beep tone while recording, use "No Talk" service observe. This
option, available from CM 3.1 onwards, does not use up a time-slot. Otherwise you must
design for an additional time-slot being used to carry each call that is actually being
recorded. If you need to inject beep tone you must use "Talk/Listen" service observe.
"Listen Only" service observe should not be used with CSCM.
Recording Methods
Single-step Conference
Single-step conferencing as used by Conferenced mode recording and Externally
controlled (single-step conference option) has the following issues:
CTI Requirement
When recording in this mode, the recorder uses JTAPI which in turn uses Avaya CT (also
known as TSAPI). This requires not only that the CTI feed be available but also that
sufficient licenses are available for it. See TSAPI Simultaneous User Licenses on page 60
for details.
Timeslots
Single-step conferences do not require an additional timeslot on the switch if beep tone is
NOT used.
Call Segmentation
The recorder continues to record only so long as the real parties on the call are connected.
If the call is on hold, recording stops.
Bridged Lines not Supported

Because of an inherent limitation in the underlying call/connection model, conferenced
recording does not support bridged line appearances. You must not attempt to record
bridged lines in this mode.
Beep Tone
If you choose to inject beep tone on a single-step conferenced recording, the recorder
becomes a full member of the call rather than a listen-only member and therefore:
● Becomes visible to the agent, who can see that the call is conferenced
● Uses an extra timeslot on the switch
Conference
In some modes a user or an external application will dial a port on the recorder. When this
happens, the recorder answers the call and is therefore a normal party on the call.

System Overview
Timeslots
As "just another party" on the call, the recorder port will use the single additional timeslot
that any other phone would use when added to a call.
Call Segmentation
As the recorder port is a normal party on the call, it is still connected even if one or more
other parties on the call places the call on hold. It will receive the same audio that the
parties remaining on the call receive. This may include music on hold or silence.
Miscellaneous
Miscellaneous
Beep Tone
Some states and countries require that both parties on a call be made aware that the call is
being recorded. One way to do this is to apply a tone to the line. Note, however, that most
telephone users are unaware of the significance of this tone and might, in fact, regard it as
a fault. There are more effective means of informing the user that the call is being
recorded. For example, you can inform users on advertising and in contract literature or
play a recorded announcement before the call is connected. Check the legal position in
your jurisdiction and apply the appropriate settings.
See the discussions of how each recording mode works earlier in this chapter to determine
whether or not you need to turn on this warning tone within the recorder. In most cases, it is
recommended that you set this option to No and use the other mechanisms described
there.
International support
The recorder's browser-based Administration and Search/Replay interfaces are provided
in several languages. Check with Verint for availability of specific languages. International
support includes:
● Time zone and DST support. All dates and times are stored in the database in
Coordinated Universal Time (UTC). However, when you view records using the search
and replay application, these are converted to your local time. If you view the records
using a database query tool, the times will be shown in the time zone of the client
machine, which may be different from the server time. Note that the XML files relating to
the recordings include ISO standard timestamps, giving both UTC and offset from
Greenwich Mean Time (GMT).
Liability
Liability of Verint for failure to record any calls is limited under the terms of supply.

System Overview
Chapter 2: Planning and Prerequisites
This chapter gives details of the prerequisites for a Verint ContactStore for Communication
Manager system. You should also review Chapter 6 System Security as some of the
optional elements described there may also require additional cost and/or effort.
The main sections in this chapter are:
z Introduction on page 40
z Audio Format on page 41
z Storage Requirements on page 43
z Server Platform on page 46
z Network Issues on page 51
z Recorder Licensing on page 52
z Avaya system prerequisites on page 54
z Topologies on page 62
z Integrating with other systems on page 67

Planning and Prerequisites
Introduction
Unfortunately, there is no one "right" order in which to plan a system. A number of the
requirements and pre-requisites are inter-dependent and it may be necessary to iterate
through this section several times, refining your plans each time. This Chapter assumes
that you have read the previous one and know the type and quantity of recording, replay
and live monitor that you require - and hence your total license requirements.
This Chapter will now guide you through:
z identifying what is to be recorded, the options for doing so and the relevant license
requirements for the preferred option
z determining the storage needed to hold your recordings
z sizing the servers needed for each application
z quantifying the network bandwidth needed
You can then determine an appropriate system topology.
Audio Format
Audio Format
In all cases, calls are recorded using resources on the Avaya Communication Manager to
conference a recorder port into a live telephone call. One of the benefits of this approach is
that the recorder can ask to receive audio in a format that suits it - without impacting the
experience of the parties actually speaking on the call.
The recorder can request either G.711 (µ-law), or compressed G.729A data streams. You
must choose which is appropriate for your recording needs, network and server sizing.
G.711
If the recorder receives audio in G.711 (µ-law) rather than G.729A, bulk recordings will be
normally be compressed by the recorder and stored as G.726 (16kbps) files. (Compression
can be disabled by setting cscm.disablecompress=true in the properties file.) Selective
Quality recordings controlled by eQuality Balance will be passed to it unchanged.
Note:
Note: When using G.711, the recorder always requests µ-law, never A-law. This
applies to all countries regardless of that country’s national preference.
Therefore, if you select G.711 rather than G.729A the implications are:
For All Recording

z Several times more bandwidth will be needed between the media processing resources
and the recorder.
z Less Avaya media processing resource will be needed.
For Bulk Recording

z The capacity of a given server will be about halved
z The quality of recordings will be a little lower
z The storage space used will be doubled
For Quality Recording

z The quality of recordings will be a little higher
z The storage space used will be eight times higher (This is for voice recordings. Note that
when combined with screen recordings, the overall impact is proportionately less).

G.729A
Conversely, if receiving G.729A, the recorder will not have to perform any compression
tasks - for either bulk or quality recording. The pros and cons are reversed from those
shown above.
Therefore, if you select G.729A rather than G.711 the implications are:
For All Recording

z A fraction of the bandwidth will be needed between the media processing resources and
the recorder.
z More Avaya media processing resource will be needed.
For Bulk Recording

z the capacity of a given server will be about doubled
z the quality of recordings will be a little higher
z the storage space used will be halved
For Quality Recording

z the quality of recordings will be a little lower
z the storage space used will be one eighth (This is for voice recordings. Note that when
combined with screen recordings, the overall impact is proportionately less).
Storage Requirements
Having determined the type, size and location of your recording capacity, you must now
determine how much storage is required within the system. Storage requirements of many
terabytes are not uncommon.
Storage is required for:
z the operating system and applications installed on each server
z the audio content of recordings
z the database holding the searchable details of these recordings
In each case you should consider requirements locally at each recorder and centrally.
To do this, you will need to know:
z how many recordings will be made on the recording channels already identified
(typically expressed in channel hours per day). In other words, how many hours of audio
you expect to record on each recorder every day.
z how long you wish to retain recordings (typically expressed in days).
z the average duration of a call (typically expressed in seconds).
Storage at Each Recorder

In addition to the operating system, installed software and its configuration data, each
recorder stores:
z details of the recordings it has made in a local database
z the recordings it has made as files on its hard disk (except for those controlled by
eQuality Balance which are copied off to that server's storage as the call completes).
In all cases, the system automatically notes the location of the recordings so that when a
user wishes to replay a call that is no longer in local online (disk) storage, the application
prompts them to insert the appropriate removable media (if any).
Recordings are stored in G.726 compressed form at 16kbps (if received as G.711) or as
G.729 files (8kbps) if received as such from the media processing resource on the switch.
This allows for high volumes of recordings to be stored on the available disk space.
Verint strongly recommends RAID arrays or fault tolerant Storage Attached Network (SAN)
devices for online storage of recordings on the recorder platform.
The recorder automatically manages the available recording storage space. This is used
as a circular buffer providing instant access to the most recent recordings and deleting the
oldest, as space is required for new recordings.

Rather than storing bulk recordings on RAID arrays in each recorder, many customers
prefer to use Storage Attached Networks. These must be connected directly to recorders.
Hierarchical File Storage (HFS) Systems and Network Attached Storage (NAS), however,
can only be supported via the separate Archive application.
The table below summarizes the requirements for Bulk Recording on a Verint ContactStore
for Communication Manager.
Details of Recordings Content of Recordings
Stored In local postgres database Stored As WAV files on local disk in

a hierarchical folder structure
Purged Nightly, after user defined period 100 at a time as disk space is
(default 60 months) needed for new recordings
Volume Approximately 2KB per recording . Approximately 7.2MB (G.726) or

3.6MB (G.729) per channel hour of
audio.
So total ~ 2KB x recordings/day x So total ~ 7.2MB (or 3.6MB) x
days retained channel hrs/day x days retained
Location /var (in addition to the normal /calls

contents of /var)
Type RAID 1 (mirrored) or 5 (striped) RAID 1 (mirrored) or 5 (striped)
strongly recommended. strongly recommended.
Local hard drive or SAN, not NAS. Local hard drive or SAN, not NAS.
Quality Monitoring
When used to provide selective audio recording services for the Verint eQuality Balance
application, the Verint ContactStore for Communication Manager only uses a small amount
of local storage for voice recordings (less than 1GB). As each Quality Monitoring recording
is completed, the recorder transfers it to the Quality Monitoring application's storage area.
As the volume of recordings made by a selective, quality monitoring application is normally
much less than that of a bulk recording system and is comparable to the screen recording
volumes, audio is normally stored in G.711 64kbps PCM µ-law format (approximately
27MB per channel hour).
This application also has to store:
z screen content recordings
z the agent configuration and evaluation data

See Verint eQuality Balance (V5 or V7) Server Infrastructure Guide for further details of
how to size the storage requirements for the Quality Monitoring application.
Central Database Storage

You should deploy Viewer or a Central Replay Server for search and replay if any of the
following apply:
z you have more than one recorder and wish to search and replay from any recorder
through a single query.
z you anticipate more than five concurrent users of the integral search and replay
application on a Verint ContactStore for Communication Manager.
If you wish to use Archive, you must install Viewer rather than a Central Replay Server.
If you choose to deploy Viewer, you will need to size the storage it requires in accordance
with Viewer Installation Guide to hold the call details from all the recorders that are
contributing to it. For a Central Replay Server, follow the guidance above to size the
storage needed for the call details using the total call volumes across all recorders that are
feeding into the Central Replay Server.
Archive Call Storage

Where you wish to retain calls for longer than it takes to fill the hard disk storage on a
recorder, two options are available:
z connect a DVD+RW drive to the recorder. All recordings made on the recorder will
automatically be archived to the removable media on this (single) drive. This requires
only 1GB of buffer space on the recorder. It is only suitable for systems recording less
than 4GB per day.
z add Archive (an extra cost option) to the system. This application supports selective
archival of recordings to multiple destinations. See Archive Insallation Guide for
guidance on sizing of the storage for such systems.
Backup Storage
See Backup/Restore on page 141 for a discussion of Backup and Restore options. You
should determine whether additional storage space is required in your corporate backup
system to accommodate the new recording system.

Server Platform
Taking the above factors and the potential location(s) of your recorders into consideration,
you must determine how many channels of each type of recording you wish to deploy and
on which site.
Having decided the total recording capacity at each of your locations, you must translate
this into one or more server platforms capable of handling the load identified.
Dedicated server(s) must be provided with no other applications running on them.
Sizing
The table below shows the maximum capacity of four server types that are used as
benchmarks for planning recorder hardware.
Server Type Server Specifications

See also Platform Prerequisites on page 81 for detailed
configuration pre-requisites.
Minimum 2.0GHz single-processor Pentium IV with:

z 500MHz front side bus
z 512K of L2 cache
z 1GB RAM·Hard disk with at least 7200rpm rating
z 100MBps full duplex Ethernet NIC port
z CD-ROM compatible drive for installation of software
z DVD+RW drive if local archiving is required.
z Sound card recommended.·
z RedHat Enterprise Linux 4.0 or 5.0
A Same as Minimum, but with:

z 3.0GHz single-processor Pentium IV Xeon with
hyper-threading
z Dual 100Mbps full duplex Ethernet NIC ports
z RAID 1 or 5 strongly recommended
B Same as A, but with dual processor

C Same as A, but with quad processor
Server Platform
Recorder
The following table shows the maximum channel counts supported for the above servers.
This table takes the following factors into account:
z Recording G.711 calls and compressing these to G.726 loads the server more heavily
than receiving and storing G.729A calls.
z The CTI feed required for Conferenced mode increases the load on the server
z Using recorder ports for replay via the telephone or live monitor not only loads the
recorder but these ports are more susceptible to packet jitter and therefore the recorder
cannot be loaded as heavily when using these modes.
z These figures assume NO encryption is being used. If the audio to and from the Avaya
system is encrypted, reduce capacities by 10%. If using a Key Management Server and
encrypted storage, capacities are futher reduced. Exact figures are not yet available.
Maximum Channel Count
G.711 from Communication G.729A from Communication

Manager Manager
No Telephone With Telephone No Telephone With Telephone

Replay or Live Replay or Live Replay or Live Replay or Live
Monitor Monitor Monitor Monitor
Server Conf Others Conf Others Conf Others Conf Others
Min. N/R 60 N/R N/R N/R 120 N/R N/R

A 100 120 80 90 150 200 120 160
(Single
Xeon)
B 200 250 170 220 400 500 320 400

(Dual
Xeon)
Conf = Conferenced mode recording ports.

Others = All other modes.
N/R = Not recommended.
Sizings assume an 8 hour working day. For 24 x 7 running, provide one server level
higher (for example, use server C instead of B).
Sizings assume moderate replay load. See below for further details.
Sizings assume an average of a three minute call handle time.

Maximum Channel Count
G.711 from Communication G.729A from Communication

Manager Manager
No Telephone With Telephone No Telephone With Telephone

Replay or Live Replay or Live Replay or Live Replay or Live
Monitor Monitor Monitor Monitor
Server Conf Others Conf Others Conf Others Conf Others
C 300 450 250 300 550 990 400 550

(Quad
Xeon)
NOTES
Conf = Conferenced mode recording ports.
Others = All other modes.
N/R = Not recommended.
Sizings assume an 8 hour working day. For 24 x 7 running, provide one server level
higher (for example, use server C instead of B).
Sizings assume moderate replay load. See below for further details.
Sizings assume an average of a three minute call handle time.
Replay Load
When using Replay Ports there are additional considerations.
z The above sizing assumes a moderate replay load of up to 5 concurrent replays. If using
telephone replay, up to 8 ports can be supported (of which rarely will more than 5
actually be playing concurrently) without compromising these loading limits.
z To support up to 10 concurrent replays, increase the server specification one level (for
example, use server C instead of B).
z For higher loads, separate replay onto a dedicated Viewer or Central Replay server.
Central Replay Server

Use server type:
z B for a Central Replay Server supporting up to 1000 recording channels on the
recorders feeding it and up to 40 telephone replay ports.
z Consult Verint in-region sales for larger scale topologies.
Server Platform
DVD+RW Drive
A wide range of DVD drives has been used successfully and Verint is not aware of any
specific model limitations at this time. Only single-layer 4.7GB DVD+RW media are
supported.
Note:
Note: Most drives are advertised as supporting a wide range of subtly different
media types (e.g. DVD+R, DVD-R, DVD+RW, DVD-RW, single and double
layer etc.). Regardless of which media the drive supports, the recorder
ONLY supports writing of DVD+RW single layer media. You must confirm
that the drive supports this media type, works under the version of operating
system you are using and that you only insert this type of media.
RedHat 5
RedHat Enterprise Linux 5 incorporates new features as part of it’s Hardware Abstraction
Layer. You need additional steps to turn off the HAL’s media detection service, which
interferes with the archiver.
To disable the media detection service follow these steps:
1. Execute the "lshal" command and direct its output to a file
lshal > hal.txt
2. Open the created file and look for the string storage.drive_type. It will list all
devices, local and external, so you will find the floppy, cdrom etc.
3. Locate your device. It would be either cdrom, cdrom1..etc, or dvd.
The adjacent lines for storage model and vendor will contain specific details regarding
the device, for example:
storage.serial = 'HL-DT-ST_DVDRAM_GSA-E50L_P01070913222527'
(string)
storage.vendor = 'HL-DT-ST' (string)
storage.model = 'DVDRAM GSA-E50L' (string)
storage.drive_type = 'cdrom' (string)

4. Scroll up until you find the line starting udi =

The udi is likely to be in the form
/org/freedesktop/Hal/devices/storage_serial_vendor_model_serial
number
For example:
Udi='/org/freedesktop/Hal/devices/storage_serial_HL_DT_ST_DVDRA
M_GSA_E50L_P01070913222527'
5. Test the value of the media_check_enabled flag using the hal-get-property command.
Substitute your actual UDI into the commands:
hal-get-property --udi /org/freedesktop/Hal/....... --key
storage.media_check_enabled
the result is likely to be "true" - it needs to be "false"
6. Set it to false using hal-set-property command:
hal-set-property --udi /org/freedesktop/Hal/....... --key
storage.media_check_enabled --bool false
7. Re-check that the value is now "false" using the command in step 5.
8. Add the command from step 6 into the file /etc/rc.local (This will run the
command automatically for you every time the machine reboots.)
9. To find the soft link created for the device return to the text file, and look for the value
for "block.device" in the Device's settings. It will look something like this:
block.device = '/dev/scd0' (string)
10. Enter the location in the recorder under Local DVD+RW Drive (on the Server page)
Network Issues
Network Issues
In planning the network that will support your Verint ContactStore for Communication
Manager system, you must consider:
z the additional load imposed on the network
z the IP ports used - so that firewalls can be configured appropriately
Load
You must design your network topology to accommodate the additional traffic created by
the recording system. Each active recording port places the same load on the network as
an IP phone carrying the same call. Note that the recorder uses a 60ms packet interval,
which significantly reduces the total network load.
Ports Used
The components of the system use a number of IP ports to communicate:
z between each other
z with various Avaya Communication Manager components
z with end-users and administrators
Recorder Interfaces on page 199 provides a diagram and table listing all of the interfaces
to and from the Verint ContactStore for Communication Manager software. Your network
and firewalls must be configured to permit traffic to pass over these links.
Network Address Translation Routing

The IP address of a Verint ContactStore for Communication Manager is sent to the AE
server as part of the softphone registration. It is therefore essential that when the media
processing resources transmit to this address, the packets are routed correctly to the
recorder. The recorder must therefore be visible to the media processing resources on the
IP address that it uses itself. Additionally, if the recorder has two NIC cards it is imperative
to ensure that all VoIP packets are transmitted over the same NIC card (i.e. the network
route to all Media Processors must be the same).

Recorder Licensing
Verint ContactStore for Communication Managers are licensed through an Activation Key
which is tied to the hardware for which is it is issued. The license key contains details of
server type and port counts. It may also contain an expiry date.
Server Licenses
You will need:
z a server license for each Verint ContactStore for Communication Manager
z a Central Replay Server license for each Central Replay Server
Port Licenses
You will need an appropriate "port license" for each port on a recorder. There are five types
of port licenses as discussed under Recording Functionality on page 25. The table below
summarises which licenses are required for each recording mode.
Port Allocated to Port License Required
Station Executive Recording Part-time Recording

(though a full-time recording license can
be used if available)
On Demand Recording
Meeting Recording
Station Bulk Recording Full-time Recording
Externally Controlled Recording
eQuality Balance V5 Quality
(Recording or Replay)
Telephone Replay Replay
Live Monitor Live Monitor
Recorder Licensing
Other Licenses
You will also require Avaya licenses as described in DMCC (IP_API_A) Licenses on
page 60 and TSAPI Simultaneous User Licenses on page 60.
Timed Trials
Verint can, at its discretion, issue an activation key which includes an expiration date. This
allows for timed trials of any combination of features and capacity. As the expiration date is
fixed within the license, the server will stop operating at that date regardless of when you
enter the license key.
To extend a timed trial of to upgrade to a full license, contact Verint for an updated license
activation key.
A five day trial license is available automatically from the license key entry page.
! CAUTION:
CAUTION: The five day trial license must not be used for production recording. When a
full license is installed, any trial recordings become unplayable

Avaya system prerequisites

To use the Verint ContactStore for Communication Manager system you will need to
ensure that your Avaya system meets the following requirements. This section discusses
the various hardware, software and configuration requirements.
Verint ContactStore for Communication Manager requires AE Services and hence is only
supported on the models and versions of Communication Manager that support this
platform.
Model
The Avaya media server running Communication Manager must be an S8xxx system
(currently S8300 through S8720).
Station Count
Each port on a recorder is an additional IP phone on the switch. Do not exceed the total
station count for the switch in question.
Loading
Each recording adds as much load to a switch as a normal call. Hence you can only record
100% of calls if your switch is running at no more than 50% of its design load. For example,
S8700 switches running at up to 20,000 BHCA (complex call center call types) can have all
calls recorded. Higher loads would require an S8710 or S8720.
Software Version
Verint ContactStore for Communication Manager requires one of the following:
z CM 3.1- CM 4.0
Please check on http://support.avaya.com for more recent loads.
Gateway Resources
These house the media switching components of the Avaya system. You must ensure that
the system has, or is expanded to have sufficient:
z Card Slots for the C-LANs and Media Processing Resources described below
z Time-slots for the original calls and, where needed, the recording channels.
Card Slots
Each C-LAN and Media Processing card must be located in the appropriate gateway and
therefore in an available card slot alongside the existing cards.
Time-slots
The recorder conferences into calls in order to record them - using Single-step conference
(conferenced modes only) or service observe.
z ANY recording in which the recorder is injecting beep-tone will require one additional
time-slot per concurrent recording.
z With Service Observe modes, use the "No-talk" option (if your switch supports it) in
preference to the "Listen-only" variant. The latter requires an additional time-slot per
concurrent recording - even if you do not inject beep tone.
Where additional timeslots are needed, the total timeslot count must not exceed the
maximum available on that port network (484). Therefore, for a 100% recorded system,
using service observe on a switch that does not support "No-talk" observe, or beep tone
injection do not equip any port network with more than:
z 6 x T1s (=144 calls, 432 timeslots)
or
z 5 x E1s (=150 calls, 450 timeslots)
These guidelines allow for reasonable additional timeslot usage for conferences with other
port networks, shared tones and so on.
Rebalance port networks or add new ones to reduce the timeslot requirement on each to
this level.
AE Services
Each port on a Verint ContactStore for Communication Manager uses a DMCC softphone.
When using Conferenced recording mode (or when an external controller needs to instruct

the record to single-step conference into calls) the recorder also makes use of Avaya CT
services.
From Communication Manager 3.0 onwards, these are both provided by Avaya's AE
Services platform.
Loading
Note:
Note: To avoid overloading an AE Server, do not attempt to record more than
20,000 calls per hour through each AE Server. (20,000 BHCA).
Note:
Note: You must not use more than 1,000 softphones (recorder ports) through a
single AE Server.
If several small recorders are used, you may connect them to a shared AE Server, but only
if the total load on the AE Server does not exceed this figure. If the load imposed by a
single recorder exceeds this figure, you must split the load across multiple smaller
recorders, spreading the load across multiple AE Servers.
Multiple AE Servers
Most Communication Managers can support up to 15 AE Servers but this is a total count -
not just those associated with recording. You may have othe AE Servers associated with
other CTI applications.
Location
In a multi-site system, you should always aim to install an AE server on the same site as
the recorder(s) that is (are) using it. This minimizes the chance of system failure due to
loss of connectivity between recorder and switching system.
Vintage
Verint ContactStore for Communication Manager 7.8 has been tested against AE Services
4.0. Ensure you are running the latest recommended load of AE Services.
Expansion Interface Boards (TN570)

All Expansion Interface Boards must be TN570C Vintage 3 or later.
C-LAN
C-LANs (TN799 DP) are used for two purposes:
z CTI information is passed through them
z DMCC softphones register through them.
Number of C-LANs
To ensure that a C-LAN does not become a single point of failure in a recording system,
you should always provide at least two C-LANs for each AE server. As the CTI load and
channel count increases, you should provide more C-LANs as shown below.
C-LANs per Maximum BHCA through the Maximum Recording

AE Server AE Server Channels through the AE
Server
1 NOT SUPPORTED NOT SUPPORTED

2 12,000 200
3 20,000 400
4 20,000 (AE Server limited) 600
5 20,000 (AE Server limited) 800
Location of C-LANs
For maximum resilience, spread C-LANs across multiple port networks.
To avoid bottlenecks between the port network and the switch, do not connect more than
400 recording ports to the C-LANs in any port network.
Vintage
Refer to the switch documentation for the release of Communication Manager you are
running.
Firmware
The latest update is recommended but these cards must be at least at Firmware update
132.

VoIP Resources
Each port on the recorder will use media processing resources on the Avaya system when
it is active (recording or replaying via the telephone). You must ensure that sufficient media
processing resources are available for the recording and replay load - in addition to any
existing use of these resources within your system.
Resource Requirements
G.711 recording or replay uses less resource than G.729A recording. Note that replay is
always performed using G.711. Depending on the type and version of your Communication
Manager, you may require one or more of the devices shown in the table below.
Resource Type Comments G.711 Recording G.729A

or Replay Ports Recording Ports
Media 64 32
Processing
Resource
TN2302AP
MM760 VoIP Included within 64 32
Module S8300
Media 320 280
Processing
Resource
2602AP
These requirements are solely for the recorder’s ports and are in addition to any used by
other IP phones or other switch components.
Note:
Note: It is not recommended to dedicate Media Processing resource to recording
so it is important to over- rather than under-provide as other users of this
resource could otherwise starve the recorder of this capability. On the plus
side, you may use existing spare capacity in the switch for recording - but
check the location of the resource as well as the amount.
Location of Resources
When adding recording to systems with multiple port networks, it is vital to check that the
recordings do not overload the interconnects between port networks.
If a call cannot be recorded using VoIP resources within a port network that the call would
have been routed through anyway, then the call must be routed to another port network to
reach the VoIP resource. This varies according to whether the phones are digital (DCP) or
IP and, with IP phones, whether the system is IP- or Multi-connect based.
Site the VoIP resources according to the table below.
Recording DCP Phones IP Phones IP Phones

System Multi-Connect IPConnect
Station-side - High Same Port Same Port Network as the trunks

% of calls on Network as the carrying the calls.
trunks recorded. Phones being
(Station Bulk, recorded.
Station Executive
or Unify applying N x VoIP resources N x VoIP resources per trunk channel on
station-based per phone being that port network that could be recorded
rules.) recorded. concurrently.
Station-side - Low Dedicated port network(s). As above

% (<25%) of calls
on trunks VoIP resources equal to N x total number
recorded.(Station of trunk channels that could be recorded
Bulk, Station concurrently
Executive or Unify
applying
station-based
rules.)
Where, N=1 for G.711 recording and N=2 for G.729A recording.
Vintage
Refer to the switch documentation for the release of Communication Manager you are
running.
Firmware
The latest update is recommended but Media Processors must be at least at Firmware
update 105.
Fault Tolerance
You should consider providing one additional board per port network. In the event of a
board failing, a spare is then available to handle the full recording load, without having to
look outside the Media Gateway - which could introduce sub-optimal use of back-plane
timeslots and potentially impact recording in other gateways.

Further Information
For more information, refer to Chapter 2: Administering C-LAN and IP Media Processor
circuit packs, in the Administering Converged Networks section of the Administration for
Network Connectivity for Avaya Communication Manager manual.
Multi-Connect Capacity
Keeping in mind the number and location of recorders and VoIP resources as defined
above, confirm that the capacity of the Multi-connect switch (if present) is not exceeded.
DMCC (IP_API_A) Licenses

Each port on a recorder requires one Avaya IP_API_A license.
TSAPI Simultaneous User Licenses

All recording modes require that Avaya CT be enabled, but only one license is needed.
(DMCC’s Call Information Services are provided by Avaya CT.)
Conferenced recording modes (only) require Type A licenses as follows:
Purpose Licenses Required
For AES to 1
connect
To record Maximum Number of concurrent recordings
calls
To track Number of Skill groups being observed to track logins
agent logins
To observe If recording If recording If recording If recording
stations specific specific specific skill specific
currently of Stations agents groups VDNs
interest
Number of Maximum Maximum 0
stations to number of number of
be recorded these these
agents agents
logged in logged in
concurrently concurrently
Note:
Note: If an external controller is tracking CTI events, then it may require Avaya CT
licenses. In addition to its requirements, the recorder will still need one
license per concurrent recording and AES will still need its 1 license.
VoIP Network Design

The recorder hosts the equivalent of 1 x Avaya 4624 IP Phone per port - whether
recording, replay or live monitor. You must therefore design the connectivity between it and
the rest of the Avaya switch infrastructure as if there were a bank of this many IP phones at
the location of the recorder. Follow Avaya's network design guidelines for this number of IP
softphones operating in either G.711 or G.729A mode, but with 60ms packet intervals.
If the bandwidth between recorder and the media processing resources it uses is less than
LAN speeds (100Mbps full duplex) then you should use G.729A recording only.

Topologies
This document has so far discussed functionality in terms of "applications" without being
specific as to the physical location of these. As the individual components of the recording
system interconnect using IP-based mechanisms, you may distribute the components
across your Enterprise's network in a wide range of topologies.
In small systems, a single server can perform recording, storage and retrieval but in larger
systems, you can separate these tasks onto different servers in a variety of ways as
discussed below. A number of the basic topologies have already been shown in the
diagrams of Recording Methods on page 31. The following paragraphs define the rules
under which each of these basic topologies can be used and introduce the more advanced
topologies required for larger and more fault tolerant systems.
Bulk Recording System

At its simplest level, a Verint ContactStore for Communication Manager system consists of
a single linux server running the Verint ContactStore for Communication Manager software
and configured as a Master. This application provides:
z system administration functions via a browser
z voice recording of Avaya phonesets using DMCC IP softphones
z integration to the Avaya Communication Manager via Application Enablement Services
(AES) for real-time control and tagging of recordings
z integral search and replay capabilities.
See the diagram in Introduction on page 16 for this basic Bulk Recording Topology
For most small to medium sized bulk recording requirements, this single server is all that is
required.
Sampled Quality Monitoring (only) System

When used for sampled Quality Monitoring, an additional software application, Verint
eQuality Balance, is required.
While the underlying Verint Contact Recording Master provides the voice recording
capabilities of the overall system, this optional application adds:
z evaluation form design
z various call selection methods
Topologies
z screen content recording and storage

z agent performance recording and reporting
Bulk Recording + Quality Monitoring System

Bulk recording and Quality recording can be combined in two different ways:
1. Selective Quality Recording + Bulk - in which the quality application chooses to record
specific calls. These recordings are made in addition to the bulk recordings and the
resultant files are copied across to the Quality Monitoring application's file share as
each recording ends. If the bulk recordings are being made using Service Observe, the
Quality recordings are "split" off from these within the recorder - so as NOT to load the
Avaya system any more. If, however, the bulk recordings are mad using conference
methods, the selective quality recordings (which always use service observe) will also
load the Avaya switch.
2. "ContactStore Plus" mode - this is where the bulk recording system advises eQaulity
Balance (R7 only) of all the calls that it has recorded. The Quality application does not
ask for additional recordings to be made. It will simply identify a subset of the bulk
recordings that it will make available for evaluation. In this case the recordings used for
evaluation are the same recordings made for bulk recording.
Large Bulk Recording Systems

Where your recording load exceeds the capacity of a single server, or where a distributed
topology is more appropriate, you may deploy multiple recorders - in one of two ways.
Partitioned Systems
If your recording requirements can be completely separated, you can deploy multiple
independent recorders, each unaware of the others. This applies if you are using:
z Station Bulk mode
z Station Executive mode
z Conferenced mode to target specified stations.
Master/Slave Recorders
Where you are using Conferenced mode to record on the basis of Agent ID, Skill hunt
group or VDN, there must be a single recorder in charge of all recordings.
To support large systems, you can split the control of recording from the actual recording of
the audio stream.

In such cases, the "Master" recorder is connected to the Avaya CT feed and is aware of
the recording rules - and of the type and locations of the other, "Slave" recorders as shown
below.
The Master recorder communicates with each Slave via a TCP/IP link. It instructs the
slaves to tag the required data/voice streams with the details it learns from the Avaya CT
link.
Recorder Type and Location

You may distribute recorders across your network. This lets you trade off network load
versus security of storage.
For example, if you wish to record calls on an overseas site to which you have limited
bandwidth, you can locate a recorder and media processing resources on that site.
High Availability Systems

Because the recording system is based on industry standard PC hardware, you can spend
as much or as little as you like on fault tolerant hardware to increase the reliability of each
server. Verint recommends that you use fault tolerant, hot-swappable RAID disk arrays,
redundant power supplies and fans as a matter of course.
Topologies
Bonded and/or dual NIC cards are strongly recommended. Where two independent NIC
cards are used, these should be connected to independent Ethernet switches to provide
maximum fault tolerance. (Never give the two interfaces IP addresses in the same subnet.)
For still higher availability, you can provide one or more standby recorders. Advanced
configuration options let the recorders match your switch failover modes - supporting both
ESS and LSP.
This is a complex topic and a brief summary of the options is given below. (See also Fault
Tolerant Systems on page 251 for further details of how to design and configure fault
tolerant systems)
Master/Standby Recorders
In small systems, you can install and configure a Standby recorder to "shadow" another
recorder - referred to as the Master. The standby copies configuration details automatically
from the primary via a fault-tolerant TCP/IP connection between the two. Over these same
links, the two recorders exchange "heartbeats" every second. The standby will take over
should the heartbeat fail or should the master request that it do so. This latter case will
occur, for example, due to the master’s disk filling or connection to the AE Services failing.
The figure below shows a typical configuration.

Note:
Note: It is normal to provide Viewer or a Central Replay Server so that all
recordings, whether made by Master or Standby, are visible to end users in
a central database. Otherwise, users would have to know which recorder
was active at a given time and search against its database to be sure of
finding a recording.
In larger systems, where a Master recorder controls one or more "Slave" recorders, the
use of a Standby recorder ensures that the Master does not become a single point of
failure for the whole system. In this case, each of the slaves permanently connects to both
Master and Standby. This allows the Standby to take over rapidly should the Master fail.
Fault Tolerant Pools

When using Conferenced mode with a Master and one or more Slave recorders,
recordings can be directed towards any available recorder. Therefore a "pool" of Slave
recorders can be equipped with one more recorder than is needed for the projected
loading. In this configuration, if one Slave fails, the Master recorder will reassign the calls
being recorded by the failed unit across the remaining slaves.
Centralized Applications
In any system with more than one recorder, you may wish to deploy central search and
replay (Viewer or Central Replay Server) and/or Archive applications. In this case, each of
the recorders:
z provides details of its recordings to the central database
z responds to requests from the central applications for replay of recordings that it is
holding.
See Viewer Installation, Archive Installation and Archive Administrative Guides for sizing
and planning of these servers. Note that the recording system is designed to continue
recording regardless of the state of these central servers. This means that the availability
of the central applications cannot affect the reliability of the recorders themselves.
Integrating with other systems
Integrating with other systems

The Verint ContactStore for Communication Manager system supports integration with a
wide variety of other applications including other CTI feeds, third party and customers' own
applications. There are three options, depending on the complexity of the integration
required.
Supplementary Tagging of Bulk Recordings

It is a common requirement to "tag" recordings with additional details, such as customer
account numbers, trouble-ticket numbers etc. This information is often held in a third party
or custom application and is known while the call is in progress.
The recorder implements a TCP/IP socket based interface. Applications can:
z receive events as recordings start and end
z send "tag" information to the recorder to be associated with the recording, alongside the
basic call details.
Users can then search for recordings based on this enhanced set of call details.
This Recorder Control Protocol is very simple as the recorder makes contact with the
application at an IP address specified in the recorder's administration pages. The recorder
advises the application as recordings start and stop on each station. Each start message
contains a unique reference (or "INUM") to that recording. If the application wants to tag a
call, it sends a "TAG" command specifying the INUM and station number plus the data
formatted as XML.
In this mode, the recorder continues to be responsible for recording rules. Should the
tagging application have a problem, the worst it can do is tag calls incorrectly.
This interface is included within the core Verint ContactStore for Communication Manager
software and you may use it immediately. A typical tagging project requires the purchase of
2-3 person-days of services to learn how to interface to and control the recorder.
See External Control Interface on page 233 for further details of the Recorder Control
Protocol, associated APIs and example applications that are available.
Explicit External Control of Bulk Recording

In normal operation, the recorder is given some basic Station/Agent/Skill Group/VDN rules,
that it uses to determine which calls are to be recorded. However, you can build more
sophisticated recording systems in which the decision to record is made by an existing call

routing/handling application. By including control of recording as an integral part of call

flow, such applications can, for example,
z automatically stop recording a call during security questions and resume once the
customer has been validated
z automatically start recording as the agent accesses the "customer complaint" form
z add additional tagging as the agent traverses a series of menus
Applications can control recording using the same interface as described under
Supplementary Tagging of Bulk Recordings on page 67. By using the additional
commands "START", "STOP" and "BREAK", they can override the rules applied by the
administration interface.
This interface is included within the core Verint ContactStore for Communication Manager
software and you may use it immediately. A typical external control project requires the
purchase of five person-days of services to learn how to interface to and control the
recorder.
Enhanced Quality Monitoring

The above sections primarily relate to Bulk Recording operation. To extend the scope of
the recording rules within the Quality Monitoring application, additional inputs are
connected in two ways.
Using the "eQConnect" toolkit, you can provide additional data fields from other
applications to complement those available from Avaya CT.
Where a layered product (such as Genesys T-server) manages all calls, you can connect
its CTI feed as an ALTERNATIVE to the CTI information received from Avaya CT. In this
case, the Quality Monitoring application connects via Verint Systems' other CTI
integrations to the CTI feed (in this case the Genesys T-server). The decisions about what
to record for quality purposes and how to tag the calls are made by the Quality Monitoring
application on the basis of this CTI feed - without input from the Verint ContactStore for
Communication Manager Master. The Recorder simply acts as a "dumb" recording
mechanism, recording the stations that the Quality Monitoring system requests in
real-time.
Both of these approaches incur additional licensing charges as well as the costs for the
services required to install and configure these options.
Chapter 3: Installation
This chapter gives details of the steps to install a Verint ContactStore for Communication
Manager system.
z Overview on page 70
z Avaya System Configuration on page 71
z Order in which to Install Applications on page 80
z Platform Prerequisites on page 81
z Installing Verint ContactStore for Communication Manager on page 86
z Installing Verint eQuality Balance (V5) on page 88
z Installing Verint eQuality Balance (V7) on page 89
z Installing Viewer on page 90
z Installing Archive on page 94

Installation
Overview
Installation of a complete recording system requires:
z Configuration of your Avaya system to support recording.
z Planning the order in which to install the application servers
z Preparing each server
z Installing the Verint software on each server
z Installing client software for users of the Quality Monitoring application (if required).
Avaya System Configuration

Before installing any Verint ContactStore for Communication Manager components, you
must ensure that your Avaya Communication Manager is correctly configured and, where
necessary, upgraded to support the recording system. As you complete these steps you
will be asked to note a number of details which you will need later when configuring the
recorder.
Prerequisites
Refer to the appropriate Avaya documentation to apply any prerequisite upgrades and/or
additional licenses as detailed in Avaya system prerequisites on page 54.
Communication Manager Configuration

Use the Avaya administration interface to configure the following items:
Customer-options
Set the required system parameters and check that IP_API_A licenses have been installed
as follows:
1. Run the following command
display system-parameters customer-options
2. On page 4, verify that Enhanced Conferencing is set to y.
3. On page 6, verify that Service Observing Basic and Service Observing Remote/By
FAC are set to y.
Note:
Note: Station Bulk, Station Executive, Quality and (optionally) Unify Recording
modes use Service Observe.
4. On page 9 or 10 (it varies depending on your system), check the Maximum IP
registrations by product ID field. This field tells you how many IP_API_A licenses
you have and how many you have already used. If you do not have enough IP_API_A
licenses or if IP_API_A product ID is not listed on this screen, contact your Avaya
representative for more licenses.

Installation
Features
You must set the following system-wide CM parameters.
1. Enter the following command line:
change system-parameters features
2. On Page 5, set Create Universal Call ID (UCID) to y and allocate a number to the
switch if it does not already have a unique reference. If there is only one switch, set it
to 1.
3. On Page 12, set Send UCID to ASAI.
Adding IP softphones
You must add a station on the Communication Manager for each port on the recorder.
Create all stations identically. You will subsequently use the recorder's Administration
pages to assign them to the various modes.
1. Use the add station command to add as many stations as there are ports on your
recorders. Note the station numbers as you will need to enter these into the recorder
later.
2. Run add station xxxx, where xxxx is the new station's extension that you want
to administer.
3. For Station Type, enter 4624.
4. For Security Code, enter the numeric security code the recorder must use to register
softphones with the Communication Manager. Note this security code as you will need
to enter it into the recorder later.
5. Verify that the Class of Restriction (CoR) you choose has Can Be A Service Observer
enabled and Can Be Service Observed disabled when you intend to use any of the
following recording modes:
a. Station Bulk
b. Station Executive
c. Quality
d. Unify/External Control (Service Observe)
! Important:
Important: Verint recommends that you use a dedicated CoR for the stations being
used as ports on the recorder.
The CoR of the stations you want to record with the recording modes mentioned above
must have Can Be Service Observed enabled. For more information on how to
administer Class of Restriction and Service Observing, refer to the following resources:
z Class of Restriction Section, in Chapter 19: Screen Reference of the Administrator

Guide for Avaya Communication Manager.
z Service Observing topics, Feature Access Code and Feature Related System
Parameters Sections, in Chapter 19: Screen Reference of the Administrator Guide
for Avaya Communication Manager.
z Service Observing, in the chapter titled ACD contact center features of the Avaya
Communication Manager, Release 3.0, Call Center Software Automatic Call
Distribution (ACD) Guide.
! WARNING:
WARNING: Since a station can only be service observed once (CM 3.x) or twice (CM
4.x), it is important that you disable, through the use of appropriate CoRs,
the ability of anyone else to service observe those stations that must be
recorded at all times. If you must listen to these stations while they are being
recorded, use additional Live Monitor ports on the recorder.
6. In the IP Softphone field, enter y.
7. Set Display Language to english.
8. On Page 2:
Set IP-IP Audio Connections and IP Audio Hairpinning to n
9. On Page 3:
a. Assign the following feature buttons in addition to the three default call
appearances:
i. Button 4: conf-dsp
ii. Button 5: serv-obsrv
b. Clear the call-appr setting on Button 3.

Installation
change station 11001 Page 1 of 4

STATION
Extension: 11001 Lock Messages? n BCC: 0

Type: 4624 Security Code: 12345 TN: 1
Port: S00081 Coverage Path 1: COR: 1
Name: CCE Line 01 Coverage Path 2: COS: 1
Hunt-to Station:
STATION OPTIONS
Loss Group: 19 Personalized Ringing Pattern: 1
Message Lamp Ext: 11001
Speakerphone: 2-way Mute Button Enabled? y
Display Language: english
Survivable GK Node Name:
Survivable COR: internal Media Complex Ext:
Survivable Trunk Dest? y IP SoftPhone? y
IP Video Softphone? n
change station 11001 Page 3 of 4

STATION
SITE DATA
Room: Headset? n
Jack: Speaker? n
Cable: Mounting: d
Floor: Cord Length: 0
Building: Set Color:
ABBREVIATED DIALING
List1: List2: List3:
BUTTON ASSIGNMENTS
1: call-appr 7:
2: call-appr 8:
3: 9:
4: conf-dsp 10:
5: serv-obsrv 11:
6: 12:
Administering hunt groups

If you want to use On Demand Recording or Meeting Recording modes, consider grouping
these ports into one or more hunt groups for each mode. Users can then access the
recording functionality through the number of the hunt group rather than through individual
ports.
You could assign all ports for a recording mode to a single hunt group to provide a single
shared pool of ports, available on a "first come, first served" strategy. Alternatively, you
could split the pools into several independent hunt groups - or even leave some individual
ports. For example, a dedicated port to be used only by the conference phone in the board
room would ensure that meetings there could always be recorded.
For more information about hunt groups, refer to Managing Hunt Groups in Chapter 7:
Handling incoming calls, in Volume 1 of the Administrator Guide for Avaya Communication
Manager.
Note which ports you have assigned to which hunt groups.
Configuring tone detection

The recorder uses tone detection for Meeting Recording, Station Executive Recording
and Live Monitoring. If you specify a delete command for Station Bulk recording, it is also
used in this mode. To configure tone detection:
1. Type change system-parameters ip-options.
2. In the Intra-System IP DTMF Transmission Mode field on Page 2, enter
rtp-payload
change system-parameters ip-options Page 2 of 2

IP-OPTIONS SYSTEM PARAMETERS
Always use G.711 (30ms, no SS) for intra-switch Music-On-Hold? n
IP DTMF TRANSMISSION MODE

Intra-System IP DTMF Transmission Mode: rtp-payload
Inter-System IP DTMF: See Signaling Group Forms
Network Region setup

The recorder requires the following:
z The DMCC softphones used as recorder ports must be in an IP network region that
supports G.729A and G.711MU with 60ms packet intervals and NO OTHER CODECS.
z There must be a media gateway or a media processor resource in the same network
region or in an interconnected network region.
To set up a network region as above:

Installation
Create a Codec Set

Create a new codec set specifically for the recorder(s) as follows:
1. Choose an unused codec set number for the recorders
2. Use the change ip-codec-set setnumber command to create a codec set that
uses G.729A and G.711MU
3. Verify that G.711MU and G.729A are the ONLY codecs in the codec set.
4. Set Silence Suppression to n.
5. Set Frames Per Pkt to 6 - which will show a Packet Size of 60 (ms).
6. Set the first two Media Encryption options to none and aes respectively.
7. Ensure that FAX, Modem, TDD/TTY are all set to off.
8. Ensure that Allow Direct-IP Multimedia is set to n.
change ip-codec-set 4 Page 1 of 2
IP Codec Set
Codec Set: 4
Audio Silence Frames Packet

Codec Suppression Per Pkt Size(ms)
1: G.711MU n 6 60
2: G.729A n 6 60
3:
4:
5:
6:
7:
Refer to the following for further information:

z For an explanation of administering IP codec sets, refer to the Administering IP Codec
Sets Section, in Chapter 4: Network Quality Administration, of the Administration for
Network Connectivity for Avaya Communication Manager guide.
z For a screen reference, refer to the IP Codec Set Section, in Chapter 19: Screen
Reference, of the Administrator Guide for Avaya Communication Manager.
Create a network region

Create a new network region and assign the previously created codec set to it as follows:
1. Choose an unused network region for the recorders’ softphones
2. Type change ip-network-region region where region is the number of the
chosen network region.
3. Specify the Codec Set created in the previous step.

4. Set the two IP-IP Direct Audio options to No.
5. Set IP Audio Hairpinning to n.
6. Complete the region-region codec table with the number of the Codec Set created in
the previous set.
Refer to the following for detailed information:
z Administering IP Network Regions Section, in Chapter 4: Network Quality
Administration of the Administration for Network Connectivity for Avaya Communication
Manager guide.
z For a screen reference, refer to the IP Network Region Section, in Chapter 19: Screen
Reference of the Administrator Guide for Avaya Communication Manager.
Assign softphones to the network region

To ensure that the recorder's softphones register in the network region created above, use
the change ip-network-map command.
Add the IP address of the AE Server to the new network region so that all the recorder's
softphones - which register via that AE Server - are created in this network region.
The following example shows how to complete the form. In this example, the AE Server is
using network region 1 with an IP address of 192.168.2.100.
change ip-network-map Page 1 of 32

IP ADDRESS MAPPING
Emergency
Subnet Location
From IP Address (To IP Address or Mask) Region VLAN Extension
192.168.2 .100 192.168.2 .100 1 n
. . . . . . n
. . . . . . n
. . . . . . n
Recording of Bridged Line Appearances

Where a bridged line appearance appears on multiple physical sets, and calls made on the
bridged line need to be recorded, you must enable Service Observe Physical Set. This is
also know as the SA7900 Special Feature.
Configure Station Bulk recording to target the main line appearance on each of the
physical sets. When a call is answered on the bridge line, the special feature allows service
observe to observe the bridged appearance call.

Installation
AE Server Configuration
The AE Server provides the recorder with DMCC client services and, (if using Conferenced
recording or eQuality Balance in selective mode) TSAPI services. The following
instructions assume the AE Server has been installed specifically for recording and that
OAM and User Management administrative accounts have been created. If the AE Server
to be used is already in use for other purposes, check these settings and confirm that its
current configuration is appropriate.
Administer C-LANs
Use the AES Administration Screens to add the C-LANs that will be used to register
softphones.
1. From the CTI OAM Admin OAM main menu, select:
Administration > Switch Connections
2. Click Edit H.323 Gatekeeper. OAM displays the Edit H.323 Gatekeeper - Switch1
page.
3. In the Name or IP Address field, type the hostname or IP Address of the switch
C-LAN, and then click Add Name or IP.
TSAPI Configuration
Even if there are no explicit CTI clients, you must configure TSAPI (also known as Avaya
CT). Do this after AE Services have been configured as above.
User Account
If you are using the Security Database for Authentication, create a CTI user account on the
AE Server as follows:
1. Go to: User Management > Add User
2. Complete all of the required fields (indicated by an asterisk).
3. Select userservice,useradmin from the Avaya Role drop-down menu.
4. Select Yes from the CT User drop-down menu.
5. Ensure that the new CTI user has access to all TSAPI-controlled devices by going to:
Administration > Security Database > CTI Users > List All Users
6. Click Enable next to the Unrestricted Access option.
Test Phonesets
You should provide three Avaya phones close to the recorder - in the same or a
neighboring rack. Configure these phones with all of the features in use on the phones that
you intend to record. You can then use these to place test calls while working on the
recorder.

Installation
Order in which to Install Applications

In many cases, the recording system will consist solely of a single Verint ContactStore for
Communication Manager Master server. However, in topologies that are more complex it is
important that you install the basic recording infrastructure first and then layer the other
applications on top.
Note:
Note: If you have multiple independent recorders, install one recorder first, then
the central applications. Finally install the other recorders. This way you will
address any problems with the recorder installation before you have
repeated them many times.
Install server components in the order shown below:
1. Install one Verint ContactStore for Communication Manager (the Master if using
Master/Standby or Master/Slave). Follow the procedures in Platform Prerequisites on
page 81 and Installing Verint ContactStore for Communication Manager on page 86.
2. Install Verint eQuality Balance (V5 or V7) (if required) in accordance with the
appropriate Installation Guide and configure the Verint ContactStore for
Communication Manager to support it as described in IInstalling Verint eQuality
Balance (V5) on page 88 or Installing Verint eQuality Balance (V7) on page 89.
3. Install Viewer as described in Installing Viewer on page 90.
4. Install Verint Contact Archive (if required) in accordance with Archive Installation Guide
and configure it to support the recorders as described in Installing Archive on page 94.
Platform Prerequisites
Before installing the Verint ContactStore for Communication Manager (Master, Slave or
Standby) software on the designated server(s), you must prepare each server as
described below.
Note:
Note: For the other server components refer to the relevant product manual rather
than this one.
Operating System
Version
The operating system must be RedHat Enterprise Linux Version 4.0 Update 5 or higher or
Version 5.0. RedHat Enterprise Linux Version 3.0 is not supported for new installations but
existing systems running on it can be upgraded to CSCM 7.8. The operating system must
be installed using the RedHat kickstart process. Verint supplies a tool to generate the
kickstart script automatically. If you cannot use the kickstart process you must contact
Verint Professional Services for guidance.
Disk Storage
You must plan the partitioning of your server's disk(s) in line with the storage needs
outlined in Storage Requirements on page 43. The kickstart script supports servers with
one logical disk or two physical disks. If using RAID, use the RAID utility to make one large
logical volume.
The kickstart script partitions the disk(s) as follows:
Mount Point Use One Physical or Two Physical Disks

Logical Disk
/boot Bootstrap 100 MB 100 MB on first disk

/ Linux and Verint 6.0 GB 6.0 GB on first disk
Software
Swap Virtual Memory Twice RAM Twice RAM on first
disk

Installation
Mount Point Use One Physical or Two Physical Disks

Logical Disk
/var Linux /var and the 1 - 99 GB Remainder of the first

database (configurable) disk
/calls The recordings Remainder of the Whole of the second
disk disk
! WARNING:
WARNING: The size of /var must be calculated carefully based on the number of
recordings per month and how many months the database records will be
retained.
Creating the kickstart script

The preferred method requires a Windows machine with a CD drive and a floppy drive.
The recorder must also have a (non-USB) floppy. If these requirements cannot be met
follow the alternative process described under Performing a kickstart install without a
floppy on page 83.
1. Format a floppy and insert it into the floppy drive of the Windows machine.
2. Insert the installation disk into the CD drive of the Windows machine. The kickstart
script generation tool will start automatically.
3. Fill in the form with the following information:
a. Select the keyboard layout of the recorder server
b. Select the timezone of the recorder server
c. If you have a corporate NTP server specify its IP address or fully qualified domain
name. If you leave the entry empty the tool picks a suitable public server.
! WARNING:
WARNING: Because good time synchronisation is so vital you must take care with this
setting and test that it is working after installation.
d. Fill in the address and netmask for the first NIC.
e. Specify the address of the default router, which must be in the same subnet as the
address that you specify for the first NIC.
f. Specify the hostname, preferably as a fully qualified domain name (e.g.
cscm1.bigcorp.com)
g. Specify the IP address of a DNS server
h. If there is a second NIC and you want to enable it, fill in the address and netmask.
This address must not be in the same subnet as the first NIC.
i. For one logical drive or physical disk, select 1 disk, and specify the size of the /var
partition.
For two logical drives or physical disks, select 2 disks, and specify the device
names of the two devices. If you do not know the names of the two devices, follow
the first few steps of a normal interactive RedHat installation. After selecting the
keyboard type, the RedHat installer will display a menu showing the names of the
devices.
j. Select the number of Linux tools to install. Minimal chooses just those parts of
Linux needed to run the recorder (Choose this option only if you are familiar with
Linux and will perform all installation from the command line.) Recommended
installs other useful management tools and the Linux windowing system.
4. Click Generate Floppy.
5. Eject the floppy.
Performing the kickstart install

1. Have the floppy just created ready.
2. Boot the target server using the first disk of the RedHat distribution.
! WARNING:
WARNING: Make sure that the disks are the Update you require. RedHat and other
vendors still sometimes supply Update 0 disks.
3. Wait for the boot: prompt.
4. Insert the floppy.
5. Type linux ks=floppy
6. Wait for the automated install to complete and insert the other disks when requested.
Performing a kickstart install without a floppy

If it is not possible to perform the recommended floppy-based kickstart installation, this
HTTP based procedure may be used instead.
1. Follow the same procedure to fill in the kickstart script generation tool, except at the
last step click Generate File. Use the file chooser to specify the location where the file
will be saved. Make sure the file is called ks.cfg.
2. Copy the ks.cfg to a web server and make it available as a web URL. The web server
must not use a non-standard port. It must use port 80. Be certain when copying the file
that it receives no text processing. If copying it using FTP be sure to use BINARY
mode. Do not edit the file. Test that you can view the file using a regular browser.
3. Ensure that the target server will be able to obtain a dynamic IP address using DHCP
on its first NIC. If necessary, temporarily cable its first NIC to a corporate LAN segment.
Ensure that it will be able to access the web server.

Installation
4. Boot the target server using the first RedHat disk, and wait for the boot: prompt.
5. Type linux ks=http://webserveripaddress/path/ks.cfg
Expert kickstart options

Great care must normally be taken to ensure that the ks.cfg file is not corrupted or edited.
The process of creating this file directly to a floppy and taking that floppy directly to the
target server is the recommended approach.
If the file must be edited to access expert options this should be performed on a Unix or
Linux computer. Windows editors introduce additional control characters that prevent the
script from working properly. Unfortunately the installation of Linux appears to work, but the
Verint software will not install correctly.
There is almost no reason to edit the automatically generated file, but details of the script
options are available at
http://www.redhat.com/docs/manuals/enterprise/RHEL-5-manual/Installation_Guide-en-U
S/ch-kickstart2.html
DVD Drive
Install and test your DVD+RW drive by writing a test file to the disk before attempting to
use it with the recording system.
Network Connectivity
Domain Name Server (DNS) Entries
Ensure that the IP node names of all servers that make up the recording system are stored
in the appropriate Domain Name Servers. Subsequent configuration can then be done by
using the host name rather than having to use numeric addresses.
Network Routes
You should ensure that valid IP paths exist between each of the servers and from servers
to the AES, CLANs and all media processing resources. See Recorder Interfaces on
page 199 for details of ports used.
Fault Tolerant Paths

If you are deploying Master/Standby pairs of Verint ContactStore for Communication
Manager Servers on the same site, it is imperative that you provide two totally independent
network paths between each Master and its standby. Failure to do so will leave the system
vulnerable to failure should a common component in the network paths between the two
fail. In such a case, both recorders will attempt to take control of all recording, with
unpredictable results.
To ensure this,
1. Assign IP addresses in one sub-net (call it A) for the first NIC on the Master and the
first on the Standby.
2. Assign IP addresses in a different sub-net (call it B) to the second NICs on the Master
and Standby
3. Connect the NICs in sub-net A to one Ethernet switch and those in sub-net B to a
different switch. (If you do not have two Ethernet switches you can use a cross-over
cable directly between the two NIC cards)
4. Log on to the Standby recorder and ping both IP addresses of the Master.
5. Remove the NIC cable from the first NIC (sub-net A) on the Standby server. Confirm
that you can only ping the Master on its sub-net B address. Replace the cable.
6. Remove the NIC cable from the second NIC (sub-net B) on the Standby server.
Confirm that you can only ping the Master on its sub-net A address. Replace the cable.
7. Log on to the Master recorder and ping both IP addresses of the Standby.
8. Remove the NIC cable from the first NIC (sub-net A) on the Master server. Confirm that
you can only ping the Standby on its sub-net B address. Replace the cable.
9. Remove the NIC cable from the second NIC card (sub-net B) on the Master server.
Confirm that you can only ping the Standby on its sub-net A address. Replace the
cable.
If any of the above tests fail, adjust your network routing and/or firewall settings until you
can be sure that there are two independent paths between Master and Standby and that
these are being used rather than just being theoretically available.

Installation
Installing Verint ContactStore for Communication

Manager
Note:
Note: If you are installing a Central Replay Server, a Standby recorder or Slave
recorder, you should first read the appropriate section in Advanced
Configuration on page 173.
To install Verint ContactStore for Communication Manager on Redhat 4:
1. Log onto the server at the console as root.
2. Type startx to launch the windowing system.
3. Insert the installation CD. Linux should detect the CD and bring up one or two
windows. Minimise any window just showing the contents of the CD. You will be asked
if you wish to autorun.
4. Autorun brings up a small window showing any versions already installed. Click the
install button against each package. The button will change to saying Up-to-date.
5. Reboot the server.
6. Continue the system configuration as described in Configuration on page 95.
To install Verint ContactStore for Communication Manager on RedHat 5:

1. Log onto the server at the console as root.
2. Insert the installation CD. Navigate through the Computer icon to the CD Drive icon.
You will see the contents of the CD.
3. Open a Terminal Window from the Applications menu.
In the Terminal window type:
cd /media/cdromlabel
sh ./autorun
4. Autorun brings up a small window showing any versions already installed. Click the
Install button against each package. The button will change to saying Up-to-date.
To install Verint ContactStore for Communication Manager on RedHat 4 or 5 from the

command line
1. Log onto the server as root.
Installing Verint ContactStore for Communication Manager
2. Mount the installation CD

On RedHat 4 type:
mount /mnt/cdrom
On RedHat 5 type:
mount -r -t iso9660 /dev/cdrom /mnt/cdrom
It may be necessary to create the /mnt/cdrom mount point using the command
mkdir /mnt/cdrom
If the CD drive is known by a different name on your computer (e.g. cdrom1) replace all
reference to cdrom with your device name.
3. Change directory to the mount point using the command
cd /mnt/cdrom
4. Locate the three rpm files on the CD using the command
ls
For each of the RPMs (jre, cscm, cs-service) type
rpm -Uvh rpmfilename

Installation
Installing Verint eQuality Balance (V5)

Install and configure this application as described in Verint eQuality Balance (V5)
Installation Guide.
Then, on the Verint ContactStore for Communication Manager, complete the settings on
the Port Allocations > Quality page as described in Quality Server Ports (for Selective
Quality Recording) on page 130.
Look on the Status > System Overview page on the Master to confirm that the link
between the recorder and the Quality monitoring application is UP. If it does not connect
within 1 minute, check the Alarms page and correct any problems shown there.

Installation Guide. Make sure you have installed the eQConnect component.
Installation Guide. Then,
If using selective quality recording

On the Verint ContactStore for Communication Manager, complete the settings on the Port
Allocations > Quality page as described in Quality Server Ports on page 129.
between the recorder and the Quality monitoring application is UP. If it does not connect
within 1 minute, check the Alarms page and correct any problems shown there.
If using "ContactStore Plus" style quality recording

Make sure you have installed the eQConnect component on the eQuality Balance server.
On the Verint ContactStore for Communication Manager, enter the IP address of the
eQConnect service on the System Settings > Server page.
between the Master and the eQConnect interface is UP. If it does not connect within 1
minute, check the Alarms page and correct any problems shown there.

Installation
Installing Viewer
The installation and configuration process for this server component is quite complex and
has many options. The instructions below describe a basic approach that is appropriate for
most installations. They complement the Viewer Installation Guide as they highlight easily
overlooked items. You should print out the guide for ease of reference as you work through
the installation.
Prerequisites
Before starting to install Viewer, double-check that:
z you have installed the Windows components IIS and Message Queuing services.
z you are not trying to access the server using Internet Explorer 5.0 or earlier. You must
use 5.5 or higher.
z you have either installed SQL Server 2005 Service Pack 2 OR SQL Server 2000 plus
Service Pack 4 and applied the Hotfix as specified in article 899761 of the Microsoft
Knowledgebase.
Follow the Viewer install guide meticulously. It is critical that you set up mixed mode
security and note the System Administrator (sa) password. Consider using "sa" as the
password temporarily and changing it once the system is fully configured. This helps
support personnel access the system to address any initial setup problems you may have.
z check that all the servers in the recording system exist in DNS correctly so that you can
ping each by name rather than just numerical address. You are less likely to make
errors entering names than numbers. Check that you can ping the Viewer server by
name from the recorders and vice versa.
Installing Viewer
You must install this application while logged in as an administrator.
Note:
Note: Do not forget to install the Core components first. Drill down into the
Installation\CoreInstall folder and run setup.exe.
When asked for a username and password for the Viewer account, consider leaving it at
the default username and setting the password temporarily to "W1tn355" and changing it
once the system is fully configured. This helps support personnel access the system to
address any initial setup problems you may have.
Installing Viewer
Create an MMC Console

Viewer is easier to administer if you create a custom MMC console with the three snap-ins
you will need regularly. To do this:
1. Click Start > Run > mmc
2. Select Console > Add/Remove Snap-in
3. Click Add…
4. Select Component Services, Services and Event Viewer and check the Local
Machine checkbox.
5. Save the console as Viewer.msc to your desktop.
Configuring Viewer
Follow Configuring the System in the Viewer Installation Guide, noting the corresponding
points below as you go. These will provide a basic working setup that you can then return
to as you become familiar with the application.
1. Using the console created above, add the local Administrator user to all four roles.
2. Open a browser on the server or any other networked computer and access
http://servername/avaya. If asked, log in as Administrator.
Note:
Note: If you see errors, try adding the server to your list of Trusted Sites - it is
trying to download an ActiveX control to you.
Tip:
Tip: You do not need to add locations at this stage.
! WARNING:
WARNING: You must use the server's hostname rather than dotted IP address (nor the
name localhost) when accessing the url and when entering the domain
name as part of the username if you are prompted for username and
password.
3. Add a system - with the following settings:
a. System Name = your server name
b. System Type = Contact Platform Database
c. Server Name = your server name
d. DatabaseName = EWareCalls (Case sensitive)

Installation
e. Location = Default Location

4. For each Verint ContactStore for Communication Manager, add a Capture Platform
with the following settings:
a. Name = the Verint ContactStore for Communication Manager Server's name
b. Capture Platform Type = Other Recorder
c. Machine Name = your Verint ContactStore for Communication Manager server
name
d. Location = Default Location
e. Contact Platform Database = the one you added in step 3 above
f. Serial Number = the serial number shown on the Settings > License page
g. Port = 8080
h. URL Path=replay (not adapter/replay)
5. Follow the section "Generating a Template". Make sure you log in using the account
and password you set in Installing Viewer on page 90. Add the View,
CSCMAvayaCallsView, giving it the name AvayaView (no space between), Screen
Name Avaya View.
6. Log back in as Administrator.
7. Create a Basic Query. Bring up the browser on http://servername/avaya again
and this time select the Viewer tab and click on the New icon at the top left to start the
Avaya Query Wizard.
a. Step 1/8. Select the query Avaya View that you created in step 6. Click Next.
b. Step 2/8. Skip this step. Click Next.
c. Step 3/8. Select the following fields:
- StartedAt
- Duration
- FirstParty
- OtherParty
- Agent
- Service
- UCID
Click the box at the top right to select all fields for the details tab.
Click Next.
d. Step 4/8. Select the order you wish the fields to be displayed. Click Next.
e. Step 5/8. Choose StartedAt, sort descending as the default sort order. Click Next.
Installing Viewer
f. Step 6/8. Leave the languages as the default. Click Next.

g. Step 7/8. Set the name to Basic Query and leave the remaining fields blank. Click
Next.
h. Step 8/8. Pick an icon and set 100 maximum results. Click Finish.
i. "Release" the Query by selecting the bent black arrow icon at top left.
j. Use the Administration tab to make this query accessible to users.
Configuring Verint ContactStore for Communication Manager

For each Verint ContactStore for Communication Manager Master, Standby and Slave, you
will need to do the following to configure and test the connection.
Do not do this until you have proved that each recorder is recording calls successfully and
that these are playable from the recorder's integral search and replay page.
1. Set the Viewer server's address on the System Settings > Server administration page
2. Check that the Link to Viewer is showing on the Status > System Overview page
and is UP (allow up to 1 minute for connection to be made).
3. If not, check the Alarms page and correct any problems showing there.
4. Make a new test recording.
5. Access Viewer using a browser and search for the new recording.
6. Play the recording.

Installation
Installing Archive
Do not attempt to install or configure Archive until you have proved that Viewer is working
with all recorders.
Follow the installation instructions in Archive Installation Guide.
Because Archive relies on Viewer, there is no additional configuration necessary on the
recorder itself.
Chapter 4: Configuration
This chapter gives details of the steps to configure a Verint ContactStore for
z Overview on page 96
z Accessing the System on page 97
z Licensing on page 99
z Security on page 103
z System Settings on page 105
z System Monitoring on page 113
z Port Allocation on page 114
z Search and Replay on page 136
z Backup/Restore on page 141
z Distributing User Instructions on page 145

Configuration
Overview
You must now configure the recording suite to suit your requirements. This section guides
you through the various tasks in a logical order. You should follow its steps immediately
after installation of the Verint ContactStore for Communication Manager application.
Accessing the System
Accessing the System

Before you can configure the system, you must first:
z access the administration web-interface via its URL
z accept the license terms and conditions
z log in
URL
You administer the Verint ContactStore for Communication Manager system via a web
interface.
1. Open Internet Explorer and navigate to http://servername:8080 using the name
of the server you wish to administer.
2. Read the entire license agreement. If you choose to accept, click I Accept.
3. Enter a username of your choice and leave the password field blank.
4. Click OK.
Note:
Note: The login page uses Javascript. If you see the login page but nothing
happens when you click OK, your Internet Explorer settings may be blocking
this. See ActiveX Control Download on page 138 for detailed instructions on
this and other necessary settings.
Initial User Account

The application will accept any username during the first log on attempt after installation
and will automatically create a local application account for you under that name and give it
administrator rights.
As the password of this account has not yet been set, the web application immediately
directs you to a page asking you to set the password for this account. In this instance,
leave the Old Password field blank and enter a new password of your choice into the two
other fields. Click OK.

Configuration
Key Points
Before using the System Administration pages, familiarize yourself with the following key
points.
Invalid settings
Any of the system's settings that are known to be invalid are shown in red. Use the
information in this guide to change the settings to valid values. If you change a setting, but
submit an invalid entry, a message indicates the reason that the entry is rejected and you
are prompted to re-enter it. To quit without changing a parameter, click on the Close
Window link.
< Previous and Next > links

At the bottom right of some pages are links which guide you through the main configuration
pages in the same order that they are shown in the left-hand navigation bar.
Show All links

At the top of pages that show a list of entries that spans more than one page, the Show All
link appears next to the page selection tags. When you click this link, all the search results
are presented on a single scrollable page.
Page at a Time
If you have clicked the Show All link described above, you can return to seeing one page
of entries at a time by clicking this link.
Impact of changes
When you change a setting, the window into which you enter the new setting explains the
meaning of that setting and the consequences of changing it. Read these notes carefully.
Some settings require you to restart the recorder while others may truncate current
recordings.
Licensing
Licensing
Until you enter a valid License Activation Key or select the Five day timed trial license, the
application will only show you the license entry screen.
Terminology
License Generation Key
This is a three digit number that is specific to a particular server. This is shown on the
license entry page. You will need it to obtain a valid License Activation key.
License Activation Key

This is a long (30 or more characters) string containing the serial number, server type and
channel counts that you have licensed. You must obtain this key and enter it into the
administration pages before your recorder will operate.
Recorder Serial Number

This is a unique identifier for every Verint ContactStore for Communication Manager. For
Verint ContactStore for Communication Managers, this is a 6-digit number in the range
800000 to 889999, which is allocated by Verint Systems as part of the licensing process.
The serial number is issued when the license is purchased and defines the first 6 digits of
the unique reference number given to each call recorded by the recorder. For example, the
recorder with serial number 800001 records its first call into the following files:
800001000000001.wav
800001000000001.xml
The serial number is encoded within every activation key issued. Once a recorder has
been configured with its initial activation key, subsequent keys must have a matching serial
number.
! Important:
Important: The 5-day license option uses serial number 800000. This temporary and
non-unique serial number is the only serial number that you can
subsequently override with the recorder's correct serial number, which is
included in the full license key.

Configuration
Obtaining a License Activation Key

Obtain a license activation key as follows:
1. Open another browser window (on this or another PC) and ensure that pop-ups are not
blocked.
2. Navigate to the Verint licensing website at www.witness.com/avaya.
3. Enter your Username and Password.
4. Click Log in
5. Click License Activation
6. Choose the Serial Number from the drop-down list.
7. Enter the three-digit License Generation Key from the License page in the
Administration application.
8. Enter the appropriate information for the end user.
9. Click Generate Key.
10. Your license activation key is:
a. Displayed on the screen
b. Sent to you through email
Activating the License

1. Return to the Enter License Key Administration page that you have open in another
browser window.
2. Enter the License Key. The license activation key is not case-sensitive, and you can
omit the dashes. If you use a browser on the same machine to obtain the activation
key, you can copy and paste the number between the browser windows.
3. Click Enter.
The page displays the licensed serial number, server type and channel capacity.
Make a note of the license key and store it safely in case you need to reinstall the
application on the same server - in which case you will be able to reuse the key. To reinstall
on a different server, you will need a new key, because the MAC address, to which it is tied
via the three digit license generation key, will be different
Once you have successfully entered a license key, you will be able to access the other
pages of the administration interface.
Licensing
Adding additional licenses

Follow the procedure in the previous section titled "Activating the license" for the
installation of additional licenses. Each activation key you are given includes the serial
number of your recorder and its license generation key, but might not contain details of all
licenses.
For example, if your initial purchase included a Verint ContactStore for Communication
Manager server license with 20 full-time and 30 part-time recording ports, the license key
is lengthy. If you then purchase 10 additional full-time recording ports, your second key
may be shorter as it need not include the part-time licenses.
Note:
Note: Additional licenses require additional ports on the Device, Media AND Call
Control API and potentially more switch components (C-LAN, VoIP
resources, media processing boards).
Reinstalling on the same PC

If you reinstall the recorder software on a new hard disk in the same chassis, you can
reuse your existing activation keys.
If you reinstall the recorder from the installation kit, you must re-enter the activation keys.
If you have more than one license key, re-enter the activation keys in the same order they
were purchased. Otherwise, as in the previous section, "Adding additional licenses," your
number of replay licenses (included in both the keys) will end up at the lower figure if the
original activation key is entered last, overwriting the higher figure (30) included in the later
key.
! WARNING:
WARNING: If you reinstall the software, you must restore the database as described in
Restoring data to a new PostgreSQL database on page 142 before starting
recording. Otherwise the recorder will reuse recording identifiers that have
already been used.
Reinstalling the Recorder on a new PC

For license security, the installation is tied to the first Network Interface Card (NIC) in the
server on which it is installed. To reinstall the recorder on another server, do one of the
following:
Move the first NIC to the new server and use your existing activation keys
Issue 5 May, 2008 101

Configuration
Note the license generation key on the new server and request new activation keys as
outlined in Obtaining a License Activation Key on page 100.
! WARNING:
WARNING: If you reinstall the software, you must restore the database as described in
Restoring data to a new PostgreSQL database on page 142 before starting
recording. Otherwise the recorder will reuse recording identifiers that have
already been used.
Security
Security
Security of recordings is very important and is discussed at length in System Security on
page 163. At this stage in the configuration of your system you should immediately create
appropriate user accounts as described below.
Note:
Note: If you have installed Viewer or a Central Replay Server then you should only
configure Administrator accounts on each recorder. Create and control end
user replay rights on the replay server.
Securing the System

The system automatically creates an initial administrator account as you log in for the first
time. If you wish to create additional administrator accounts, you should do so now. Until
you do so, the only means of accessing the system is with the initial account.
To create a new user account:
1. Click on Security > Users at the left of the Administration screen.
2. Click on Add User.
3. Enter the user's name (with domain name if using Windows domain accounts).
4. Check the User is an administrator box if you wish to grant this user administrator
rights.
Note:
Note: The Comment field is for your own notes.
See Search and Replay Access Rights on page 136 for an explanation of how to use the Is
allowed to replay calls owned by field.
Windows Authentication
If you enter a simple username without a domain qualifier (for example, admin rather than
mydomain\admin) then the account is administered by the recorder application.
Administrators may reset this account's password and the user may change their account
password and log off from the application using the administration web pages.
However, if you specify a windows domain and username, the system will attempt to
authenticate you via your normal Windows domain controller or WINS server(s). In this
Issue 5 May, 2008 103

Configuration
case, you will not see Logoff or Change Password links on the web pages presented by
the recorder.
This feature is also known as "Single Sign On" (SSO) as users only have to log onto their
Windows workstation, but do not need to log on to the search and replay application
separately. See Windows Domain Authentication on page 165 for details of how to enable
this feature.
System Settings
System Settings
You should now configure how the system makes recordings and how it interfaces with
your Avaya Communication Manager. Follow the procedures below, clicking on the
appropriate link at the left of the Administration screen for each section.
Server
This is the section where you specify information about the recorder hardware and
environment.
Note:
Note: It is important that you review and change any of the settings that are not
correctly defaulted.
Optional Internal DVD+RW Drive

This setting shows the path of the DVD+RW drive being used to archive calls - for
example, /dev/cdrom1. If you do not have such a drive, or do not want to use it for
recording archival, leave it blank. It then shows as "Not defined".
Start DVD+RW volume labels with

The recorder sets the volume label on all DVD+RW media it produces. The label consists
of up to five characters that you can set here, followed by a media number. If you have
more than one recorder, set this differently on each one so that disks produced by each
recorder can be told apart.
Maximum total call duration (hours)

To avoid having recording channels permanently active, the recorder will reset a channel
that has been continuously recording for this many hours. Set this field to a value that you
can be sure will not occur for a real call - typically just longer than any one person would
ever be present on a shift.
Maximum recording segment duration (mins)

To optimize the playback experience, the recorder cuts long recordings into segments of
the designated length. A typical value for call centers is 120 minutes. This is the default.
However, if your switch regularly handles longer calls, you may increase this value
Issue 5 May, 2008 105

Configuration
Time between beeps (secs)

If you are using the recorder to inject warning tone onto calls as they being recorded (a
mandatory requirement in some countries/regions) you can specify the interval between
beeps.
Retain call details for (months)

At some point, the size of the call details database will become either unmanageable or will
fill the available disk space. Specify, in months, how long the system should retain call
detail records before they are purged from the system. This ensures that the database
stabilizes at a finite size. Purging is carried out at, or shortly after, 1:00 a.m. each night and
does not affect recording or replay.
Add VDN number as additional "owner" of calls

Access to recordings is normally controlled according to the station or agent that is the
subject of the recording. You may, however, choose to control access to recordings on the
basis of which VDN the call was routed by. To do this, you must first set this option to Yes.
Master Recorder to shadow

If this recorder is licensed as a standby recorder, you must enter the two IP addresses of
the master recorder. Separate the two addresses with a semicolon.
Central Replay or Viewer Server

If using a Central Replay or Viewer Server, set this field to the name of that server. In the
case of a Central Replay Server you MUST also specify the port number (e.g.
my_crs_server:8080). Once you have administered this setting, the recorder queues
call details to send to the replay server.
! WARNING:
WARNING: If you intend to install either a Central Replay or Viewer server and wish all
call recordings to be uploaded to it, set the appropriate address to a
temporary, dummy value before you start recording. Do this even if the
server is not yet ready. This way, all recordings' details will be queued ready
for upload when the server is available. The recorder will raise an alarm
daily until you install and correctly configure the server. Ignore this alarm.
eQuality Connect Server

If this recorder is to provide recordings for an eQuality Balance V7 system running in
"ContactStore Plus" mode, set the name of the server on which the eQConnect interface is
System Settings
running. By default, the recorder will contact it on port 3020. If you need to override this
default, add a colon and the port number. For example: myeqbserver:3030.
Allow Full Database Vacuum on Startup if required

Once every 6 months, the recorder must perform essential database functions. It does this
on startup, but only after warning you that this will happen. If you receive an alarm saying
that this is required but cannot afford to have the recorder do this on its next restart, you
can defer this activity by clearing this checkbox. After each restart, this checkbox is reset
and you must clear it again if you want to defer the task further. You must allow the
recorder to perform these tasks within one month of being alerted.
URL(s) of Unify/External control port(s) to connect to

Specify the IP node name of each external controller. The port number will default to 1414
but to override this add a colon then the port number. If the recorder is supporting multiple
servers, list their names separated by semi-colons.
When using the PDS Connector, for example, you must set the port number to 1415.
When configuring a Slave recorder, it should be configured to contact the Master recorder
on port 1416.
Key Management Server

If you intend to encrypt recordings, refer to Encrypted File Storage on page 169 and set the
name of your Key Management Server here.
Key Management Certificate Passphrase

If you intend to encrypt recordings, refer to Encrypted File Storage on page 169 and set the
passphrase for your Key Management Certificate here.
SMTP Mail "From" Address

To have the recorder email alarms to you or your support staff, first create an email
account that the recorder can use to send emails. Then complete this and the following
four settings. Set this one to the name from which alarm email messages should originate,
for example, recorder1@alt.bigcorp.com.
By default, all alarms are sent via email but you can adjust this using a properties file
setting as described in Properties File on page 174. This restricts the severity of alarm sent
- apart from a nightly message which is always sent to let you know the system is
functioning.
Issue 5 May, 2008 107

Configuration
SMTP Mail Server

Enter the name of the SMTP mail server on which you have established the email account
that the recorder will use to send email messages. If you leave this blank, the system will
not send email messages when alarms occur - and you can then leave the remaining
settings on this page blank.
SMTP Username
Leave this blank if your SMTP server allows unauthenticated sending. If it requires
authentication, set the username of the SMTP account here.
SMTP Password
Leave this blank if your SMTP server allows unauthenticated sending. If it requires
authentication, set the password of the SMTP account here. The password is masked
when entered in this field.
Send alarm/event emails to

Specify the email address(es) to which alarm and event messages should sent. Separate
multiple addresses with a semi-colon (;). The email recipient can be a local system
administrator, a manned help-desk and/or suppliers' support desks if you have a support
agreement that includes this facility. The system sends an email message each time an
alarm occurs or is resolved. It also sends an email once a night as a "heartbeat" to let you
know it is still operating. You should investigate any failure to receive the daily heartbeat
message as it could indicate that the server has failed.
Note:
Note: The system will batch emails for up to 10 minutes to avoid flooding users'
inboxes.
Confirm that you are receiving emails correctly after you make any changes to these
settings.
SNMP Read Community

You can use an SNMP monitoring system such as HP OpenView to monitor Verint
ContactStore for Communication Managers. To do so, set the name of the SNMP Read
Community. The recorder will then respond to SNMPV1 Get messages. The MIB files for
the recorder specific variables are provided on the installation CD.
For security reasons, recorders:
z do not allow "well known" community names like private or public
z do not respond to SNMP Gets until a community name has been set
System Settings
z do not use the usual port of 161, but instead use 2161.
SNMP Trap Destination

Enter the IP address (if any) that SNMP traps should be sent to.
This page lets you set details about the Avaya Communication Manager. The recorder
uses these settings to create softphones that it can use to record, replay and monitor calls.
The settings are described below in the order in which they appear on the screen.
Avaya Communication Manager Name

Set this to the name of the Avaya Communication Manager system you wish to record.
AE Server Address(es)
The IP address of the Avaya Application Enablement Server on which the Device, Media
and Call Control (DMCC) API is running. If you want to specify backup servers, add a
semicolon before each subsequent address. If the first address in the list is unavailable,
the recorder attempts to establish a connection with the next server specified in the list.
DMCC User Name

The user name that the recorder should use to log in to the Device, Media AND Call
Control API.
DMCC Password
The password that the recorder should use to log in to the Device, Media AND Call Control
API.
Encrypt Media Streams

Check this setting to have the audio between the recorder and VoIP resources encrypted.
You can still record calls on which the other party’s audio is encrypted without setting this
option. If you do set it, ensure that you have configured the codec set used by the
recorder’s softphones to support encryption. Also note the impact on server capacity.
Issue 5 May, 2008 109

Configuration
IP Station Security Code

All IP softphones that register with Communication Manager must provide a security code.
The code you enter must match the code entered for all the stations that you created
earlier on Communication Manager for the recorder to use. This field entry is masked for
security purposes.
Avaya CT Server(s)
Enter the IP address of the AE Server that is configured to provide TSAPI (Avaya CT)
services to the recorder. If you have one or more backup servers, you can specify further
addresses, separating them with a semi-colon. Names are attempted in order if the first
one fails.
Avaya CT Service Identifier(s)

Enter the service identifier of the Avaya CT service that the recorder is to use. If you have
specified multiple servers, specify the service name for each, in the same order. Separate
multiple service identifiers with a semi-colon. All services must support the same login ID
and password.
Avaya CT Service Login ID

Enter the login identifier that the recorder should use when accessing Avaya CT.
Avaya CT Service password

Enter the password that the recorder should use when accessing Avaya CT.
Extensions Assigned to recorder

This line shows how many stations you have entered in the list below - and hence the
number of ports available on the recorder.
Service Observe Feature Access Code

Enter the Feature Access Code configured in your Communication Manager that the
recorder should use for Service Observe. Use the code for No-Talk Service Observe,
unless you need beep tone, in which case use the Talk-Listen code.
Unassigned Capacity
This line shows how many ports are still available for allocation to a specific function before
using up your total licensed capacity. This value only refers to the recorder's own licenses.
System Settings
It does not indicate that the Communication Manager has a corresponding number of IP
softphone licenses installed.
Port Range(s)
Here you enter the ranges of stations that you configured earlier for the recorder to use.
Each port on the recorder corresponds to a station number on the Communication
Manager. Ideally, you should choose a single, contiguous range of station numbers.
However, you can choose multiple ranges.
To make subsequent administration as easy as possible, assign sufficient stations to
handle your entire licensed capacity, even if you do not intend to allocate all of these ports
immediately.
To add a port range
1. Click Add Port Range at the bottom left.
2. Enter a range of station numbers.
To edit a port range,
1. Click on the Edit link in the right-hand column.
2. Change lowest or highest port number or both.
To delete one or more port ranges:
1. Click the checkbox in the Select column for each station range you want to delete.
2. Click on the Delete selected port(s) link.
Note:
Note: You can only delete a port range if none of its stations is assigned to any of
the recording modes.
! Important:
Important: Set these port ranges up correctly before proceeding to the subsequent
pages on which you allocate these ports to the various recording and replay
tasks.
For a basic, small system, this is all that you need enter here. However, on larger systems
you may want to click on the Advanced link to set some or all of the following:
z Specify which of several servers should act as standby to a particular set of ports. Enter
the serial number of the server.
z Force the ports to register through a specific C-LAN(s) rather than allowing AES to
determine this automatically. Enter the IP address of the C-LAN they should use.
Issue 5 May, 2008 111

Configuration
! Important:
Important: Since the C-LAN address is only used when registering a softphone on
startup, if you change the C-LAN address of a softphone that has already
successfully registered, it does not re-register to the new address until you
reset it, the recording service, or the server.
Once you have set any of these Advanced options, this setting will be shown on the list of
ports.
System Monitoring
System Monitoring
The status of your recording system is shown on several web pages beneath the Status
link on the Administration interface. Here you can see:
z overall status summary for the server and its interfaces to other components
z loading levels - both current and peak
z individual recording channel detail
Having just configured the recorder's main settings, you should now review the status and
address any problems highlighted on these pages or the Alarms page.
Alarms and events occurring within the system are stored in its local database. You can
view them via the Alarms link on the left of the Administration page. If this link is red, it
indicates that there are one or more uncleared alarms. As you address the cause of an
alarm, you should clear it by clicking the checkbox to the left of it and then clicking the
Clear Selected Event(s) link.
You should configure email alerting and/or SNMP monitoring of the system as described in
System Settings on page 105. The system will then be able to bring alarms to your
attention promptly.
Issue 5 May, 2008 113

Configuration
Port Allocation
After initial setup, configuration of the recorder on a day-to-day basis largely consists of
changing what is being recorded and by which ports on the recorder.
Common Settings
The following settings are used on more than one port allocation page and have the same
meaning on each. This does not imply that all of these settings are applicable to ALL
modes.
Apply Beep Tone within recorder

For each recording mode, you can specify whether or not the recorder injects beep tone.
For a full discussion, see Beep Tone on page 37.
Audio format
In most modes, you can specify whether the VoIP audio is sent to and from the recorder in
G.729A or in G.711. See Audio Format on page 41 for a discussion of compression
formats.
Stop recording if the call drops to just one other party

In some modes, users call a port on the recorder in order to record a call. Because the
recorder port is then itself a party to the call being recorded, the call will stay active if any
one of the other parties fails to hang up.
To avoid the port getting stuck in this way, you can set the recorder to hang up if other
parties on the call hang up - leaving only the recorder and one other party left on the call.
Ports Configured
This figure shows how many ports you have allocated to this recording mode - as detailed
in the table at the bottom of the page.
Unassigned Capacity
This shows how many more ports you could assign before using up the available channel
licenses on the recorder. (It does not indicate that the Communication Manager has a
corresponding number of IP softphone licenses installed.)
Port Allocation
Standby Recorder Serial Number

If you are using ESS or LSP standby configurations you can specify the serial number of
the standby recorder that provides backup services for this range of ports or stations.
Leave this field blank if you have no standby or a single standby recorder. It is only
required for systems where multiple standby recorders each take over a part of the load
during failover.
Warn when free port count falls BELOW

Where a pool of ports is used as a shared resource, this setting triggers an alarm warning
that a pool is running short of available capacity. The warning message identifies the pool
of ports using the Comment field (or the port number range if no comment was entered).
The default for this setting is zero - and as a pool can never have less than zero ports
available, the warning is effectively turned off by this setting. If you set it to 1 the recorder
will alarm when the free count falls below 1 - that is to zero. And so forth.
Recording owner
See Search and Replay Access Rights on page 136 for details of how a recording's
"owner" determines which user(s) can replay it. You can use this setting to override the
default ownership (station or agent number).
For example, if you are configuring a pool of ports for use by the human resources
department, you might find it more convenient to have all recordings made on them owned
by "hr" rather than several different station numbers, as would be the case if you left the
default ownership in place. You can then control access to all of these recordings by
assigning replay rights over "hr" to those users entitled to play them. You can enter an
alphanumeric string or a number. Leading zeroes, if entered, are significant and are
retained. All characters are stored in lowercase, so replay rights are applied case
insensitively.
Record calls that do NOT have a VDN number?

This setting specifies whether calls that do not have a Vector Directory Number (VDN) are
recorded. This typically includes all outgoing calls. When it appears at the top of an
administration page, this rule applies to the recording mode as a whole. It can be
overridden for specific stations or ports using the Advanced settings.
Filter calls by VDN and/or Skill?

This option lets you control whether or not to record on the basis of a call's VDN number or
the skill hunt group the call was answered by. It applies ONLY to calls that DO have a VDN
number. Record calls that do NOT have a VDN number determines how other calls are
handled.
Select one of the following options:
Issue 5 May, 2008 115

Configuration
z Record all calls that have a VDN number

z Record only those that are on specific VDNs or skill hunt groups. You can specify one or
several VDNs, skill groups, or ranges.
z Record only those that do NOT come in on specific VDN or skill hunt groups. You can
specify one or several VDNs, skill groups, or ranges.
! Important:
Important: Enter VDN numbers or skill hunt groups only. Not station or agent numbers.
When it appears at the top of an administration page, this rule applies to the recording
mode as a whole. It can be overridden for specific stations or ports using the Advanced
settings.
Assigning Ports
Links under "Port Allocations" let you assign these ports to any of the uses that are
appropriate given the server and channels licenses you have installed. You allocate these
ports to specific uses in two ways:
z Explicit Allocation - where you specify exactly which ports should be used. This
approach is used for all cases where users dial in to or conference in to the ports in
order to use them. An example is a pool of On Demand recording ports. The station
numbers assigned to this pool must match those you have placed in the hunt group that
you are going to tell your users they should dial for a recorder port.
z Automatic Allocation - where you specify what you want the ports to record or how
many ports are to be used. This approach is used for modes where users do not need to
dial into ports and hence it does not matter which ports the recorder selects for a given
task. An example is Station Bulk recording of a particular station. You tell the recorder
which station you want to record but don't care which of the recorder's stations is used
to service observe the "target" station so it can be recorded.
Read the text in the pop-up dialogs carefully to see whether you are being asked to specify
recorder ports or targets to be recorded.
Ranges
When assigning ports or specifying which stations or addresses are to be recorded, you
can enter single numbers or ranges of numbers. For example, entering "4000-4099" is
much easier to enter than 100 different numbers. Typically you need to enter some
contiguous ranges and some individual port numbers.
A set of recorder ports is referred to as a "pool" and is configured on the appropriate page
of the administration application by entering one or more port ranges. As you enter port
numbers, keep the following in mind:
Port Allocation
z A range can be a single number or a set of contiguously numbered ports or stations.

z Within a Communication Manager system, it is possible to have port or station numbers
that have one or more leading zeros. It is therefore important that you enter any leading
zeros. The start and end ports in any port range must have the same number of digits
even if some of these are leading zeros.
z The lowest and highest port numbers in a range must have the same number of digits.
Therefore, if you have some 4 and some 5-digit ports, you must enter these as two
separate ranges.
Entering a range
Enter a range of ports as follows (station and address ranges are entered in the same
way):
1. Click the Add port range link. The Station Range dialog is displayed.
- To enter a single port, type the number in the top text box.
- To enter a range, type the number of the first port in the range into the top text box
and the final port in the range into the second text box.
2. Add a Comment (optional).
For pooled modes, you can use this field to name a range of ports and to note any hunt
group number that you assigned to these ports. The text you enter appears in status
reports and warning messages as labels for the specified range. For more information
about pooled modes, see Using pooled port modes on page 118.
3. To quit without entering the port numbers, click Close Window.
To enter the port numbers and keep the window open to specify additional port
numbers, click Enter and Stay Open.
To enter the port numbers and close the window, click Enter and Close.
Port ranges are checked for consistency with other ranges and with license conditions as
they are entered. If an entry is invalid, a message indicating the error is displayed; you can
change the information as necessary.
Editing ranges
You can either change or delete the ranges listed. You can change the following fields
without interrupting any recordings that might be active on the port(s) affected:
z Comment
z Prompt User in...
z Warning Level
z Recording Owner
z VDN recording rules
Issue 5 May, 2008 117

Configuration
Some changes require that the softphone be reset and hence active recordings will be
truncated. These are:
z C-LAN address
z Standby Recorder Serial Number
z Any change to the number of ports in the range
z Any change to the port numbers in the range
z A change to the Codec
Deletion terminates any recording activity on a deleted port.
To edit a range:
1. Click the Edit link to the right of the range you want to alter.
2. Edit the range in the port entry form.
3. Click Enter.
Deleting ranges
Delete one or more ranges as follows:
1. Click the checkbox in the Select column for each range that you want to delete.
2. Click Delete selected port range(s).
Advanced settings
To implement advanced settings on a range of ports, stations or addresses:
1. Add or edit a range as above.
2. Click the Advanced link.
3. Enter data in the Advanced fields.
4. To quit without saving the settings, click Close Window.
To enter the port numbers and keep the window open to specify additional port
numbers, click Enter and Stay Open.
To enter the port numbers and close the window, click Enter and Close.
Using pooled port modes

If you use On Demand Recording, Meeting Recording, Unify/Externally Controlled
(Conferenced), Phone Replay, or Live Monitor, you can manage each range of ports
dedicated to a recording mode separately. For example, if you assign two ranges of ports
to Meeting Recording, you can use one for English speakers and one for French speakers.
You can also track usage of individual pools. You could, for example, set up two different
On Demand pools, one for a particular department on one hunt group and one for
Port Allocation
everyone else on another hunt group. You can track the status of these pools through the
Status pages. For each port range assigned to a mode, you can view activity for that port
range in the System Overview and Peak Activity pages.
On Demand Recording
This type of recording uses a pool of ports on the recorder that you can access via one or
more hunt groups.
Mode Setup
At the top of the page are the following settings:
z Apply Beep Tone within recorder
z Audio format
z Stop recording if the call drops to just one other party
z Ports Configured
z Unassigned Capacity
All of these are explained in Common Settings on page 114.
Ports
The table at the bottom of the page lets you assign specific recorder ports to this recording
mode. Refer to Assigning Ports on page 116.
In most cases, you will want to place each of these ports in a hunt group and make your
users aware of this hunt group number and/or set up a key on their phones to access it.
Advanced Settings
Click on the Advanced link when entering or editing a port range to set:
z Free port count warning level
z Recording owner
All of these are described under Common Settings on page 114.
Audix
To allow a user easy access to On Demand Recording from a station, combine On
Demand Recording with the "One-Step Recording via Audix" Communication Manager
Issue 5 May, 2008 119

Configuration
feature. To do this, enter the hunt group used for some On Demand ports as the parameter
for the Audix-rec feature button you assign to the user's station.
When using the Audix-rec feature, keep in mind:
z Recording beep tone can only be applied by the Communication Manager, not by the
recorder
z The user pressing the button will not hear the beep tone.
z Calls recorded using this feature are only indexed with the party that pressed the button,
not the other party on the call.
For more information, refer to the One-Step Recording via Audix topic, Feature Related
System Parameters Section, in Chapter 19: Screen Reference of the Administrator Guide
for Avaya Communication Manager.
Meeting Recording
This type of recording uses a pool of ports on the recorder that you can access via one or
more hunt groups.
Mode Setup
z Stop recording if the call drops to just one other party
z Ports Configured
Note that all Meeting recordings are performed in G.711 to make the production of custom
voice prompts simpler.
Ports
In most cases, you will want to place each of these ports in a hunt group and make your
users aware of this hunt group number and/or set up a key on their phones to access it.
Advanced Port Settings

Port Allocation

z Recording owner.
Tip:
Tip: On Meeting Recording ports, a user can enter a list of owners manually
using the dial pad .
All of the above are described under Common Settings on page 114.
z Prompt users in - Sets the language for spoken prompts. If using multiple languages,
you should configure a hunt group to correspond with each range of ports that uses a
different language.
Custom Prompts
If the language you require is not offered, select Customer defined prompts and provide
your own set of prompts. These files are located in /wav beneath the folder into which you
installed the recorder and are called:
z welcome_custom.wav
z owners_custom.wav
z recording_custom.wav
z help_custom.wav
You can replace these four files with your own recordings. Listen to the corresponding file
in a language you understand (for example, welcome_english.wav) and record the
equivalent messages in your own language.
! WARNING:
WARNING: You MUST use the same G.711 µ-law encoding that the supplied files use.

Unlike the previous recording modes, this type of recording dedicates a port on the Verint
ContactStore for Communication Manager to each station it records. You do not specify
which ports the recorder should use. You simply list the stations that you want to record.
The system chooses from the available ports.
Mode Setup
At the top of the page are the following settings, most of which are are explained in
Common Settings on page 114:
Issue 5 May, 2008 121

Configuration

z Audio format
z Delete Recording by entering - You can specify a digit string that the user can dial to
delete the recording of the current call. When this delete command is issued, call
segments from the past 24 hours with the same UCID and owner as the current call are
deleted from the local database and the audio (wav) and details (xml) files are removed
from the hard disk. If you enable this option, be sure to inform the users of the selected
string. When choosing the string for this command, consider the following:
- A very short string, such as *, can be entered by accident.
- A longer string is more difficult to remember and enter.
Consider using the alphanumeric equivalent for a digit key, for example, **D is really **3
but D for delete is easier to remember.
To disable this option, set this parameter to an empty string - which will show as Not
Defined and the feature is disabled.
Viewer and Central Replay Server do not support call deletion. Since audio files are stored
on the local recorder hard drive, the wav file is deleted when the command is given, but the
call detail record remains on the central database. So, the deleted calls can be searched
for but not replayed. Also, if any of the archive mechanisms copy the recording to an
archive before it is deleted, then this copy will still be present.
z Record calls that do NOT have a VDN number?
z Filter calls by VDN and/or Skill Hunt Group?
z Ports Configured
Stations
The table at the bottom of the page lets you specify which stations are to be recorded. The
system determines which ports will do the recording. It does this to minimize the impact of
changes, leaving previously assigned ports untouched if possible to allow updates to
configurations to be done during business hours rather than having to wait for close of
business.
! Important:
Important: As the recorder picks the ports to use for this mode you must be careful
when YOU choose to assign ports to the other modes. Use the Status >
Port Details page to check if the ports you are about to assign are in use for
this mode before doing so. You can still assign them to the other mode but
any recording in progress will be truncated before being re-established on
another port.
For complete instructions on adding, editing and deleting stations, refer to the earlier
section "Port(s) and station(s)."
Port Allocation
! WARNING:
WARNING: Do NOT enter Agent, Skill hunt group or VDN numbers. You can only record
STATIONS with this recording mode.

z Recording owner
z Filter calls by VDN and/or Skill?
All of the above are described under Common Settings on page 114.

As with Station Bulk recording, you do not specify which ports the recorder should use. You
simply list the stations that you want to record. The system chooses from the available
ports.
Mode Setup
At the top of the page are the following settings, most of which are are explained in
Common Settings on page 114:
z Apply Beep Tone within recorder - note that a third option is available in addition to
the normal Yes and No. You can choose to have beep tone start only once the user has
entered the retain command.
z Audio format
z Retain Recording by entering - You can specify a digit string that the user can dial to
retain the recording of the current call segment. Normally, the recording is deleted by
default as soon as the user ends the call segment, but entering the Retain digit string
during the course of the call instructs the recorder to retain the recording. If you enable
this option, ensure that you inform the users of the selected string. When choosing the
string for this command, consider the following:
- A very short string, such as *, can be entered by accident.
- A longer string is more difficult to remember and enter.
- You can use letters instead of numbers. For example, **R is actually **7 but R for
retain is easier to remember.
z Ports Configured
Issue 5 May, 2008 123

Configuration
Stations
The table at the bottom of the page lets you specify which stations are to be recorded. The
system determines which ports will do the recording. It does this to minimize the impact of
changes, leaving previously assigned ports untouched if possible to allow updates to
configurations to be done during business hours rather than having to wait for close of
business.
! Important:
another port.
For complete instructions on adding, editing and deleting stations, refer to the earlier
! WARNING:
WARNING: Do NOT enter Agent, Skill hunt group or VDN numbers. You can only record
STATIONS with this recording mode.

z Recording owner
Both of the above are described under Common Settings on page 114.
Delayed Call Deletion

Optionally, you can configure the recorder to wait before deleting a recording made in this
mode. If a user decides to retain a call after it has completed, he can do so by ringing a
specific number (the "Retain Port") on the recorder.
Note:
Note: This feature requires one additional port on the recorder.
To implement this configuration, you must add the following lines to the properties file:
execmode.deletedelaymins=NN
execmode.retainnumber=NNNNN
Port Allocation
Where
z NN is a number of minutes from the end of the call within which the user can call the
"retain" number to retain the call
z NNNNN is the station number of an otherwise unallocated port on the recorder that will
be dedicated to receiving Retain commands after hang up.
If you set these properties, call segments recorded in Station Executive mode are not
deleted until the time specified has elapsed after the end of that recording segment. A
retain command entered during or after the recording ends (within the specified period) will
preserve the calls. When determining an appropriate value for this delay, consider the
following:
z You want to maximize the chance of retaining all segments of a call that the Station
Executive user chooses to Retain, but
z You should minimize the time in which unwanted or unauthorized recordings are
available for replay.
The Retain command applies only to the most recent call on a station. So, if a call is
placed on hold and a consultation call is made, this consultation call is now the most
recent.
Calling the retain number when the previous call is still on hold results in the consultation
call being retained. However, if the user resumes the held call, hangs up, then dials the
retain number, the original (and final) call is retained. In this case, the segment before the
call was placed on hold is retained, as long as it ended less than NN minutes before the
retain command was given.
This type of recording uses single-step conferencing to record calls.
It offers a flexible alternative to Station Bulk and Unify/External control for rules-based
recording. Instead of relying solely on DMCC, this mode uses the more powerful CTI
capabilities of TSAPI (also known as Avaya CT), an extra cost option, to track call activity
and control recording.
The recorder does not dedicate a softphone to each recording target. Nor does it suffer
from some of the limitations noted with respect to recordings made via the service observe
feature of Communication Manager.
Recording is triggered by rules that can be applied to four types of "target" addresses:
z Stations
z Agent IDs
z Skill hunt groups
z VDNs
Issue 5 May, 2008 125

Configuration
! Important:
Important: There are a number of prerequisites (including additional Avaya licenses)
and settings that are required for this recording mode.
! Important:
Important: If you are using a Security Database as part of your Avaya CT setup, you
must ensure that the recorder is granted access to all the addresses
(stations, VDNs and skill hunt groups) that it will need to observe.
Mode Setup
At the top of the page are the following settings. Those that are simply listed here are
explained in Common Settings on page 114. The other settings are unique to Conferenced
mode. You must set ALL of these parameters in accordance with the instructions below
before you specify which addresses are to be recorded.
! Important:
Important: Setting this to Yes means that every Conferenced recording will use an extra
timeslot on the port network handling the call.
z Audio format
z VDN(s) to Observe - If you want the recorder to tag and/or control according to the
VDN(s) through which a call is routed, you must enter all of the VDNs in use here.
Separate VDNs with a semi-colon. You can enter ranges of VDNs - but only if all values
within the range are valid VDNs.
Tip:
Tip: If you have ranges of VDNs within which some numbers are not used,
consider creating these as VDNs to allow you to enter the whole range here.
The next two settings operate as described in Common Settings on page 114 but note that
these are secondary tests on a call. Calls must match an address shown at the bottom of
the screen before they are even considered for recording. These tests are then applied to
filter the calls further. These settings should only be used if your main recording rules (at
the bottom of the screen) target stations or agents.
z Test VDN filter against which VDN? - A call may be routed through several VDNs. If
you have specified a rule in the previous setting, you can use this setting to tell the
system whether your rule refers to the first, last or any VDN that the call has gone
through. The default is to use the first VDN. Note that the recorder will only be aware of
VDNs that you listed above. You therefore have considerable flexibility should you wish
to ignore specific VDNs.
Port Allocation
z Tag calls with which VDN? - Each recording can only be tagged with a single VDN.
You can choose whether this is the first or last VDN that a call went through. As with the
previous setting, "first" and "last" are "as far as the recorder is aware" i.e. restricted to
those VDNs you tell it to observe.
z Maximum Concurrent Recordings (on this recorder) - determines how many of this
recorder's ports are assigned to this recording mode. If this recorder is controlling other
Slave recorders, their ports will be in addition to this count.
z Agent Skill Group(s) to Observe - Avaya CT does not let the recorder observe
AgentIDs directly. If you wish to record calls based on AgentID - or even to tag
recordings with Agent Ids and names, you MUST configure this setting. Enter enough
skill groups to ensure that each agent you wish to record is in at least one of these
groups. Separate skill groups with a semi-colon. You can enter ranges of skills - but only
if all values within the range are valid skill hunt groups.
Tip:
Tip: Consider creating one dummy skill hunt group and assign all agents to this
skill. You then need only enter that one skill hunt group here. This also
minimizes the number of Avaya CT licenses you need.
z Ports Configured - this should match the Maximum Concurrent Recordings figure that
you have specified. If it does not, you should provide more recorder ports.
Addresses
The table at the bottom of the page lets you specify which addresses are to be recorded.
These addresses can be station numbers, agent logon ids, skill hunt groups or VDN
numbers.
! WARNING:
WARNING: Using a mixture of different address types can give confusing results. You
should normally choose which type of address you are going to use and not
use the other three. The exception to this is if you can clearly partition your
traffic so that there is no chance of multiple rules firing for a single call.
You can target single addresses or ranges. For example, if your 45 agent identifiers are all
in the range of 7400 to 7450, you can use the range 7400-7450, even though not all these
numbers have been assigned. Use this feature sparingly though as it will take longer for
the recorder to start and recover from outages if it has to try many more addresses than
are strictly necessary.
Issue 5 May, 2008 127

Configuration
! Important:
Important: If you are targeting Agent logon IDs, the recorder notes which of these have
been identified as valid agents and does not attempt to observe them
directly. Instead, it waits to hear that they have been logged on. Should you
ever change your numbering plan so that numbers previously used for
Agent Logon IDs are now used as station numbers, you must restart the
recorder if you wish to record these stations.
The system automatically determines which ports will do the recording. It does this to
minimize the impact of changes, leaving previously assigned ports untouched if possible to
allow updates to configurations to be done during business hours rather than having to
wait for close of business.
! Important:
another port.
You can enter more addresses than you have assigned ports to this mode. The number of
ports assigned is controlled by the Maximum Concurrent Recordings (on this recorder)
setting above - not by the number of entries in this table.
For complete instructions on adding, editing and deleting addresses refer to the earlier
Advanced Settings
Click on the Advanced link when entering or editing an address range to set the following
options. Where these are simply listed below, full details can be found under Common
Settings on page 114. Those settings that are specific to this mode are described in full.
z Recording owner
The above two settings should only be used where the address range you are applying
them to is an agent or station range.
z Trigger recording on alert - If this box is ticked, a call will be recorded if it alerted the
specified address - even if it was never actually connected to this address. If the box is
cleared, calls are only recorded if they actually connected to the specified address.
Port Allocation
Note:
Note: As calls are never actually connected to VDNs or Skill Hunt Groups (they
alert on them but are connected to a station or agent) you must leave this
box checked when specifying VDN or Skill Hunt Group address ranges.
z Continue recording to end of call - If checked, a call will be recorded until the call
ends. If cleared, the recording will stop once the address that triggered recording is no
longer connected to the call. For example, if the recorder is only configured to record
calls made by Agent 1234, then if this box is cleared, the recording will end if Agent
1234 transfers the call to another agent. If the box is checked, recording will continue.
As with the previous setting this must be checked if the address range specifies VDN or
Skill Hunt Groups.
! Important:
Important: These last two settings can cause the recorder to make more concurrent
recordings than you originally planned. As calls are diverted to or transferred
off to other stations, those you have targeted in this mode can take other
calls while those calls are still being recorded.
! CAUTION:
CAUTION: This setting cannot be used if any other application is using the Avaya "Take
Control" feature to control these calls. In order to maintain a CTI observer on
a call after it has left the originally observed address, the recorder implicitly
performs this "Take Control" function itself and only one application can do
so. Note that this effectively blocks the use of recording by VDN or Skill Hunt
Group if another application is controlling call routing. In this case, you must
specify station or agent ids as recording targets and uncheck this option.
Master/Slave Configurations
Because this recording mode can target skill hunt groups, agent ids and VDNs, it is not
easy to partition a large site into several independent recorders (as you can with Station
Bulk recording, for example).
If you need to record more channels than a single server can handle, you can configure
Conferenced recording on one (the Master) and provide one or more Slave recorders to
provide the required number of concurrent recording channels.
To configure a system in this way:
1. Configure the Conferenced recording mode on the Master as normal but specify 0
Concurrent Recording Channels (on this recorder).
2. On each Slave recorder, allocate all ports to On Demand mode and specify port 1416
on the Master recorder as the external controller.
3. On each Slave recorder, add the following line to their properties file:
ondemand.ocpneeded=false
Issue 5 May, 2008 129

Configuration
If using a Standby recorder (recommended to avoid the Master becoming a single point of
failure in a large system),
1. Configure the standby as you would normally, specifying the Master recorder as the
primary recorder it should shadow.
2. On each Slave recorder, add the standby's IP address (again port 1416) to the list of
external controllers.
Quality Server Ports (for Selective Quality Recording)

The Quality Server Ports page is available if you have purchased one or more Quality
Recording channel licenses. On this page, you specify the settings required for the
recorder to interact with one or more eQuality Balance Servers and which ports on the
recorder are available for it/them to use.
Mode Setup
explained in Common Settings on page 114. The other settings are unique to this mode.
z Audio format
! Important:
Important: If you want to use these quality ports to record stations that are also being
recorded in other service observe based modes, you must use the same
codec as these other modes.
z URL(s) of Quality server(s) to connect to - Specify the IP node name of the Quality
server. The port number will default to 1415. If the recorder is supporting multiple
servers, list their names separated by semi-colons.
z Username/Password for file-share on Quality server - Recordings made under the
control of this Quality application are only held temporarily in the recorder's bulk
recording call storage area. As each call completes it is copied onto a location
maintained by the Quality application. If you have configured eQuality Balance to store
recordings in a file-share, you must enter a username and password here so that the
recorder can write recordings to and read recordings from the file-share on the quality
server..
Note:
Note: When you use more than one Quality server, you must use the same
username and password on each one's fileshare.
z Ports Configured
Port Allocation
Ports
For this mode, you typically add a single range of ports - large enough for the total number
of recording and replay ports. The split between recording and replay is set up on the
Quality Server not the recorder.
Advanced Settings
This is described under Common Settings on page 114.
Unify or Externally Controlled Ports

! CAUTION:
CAUTION: Since Version 7.7, this mode is no longer required for most external control
applications. Ports in the other recording modes can now be controlled
and/or tagged by external controllers such as Unify. There is therefore no
need to use this mode. Existing users should consider reconfiguring their
recorder to remove reliance on this mode.
Allocate ports to this mode if you want them to be controlled by or have additional tagging
applied by an external application. This may be Unify, a Verint supplied CTI interface (for
example PDS or AIC connectors) or a third party application.
Mode Setup
explained in Common Settings on page 114. The other settings are unique to this mode.
Note:
Note: There are several additional properties file settings that can be used to
modify how the recorder controls these ports.
z Audio format
z Record Calls Using - Use this option to specify how the recorder interacts with the
external controller(s).
- If you select Service Observe, this mode is configured and acts like Station Bulk
recording. You specify the set of stations to be recorded; the recorder will service
Issue 5 May, 2008 131

Configuration
observe each of them. According to a properties file setting, the recorder either
starts and stops recordings or waits for the external controller to tell it to start and
stop.
- If you select Single-step Conference, Unify or the external server must determine
what to record - either by establishing a single-step conference with a recorder port
or asking the recorder to do so for it. In this case, the external server must provide
most of the call information.
! Important:
Important: If you change this setting you must restart the recorder.
The next two settings are only used if Service Observe is selected above - in which case
they act in the same way as defined for Station Bulk Recording on page 121.
z Filter calls by VDN and/or Skill Hunt Group
z Ports Configured
Ports or Stations
If you have selected Record calls using Single-step Conference, the table at the bottom
of the page lets you assign specific recorder ports to this recording mode. Refer to
Assigning Ports on page 116. You would normally assign a single range of ports in this
mode.
If you have selected Service Observe, this table lets you specify the stations to be
recorded - as described for Station Bulk Recording on page 121.
Advanced Settings
If you have selected Record calls using Single-step Conference, click on the Advanced
link when entering or editing a port range to set:
z Recording owner
If you have selected Record calls using Service Observe, click on the Advanced link
when entering or editing a station range to set:
z Recording owner.
z Filter calls by VDN and/or Skill Hunt Group?
Port Allocation
All of these are described under Common Settings on page 114.
Phone Replay Ports

The Phone Replay Ports page is only available if you have purchased one or more
telephone replay port licenses.
Mode Setup
This mode requires no manual settings. The top of the page simply shows:
z Ports Configured
These are explained in Common Settings on page 114.
Ports
For this mode, you typically add a single range of ports - enough to handle the maximum
concurrent number of replay users the recorder is to support.
! Important:
Important: Replay ports impose a significant load on the recorder. Be sure that you
have specified a powerful enough server.
Advanced settings
These are described under Common Settings on page 114.
Live Monitor Ports

The Live Monitor Ports page is only available if you have purchased one or more live
monitor channel licenses.
Issue 5 May, 2008 133

Configuration
Mode Setup
z Audio format
Note:
Note: This must match the format used to record the stations you wish to live
monitor.
z Ports Configured
Ports
mode. Refer to Assigning Ports on page 116
For this mode, you typically add a single range of ports - enough to handle the maximum
concurrent number of live monitor users the recorder is to support.
Advanced settings
These are described under Common Settings on page 114.
User Account Setup

If you want to use this feature, you must setup an additional user account for each agent or
station that you want to authorize to use it:
z The "username" is actually a number - and must match the agent number (if logged in -
otherwise the station number) that the caller rings in from. (Or the Automatic Number
Identification (ANI) or Calling Line Identifier (CLI) if you want to let the caller ring in from
an outside number.)
z Give the user account replay rights over the station(s) that you want to let that phone
live monitor.
Port Allocation
Hunt Groups
You might want to configure a hunt group on the switch to include all of these live monitor
ports; this configuration makes them accessible through the single hunt group number on a
"first come, first served" basis.
Issue 5 May, 2008 135

Configuration
Search and Replay

If you are deploying a centralized search and replay system using Viewer, you should see
Viewer Installation Guide for instructions on how to customize, deploy and use the
application. However, you should still prove that you can search for and replay calls directly
from the recorder because:
z if you cannot, then Viewer will not be able to play calls either.
z in the event of server or network failure you may need to access recordings directly from
the recorder.
The instructions below relate to the integral search and replay application that is part of the
recorder itself. The accompanying User Guide provides end user instructions on this
application. As the system administrator, however, you may wish to:
z control which users can search and replay recordings
z customize the search fields and/or display
z ensure that the ActiveX control used can be downloaded to your clients' PCs
z modify default behavior via the properties file
The second of these is rarely required and is described with other advanced topics in
Customizing Search and Replay on page 186. The others are needed in most cases and
are described below:
Search and Replay Access Rights

When a recording is made, the recorder assigns its "owner" or, in some cases "owners" as
follows:
z If the Recording owner field is set on the Advanced settings for the address being
recorded, the owner will be the number or name specified there.
z If an owner is not specified but the call was made by a logged in Agent, then the owner
is that Agent's ID.
z If an owner is not specified and the call was not made by a logged on Agent, then the
owner is the station that was recorded.
z If Add VDN as additional "owner" of calls is set on the System Settings > Server
page, then calls routed via a VDN will have a VDN as another owner. In the case of
Station Bulk recording, this is the last VDN the call went through. In the case of
Conferenced Recording, it is either the first or last according to how you have
configured the recording mode.
Search and Replay
Note:
Note: Conferenced Recording is slightly unusual. Where an agent is logged in, the
recording will be "owned" by the station the call was taken on AND the agent
who was logged on.
Meeting Recording however, does not follow the above rules. In this mode, the voice
prompts advise the caller to enter one or more owners.
Tip:
Tip: Use dummy station identifiers to allocate owners to calls made in Meeting
Recording mode. All members of a particular team can be configured with
replay rights for a particular number, even though this is not a valid station
number. When prompted at the start of the call, mark meetings recorded for
and by a team with this "owner" so that all members of the team can access
the recording.
You control which recordings your users can search for and replay by adding a user
account for each person that needs to use the Replay page and specifying which range(s)
of owner they are entitled to see.
To add a user account, follow the same procedure that you used in Securing the
System on page 103 but do not check the User is an Administrator box.
In order to search for and replay a call, the user's replay rights must include at least one
owner of that call. Each user's rights are shown on the Security > Users page and are set
when adding the user account. You can change these by clicking on the Edit link next to
them.
The initial administrator's account is automatically given access rights to all number ranges
up to 10 digits. As you add other users you must specify which ranges of owners each user
is entitled to replay. The number of digits is significant. A user with replay rights over
0000-9999 cannot replay calls made by and "owned by" agent 567 though they could play
calls owned by agent 0567. In this example, you might grant the user replay rights over
0000-9999, 000-999.
Each user account is automatically entitled to replay any calls where the owner field is the
same as their user name. This makes it easy to assign each station to the person who
uses it by setting their username on the Advanced tab of the Port Allocation > Station
Bulk entry for that station.
Typical examples of how to use replay rights are:
z A user allowed to play calls made on his own station (1234) would be given replay rights
1234.
z An Agent who logs on as AgentID 5012 and is allowed to replay his own calls may be
given replay rights 5012.
z A supervisor who logs on as AgentID 5050 and manages AgentIDs 5010-5019 and
5025-5028 may be given replay rights 5050,5010-5019,5025-5028.
Issue 5 May, 2008 137

Configuration
z All recorded stations used by the HR staff have their Recording Owner set to HR on
the Advanced settings. The Human Resources Manager, who uses station 5678 may be
given replay rights HR, 5678
ActiveX Control Download

The Replay page downloads two ActiveX controls, which it uses to decompress the audio
for replay and export. If your default browser security settings prohibit the downloading of
such controls, you need to provide a means of getting the ActiveX control to your users'
desktops.
Internet Explorer (IE) determines rights by putting web servers into zones and then
granting those zones specific rights.
To access and use the Replay page with its ActiveX controls, the recorder must reside in a
zone with the following rights:
z ActiveX controls and plug-ins
- Download signed ActiveX controls
- Run ActiveX controls and plug-ins
- Script ActiveX controls marked safe for scripting
z Scripting
- Active scripting
On Windows Vista you must also clear the "Enable Protected Mode" checkbox.
Your Intranet zone and/or the Trusted Site zone may already have these rights. If so, you
need to verify that the Verint ContactStore for Communication Manager is in one of these
zones. If the recorder is not in the local Intranet Zone or is not a Trusted Site, you can add
it as follows:
1. In the Control Panel, open Internet Options.
2. Click the Security tab.
3. Click the Trusted Sites icon.
4. Click Sites.
5. Uncheck the require HTTPS box (unless you are forcing users to use https)
6. Enter the URL of the Verint ContactStore for Communication Manager server and click
Add.
Note:
Note: Internet Explorer does not recognize that a certain Fully Qualified Domain
Name (FQDN) and IP address are the same; you must add the URL to the
list exactly as your end users are expected to type it in the address bar.
Search and Replay
Before you advise end users of the URL of the recorder, you should make sure that your
users can access the Replay page through your network and that the ActiveX controls
download successfully. To test this:
1. Create a non-administrator account and assign it some replay rights
2. From a typical client machine, enter the URL for the recorder in the form:
http://myservername:8080/ (using the recorder's IP address or hostname -
assuming you have entered it in your DNS server). If using https, replace 8080 with
8443.
3. When prompted, enter the account's Username and a blank Password.
4. Set a new password as directed.
5. Confirm that the Replay page displays correctly and that the ActiveX control is
downloaded.
Tip:
Tip: You may wish to copy steps 2 to 4 above, fill in your URL and send them out
as instructions to your end users.
Modify Default Behavior

You can add or modify lines in the properties file as described in Properties File on
page 174 to change the behavior of the Replay page to:
z limit the number of results returned
z save filter settings from session to session
z restrict access to the export functions:
These settings affect all users and are only updated when the Verint ContactStore for
Communication Manager Service starts.
Limit results returned

Add the following line to the properties file to set the maximum number of calls returned
from a query:
viewerx.limit=nnnn
where nnnn is the maximum number of calls to return. Increasing this maximum over the
default of 100 results in more CPU use and higher network traffic if users choose to view or
accidentally request a large number of calls.
Issue 5 May, 2008 139

Configuration
Save filter settings from session to session

Normally, the entries in the search filter pane are blank when you first access it. If you wish,
the application can remember the last used settings and apply these instead, This feature
is controlled by a setting in the properties file:
viewerx.savesettings=true
Restrict access to the Export function

Access to the individual and bulk call export features is controlled by two entries in the
properties file:
Allow administrators to export calls (Default = true)
replay.export.admin
Allow non-admin users to export calls (Default = false)
replay.export.all
By default, only Administrators can export calls.
Note:
Note: The replay.export.all setting overrides the replay.export.admin setting, so
if replay.export.admin=false but replay.export.all=true, everyone, including
administrators can use the export feature.
Backup/Restore
Backup/Restore
Due to the huge volume of new files created every day, a voice recorder is not backed up
in the same way as most application servers. This section guides you through the issues
around backing up the application, the call details database and the recordings.
Application
The recorder's configuration is stored in its database (using PostgreSQL), alongside the
details of the call recordings. To preserve the configuration of the server, back up the
database frequently as described below.
If you have not installed other applications on the server, there is no need to backup the
operating system or the recorder software. It is faster to reinstall these server components
in the event of disk failure. You should therefore retain the installation media and license
key that you used.
Backing up the Database

You can back up your recorder's database using a command line procedure. (See Local
Archive on page 158 for an automated method of backing up the database if you are using
DVD archive.) The procedure uses the PostgreSQL pg_dump command to extract data
from the database. It must be executed while the database is running. Do not stop the
Verint ContactStore for Communication Manager service or the Postgresql service before
proceeding.
To back up your postgres database:
1. Log on as root.
2. Become the database owner by typing su - postgres
3. Create a backup file by entering the command:
pg_dump --format=c --compress=5 eware > backupfile
You should specify a full path for the backupfile, and consider moving the resulting
backup file to external media or another machine.
Please observe the following guidelines concerning the compression factor:
z 5 is a modest compression factor.
z using a higher number (maximum is 9) makes the backup slower and uses more
resources. However, it results in a smaller backup file.
Issue 5 May, 2008 141

Configuration
z using a smaller number makes the backup faster and uses fewer resources. However, it
results in a larger backup file.
Restoring data to a new PostgreSQL database

Note:
Note: You can only restore data to the server from which you dumped it because
the dump file stores the software serial number and license key information.
These are tied to a MAC address on the recorder. Unless you can move the
original NIC into the new server, you will need to obtain a new license key if
you wish to restore to different hardware.
The following process erases the default database that exists after a complete
re-installation and replaces it with the database that you have backed up.
To restore the database:
1. Re-install the operating system.
2. Log on as root and install the recorder as described in Installing Verint ContactStore for
Communication Manager on page 79.
3. Stop the Verint ContactStore for Communication Manager service.
4. Become the database owner by typing su - postgres
5. Drop the existing database by entering the following command:
dropdb eware
6. Create an empty copy of the postgresql database by entering the following command:
createdb eware
7. Restore the data by entering the following command:
pg_restore --dbname=eware --use-set-session-authorization
backupfile
Backing up Voice Recordings

The Verint ContactStore for Communication Manager stores voice recordings in the
/calls partition. This partition quickly fills up with thousands of directories and millions of
files. When the partition is nearly full, the recorder maintains only a tiny amount of free
space on the partition by deleting batches of 100 recordings (and the directory that
catalogued them) at a time, as it requires space for new recordings. This causes a huge
churn of files every day.
Backup/Restore
Limitations of full and incremental backup procedures

On a Verint Contact Recorder server, two issues make it difficult to back up voice files:
z the file size
z the rate of change of the voice recording files
Together these issues make most traditional backup strategies for the voice recordings
ineffective. Traditional full backups are required more frequently than normal, which wastes
backup media, and incremental backups are larger than expected because of the large
churn of creations and deletions. For a backup strategy to be successful, it must be easy to
restore the data if necessary.
Traditional "full plus incremental" backup solutions are ineffective because these backup
solutions cannot complete fully. In the event of a complete disk failure, the process
restores the full backup, then the increments in chronological order. This procedure
immediately overflows the disk when the restore program tries to create the increments
because the partition holding the call is almost at capacity to begin with. The full plus
incremental backup will fail because it runs out of disk space before it has processed the
"removals" part of the procedure.
Traditional restore procedures are also ineffective. If you use this solution to review a
recording that has been deleted because of age, the recorder immediately deletes any
restored file as part of its disk maintenance.
Finally, traditional backup solutions often require locks on the disk while they work. This
can seriously disrupt the working of the recorder.
Two suitable strategies for audio backup
DVD+RW archive
This simplest and cheapest strategy is to use the built in DVD+RW archive mechanism.
This is not only fully integrated with the workings of the recorder and its search and replay
mechanism, but also is well suited to the incremental recording required for a recorder. As
recordings are added to the calls path they are copied to DVD in an efficient manner. Even
when they have been deleted from the hard disk, the recorder is still able to play them
because it knows which DVD they are on and can replay directly from DVD, without an
intervening 'restoration' step. Each DVD holds about 4GB, which means it can hold about
150 channel-days worth of recordings from a busy system. For less than a dollar a day,
even a busy system can have limitless backup.
Archive Server
The second most effective strategy is to implement the Archive system. This is a
rules-based system. It copies audio files from the Verint ContactStore for Communication
Manager onto different, centralized disks. The data on these centralized disks is
z organized in a more permanent way
Issue 5 May, 2008 143

Configuration
z subject to less "churn"

It is possible to pause the Archive Manager when required, so, if a backup process
requires a disk lock, the downtime does not cause a problem with the server's operation.
This pause feature, together with the way Archive organizes the audio on disk, makes this
data much more appropriate for traditional full/incremental backup solutions.
Distributing User Instructions
Distributing User Instructions

Once you have configured the recorder, you should ensure that the end users know how to
use it. Some users may need to know how to use and control the recording modes. Some
will need to know how to search for and replay recordings.
Those Using Recording or Live Monitor

You will need to advise users of some or all of the following:
Mode If you… You should This information

tell…
On Demand Use this mode at All potential The station and/or

Recording all users of On hunt group number(s)
Demand to dial to reach an
recording appropriate On
Demand recording
port.How to use this
mode.
Configured any Users of these How to use the
Audix-rec stations Audix-rec button
buttons
Meeting Configured any All potential The station and/or
Recording Meeting users of Meeting hunt group number(s)
Recording ports recording to dial to reach a
Meeting recording port
(with prompts in the
appropriate language)
Station Bulk Enabled the Users who may The digits to dial
Recording delete command need to delete during a call to have a
recordings recording deleted
during a call
Issue 5 May, 2008 145

Configuration
Mode If you… You should This information

tell…
Station Use this mode at Users whose The digits to dial

Executive all stations are during a call to have a
Recording targeted by this recording retained
mode
Have configured Users who may The number of the
delayed retention need to retain a retain port and explain
call after the call that only the previous
has ended call will be retained.
They must retain the
call before making
another call from the
same station.
Live Monitor Use this mode at Supervisors that The station and/or
all need to use it. hunt group number(s)
to dial to reach an
appropriate port.
The station or external
phone(s) that they can
dial FROM.
To enter the station
number that they want
to monitor.
Those entitled to replay calls

Note:
Note: This refers to the recorder's integral search and replay. If using Viewer, refer
to the manuals for this product instead.
The integral search and replay application is very straightforward and the online help within
it is rarely needed. If you do not distribute the manual to all users you should still advise
them of the following:
z The url (http or https) they should use to access search and replay
z Their username and how to log in for the first time (unless you are using Windows
authentication)
z Any tips on which data fields would be particularly useful for them to search on
(especially if you have populated any user defined fields).
Configuring Avaya Support Remote Access
Configuring Avaya Support Remote Access

To allow support by Avaya, you need to set the unique RAS server IP address assigned by
the Automatic Registration Tool (ART) when you registered the system.
When ART generates the unique RAS address, it sets the associated client address to the
same as the server address, except with the last octet incremented by one.
To set the RAS addresses:
1. Log onto the CSCM as root
2. Change directory to the /opt/witness/bin directory.
3. Enter the command:
bash SetRASIP serveripaddress
You do not need to enter the client IP address, because the script calculates this for
you automatically.
4. Verify that the changes were made by entering the command:
bash SetRASIP
Issue 5 May, 2008 147

Configuration
Chapter 5: Operations, Administration &
Maintenance
This chapter provides details of regular maintenance required for a Verint ContactStore for
z Introduction on page 150
z Status Monitoring on page 151
z Local Archive on page 158
z Preventative Maintenance on page 160
Issue 5 May, 2008 149

Operations, Administration & Maintenance
Introduction
In addition to initial configuration, there are a number of tasks that need to be performed on
an ongoing basis. This section discusses
z the use of the Status monitoring pages
z the Audit Trail
z preventative maintenance tasks that should be carried out on a regular basis
Status Monitoring
Status Monitoring
Status is shown over five pages that are accessed under the Status heading at the bottom
left of the Administration web interface. These show the current:
z Alarms and Events
z System Overview - links to other components, local archive and storage
z Port States - the state of each recording channel on this recorder.
z Peak Activity - the peak concurrent load on the recorder (and, if it is the Master, on any
Slaves it is controlling)
z Audit trail of all significant configuration and usage events.
Alarms and Events

The Alarms and Events page shows system warnings, alarms and events. The recorder
stores alarms in its database. It deletes them when they are more than a month old.
Check the box at the top of the page if you want to see all Alarms, including those that
have been "cleared". The default is to show only those alarms that have not yet been
cleared.
The radio buttons at the top of the page let you select the minimum severity of alarms that
are shown. The Alarms page will only show the most recent 1000 alarms (in 20 pages of
50 alarms each). If there are Informational or Warning level messages, you may need to
restrict your view to Minor or Major alarms if you want to see further back in time.
Use the links above or below the table to refresh the page and to clear some or all alarms.
Note:
Note: As long as you have set up mail account information on the System
Settings > Server page, an email message detailing alarms and events will
be sent to the address(es) listed there.
Refer to Appendix C: Alarms, for a list of alarms and events that may be generated and
what to do about them.
Viewing alarms and events

The default on the page is to show all alarms and events that are uncleared. You can see
any new or outstanding issues on first viewing the page. To change the set of events
shown:
1. Click on the check box and/or radio button to specify your preferences.
Issue 5 May, 2008 151

2. Click on the Refresh link above or below the table.
Clearing specific events

New alarms and events are initially "active". To "clear" an individual alarm or event so that
it no longer shows:
1. Click the check box to the left of the event.
2. Click the Clear selected events link.
Clearing all events

Be careful using the Clear All Events link. Clearing an alarm without fixing the problem
may lead to system problems being "hidden" without your knowledge.
System Overview
This page shows summary information about the current state of the recorder and should
be checked at least daily.
Overall Status
The title of this page includes the words:
z ACTIVE if the recorder is actively recording. During startup and when another recorder
has taken over (e.g. a standby) this will not be present.
z *** NOT VIABLE *** if the recorder has identified one or more problems that prevent it
from recording (e.g. disk full, cannot connect to DMCC etc.)
Links
This shows the status of each link that the recorder has with other components in the
network. These include, DMCC, Avaya CT, Viewer, eQuality Balance (V5 or V7), Standby,
Slave and Central Replay servers, The status of all these links should be UP.
See Recorder Interfaces on page 199 for a comprehensive list of interfaces.
Current DVD media

Use this entry to confirm that the DVD drive has sensed, and is using, a valid disk. It will
show one of the following values:
z active if it is able to record to the disk
z foreign if it cannot use or recognize the inserted disk
Status Monitoring
z old if a previously closed disk has been reinserted

z empty if there is no disk in the drive.
Unless it is showing Active, or you are replaying from a previous disk, you should insert a
new or the latest partially filled disk so that archiving can continue.
Note:
Note: The DVD entries are only shown if you have configured a DVD+RW drive on
the System Settings > Server page.
Current DVD Label

Use this value to label the current disk. Disks are sequentially numbered. You should label
the next disk (with the next available number) before inserting it.
Free Space on current DVD media

Use this value to determine when you will need to change the DVD media. A blank media
shows 4 GB free space.
Total call segments recorded to date

This value shows the total number of call segments ever recorded by the system.
Total call segments recorded today

Use this value to confirm that recordings are being made today. If you have restarted the
server today, this will show the number of call segments recorded since that restart. This
includes recordings made on behalf of the Quality Monitoring application.
Date of oldest call held on disk

Until your disk has filled for the first time, you should monitor the available space on the
drive. Check that the rate at which space is being consumed is in line with your predictions.
You should be able to estimate when the disk will reach capacity and when the first calls
recorded will be deleted to make way for new calls. This occurs when the free space drops
below 1GB.
Once the disk has started to "wrap" and calls are being deleted daily, use this figure to
monitor the online retrieval capacity. If the figure starts to fall, recordings are using up your
disk space more rapidly than before. The recording volumes are increasing, so you may
need to expand the disk capacity before the duration of calls it can hold falls below your
minimum requirement.
Issue 5 May, 2008 153

Port Usage Summary

The second table on the page shows summary information about the current state of
various port pools on the recorder. The table only shows the modes that you use. The first
rows of the table show the number and percentage of ports in each of the states for each
mode. The final line shows the overall figure for all modes.
The possible states are:
Faulty
These ports are incorrectly configured or have experienced an error. They will be
reregistered two seconds after the initial problem, in an attempt to recover them. If the
problem persists, they will back-off, doubling the time between retries until this reaches 1
minute. They attempt to reregister every minute thereafter.
Starting
These ports are registering or queuing to register with the Device, Media AND Call Control
API.
Idle
These ports have registered successfully, but are on-hook and not in use.
Setup
These ports are off-hook and are:
z Attempting to establish a Service Observe session
z Placing or answering a call
z Receiving instructions from the caller, for example, with Meeting Recording or Live
Monitor
! Important:
Important: A port used for Station Bulk or Station Executive Recording that stays in the
Setup state indicates that Service Observe cannot be established on the
target station. The station might not be configured to allow Service Observe
or another station might already be observing it. Recording is not possible in
either case.
Connected
These ports are in one of the following states:
z Have established a service observe connection but the port being observed is idle
z Are replay ports that have placed a call but are not currently playing a file
Status Monitoring
Active
These ports are:
z Recording ports that are recording a call
z Replay ports that are actively playing a file
z Live Monitor ports that are connected to a station
Port States
The Port States page shows the current state and configuration of each channel on the
recorder:
z The station number of the softphone port on the recorder
z Its recording mode
z the external controller (if any) that is controlling the port
z whether or not a media (audio) stream is flowing to and from the port
z whether or not recording is enabled on this port
z the state of the channel (see Port Usage Summary on page 154 for an explanation of
channel states)
z which station it is recording
To reset an individual port, select the Reset link to the right of the port. This action stops
any current recording on that port.
To force a reset on all ports in quick succession, select the Reset All link at the upper left.
The Recorded Station column for ports in On Demand and Meeting modes shows the
port number that invoked the On Demand Recording and the owner of the Meeting
Recording. (It shows the first one entered if multiple owners were specified at the start of
the Meeting Recording.) Conferenced mode ports show multiple addresses if more than
one address rule has triggered a combined recording.
To update the page, click the Refresh link.
Peak Activity
This page shows the peak loading levels of the recorder. For each recording mode
configured, it shows
z how many channels have been allocated
z the maximum number that have been active concurrently since midnight
Issue 5 May, 2008 155

z the maximum number that have been active concurrently since the date and time shown
above the right-hand column
To reset the monitoring period, select the Restart Peak Activity Counts link. For example,
if your business has a weekly cycle, you may want to reset the monitoring period at the
start of each week. Use this page to predict when to expand your recording capacity.
For more detail of contention and "busy" events on the hunt groups associated with the
pooled recording modes, use the Communication Manager's call detail recording tools.
These tools might be useful if, for example, you are required to provide 98% availability of
Meeting Recording ports.
Audit Trail
The Status > Audit Trail page shows administrator and user actions over a specific
period. The default reporting period is the current day. You can also filter this report
according to Event Type and Username. To generate a report for a different period, enter
the date range in the calendar controls, and click the Refresh link.
The Audit Trail functions track the following user actions:
z Successful user logins
z Failed user logins
z Password changes (although, for security reasons, the actual password is not stored)
z End user searches on the database
z Replay requests
z Live Monitor requests
It also tracks all administrator actions that affect recording, such as configuration changes ,
manual port resets and creation or deletion of user accounts.
Note:
Note: Editing a station range is logged as a deletion followed by an addition.
The Detail column includes the SQL statement used in searching for calls. It also uses the
internal name of a setting rather than the user-friendly, localized name. This avoids any
change of meaning that could occur in internationalization.
Each report is restricted to a maximum of 1000 audit records. To report on more, break
your reporting period into a number of smaller date ranges.
Use your browser's print, save, or email features to provide a permanent record of the
details. To create a summary that presents all results on a single page, click the Show All
link at the top. The Show All and Page at a Time links are not shown if the list of audit
entries is less than one page long.
Status Monitoring
Configuration records (which include the audit trail) are retained for 13 months. Each night
after that period has elapsed, a background job deletes any records older than 13 months.
If you want to retain the records longer, back up the database as described in Backing up
the Database on page 141. You can change this default value of 13 months using the
audit.purgemonths property as described in Properties File on page 174.
Issue 5 May, 2008 157

Local Archive
This section explains how to use a local DVD+RW drive to archive recordings,
The recorder writes both the audio (WAV) and call detail (XML) files to DVD in batches as
calls are recorded. Recording files are copied to the DVD disk when one of the following
occurs:
z 14 hours of recordings have been made (equivalent to 100MB when recording in G.726
or 50MB in G.729A mode)
z 24 hours have passed since the last write to DVD
The administration screens have an option to backup the calls details database to the end
of the DVD. You should consider changing the DVD before it is completely full to leave
space for the database backup.
DVD+RW media control and care

To find calls easily and reliably, ensure that all media are labelled, handled and stored
correctly. To ensure maximum reliability and consistent high quality recording and
playback:
z Follow the drive manufacturer's guidelines for cleaning the drive's lens.
z Follow the media manufacturer's guidelines for storage and handling. Check that the
maximum shelf-life of the disks is adequate for your purposes and, if not, plan to copy
the contents to new media within this period.
Loading a disk
To load a new disk:
1. Insert a blank DVD+RW disk in the drive.
2. Wait for it to spin up and be recognized.
! Important:
Important: Do not close the rack door as the tray will eject and automatically re-insert
during the load process.
3. In the recorder Administration application, click on the System Overview link to
confirm that the recorder has recognized the disk.
Local Archive
The Current DVD media field shows that the disk has been recognized and is being
written to; it also shows the volume label given to the new disk. The Free Space on
current DVD media line should show 4GB or more available on a clean disk.
Changing disks
The System Overview page shows when the disk is full and an alarm is raised.
To eject a disk that is still being recorded in order to retrieve calls from another disk,
1. Click on the Eject link on the System Overview page.
2. Insert the required archive disk.
3. When you have finished retrieving calls, replace the partially full disk so recording can
continue.
! Important:
Important: The archival process never writes to disks out of sequence. For example, if
you eject a partially full disk, then insert a blank disk onto which the next set
of calls is then written, you cannot then reinsert the previous disk.
Tip:
Tip: If you have enough disk space left when you change from a nearly full disk
to a new disk, you can automatically backup the database to the end of the
disk using the Backup database and Eject link
Labeling a disk
As you eject each disk, label the disk itself using an approved indelible marker. Your label
should indicate:
z The identifier of the recorder producing the disk. The identifier is Calls by default; you
can change this value on the System Settings > Server page of the System
Administration application.
z The sequential serial number of the disk.
z The date/time the disk became full.
Issue 5 May, 2008 159

Preventative Maintenance
This section highlights a number of administrative tasks that should be performed on a
regular basis to ensure the system continues to operate smoothly.
Daily
Unless you have fully automated alerting of these conditions, you should carry out the
following procedures at the start of each day:
Alarms
Check the Alarms page for new problems.
Disk capacity
Check the available disk space. The disk where recordings are stored will appear to be at
or near capacity. However, the system consistently maintains a level of 1 GB of free space
by deleting older files. This maximizes the number of recordings that are available online to
you. The Verint ContactStore for Communication Manager's disk manager thread deletes
files on a FIFO (First In First Out) basis. Check the contents of the log files as described in
Troubleshooting on page 205 and examine any errors logged since the previous check.
Look at all error and warning messages, not just those generated by the Verint
ContactStore for Communication Manager services.
System Status
It is difficult to detect some problems automatically. Check the system status regularly via
the Status > System Overview page and verify that all figures are in line with expectations
as described in System Overview on page 152.
Confirm channel status

Use the Status > Port States page of the Administration application to confirm that the
recording ports are in the appropriate states.
Confirm recording and replay

To confirm recording and replay:
z Verify that calls are being uploaded into the database.
Preventative Maintenance
z Use the Replay page to select the most recent calls to verify that calls are accessible.
z Confirm that the start time of these calls matches expectations. Verify that the start time
corresponds to the most recent calls made on the extensions being recorded.
z Confirm that these calls are playable and that audio quality is good.
Archive
If using DVD+RW archive, check the current disk's available capacity. Change the disk
when it fills.
Weekly
As you become comfortable with the normal operation of your recorder, you can reduce the
frequency of the daily tasks. For example, if you know that the rate at which your disk is
filling is not going to fill the available space for several months, you can check it weekly.
Perform the following tasks each week:
Disk capacity: main recording store

When your recorder is first installed, the disk is almost empty. As it gradually fills, you
should note the rate at which it is being used (at least weekly) and extrapolate to estimate
when the disk will be full. At this point, the Verint ContactStore for Communication Manager
will begin deleting the oldest calls to make room for new ones. If this happens to calls that
are younger than planned, check the configuration of the recorder to ensure that only the
anticipated calls only are being recorded. Add additional disk capacity to the partition
before it fills.
Disk capacity: other partitions

Check the available space on any other disk partitions. Verify that these other drives have
sufficient space. The recorder will warn you if they fall below 500MB of free space.
Accumulated temporary files or log files can account for this drop in available space. You
may need to purge them manually.
! Important:
Important: When you are purging files, remember that files you delete go to the
Recycle Bin and that the space they occupy is not freed until you empty it.
Call detail database purging

If you have enabled automatic purging of aged call detail records, you should still monitor
the size of the calls database during the first few months of use. You can then predict how
Issue 5 May, 2008 161

large the database will get by the time old records begin to be purged. Many customers
plan never to purge call detail records, but choose instead to add disk capacity every year
or two as the database grows. If you do this, you should upgrade your server every few
years to compensate for the increasing size of the database and the reduction in search
and update speed.
Configuration Backup
Changes to system configuration that affect user access rights are stored in the
PostgreSQL database. This means that the system configuration is backed up whenever
the call detail records are. See Backing up the Database on page 141.
Monthly
Check the following aspects of the system on a monthly basis:
Loading trends
Note the total call volumes recorded every month to be aware of gradually increasing traffic
trends. To do this:
z Note the number of calls recorded at the end of each month and compare with previous
month's accumulated total.
z Note the age of the oldest call on the disk (only applicable once the disk has filled for the
first time)
z Note the CPU load during busy hour
If it appears that the load is increasing, consider purchasing extra licenses if required
and/or increasing server specification or disk space.
Every Six Months

The recorder must perform a full vacuum of the database approximately once every six
months. As this interval is reached, the recorder issues a daily warning message. This tells
you that it will do a full vacuum on next restart - unless you postpone it by clearing the
checkbox on the System Settings > Server page. You must restart the recorder and allow
it to perform this database maintenance task within one month of being warned about it.
Chapter 6: System Security
Security of customer recordings is very important. This Chapter discusses the various
features - some optional - that you can use to ensure the safety and integrity of recordings.
This chapter assumes that you have suitable firewall, antivirus software and physical
access procedures in place.
The mains sections in this chapter are:
z Use of SSL on page 164
z Windows Domain Authentication on page 165
z Blocking Replay from the Recorder on page 166
z Changing Passwords on page 167
z Encrypted File Storage on page 169
z PCI Compliance on page 170
Issue 5 May, 2008 163

System Security
Use of SSL
You should consider whether you wish to enforce the use of Secure Sockets Layer (SSL).
By default, users can access the recorder via http (on port 8080) or by encrypted https (on
port 8443). You can force users to use the secure https port, by setting Allow
unencrypted (http) access? to No on the Security > Users administration page. When
you do this, any user who attempts to access the recorder through the unsecured (http)
route is automatically redirected to the secure (https) address.
! WARNING:
WARNING: You should not force the use of https if you use Central Replay Server or
Telephone Replay.
The application is distributed with an SSL certificate that is valid for 3 years from the date it
was issued. The certificate makes it possible to give users secure access to the server.
When users access it through this secure https port, the traffic between their browser and
the recorder is automatically encrypted.
However, Internet Explorer will warn your users that the name on the certificate does not
match the name of the server using it. You can either advise your users that this is
acceptable and should be ignored or, for greater security, you may acquire and install your
own SSL certificate as explained in Installing a Signed SSL Certificate on page 303.
Windows Domain Authentication
Windows Domain Authentication

You can create local user accounts within the recorder application itself. However, it is
more secure to use Windows domain accounts and you may wish to enable this feature -
or even restrict access so that only windows domain accounts have access to the system.
To ENABLE Windows Domain Authentication:
1. Create a user account (as described in Securing the System on page 103) who's
username is domain\username - for example, CORP1\JSmith. Note that the
username is case sensitive and must match exactly the case of the username stored in
the domain controller.
2. Add properties to the properties file to define either your domain controller or WINS
controller as follows:
sso.dc=IP address of domain controller
OR
sso.domain=domain name to use
sso.wins=IP address of WINS server to use
To ENFORCE Windows Domain Authentication only:
1. Enable Windows Domain Authentication as above.
2. Log in as an Administrator using a domain account
3. On the Security > Users page, set Allow local user accounts? to No.
Tip:
Tip: If users are prompted for their domain passwords when they access the web
interface, make sure that the recorder is either part of the intranet zone, or
make it a trusted site and configure Internet Explorer to automatically log on
to trusted sites.
Issue 5 May, 2008 165

System Security
Blocking Replay from the Recorder

If you are using Viewer to replay calls, you may want to block users from replaying calls
directly from the recorder itself. To do this:
1. Log in as an Administrator
2. On the Security > Users page, set Allow search and replay from this server? to
No.
Changing Passwords
Changing Passwords
The recorder and related applications use a number of user account settings that are
installed with a hard-coded default. You can change these as follows:
Linux User Accounts

The root and witness user accounts are installed with default passwords. Change these by
logging on to the server and typing passwd.
Be sure to note the new passwords securely as remote support staff will need these to
maintain your system. Should you lose these passwords, your system will become
completely unmaintainable.
Postgres Database Owner

You may wish to change the password used by the recorder to access its local postgres
database. Before doing so, you must first enable 256 bit encryption as described in
Installing Unlimited Strength Encryption on page 302 and obtain the Verint encryption tool
as described in Encrypting Properties File entries on page 308.
1. Log in as root
2. Switch to the postgres user account by typing
su - postgres
3. Access the database by entering
psql
4. Change the password by entering
alter user eware with encrypted password ’newpassword’;
5. Quit the database by entering
\q
6. Encrypt the new password using the WitsBSUserCredentials tool.
7. Add the new password (in encrypted form) to the properties file as
db.password=encryptednewpassword
Issue 5 May, 2008 167

System Security
Recorder Account on eWare/Viewer SQL Server

If you change the username and/or password on the eWare database's SQL server, you
must also set these on the recorder so that it can upload recordings into that database. To
do this:
1. Log in as an Administrator
2. Enter the following url
http://servername:8080/servlet/cscm?cmd=mtce
3. Enter the username and password on the web page shown.
Encrypted File Storage
Encrypted File Storage

Recordings (WAV files) and their associated XML data files are not normally encrypted but
can be - using the AES256 algorithm and RSA Security's Enterprise Key Manager.
Encrypted files are also "fingerprinted" to avoid tampering.
See the manual Impact 360, Full-time Recorder, Enterprise Security Administration Guide
Release 7.8 for details of how to deploy an RSA Key Management Server (KMS). To then
enable encryption on the recorder:
1. Install the Unlimited Strength policy files as described in Installing Unlimited Strength
Encryption on page 302.
2. Create a KMS client certificate (a PKCS#12 file) for the recorder. This will be protected
by an passphrase to preserve the confidentiality of the private key.
3. Install this certificate on the recorder by logging in as witness and copying the file to the
/opt/witness/keystore directory.
4. Back at the KMS, convert the client certificate into a PEM file. The PEM file must be
imported into the KMS to identify the recorder as a valid client. To convert the
PKCS#12 file to PEM format use the command:
OpenSSL pkcs12 -in svr_cert_key.p12 -out svr_cert_key.pem
5. On the System Settings > Server page of the recorder, enter the IP address of the
Key Management Server and the passphrase for the certificate. You must restart the
recorder after making any changes to the KMS settings, certificate or passphrase.
Note:
Note: You must also configure Viewer and Archive (if used) as described in Impact
360, Full-time Recorder, Enterprise Security Administration Guide Release
7.8
Issue 5 May, 2008 169

System Security
PCI Compliance
To make your system compliant with PCI recommendations, you should adopt all of the
above features. In addition:
1. Do not store sensitive data in the User Defined Fields of recordings.
2. Use the PAUSE and RESUME recorder control features to avoid recording sensitive
information. This will require integration with your other systems.
3. Do not reduce the default 13 months for which audit records are kept.
4. Use encrypted audio to and from Avaya Communication Manager (setting on System
Settings > Communication Manager page).
5. Force users to use Viewer rather than replaying calls from the recorder.
6. If using Quality recording, use Version 7.8 or higher. Version 5.x does not support
encryption.
7. Ensure all other components of your system (Viewer, Archive etc.) are configured in
accordance with the Enterprise Security Administration Guide.
8. PCI dictates a session timeout of 15 minutes. When using Windows Domain
Authentication (as required to meet other PCI requirements) the recorder will accept a
valid Windows logon (on an appropriate account) as sufficient to gain access to the
recorder's administration pages. You should therefore ensure that all users' PCs are
configured to launch a screen saver after 15 minutes of idle time and have Password
Protection on Resume enabled. To ensure this, domain administrators should lock
down these settings in the group policies of the domain controller.
9. Review and adjust your Windows Domain policies for user accounts if required. For
instance, the PCI specification mandates the following rules:
z Immediately revoke access for terminated users.
z Remove inactive user accounts every 90 days.
z Do not use group, shared or generic accounts and passwords.
z Change user passwords at least every 90 days.
z Passwords should be at least 7 characters long and should include both numeric
and alphabetic characters.
z Do not allow individuals to submit a new password that is the same as any of the
last 6 they have used.
z Lock out the user account after not more than six unsuccessful access attempts
(have a lockout duration of 30 minutes or until admin enables the account).
10. To ensure no recorded data is stored anywhere in an unencrypted format, including on
the supervisor's PCs, Internet Explorer's Advanced Security Do not save encrypted
pages to disk and Empty Temporary Internet Files folder when browser is closed
PCI Compliance
settings on the supervisors' PC must be enabled and locked down. (Internet Options
> Advanced > Security).
11. To ensure proper authentication when SSL is enabled in a recording system, the
following advanced Internet Explorer security settings on Supervisor PCs must be
enabled
z Check for the publisher's certificate revocation.
z Check for server certificate revocation.
z Warn about invalid site certificates.
z Use SSL 2.0.
z Use SSL 3.0.
z Use TLS 1.0
Issue 5 May, 2008 171

System Security
Chapter 7: Advanced Configuration
This chapter provides an overview of the more complex and rarely used options for a
Verint ContactStore for Communication Manager system.
z Properties File on page 174
z Standby Server on page 182
z Central Replay Server on page 183
z User Defined Fields on page 184
z Customizing Search and Replay on page 186
z Usage Report on page 191
z Selective Record Barring on page 193
Issue 5 May, 2008 173

Advanced Configuration
Properties File
A number of system settings can be changed from their default values by placing entries in
the properties file as described below:
z this is a plain text file, located in the installation path and with filename:
/opt/witness/properties/cscm.properties
z you should edit this file using the vi text editor when logged on as witness - NOT root.
z the Verint ContactStore for Communication Manager service reads this file as it starts,
so any changes made to the file will not take effect until you next restart the service.
z the file is normally empty as all settings default as shown in the table below.
Most of these entries are discussed elsewhere in this manual, in the appropriate context.
The table below provides a summary of the available settings.
Entry (case sensitive) Default Meaning
audioserver.inactvitytimeout 30 The number of seconds of inactivity

before an audioserver port is released
back to the pool of available ports.
audioserver.viewerbase64 not Use this setting to specify the base64
disclosed encoded name that should be included
in requests to Audio Server. You only
need to set this if you have changed
Audio Server from its default.
audit.purgemonths 13 The number of months to keep audit trail
entries in the database.
conf.hanguponmediastop false Conferenced mode recording ports no
longer hang up (by default) when the
media flowing to them stops. If you are
experiencing "stuck" or "zombie" ports
left up long after calls should have
finished, consider setting this. However,
it will make the system more sensitive to
shuiffling and may impact other
applications that are watching the CTI
events for these calls.
Properties File
conferenced.defaultrecord true Whether or not conferenced mode ports

will start to record as soon as they have
been conferenced into a call. Rather
than setting this false, consider using On
Demand ports and having the external
controller determine which port is used.
(It is difficult to know that a conference
mode port has connected to a call
unless it sends a STARTED message -
and it will not do this if it does not start to
record when it connects to the call.)
conferenced.ocpneeded true Whether or not Conferenced recordings
should detect other call parties using the
conference display feature. Set false if
an external controller is to provide this
information instead.
cmapi.timeout 65 How long (in seconds) the recorder will
tolerate failure of DMCC.
cscm.disablecompress false Normally, recordings made in G.711 will
be compressed to G.726 16kbps. Set
this to true to disable compression.
cscm.diskmanager true Whether to delete oldest calls when
needed to create space for new
recordings.
cscm.localport 8080 The http port used by the recorder. Must
match the entry in server.xml.
cscm.logkeepdays 30 Number of days log files to retain before
purging.
disk.stopatMB 10 Stop writing to disk if free space is less
than this many MB
disk.warnAtMB 500 Raise warning if free disk space falls
below this many MB
dmcc.port 4722 The port number on the AES for DMCC.
dmcc.secure true Whether to connect to DMCC using
SSL.
dmcc.trustall false Allows different SSL certificates to be
installed on the AES.
dnis.fieldname spare2 The field name to tag DNIS information
under. Applies to Conferenced mode
recording only.
Issue 5 May, 2008 175

email.minalarmlevel 0 Alarm level at or above which, emails

will be sent . Default, 0 is INFO level. Set
to 1, 2 or 3 to restrict emails to
Warnings, Minor and Major Alarms
respectively.
eqc.noagents false To use eQuality Balance 7.x when
agents do not log in, set this property to
true. The recorder will then spoof an
agent login on the first call taken on a
recorded device.
eqconnect.agentdevice Avaya Agent Device string sent to eQConnect.
eqconnect.connecttimeout 120 Seconds to keep trying to connect to
eQConnect on startup before reporting
failure.
eqconnect.heartbeatinterval 2 Seconds between sending heartbeats to
eQConnect
eqconnect.inactivitytimeout 10 Seconds of inactivity after which to
report eQConnect failure.
eqconnect.systemdevice Avaya System Device string sent to eQConnect
eware.forcesameserver false Normally, the recorder looks in the
SystemSQLServerView in the eWare
Site database to determine which SQL
server to use for its calls database.
When using SQL server clusters, you
may need to set this property to true.
The recorder then assumes that the
Calls database is on the same server as
the Site database.
eware.recorderhostname no Name by which the recorder is entered
default into Viewer rather than its fully qualified
value IP name. Should not be needed after
Viewer 7.6.1
eware.sqlserverport 1433 IP port to use when connecting to SQL
server.
execmode.deletedelaymins 0 Number of minutes to wait before
deleting Station Executive mode
recordings.
execmode.retainnumber no The station number of the recorder port
default to be used as the Station Executive
value "Retain" port.
Properties File
jtapi.required true Normally, if JTAPI (over Avaya CT) has

been configured and is in use, it is
considered a critical error if it fails. If a
standby recorder is present it will be
asked to take over. This can be stopped
if JTAPI is not critical to the recorder.
jtapi.retries 3 Number of times to keep trying to
establish single-step conference.
jtapi.retrydelayms 500 Interval in milliseconds between
single-step conference attempts.
jtapi.timeout 65 How long (in seconds) the recorder will
tolerate JTAPI failure to respond to
heartbeats.
master.localport 1416 TCP port on which the Master recorder
listens for slaves
meeting.defaultrecord true Whether or not Meeting recording ports
will start recording automatically at the
start of a call or wait for a START
command from an external controller.
meeting.ocpneeded true Whether or not Meeting recordings
numberplan.maxlength 7 Number of digits above which a
telephone number is assumed to be an
external number. Used in Conferenced
mode only.
ocp.alwaysrefresh false In recording modes where the
conference display button is used to
determine who is on the call, this
process will be restarted if the display
indicates that a party has joined the call.
In some cases, however, an existing
party may be updated e.g. a trunk
name/number may be replaced by a
(late arriving) ANI. Set this property true
to have ANY change to the display
trigger a refresh of all parties on the call.
Issue 5 May, 2008 177

ocp.strictparsing false When forwarding via coverage answer

groups, the conference display may not
contain a single numeric field in the end
of the display. Previously this would lead
to the recorder continually pressed the
Conf Disp button - resulting in the party
being removed from the call. Now
default is not to check for spaces or valid
number in this string. To restore previous
behaviour if this causes problems, set
this property = true.
ondemand.defaultrecord true Whether or not On Demand recording
ports will start recording automatically at
the start of a call or wait for a START
ondemand.ocpneeded true Whether or not On Demand recordings
an external controller (such as a Master,
if this is a slave recorder) is to provide
this information instead.
ondemand.ocpblocked false If enabled, prevents OCP. Must be
enabled for slave recorders.
origucid.fieldname spare3 The field name to tag original call UCID
information under (i.e. the UCID of the
call that a consult call relates to). Applies
to Conferenced mode recording only.
postgres.password not Encrypted password to use for database
disclosed vacuum.
primary.localport 1414 TCP port on which the recorder listens
for standby recorders.
purge.hourofday 1 Hour at which database and log file
purges will take place. Default is 1am.
Valid range 0 - 23
qcm.keepwav false Whether to retain Quality Monitoring
recordings in the recorder's call path.
queue.xxx.threads 1 Number of threads to use for job queue
xxx.
Properties File
rec.keepnullucids 5 Whisper announcements are recorded

in service observe modes. These result
in (largely useless) recordings of a few
seconds duration that can be identified
as such by the lack of a Universal Call
ID.(UCID) The recorder therefore
deletes recordings without a UCID that
are shorter than the number of seconds
specified here. To turn the feature OFF,
set the value to 0.
rec.maskallowed false You must set this to true to enable the
"pause/resume" features of the
Recorder Control Protocol.
rec.mincallduration 250 Calls shorter than this many
milliseconds are deleted.
replay.export.admin true Whether administrators are allowed to
export recordings.
replay.export.all false Whether all users are allowed to export
recordings. (If set true, overrides
replay.export.admin=false)
rtp.packetlog false If true, log details of every RTP packet
received
slave.inactivitytimeout 10 How long the master will tolerate lack of
heartbeat with slave recorder before
reporting error.
snmp.port 2161 Port on which recorder listens for SNMP
sso.dc no The IP address of the domain controller
default to use for Single Sign On. (Incompatible
value with sso.domain.)
sso.domain no The domain name to use for Single Sign
default On. (Incompatible with sso.dc.)
value
sso.wins no The IP address of a WINS server to use
default to locate domain controllers. (Mandatory
value with sso.domain.)
standby.backoffdetect true Whether the Master recorder checks for
a standby taking over.
standby.connecttimeout 120 Seconds a recorder will wait on startup
before reporting that a previously
connected primary/standby has failed to
connect.
Issue 5 May, 2008 179

standby.inactivitytimeout 60 Seconds the Master will wait between

polling messages before reporting a
Standby as having failed.
standby.localconfig false Whether the standby is locally
configured. If not, it copies config from
the Master.
standby.reconnecttimeout 10 Seconds the standby will attempt to
reconnect before taking over.
standby.sscdelay 5 Number of seconds standby waits for
Master to act on single step conferences
before taking over.
stnbulk.defaultrecord true Whether or not Station Bulk recording
ports will start recording automatically at
the start of a call or wait for a START
stnbulk.ocpneeded true Whether or not Station Bulk recordings
stnexec.defaultrecord true Whether or not Station Executive
recording ports will start recording
automatically at the start of a call or wait
for a START command from an external
controller..
stnexec.ocpneeded true Whether or not Station Exedcutive
recordings should detect other call
parties using the conference display
feature. Set false if an external controller
is to provide this information instead.
unify.defaultrecord false Whether Unify/external controlled ports
will record by default or only when
instructed to by the controller.
unify.ocpneeded false Whether Unify/external controlled ports
will attempt to provide call details via
DMCC.
usage.reporting false Whether to track actual usage to the
ConfigHistory table and show the Usage
report
unity.required false Whether a link to Unify is required for the
recorder to be viable.
Properties File
uui.fieldname no User defined field name to store Avaya

default User to User data. If NULL, this data is
value not stored. Applies to Conferenced
mode recording only.
vacuum.interval 180 Number of days between full database
vacuum.
vdn.mapsize 10000 How many entries to hold in the
UCID/VDN map. The recorder holds two
maps and alternates between them.
viewerx.limit 100 Maximum number of recordings
returned for each search
viewerx.savesettings false Whether to save and reuse users’
previous entries in search filter pane.
Issue 5 May, 2008 181

Standby Server
First, determine the number and type of standby servers required using the Topology
tables in Standby Recorder Options on page 254
Standby servers are installed in the same way as a standalone recorder (seeInstalling
Verint ContactStore for Communication Manager on page 86 - steps 1-7). The license key
determines that the server will act as a standby.
Follow the configuration guidance given in the appropriate table:
z Where a Standby is to copy the configuration of the Master recorder (the default) then
you need to configure very few other details. Work through the Configuration chapter of
this manual configuring just those fields with "Edit" links next to them.
z Where specific standby recorders need to be specified, ensure you have set the
Standby Recorder Serial Number for each set of softphones and each set of stations on
the Station Bulk or Conferenced Mode settings page.
z Where a standby recorder is to be locally configured,
1. Add the following line to the properties file (See Properties File on page 174 for
detailed instructions)
standby.localconfig=true
2. Restart the Verint ContactStore for Communication Manager service
3. Configure all settings as you would on a standalone recorder.

You can deploy the recorder application onto an additional server which collects details of
recordings made on one or more other servers and therefore can act as a dedicated
search and replay server. Users can then search for recordings from any of the recorders
feeding it with a single url and a single search.
There are differences in the functionality between the Viewer application and the recorder's
integral replay application as used when a recorder is deployed as a Central Replay
Server. Refer to the Viewer documentation to compare it with the functionality described in
the User manual for this recorder.
Installation
Install a Central Replay server in the same way as the Master recorder (see Installing
Verint ContactStore for Communication Manager on page 86 - steps 1-7). The license key
determines that the server will act as a Central Replay Server.
Configuration
Work through Chapter 4: Configuration configuring just those fields with "Edit" links next to
them. Then configure the user accounts and replay rights for those entitled to use the
application.
Issue 5 May, 2008 183

User Defined Fields

The recorder can store any number of user-defined fields (UDFs) in its internal
PostgreSQL database. You can search on these fields using the Search and Replay web
application.
Storing UDFs
Use an external controller to TAG calls using the interface described in Appendix
D: External Control Interface. The database automatically notes new UDFs and stores the
call's details.
Search and Display UDFs

To search on and display UDFs, copy the browse query (named "browse") in the postgres
database to a new file that you will use to create a new query. Name this "browse2" or
something more descriptive.
Use the stored procedure udf_value(inum, udfid) to add the chosen UDFs to the
new view.
For example, the following fragment of SQL inserted into the SELECT statement's list of
fields adds an additional column called spare1:
…… udf_value(calls.inum, 1) AS spare1
The udfid is determined by searching the udfnames table for the relevant udfname. The
spare1 column is then available to be used in a customized search and replay view as
described in Customizing Search and Replay on page 186.
UDF Support in Viewer

When one compares UDF support in this recorder (and hence Central Replay Server) to
UDF support in eWare, this recorder has the following advantages:
z Unlimited number of user defined fields.
z Unlimited length of each user-defined field.
z User defined search forms created by editing the xml templates rather than the end-user
interface available within Viewer.
User Defined Fields
z Better support for sparsely populated data sets in which there are more user-defined
fields, but not every call has every field present.
Viewer Compatibility
If you name your user defined fields spare1 through spare8 you can upload them to a
Viewer database as well as have them in the local database. The name by which the users
see these fields can be set in the template files.
Issue 5 May, 2008 185

Customizing Search and Replay

By renaming and modifying the layout files provided with the Search and Replay
application, you can define search forms and results display list formats. You can also
provide alternate language translations for the fields and columns shown.
Default Configuration Files

The default layout is determined by two files located
/opt/witness/properties/layouts:
z lcscm.xml is the default layout
z idefault.xml is the default internationalization file
! WARNING:
WARNING: lcscm.xml and idefault.xml must not be modified. They are overwritten
during upgrades. You will make your changes in files with different filenames
(of your choice).
Note:
Note: The web server picks up changes that result from additional files being
placed in this folder within a minute. There is no need to restart. However, it
makes sense to make these changes when there are no users on the
system.
Custom Layout Files

Layout files are very straightforward. Each file specifies a single "layout". The filename
must be of the form: lsomething.xml. The default, lcscm.xml is shown below:
<?xml version="1.0" encoding="UTF-8"?>

<layout name="cscm" display="ldefault">
<view>browse</view>
<filters>
<filter type="date" dbName="startedat"
display="fdate"/>
<filter type="string" dbName="partynames"
display="fparty"/>
<filter type="string" dbName="partyagentnames"
display="fagent"/>
<filter type="duration" dbName="duration"
display="fduration"/>
<filter type="string" dbName="vdn"
display="fdnis"/>
<filter type="ucid" dbName="callid"
display="fcallid"/>
<filter type="callset" dbName="csnames"
display="fcallset"/>
</filters>
<columns>
<column dbName="startedat" columnWidth=""
formatType="DATE" display="cdate"/>
<column dbName="duration" columnWidth=""
formatType="TIME" display="cduration"/>
<column dbName="partyagentnames" columnWidth=""
formatType="" display="cagent"/>
<column dbName="partynames" columnWidth=""
formatType="" display="cparty"/>
<column dbName="vdn" columnWidth=""
formatType="" display="cvdn"/>
<column dbName="callid" columnWidth=""
formatType="UCID" display="ccallid"
enableLikeSearch="true"/>
</columns>
</layout>
The outer tag is <layout>, within which the internal name of the layout is given by the
name parameter and the display name token is given by the display parameter.
It makes sense to name the file linternalname.xml but this is not mandatory.
The web application presents the layouts in alphabetical order by internal name in the
layout name drop list box.
There are 3 inner tags:
z View: The name of the query on the database
z Filters: The filters for the left panel
z Columns: The columns for the grid display
Issue 5 May, 2008 187

Inside the outer filters tag, there are many filter tags. Each must have the following:
z a type, which can be date, string, duration, ucid or callset
z a column name matching one of those in the specified database view
z a display token which is looked up in the internationalization file.
Inside the outer columns tag there are many column tags. These must have a column
name, width, format and display. They may optionally have an enablelikesearch flag. See
the default above for all valid display types.
Widths may be valid HTML table widths (that is, nnnpx for pixels or mm% for percentages).
However, it is better to leave these blank and let the browser decide.
If enablelikesearch is true the application displays this field as a hyperlink that, when
clicked, executes a search for all calls (+/- one day) which have exactly the same data in
this column as the call clicked upon.
(The example use of this is for universal call id - which you can click on to find all recorded
segments of this call).
Internationalization
The file idefault.xml contains translation strings that allow you to format the output in any
language. The current default is shown below.
<?xml version="1.0" encoding="UTF-8"?>

<tokenset>
<language lang="en">
<token tid="ldefault">Default Layout</token>
<token tid="fdate">Call Start Range</token>
<token tid="fparty">Parties</token>
<token tid="fagent">Agent</token>
<token tid="fduration">Length</token>
<token tid="fdnis">Service</token>
<token tid="fcallid">Universal Call ID</token>
<token tid="fcallset">Call Set</token>
<token tid="cdate">Call Start</token>
<token tid="cduration">Len</token>
<token tid="cagent">Agent</token>
<token tid="cparty">Parties</token>
<token tid="cdnis">Service</token>
<token tid="ccallid">Univ. Call ID</token>
</language>
+ <language lang="es">
+ <language lang="it">
+ <language lang="ja">
+ <language lang="zh">
+ <language lang="fr">
</tokenset>
It has an outer tag <tokenset> with inner tags <language>, which in turn have inner tags
<token>.
Note:
Note: The naming convention adopted is as follows: the display tokens for layouts
begin "l", for filters "f" and for columns "c". This is only a convention.
When the application wants to display the name of a layout, it looks up its display token in
the internationalization maps by name, according to the user's language preferences set in
the browser.
For example, lcscm.xml has the display token ldefault, the string for which (in
English) is "Default Layout".
To change/add strings you can supply a file called icustom.xml.This file must contain a
tokenset, as described above. It is merged with the tokenset in idefault.xml. Because
the file icustom.xml is read after idefault.xml, any duplicates will use your values
rather than supplied defaults.
For example, if your French users dislike the default translation of call ID (ID d'appel) and
would prefer "ID", then supply an icustom.xml with the following:
Issue 5 May, 2008 189

<?xml version="1.0" encoding="UTF-8" ?>

<tokenset>
<language lang="fr">
<token tid="ccallid">ID</token>
</language>
</tokenset>
This text provides the override of the original value, and the column displaying call id is
now labelled ID.
If you are designing additional layouts, you may use any or all of the supplied token ids and
these will be translated from the default file.
If you wish to label things differently, you may think of additional token ids, use those in
your layouts and provide translations for them in your icustom.xml. You only need to
provide translations for your own languages.
However, if you wish to override the default layout, you may create an lsomething.xml
file with a valid layout in it, and give that layout the name "cscm". This file replaces the
default layout.
Note:
Note: Do not call your file lcscm.xml. If you do, your file will be overwritten during
upgrades.
Examples
To change the labels shown

Create a icustom.xml file and override the ones you do not like.
To change the order of the columns

Copy lcscm.xml to lcust.xml. Edit the latter and delete columns or change their order.
As you have not changed the name of the layout, it replaces the default one and you still
have only one layout.
To create additional layouts

Copy lcscm.xml to lsomething.xml. Edit the latter and change the <layout> tag to
give it a new name and a new display name. Make other changes as necessary. The new
layout appears in alphabetical order by internal name. Create an icustom.xml. Add a
translation for the display name token that you gave your new layout.
Usage Report
Usage Report
The recorder is occasionally sold to service providers with a "per use" license. Charges are
levied based on the actual usage made of a recorder each month. This report can be
enabled in such cases to provide the necessary billing information.
Enabling the Report

To enable this feature, you must add the following line to the properties file:
usage.reporting=true
Content
The Usage Report page shows a summary of the recording modes that have been used
over a specific period. The default reporting period is the previous calendar month.
To generate a report for a different period, enter the date range in the calendar controls,
and click Refresh.
Usage of most modes is reported straightforwardly but the Bulk recording row is a count of
those stations that have been
z Configured for Station Bulk Recording
and/or
z Recorded using a Verint Quality Recording port at any time during the reporting period
The number in the Usage column shows:
z the maximum size of the pool for On Demand, Meeting, Telephone Replay, Live
Monitoring, Conferenced, and Unify/External using single-step conference.
z the total number of different stations that have been recorded in Station Bulk, Station
Executive, Unify/External modes during the reporting period.
The Stations Recorded column shows the individual stations and/or station ranges that
were targeted.
Configuration records are retained for 13 months. Each night after that period has elapsed,
a background job deletes any records older than 13 months.
Issue 5 May, 2008 191

Accessing through URL:

This report can be accessed directly, without having to use the normal system
administration pages.
To access the report by way of its URL, enter the following line in the navigation bar of your
browser:
http://myservername:8080/servlet/report?from=t1&to=t2
Where t1 is the start time in UTC seconds and t2 is the end time in UTC seconds.
Note:
Note: These times are in seconds not milliseconds and do not attempt to correct
for leap-seconds. Unless an administrator made configuration changes
within a few seconds of midnight, this will not affect reporting on monthly
boundaries.
The names of recording modes and other information that can be localized are returned
according to the language preferences established for the interface. If localized terms are
not available, the returned values are in English.
Accessing the Usage report in a log file

If you request the usage report by way of its URL and the request is successful, the
recorder writes the usage data to a log file called usage.log in the
/opt/witness/logs directory beneath the install path.
You can access and view this log file in any text viewer. Its content is the same as the
report returned for the URL request. Each time you request a usage report, a new log is
created that overwrites the previous one.
If a request for the Usage report is not successful, no log file is written for it. You should
examine the return value of the URL request for an indication of the error conditions (bad
time parameters, configuration has been tampered with).
Selective Record Barring
Selective Record Barring

Recent legislation changes mean that some customers may need to block recording of
calls to or from specific area codes (e.g. California).
Configuration
It is possible to bar recording of calls to or from certain numbers. To configure such a
“recording bar”, add the property file entry:
recording.barred=<regular expression>
Where <regular expression> is a regular expression that will match the digit strings to be
barred from recording.
Tip:
Tip: See http://java.sun.com/docs/books/tutorial/essential/regex/index.html for
instructions on how to form a regular expression.
Example
The following example shows how to bar calls to or from area codes 234, 567 and 890 –
where the recorder is situated in area code 234 (which therefore does not have a 1 in front
of it, unlike the others which may or may not). The trailing periods (there are seven of
them) are important – as this forces the pattern to match only numbers with 7 digits
following the area code.
recording.barred=((234)|(1?567)|1?890)).......
Any recording that is barred due to matching the digit pattern specified will cause an INFO
level message to appear in the log file.
Limitations
1. This feature applies to Station Bulk, Station Executive, Unify/External Control (Station
Bulk) and Conferenced modes only (On Demand, Meeting and Externally controlled
Conference mode are not affected).
2. To bar incoming only or outgoing only, first determine the digit patterns that are used.
You may be able to change the outbound dial plan to always prefix with a ’1’. The "1?"
Issue 5 May, 2008 193

matches calls with or without a preceding ’1’. Remove the ’?’ to require a ’1’ before the
pattern is matched.
3. Behavior differs slightly between the Conferenced and Service Observe modes. The
former break each recording into separate segments whenever the parties on a call
change e.g. consultation call becomes 3-way conference call. The record/bar decision
can therefore be applied on each segment and an internal consult call will therefore be
recorded even though the resultant conference call (involving a barred party) will not.
Service Observe, however, does not break the media stream at this point and hence a
single recording segment containing the consult call and the three-way call is the norm.
In this case, if a barred party is present at all (i.e. would appear in the list of parties
tagged on the recording) then the entire segment – in this case including the consult
call – is deleted rather than stored.
4. Note that the Service Observe modes may record calls (as Station Executive mode
does for all calls) that are in progress but then delete the recording and associated
data file at the end of the call.
Appendix A: Technical Reference
This appendix provides technical details about the Verint ContactStore for Communication
Manager system.
The main sections in this appendix are:
● Recording files on page 196
● Internal Database on page 197
● Recorder Interfaces on page 199
Issue 5 May, 2008 195

Technical Reference
Recording files
Call segments are stored in an industry standard wav file. When each call is completed
and as each recorded call segment becomes available, the recorder updates its local
database with a record of the call segment. These files are stored in a hierarchy of folders
beneath /calls.
Every recording results in one or more:
● wav file
● xml file
WAV files
The wav files contain the actual audio of the recording. You can double-click some wav
files to play them directly. Others are in audio formats that are not directly supported by
Microsoft's Media Player. This applies to most recordings made by this recorder. These
must be converted into a supported format before they can be played. Since the recorder's
Search and Replay application does this conversion automatically, you do not need to
access these files directly.
XML files
The xml files contain details about the recorded call segments. Although most users
typically search against the recorder's database of calls, you can view these files directly in
a browser if required.
Within each xml file there is:
● All the details known about this recording. Most of the information, but not all, is inserted
into the calls database. Some of the information is only of interest for diagnostic and
maintenance purposes.
● Start and end time in ISO format giving local time and offset from GMT.
Internal Database
Internal Database
If you have retained all of the xml and wav files as described above, then you have kept all
of the details about the recordings you have made. However, the system uses an industry
standard database (PostgreSQL) to hold this information in more readily accessible forms.
This database is located on the recorder itself. The database stores details of the
recordings as well as details of the recorder's configuration.
Recording details
The call details database uses approximately 1KB per call (in the absence of user defined
fields).
To allow you to search for calls easily, the details of recordings (except those made for and
copied off to a Quality monitoring server) are inserted into this database. It contains one
record for each call segment recorded and additional records for each party on the call and
each owner of the call. The information stored for each call is:
● A unique reference for the recording
● The start date and time
● The duration of the recording
● The recording mode that was used to make the recording
● The Communication Manager's Universal Call Identifier (UCID)
● The name and number of the parties on the call-where this was available to the
Communication Manager (through ANI or CLI) at the time of the call
● The agent number and name of an agent involved in the recording if one was logged on
at the station being recorded
● The direction of the call (incoming or outgoing)
● The owner(s) of the recording
● The service or vector directory number (VDN) name where available
● Additional fields provided by external controllers
● The DNIS if present (Conferenced mode only). Stored as user defined field "spare2"
Issue 5 May, 2008 197

Technical Reference
Configuration details
Several tables hold details of system configuration, such as port assignments, file paths,
timeouts and user authorization rights.
Recorder Interfaces
Recorder Interfaces
The interfaces supported by the recorder are described below (working clockwise round
the diagram, starting at top left).
HTTP Interfaces Offered

The recorder uses the tomcat (see www.apache.org) web servlet container to offer a
number of services via HTTP and/or HTTPS (on ports 8080 and 8443 respectively). These
provide
Administration Interface
This provides administrators with access to configuration and status monitoring pages.
Issue 5 May, 2008 199

Technical Reference
Search Interface
End users access this to search for call recordings that match specific criteria.
Replay (Retriever) Interface

End users and other applications (such as Viewer, Vision and Archive) use this interface to
retrieve the voice content of a specific recording.
AudioServer Interface
If at least one port on the recorder is licensed for and assigned to the Telephony Replay
pool, then the recorder supports the AudioServer Interface.
The login page for users entitled to replay allows entry of a telephone number on which
they can be reached. If they do not enter a number, the browser replays through the
soundcard. You can also use separate Replay applications. Applications that can use this
interface include:
● Viewer and eWare
● Vision
Call Details Interface

The details about a recording can be provided on request. This is used by the bulk export
feature of the integral search and replay application and the Central Replay Server to
populate its database.
The recorder interfaces to the Avaya components via several mechanisms:
DMCC
DMCC runs on an AE Server and provides softphone registration and signalling services.
TSAPI (Avaya CT)

Internally, the Verint software uses JTAPI but this is layered on top of TSAPI and hence the
recorder server exchanges TSAPI messages with an AE Server if Conferenced recording
is used or if an external controller asks the recorder to establish single-step conferences.
Recorder Interfaces
Audio over RTP

The softphones on the recorder terminate RTP streams over which the audio to be
recorded flows. These sockets are connected to whichever VoIP resource (for example,
Media Processor) is providing the conferencing bridge through which the call is recorded.
The recorder uses ports 10000 upwards, allocating two ports per recording channel.
eQuality Balance V5 or V7 in Selective Recording Mode

Recording Command Interface
The recorder can establish a TCP/IP connection to port 1415 on one or more eQuality
Balance servers. The Quality monitoring application instructs the recorder to record
specific calls; to replay calls via the phone and live monitor calls.
Recordings Written to Disk

As the recorder completes each voice recording that an eQuality Balance server requests,
it copies the audio to a fileshare on the Quality Monitoring server. This uses a fileshare as
specified in the quality monitoring server's .INI file. Username and password for the
fileshare are configured on the recorder.
Recordings Read from Disk

As eQuality Balance instructs the recorder to replay calls, the recorder reads the audio
from files it previously placed on the Quality monitoring server's disk. This uses the same
fileshare or local path (if co-hosted) as above.
eQuality Balance V7 in "ContactStore Plus" mode

The recorder can interact with eQuality Balance V7 in one of two ways - either in the same
"selective recording" mode as described above OR in what is known as "ContactStore
Plus" mode - using the eQConnect Interface below to advise the quality application of all
the bulk recordings it has made
eQConnect Interface
The recorder can use an HTTP interface to communicate with eQConnect if this is installed
on an eQuality Balance (V7) server. (This defaults to port 3020). This is used to advise the
Balance server of agent logon/logoff and recording details.
Issue 5 May, 2008 201

Technical Reference
Replay Interface
eQuality Balance uses Viewer to replay calls made on this recorder.
Other Recorders
Recorders establish links with each other when acting in Master/Slave and
Primary/Standby topologies. These are TCP/IP socket interfaces.
Master/Slave Recorder Control

A recorder configured for Conferenced recording mode can control additional 'slave'
recorders. Each slave establishes contact with TCP/IP port 1416 on the Master.
Primary/Standby Recorder Interface

Standby recorders establish contact with TCP/IP port 1414 on the primary recorder. The
primary provides configuration data to the standby and instructs it to take over should it
sense a fatal error. Two such links must be established over separate paths. The recorders
exchange polling messages once a second.
External Control Interface

The recorder supports a simple TCP/IP command interface. This allows other applications
to control recording directly and/or add further user defined data fields to recordings. This
protocol supports START, STOP, TAG and BREAK (seamless stop and start) commands.
Database Upload Interface

When configured with either a Central Replay Server or a Viewer database, the recorder
uploads details of calls into database on the apropriate server.
To Central Replay Server

To upload call details to the Central Replay Server (CRS) a recorder uses a web service on
the CRS, to inform it that a new recording is available. The CRS in turn uses a web service
on the recorder to retrieve the details. Both of these interactions use port 8080.
Recorder Interfaces
To Viewer database
When deploying Viewer, the recorder uploads call details to a central Microsoft SQL Server
database over this link. The recorder acts as a standard Microsoft SQL Server client
application. It uses the TDS protocol, which uses TCP/IP port 1433 by default. This can be
overridden with a properties file setting.
Summary
Interface Protocol Local Remote Direction
HTTP - Admin and Replay TCP 8080 Inbound

HTTPS - Admin and Replay TCP 8443 Inbound
DMCC TCP 4722 Outbound
TSAPI TCP 450, 1050 Outbound
RTP UDP 10000 - Bidirectional
12000
Selective Quality Control TCP 1415 Outbound
Selective Quality Fileshare NTLM Outbound
ContactStore Plus Quality TCP 3020 Outbound
Inter-Recorder communication TCP 1414-1416 Towards
master
Central Replay Server TCP 8080 Bidirectional
Viewer Database TCP 1433 Outbound
Issue 5 May, 2008 203

Technical Reference
Appendix B: Troubleshooting
This appendix covers two areas: general troubleshooting tips and some specific common
issues:
● Hints and Tips on page 206
● Specific Problems on page 208
Issue 5 May, 2008 205

Troubleshooting
Hints and Tips
Where to Look for Clues

When problems occur, check the following:
● Emailed Alarms and Events. If you have been using the email settings to have alarms
and events forwarded to one or more email addresses, you should check these
carefully. As well as checking the contents of messages that you have received, also
check for days when the nightly log file purge message has not been received.
● Alarms Page. This page within the administration application provides a wealth of
information on problems that the system has detected. Review the alarms carefully. If
the problem is not immediately apparent, consider viewing all alarms, including those
that have previously been cleared. It may be that someone has cleared an alarm without
addressing it or realizing its significance.
● Log Files. Check for errors being reported in log files within the following directories
beneath your installation directory:
● /opt/witness/logs
● /opt/witness/tomcat5525/logs
Determining Current Version

When reporting problems you should state precisely which version of software you are
running. To determine this, click on the System Settings > License page.
Note the precise version number shown.
Application Logs
The Verint ContactStore for Communication Manager writes log files to /opt/witness/logs.
The current day's log file is called cscm.log. At midnight the current log file is closed,
renamed to cscm.log.<date>, and a new log file opened. These log files are
automatically purged after 30 days by default but this can be overriden with a properties file
entry.
Hints and Tips
Setting the server log level

You can set the level of messages logged by the Verint ContactStore for Communication
Manager to DEBUG, INFO (the default), WARN, ERROR or FATAL in one of two ways.
Permanently from next restart

To change the level permanently, enter the following line in the properties file
cscm.properties and restart the Verint ContactStore for Communication Manager
service to have it take effect.
log.level=DEBUG
If you change the log level in the properties file, it remains set on subsequent restarts.
When logging at DEBUG level, note that the log files grow very quickly and can overflow
the disk if left at this level.
Temporarily, immediately
To change the level temporarily, without restarting the service, simply use a browser to
request the URL:
http://myrecorder:8080/log?level=DEBUG
using the name of your server.
Alternatively, if you do not have access to the web administration screens, but do have
access via secure shell, execute the following command:
perl /opt/witness/bin/loglevel.pl DEBUG
You do not have to stop recording in order to change the logging level. To set it back again,
enter the same URL, replacing DEBUG with INFO. The command is case-sensitive. Using
this method changes the log level temporarily. It will revert to normal the next time that the
system is rebooted.
Tomcat Logs
Verint ContactStore for Communication Manager uses the Tomcat web servlet container,
which writes log files to /opt/witness/tomcat5525/logs.
Remote Access
You are strongly advised to provide remote access to the server to aid fault-finding.
Issue 5 May, 2008 207

Troubleshooting
Specific Problems
System Administration page problems

You may encounter problems as you access and use the System Administration
application. This section lists those problems and suggests steps to take to correct them.
Cannot access the System Administration pages

If you cannot access the System Administration pages, try the following:
● Ping the server to confirm that connectivity is possible. If not, trace the network
connections between client and server and double-check the server's IP address,
default gateway etc.
● Use the numeric dot notation IP address instead of the hostname. If this works, then the
hostname is wrong or cannot be translated by your DNS services. You may need to use
a fully qualified node name, such as recorder.bigco.com.
● Use the browser installed on the server itself to access the application at
http://localhost:8080.
If this works, then the problem is in the network between server and client. If it does not
work, then the problem may be with the Tomcat web server.
Cannot log in
If you have trouble logging in, double-check the state of Caps Lock and ensure the
password is being entered with the correct case.
If you can log in under another account, reset the password of the account having
problems.
If nothing happens when you click the OK button, check that your Internet Explorer settings
allow javascript to run. See ActiveX Control Download on page 138.
These symptoms have also been seen when trying to access a server with an underbar in
its node name. Note that this is not a valid IP name and should be changed.
Web pages appear "messy"

If the forms on the System Administration pages and/or Search and Replay page appear to
be out of alignment, or fields are truncated, check your desktop properties. If set to Large
Fonts (120dpi), this will impact all website accesses not just the Recorder pages.
Specific Problems
To correct display issues:

1. Right-click on your desktop and select Properties.
2. On the Settings tab, click Advanced and set the DPI Setting to 96dpi.
On new laptops, especially those with 1400x1050 resolution screens, the DPI is often set
to 120dpi. Using too small a window or too large a type (View > Text Size in Internet
Explorer) can make the columns and headings on the web pages wrap where they would
not normally.
Connectivity
Email alarm problems
Invalid entries in any one of the parameters used to define the email settings will result in
errors. To check this:
● Try the settings you are using in a standard mail client, such as Outlook. Send a
message using the account specified to prove that the settings are valid.
● If email messages have been working and then stop without any of the settings
changing, verify that nothing has changed on the mail server. This problem occurs, for
example, if your password has been reset or changed on the mail server.
● If the recorder is not sending email messages, it may be because it is not able to access
the SMTP server. Check the network connections to the recorder.
Recorder to AE Server Problems

The recorder communicates constantly with the AE server. If this link fails, alarms are
raised. If you suddenly get multiple alarms, typically those reporting problems with
softphones timing out registration, this link might have failed. Ping the AE Server to check
the link.
Recorder to Quality Monitoring server problems

If the problems reported include mention of the file-share:
1. Verify that the Quality Monitoring server is running.
2. Verify that the username and password previously in use are still valid. Try to access
the file-share from another machine using the username and password provided to the
recorder.
Issue 5 May, 2008 209

Troubleshooting
Recorder to Viewer server problems

If you have problems connecting to the Viewer server:
1. Verify that the Viewer server is running.
2. Verify that the SQL Server is active on it.
Search and Replay problems

For most problems with Search and Replay, consider the following diagnostic approaches
to narrow down the cause of the problem:
1. Search for a different call, for example, one that is more recent or older, shorter or
longer.
2. Log in as a different user with different replay restrictions
Cannot access the replay application

If you cannot get to the login page, try accessing the page from a different machine:
1. From the same side of any firewalls.
2. On the same LAN if you are having problems with WAN access
3. From the same sub-net, if having problems from a different sub-net
4. From the recorder itself, if having problems from the same sub-net.
User replay restrictions do not work

If you have given a user account replay rights over a number of addresses but the calls
from these stations are not listed when you enter a valid search that should include them,
check the recording ownership. When an agent has logged on to a station that is being
recorded, the calls recorded are "owned" by the agent number not the underlying station.
Give the user replay rights to the range of agent IDs who have been using the stations in
question. In fact, in most cases, you should restrict access to a set of Agent IDs rather than
station numbers.
Problems downloading ActiveX control

If you see error messages relating to ActiveX controls being downloaded-or blocked from
being downloaded, your security settings may be too restrictive. See ActiveX Control
Download on page 138 for more information.
Specific Problems
Problems displaying ActiveX control

If the ActiveX control downloads but displays a red cross on white background at top of
page, upgrade to Internet Explorer 6.0. These symptoms have been seen on Internet
Explorer 5.0.
Cannot log in
If you see the login page but cannot get past it:
1. Verify that Caps Lock is off and that you are entering the password with the correct
case.
2. Log in as a different user
3. Confirm the spelling of your log in name with the system administrator and check that
your account is still configured in the administration pages.
4. Ask the system administrator to reset your password. Log in with a blank password
and change your password when redirected to the Change Password page.
Search returns no calls

If you get to the search page but no calls are returned when you perform a search:
1. Broaden your search criteria to confirm that you can at least find some calls. Start by
requesting calls from any parties for today. If that shows no calls, extend the time
period.
2. Try setting the date range back to at least the time you know you have seen call
records for in the past.
3. Check that the system administrator has given you access to the correct calls. Your
search and replay restriction may be wrong or too narrow for the search you are
attempting.
4. Confirm that calls are being recorded. Follow the troubleshooting guidelines for
recording problems if you suspect that the system is not actually recording or
processing any calls.
Calls listed but cannot play them

If you can see the list of calls that matched your search criteria, but cannot actually play
them, look at the area at the top of the browser page where the "graph" of the audio
normally shows and match your symptoms to one of the following:
Issue 5 May, 2008 211

Troubleshooting
No Audio "graph"
This means that the call has not been retrieved from the recorder or DVD disk or has not
reached the client PC.
1. Check the server logs for errors.
2. Note the call's 15 digit reference number (shown if you hold the mouse pointer just to
the right of the radio button that you click to retrieve the recording. Search for that wav
file in the calls path to confirm that the recorded file exists.
3. Check connectivity and available bandwidth to the client PC.
Audio graph stops in mid call

This implies that the transfer of data from the Recorder to your client PC has been stopped
or interrupted.
1. Request the same call again. There may have been a temporary network problem.
2. Request a different call. If the problem is only with one call, you may have a corrupt file
on your hard disk.
3. Request the problem call from another PC on the same network. If the other PC can
retrieve it successfully, assess the differences between the two client PCs; the problem
is most likely at the client end.
4. Request the problem call from different sub-nets, ideally working closer to the recorder.
5. Request the call from the recorder server's own browser. If this works and the others
don't, then the problem is likely to be in the network between server and clients.
Audio graph appears but no sound

The audio file has reached the client PC successfully; the problem is most likely to be with
the PC's multimedia setup or current settings.
1. Verify that the PC has a sound card.
2. Play a wav file through Media Player or similar application to verify that that the sound
card is set up correctly.
3. Adjust any hardware volume and/or mute controls on the speakers/headphones.
4. Double-click on the icon in the system tray at the bottom right-hand corner of the
screen to verify that the PC's software volume controls are not set to mute or very low.
5. Ensure you are not running any other programs that may be locking the sound card
exclusively. If in doubt, shut down all other programs.
6. Try another similar PC. If that works, look for differences in the multimedia setup of the
two PCs.
Specific Problems
No New Recordings Playable

If you can replay old recordings but not newly made calls, there may be a problem with the
recording and/or storage components of the system. Follow these steps:
1. On the Status > System Overview page of the Recorder's System Administration
application, look at the counts for total calls recorded and calls recorded today.
2. Use Bulk recording mode to make a test recording.
3. Complete the recording and hang up.
4. Return to the Status > System Overview page and note the Total call segments
recorded today and Total call segments recorded to date (or since restart, if today).
These counts should have increased by at least one, the recording that you just made.
If the counts have increased, the recorderingis processing recordings. This is probably
a search/replay problem. See earlier sections for help.
Note:
Note: If these counts have not increased, the recording has not been successfully
compressed and stored on the Recorder or inserted into the call details
database. Do the following:
5. Look for alarm messages that indicate problems with the recording channel or with file
read/write or rename functions. The error message should indicate whether disk space
or a directory access problem is the cause. Check that .wav files are appearing in the
latest folder beneath the calls path as recordings are made.
6. Check disk space in all partitions. If any of these is 0 or less than 50MB, this may be
the problem. Check for build up of log files. Check that the call details database hasn't
exceeded the available space. Consider reducing the number of months of calls
kept-use the purge settings on the Server configuration page to adjust this.
7. Look for alarm messages that indicate licensing problems. The recorder will not
process any new calls if you have changed the MAC address, tampered with license
settings or are running on a time-expired license. In all cases, you should obtain a new
license key.
8. Check that the Tomcat server is running (cscm service).
9. Check for messages in the log files.
10. Reboot the server and watch for error messages on startup.
Poor Audio Quality on Telephone Replay

If the network configuration is correct and there are no problems with its function, the most
likely cause of this problem is that the recorder or the network is overloaded.
To look for a problem on a managed network switch, you should look at the diagnostics
and configuration details. You should see ZERO errors on all ports. Any port showing more
than 1 packet error in 10,000 is suspect and must be looked at.
Issue 5 May, 2008 213

Troubleshooting
If the C-LAN, MedPro and/or recorder ports are failing to auto-sense full/half-duplex
properly, you can force each port to either full or half duplex so as to reduce the error count
to zero.
Note:
Note: Even though a port may show an error rate of less than 1 in 100 packets, the
error counts are deceptive. A single packet error can trigger a full/half duplex
negotiation during which all packets are lost in the servers, but none of these
show as errors on the switch.
If your error counts are zero on all ports, then we must also consider overload of the
recorder as a possible cause. You should monitor the CPU load of the recorder during
busy hours. Replay and live monitor are very sensitive to overload. Recording may be
unaffected but if the CPU load is too high, audio quality on replay can suffer.
Similar problems have also been seen on multi-CPU AMD Opteron servers. This is caused
by an unstable system clock, which is addressed in RedHat Versions 4 and 5.
Recording Problems
Partial recording problems
Since no hardware component in the system is dedicated to specific ports, any hardware
problem is likely to affect all recordings equally. Therefore, if some calls are being recorded
and are playable but others are not, the problem is probably in the configuration.
1. Recording Mode versus Recording Channel? Determine whether your problems relate
to all channels of one or more recording modes or just to certain ports.
2. Check the configuration pages for the affected recording mode(s).
3. Calculate the range of stations carefully. For example, 11000 to 11010 is a range of 11
addresses, not 10.
4. Use the Status > Port Status page to observe the ports on the recorder during your
test calls. The ports go should go from idle to active and back again.
Quality Recording
Check that the Quality recording ports are set to use the same codec as any bulk recording
you are doing on the same phones.
Specific Problems
Meeting Recording
Cannot Enter Owners

The recorder requires rtp-payload signalling as described in Configuring tone detection on
page 75 in order to interpret dialled digits. Check that both IP phones and Digital phones
are configured for this signalling mode as IP phones will default to it whereas digital
phones may not.
Cannot Enter Delete Command

phones may not.

Cannot retain recordings
phones may not.
Live Monitor
Cannot monitor
1. Check that you have created the appropriate numerical user account so that the
station/phone dialling in is recognized as an authorized user.
2. Check that the live monitor ports are set to use the same codec as the ports you are
trying to monitor.
3. The recorder requires rtp-payload signalling as described in Configuring tone
detection on page 75 in order to interpret dialled digits. Check that both IP phones and
Digital phones are configured for this signalling mode as IP phones will default to it
whereas digital phones may not.
Poor Audio Quality

See Poor Audio Quality on Telephone Replay on page 213. The causes are the same.
Issue 5 May, 2008 215

Troubleshooting
Appendix C: Alarms
This appendix provides details of the alarms that can be raised by the system..
● Alarms on page 218
● Alarms Table on page 219
Issue 5 May, 2008 217

Alarms
Alarms
The recorder may generate the following Alarm or Event notification messages. These
events are:
● shown on the Status > Alarms/Events page
● sent in email messages as specified on the System Settings > Server page
● reported via SNMP
● logged to the recorder's log file cscm.log
Alarms Table
Alarms Table
Within messages, the strings XXX and YYY represent a specific parameter such as a
station number, an IP address etc. The table shows both the English text of the message
(that appears on the Alarms page) and the underlying resource string that appears in the
log file.
Entries within the table are sorted according to the text of the message.
Severity Log File Entry English Alarm text Comment What to Do
Warning alarms.binaryuui One or more User to Only text fields are Change to text
User Information fields supported. fields or stop
are in binary format. attempting to store
This is not supported. UUI
See call XXX for
example.
Major alarms.callspath.invalid Call storage path is Separate partition Set this to a valid
invalid. Please change preferred. path.
it under Settings >
Server.
Major alarms.cmapi.down Device, Media AND Major failure of Check connectivity

Call Control API not DMCC services. to AE Server.
running on YYY.
Reason: XXX
Major alarms.cmapi.up Device, Media AND DMCC services are No action required.
Call Control API restored.
running on XXX.
Major alarms.dapi.down Communication The link remains Check connectivity

Manager Data Link down unless to AE server.
DOWN. Please check advised otherwise.
the current status on Alarm will repeat
System Overview daily until resolved.
page. Reason: XXX
Major alarms.dapi.up Communication DAPI services are No action required.

Manager Data link UP. restored.
Major alarms.dapi.aesvcsconfig Cannot use DAPI. See AE Server Configure the

Please check you have manuals CLANs as H.323
added XXX to the Gatekeepers.
H.323 Gatekeeper list
on the web
administration Switch
Connections screen.
Major alarms.database.cannotinser Cannot insert details of Check disk space

t recording XXX into on the partition
database. holding the call
details database.
Major alarms.disk.full Disk full on partition 'Check log files. If Delete some files to
'XXX calls partition, make space.
check for files that
cannot be purged.
Issue 5 May, 2008 219

Alarms
Warning alarms.disknearlyfull Disk nearly full on Check log files. If Delete some files to
partition 'XXX'. Only calls partition, make space.
YYY MB free. check for files that
cannot be purged.
Major alarms.eware.diskload Error advising eWare Check connectivity

of disk insertion. to SQL server.
Reason: XXX
Major alarms.eware.linkdown eWare Server Link Check connectivity

DOWN. Reason: XXX to SQL server.
Major alarms.eware.linkup eWare Server Link UP. No action required
Major alarms.eware.tar Error adding archive Check connectivity

details to eWare. to SQL server.
Reason: XXX
Major alarms.file.copyfail Failed to copy file XXX. Check disk space,

path, security.
Major alarms.file.copyretry Retrying file copy of No action required

XXX.
Minor alarms.file.deletefailed Failed to delete file Depends on reason

XXX. Reason: YYY.
Major alarms.file.iocopyfail Failed to copy file XXX. Depends on reason

I/O Error: YYY.
Minor alarms.file.wavwritefail Failed to write file Depends on reason

XXX.wav. I/O Error:
YYY.
Minor alarms.file.compressfailed Failed to compress Depends on reason

audio file for recording
XXX. Reason YYY.
Major alarms.file.xmlwritefail Failed to write XML file Depends on reason

XXX. Reason: YYY.
Warning alarms.file.notindb XXX recordings up to Implies last Shut down service

and including YYY shutdown was not rather than kill.
were not stored to "clean".
database on last
shutdown.
Warning alarms.jtapi.backingoff Backing off Single-step If no standby Check if standby

Conference control as recorder is present, did take over. If not,
other recorder is can imply spurious report problem.
establishing calls. call details being
advised by Avaya
CT.
Major alarms.jtapi.backlogcleared Single-step Queue of requests Check system

Conference Queue is too long. loading and
Backlogged. Flushing. increase capacity if
needed.
Major alarms.jtapi.backlogged Single-step Queue of requests No action required.

Conference Backlog now acceptable
Cleared. length.
Alarms Table
Warning alarms.jtapi.conffailed Single-step Very occasional Increase number of

Conference Failed: failures are nothing retries in properties
XXX to be concerned file. Check for latest
about. Regular versions of Avaya
failures should be CT\TSAPI.
reported.
Major alarms.jtapi.configchanged Avaya CT User changed No action required.

Configuration changed. configuration
Dropping previous details,
connection.
Minor alarms.jtapi.csta44 Single-step Always means that Stop using Follow

Conference failed on another application the call options or
address XXX. CSTA is trying to control stop the other
Error 44. Check that no the call. application.
other AE Server is
controlling this call.
Major alarms.jtapi.error Avaya CT Services on Depends on error

XXX reporting error: shown.
YYY
Major alarms.jtapi.heartbeatfailed Avaya CT Services System should Report if recurring,

failed. Heartbeat lost. attempt to restart
Avaya CT services
automatically..
Major alarms.jtapi.invalidparams Invalid Service, Check all

username or password parameters.
parameters for Avaya
CT: XXX
Major alarms.jtapi.invalidskill Invalid Skill Group: Check that this is a

XXX valid skill hunt
group.
Major alarms.jtapi.invalidvdn Invalid VDN: XXX Check that this is a

valid VDN.
Major alarms.jtapi.notaskill Not a valid skill group Check that this is a

(or Avaya CT cannot valid skill hunt
reach switch) group.
Major alarms.jtapi.observationende Observation ended Report problem.

d
Major alarms.jtapi.observer Avaya CT Observation Report problem.

of XXX reports
error:YYY
Major alarms.jtapi.outofservice Out of Service Check Avaya CT

services.
Major alarms.jtapi.overflow Conferenced Increase number of

Recording Mode failed concurrent
to record a call recordings allowed
because all ports were on Conferenced
busy. mode.
Issue 5 May, 2008 221

Alarms
Major alarms.jtapi.portsavailable Conferenced No action required.

Recording Mode now
has XXX ports
available.
Warning alarms.jtapi.portslow Conferenced Increase number of

Recording Mode has concurrent
less than XXX ports recordings allowed
available. on Conferenced
mode.
Major alarms.jtapi.providerok Avaya CT Services on No action required.

XXX UP
Major alarms.jtapi.shutdown Service shut down Check Avaya CT

services.
Warning alarms.jtapi.takingover Taking over Single-step See Check standby

Conference control as alarms.jtapi.backing recorder.
other recorder failed to off
establish call.
Info alarms.logpurge Purged XXXKB of old Nightly message. No action required if

log files. Now YYYMB received. If not
free. received, check
recorder is running.
Minor alarms.monitor.codecwrong Live Monitor failed due Reconfigure

to incompatible audio monitor ports to use
formats. Monitor ports same codec as
and monitored ports other modes.
must have identical
Audio Format settings
Major alarms.master.linkerr Error establishing link "Depends on

to Slave recorder on reason shown.
XXX: "YYY
Major alarms.master.serversocket Error creating socket Depends on reason

for slave recorder on shown.
port XXX. Reason:
YYY.
Major alarms.master.timeout Slave recorder timed Check slave

out: XXX at address recorder and
YYY network paths to it.
Major alarms.primary.linkerr Error establishing link "Depends on

to Standby recorder on reason shown.
XXX: "YYY
Major alarms.primary.serversocket Error creating socket Depends on reason

for standby recorder on shown.
port XXX. Reason:
YYY.
Major alarms.primary.timeout Standby recorder timed Check standby

out: XXX at address recorder and
YYY network paths to it.
Alarms Table
Major alarms.properties.notanumbe Property XXX in Correct the entry in

r properties file is non the properties file.
numeric (YYY). Using
default value instead.
Major alarms.queue.acceptable XXX Job Queue No action required

backlog reduced to
acceptable level.
Currently YYYms.
Major alarms.queue.copy Failed to copy to XXX. Will cache files until Determine why
Reason:YYY. NOTE: problem is resolved recorder cannot
Further errors with the then copy all write to the share.
same root cause on the outstanding.
same path will only
show here once every
24 hours. Check the
log file if in doubt.
Major alarms.queue.slowjob XXX Job Queue Occasional Report frequent or

individual slow job - message not a very long delays.
took YYYms. problem.
Major alarms.queue.tooslow XXX Job Queue May indicate Consider faster

backlogged. Delay overload. hardware or reduce
currently YYYms but load.
may go higher.
Minor alarms.rtp.packetloss Unacceptable packet Reported once per Check network path
loss from XXX. day per IP address to recorder. If
but could be several different
occurring on all addresses report
recordings from that problems at the
address. same time, identify
common path. If
many addresses
report at same time,
could be CPU
overload at
recorder.
Major alarms.rtp.nopackets No audio packets See Avaya support Check TN570

received on call from site re PSN #345U - boards are correct
XXX. Call failed or no TN570C and vintage.
network path to TN570D Expansion
recorder. Interface boards log
chronic fiber alarms
Major alarms.softphone.beepstart Port XXX. Failed to Internal error Report problem.

start beeptone.
Reason: YYY
Major alarms.softphone.beepstop Port XXX. Failed to Internal error Report problem.

stop beeptone.
Reason: YYY
Major alarms.softphone.buttonlook Port XXX. Error looking Softphone Check softphone

up up button Id. Reason: misconfigured. configuration is as
YYY. per installation
section.
Issue 5 May, 2008 223

Alarms
Major alarms.softphone.buttonmissi Port XXX. Unusable as Softphone Check softphone

ng button YYY is missing misconfigured configuration is as
from softphone. per installation
section.
Major alarms.softphone.callstopped Port XXX. Call dropped Recording port has Consider setting
as it exceeded been active for recording mode to
maximum permitted several hours. release call on
duration. dropping to one
other party.
Major alarms.softphone.cmapidown Port XXX. DMCC Verify connectivity

services not available. to AE services and
that DMCC is
running.
Major alarms.softphone.endrecfaile Port XXX. Error ending Internal error Report Problem
d recording. Reason:
YYY.
Major alarms.softphone.hookswitch Port XXX. Error setting Internal error Report Problem
hook switch. Reason:
YYY.
Major alarms.softphone.inservice Port XXX restored. No action required
Minor alarms.softphone.invalidtarge Port XXX - Invalid Quality app asked Check Quality
t target for recording: to record invalid server
YYY station. configuration.
Major alarms.softphone.mediacontr Port XXX. Cannot Internal error Report Problem

ollistener add/remove
MediaControlListener.
Reason YYY.
Major alarms.softphone.notarget Port XXX. Service Check configuration

observe failed as no
target station set.
Major alarms.softphone.nullpointer Port XXX. DMCC event Internal DMCC Report occurrences
'YYY' fired with null error related to a with a copy of your
pointer. particular recorder log files.
port.
Major alarms.softphone.ownerreg Port XXX. Error Internal error Report Problem.

unregistering owner.
Reason: YYY.
Minor alarms.softphone.outofservic Port XXX out of Should recover Depends on reason

e service. within a few shown before this
seconds. alarm.
Major alarms.softphone.play Port XXX. Error playing Depends on reason

file. Reason: YYY. shown.
Minor alarms.softphone.processfail Error processing file Depends on reason

ed XXX. Reason YYY. shown.
Major alarms.softphone.recordstartf Port XXX. Failed to Depends on reason

ailed enable recording. shown.
Reason: YYY
Alarms Table
Major alarms.softphone.recordtime Port XXX. Recording Depends on reason

out failed to start. shown.
Resetting port.
Major alarms.softphone.registration Port XXX Registration Depends on reason

failed failed. Reason: YYY shown.
Major alarms.softphone.regtimeout Port XXX. Registration Check connectivity

timed out. to AE Server.
Major alarms.softphone.rtpthread Port XXX. RTP handler Depends on reason

failed. Reason: YYY. shown.
Major alarms.softphone.setuprecor Port XXX. Error setting Depends on reason

ding up recording. Reason shown.
YYY.
Warning alarms.softphone.shortpacke Port XXX. VoIP packet Default settings of Configure codec set
t interval of YYYms less 20 or 30ms are not and network region
than recommended efficient for as per installation
60ms. recording. instructions.
Warning alarms.softphone.sodropped Port XXX. Service Check for resource

observe dropped overload on switch.
unexpectedly while in
state 'YYY'.
Major alarms.softphone.sofailed Port XXX. Service Check COR of

observe failed despite softphone and
multiple retries. target. Check for
other observers.
Major alarms.softphone.soachieved Port XXX. Service No action required.

observe achieved on
YYY.
Minor alarms.softphone.sodenied Port XXX. Service See text of alarm.

observe denied trying
to observe YYY. Check
station is allowed to be
service observed, the
station is present and
no other service
observe active. Try
manual service
observe. Also check
that G729A and
G711MU codecs are
enabled for the
recorder's softphone
region.
Warning alarms.softphone.sotimeout Port XXX. Service As above.

observe setup timed
out. Trying to observe
YYY.
Major alarms.softphone.sscdroppe Port XXX. Single-step Report problem if

d conference dropped more than very
unexpectedly while in occasional
state YYY. occurrences.
Issue 5 May, 2008 225

Alarms
Major alarms.softphone.sscfailed Port XXX. Single-step Consider increasing

conference failed number of retries.
despite multiple retries.
Major alarms.softphone.ssctimeout Port XXX. Single-step Check system

conference setup timed loading.
out. Trying to
conference in to YYY.
Major alarms.softphone.startrecfail Port XXX. Recording Depends on reason

ed failed to start. Reason: shown.
YYY
Major alarms.softphone.stopplaying Port XXX. Error Depends on reason

stopping file playing. shown.
Reason: YYY.
Major alarms.softphone.stoprecordf Port XXX. Error Depends on reason

ailed disabling recording. shown.
Reason: YYY.
Major alarms.softphone.timedout Port XXX. Timed out Report if more than

when in state YYY. very occasional
occurrences.
Major alarms.softphone.ttd Port XXX. Failed to set Configure signaling

up touch tone detect. in accordance with
Check you have installation
OUT_BAND signalling instructions.
configured. Failure
Reason: YYY.
Warning alarm.softphone.brokenflutter Port XXX. Feature Check COR, switch

failed on button YYY. resources.
Major alarms.softphone.unregistere Port XXX. Can imply standby Check if standby is

d Unregistered. Reason: recorder has active.
YYY. registered this
phone.
Major alarms.softphone.unregisterf Port XXX. Error Depends on

ailed unregistering. Reason: reason.
YYY
Warning alarms.softphone.userreset Port XXX. User 'YYY' Also logged to audit No action required.
reset the port. trail.
Major alarms.softphone.voiceunitlis Port XXX. Cannot Internal error. Report problem.

tener add/remove
VoiceUnitListener.
Reason: YYY.
Major alarms.standby.invalidstns Invalid Station list "Internal error Report problem.

"YYY" received from
primary for station pool
"XXX
Major alarms.standby.noswitch No link to Check AE server is

Communication up.
Manager, All ports
have failed.
Alarms Table
Major alarms.standby.notviable Recorder fatal error. Recorder will See next error
request that message.
standby takes over
if one is present.
Major alarms.standby.primarynotok Master recorder Check fatal error

requests that causing switchover.
Standby(s) take over.
Minor alarms.unify.ioexception File I/O error updating Depends on

call details of XXX. reason.
Reason: YYY.
Minor alarms.unify.parserror Error parsing Correct external

Unify/External control controller
information. XXX is not command.
a number.
Minor alarms.unify.parseexception Parsing error updating Depends on

call details of XXX. reason.
Reason: YYY.
Major alarms.url.badport Invalid IP port number No action required.

for recording mode:
XXX, Set to: YYY
Major alarms.url.general Error on link to 'XXX': Check connectivity.

YYY
Major alarms.url.linkup Link established with 'No action required.

XXX server on 'YYY
Major alarms.url.nourl Ports are allocated to Configure

XXX recording but no appropriate link's
url is set to URL.
communicate with the
server.
Major alarms.url.socket Error connecting to Problem connecting Depends on reason

'XXX'. YYY to other server. shown. Typically
network routing or
security issues.
Major alarms.url.unknownhost Unknown host 'XXX' in Check host name is

url for YYY server correct and host is
entered in DNS and
is up.
Issue 5 May, 2008 227

Alarms
Warning alarms.vacuum It is now XXX days See text of

since the last full message.
database vacuum. The
recorder will perform a
full vacuum the next
time it is restarted. This
may take some time.
To postpone this to the
following reboot, clear
the setting on the
System Settings >
Server page. You
<b>MUST</b> restart
the recorder and allow
it to vacuum the
database at least once
every 240 days.
Major eqconnect.reason.connectti Initial connection timed Check URL and

meout out with eQConnect at ensure eQConnect
XXX is up.
Major eqconnect.reason.inactivityti Inactivity timeout with Check eQConnect

meout eQConnect at XXX is up and network
path is available.
Major eqconnect.reason.ioexceptio IO Exception with Check network path

n eQConnect at XXX. to eQConnect.
Reason:YYY
Major eqconnect.reason.linkok Successful response No action required

from eQConnect at
XXX
Major err.alarmtag.exception Error displaying alarm Internal error. Report problem.

tag.
Major err.alphas.invalidchar Invalid character ('-') in Correct the entry.

a non-numeric string.
Major err.alphas.toolong Non-numeric list is too Reduce the length

long. 10000 character of the entry.
maximum including
commas.
Major err.archive.baddisk Bad or wrong type of Insert the current

archive disk inserted. DVD+RW or a new,
blank one.
Major err.archive.error Unable to write to Depends on

archive file YYY. reason.
Reason: XXX. Check
Troubleshooting Guide.
Major err.archive.nodisk No disk available for Insert a DVD+RW

calls archive.
Major err.archive.wrongdisk An old archive disk Insert the current

(XXX) is inserted, but DVD+RW or a new,
archiver needs the blank one.
current disk (YYY) or a
blank disk.
Alarms Table
Major err.confighistory.recordsdelet One or more Do not trust config

ed configuration history history or audit
records have been records
deleted.
Major Err.confighistory.tampered One or more Do not trust config

configuration history history or audit
records have been records
altered.
Major err.cscmstatetag.exception Error displaying state Internal error. Report problem.

tag.
Major err.database.purge Error purging Depends on reason

database. Reason: shown.
XXX.
Major err.eqb.1000 Licensed port count Quality server tried Reconfigure Quality
exceeded: requested to access more server or increase
XXX, available YYY ports than are license.
licensed.
Major err.eqb.1001 Insufficient ports Quality server tried Reduce port count
allocated: requested to access more on quality server or
XXX, allocated YYY ports than have allocate more ports
been allocated to it. on recorder.
Major err.eqb.1002 Port XXX: Invalid Internal Error Report problem.

Parameter: YYY
Major err.eqb.1003 File share XXX Confirm that Open fileshare

unreachable fileshare is manually. Confirm
reachable. access details.
Major err.eqb.1004 File share XXX access Confirm account

denied has read\write
access
Major err.eqb.1005 Port XXX: File I/O error Check access to

on 'YYY', error: path shown.
Major err.eqb.1006 Port XXX: Invalid state. Report if persistent.

Currently YYY
Major err.eqb.1007 Port XXX Out of Report if port is not

Service recovered rapidly.
Major err.eqb.1008 DMCC Down Check AE Server.
Major err.eqb.1009 Port XXX: Local file I/O Check access to

error on 'YYY', error: path shown.
Major err.license.clocksetback The system clock has Obtain and enter a

been set back. Timed non-timed license.
licence is invalid.
Major err.license.create Error creating licence Internal error. Report problem.

token object
Minor Err.mail.invalidaddress Invalid email address: Correct the address

XXX
Issue 5 May, 2008 229

Alarms
Minor err.mail.send Error sending e-mail. Check all email

account entries.
Send a test email
manually to verify
settings.
Major err.mail.authentication Authentication failed Check email user

attempting to send and password
e-mail. entries.
Major err.maxusagetag.exception Error displaying peak Internal error. Report problem.

usage details.
Major err.portpool.nocurrent No port pool specified. Internal error. Report problem.
Major err.portpooltotaltag.exception Error displaying port Internal error. Report problem.

pool totals.
Major err.settingstag.exception Error in Internal error. Report problem.

IterateSettingsTag
Major err.settingstag.invalid Invalid settings group Internal error. Report problem.

requested in
IterateSettingsTag
Major err.settingtag.exception Error in SettingTag Internal error. Report problem.
Major err.settingtag.invalidfield Unrecognised field Internal error. Report problem.

request in SettingTag
Major err.softphonestatetag.excepti Error showing port Internal error. Report problem.

on state details.
Major err.stnpooltag.exception Error in StnPoolTag Internal error. Report problem.
Major err.stnpooltag.invalid Invalid station pool Internal error. Report problem.

requested in
StnPoolTag
Major err.stnrangetag.exception Error displaying station Internal error Report problem.

range details.
Major err.stnrangetag.invalidfield Unrecognised field Internal error Report problem.

name in StnRangeTag
Warning err.system.restart System restarting. No action required.
Info err.system.shutdown System shut down. No action required.
Major err.tags.misc Error displaying Internal error Report problem.

miscellaneous tag.
Major err.tags.license Error displaying licence Internal error Report problem.

tag.
Major err.tokentag.exception Error in TokenTag Internal error Report problem.
Major err.tokentag.invalid No token set in Internal error Report problem.

TokenTag
Major err.tokentag.invalidfield Unrecognised field Internal error Report problem.

name in TokenTag
Alarms Table
Major err.usertag.exception Error in UserTag Internal error Report problem.
Major errpage.label.badstuffhappen An error has occurred: Internal error. Report problem.

ned
Info info.archive.errorcleared Wrote to archive disk No action required.

correctly.
Info info.archive.rightdisk Correct archive disk Now able to write to No action required.
now inserted. disk in drive
Info info.archive.ewareok Eware connection OK. No action required.
Info info.mail.sent Mail sent successfully. No action required
Info info.queue.copy File copied No action required

successfully to XXX
Info info.standby.goingstandby Standby recorder No action required.

returning to idle.
Info info.standby.primaryok Primary recorder No action required.

requests Standby(s) go
idle.
Info info.standby.viable Recorder fatal error No action required.

resolved. OK to record.
Major link.master.linkerr Error on link to Slave Check connectivity,

recorder at XXX: YYY check other node is
up.
Major link.primary.linkerr Error on link to Standby Check connectivity,

recorder at XXX: YYY check other node is
up.
Major standby.reason.connecttimeo Initial connection timed Check connectivity,

ut out with XXX check other node is
up.
Major standby.reason.timeout Initial connection timed Check connectivity,

out check other node is
up.
Major standby.reason.inactivity Heartbeat failure via Check connectivity,

XXX check other node is
up.
Major standby.reason.noactivity Heartbeat failure Check connectivity,

check other node is
up.
Major standby.reason.primaryreque Instruction from MAster No action required.

st Recorder
Major standby.reason.reconnecttim Connection dropped Check connectivity,

eout and reconnection timed check other node is
out with XXX up.
Issue 5 May, 2008 231

Alarms
Major err.confighistory.tampered One or more Do not trust audit

configuration history records.
records have been
altered.
Info stnrange.usage.nowarn The pool of ports XXX No action required.

has YYY or more
port(s) available.
Minor stnrange.usage.warn The pool of ports XXX Consider allocating

has LESS THAN YYY more ports to this
port(s) available. mode.
Appendix D: External Control Interface
This appendix provides details of the external control protocol and Java class library.
● Introduction on page 234
● Java API Toolkit on page 237
● TCP/IP Protocol Overview on page 238
● Examples on page 242
● TCP/IP Message Sequences on page 245
Issue 5 May, 2008 233

Introduction
External applications can control or influence recording by using the Recorder Control
Protocol. You may integrate directly to a TCP/IP socket interface using the protocol
described or by using the Java package provided for this purpose. An example application
that uses the Java interface is also provided for reference.
When to Use External Control

The recorder includes sophisticated CTI interfaces and recording control but there are still
occasions when additional flexibility or further data tagging is required. You will require an
"external controller" in the following cases:
Complex Recording Rules

If your "record/do not record" requirements are more complex than the administration
interface supports. For example, if you wish to start and stop recording at specific points
within a call.
Additional Tagging
If you need to "tag" recordings with details that are not normally provided by the recorder.
Hybrid Systems
If your telephone system is controlled by another application (e.g. Genesys) then it may be
appropriate to control recording as a result of the events occurring on a CTI feed from that
application. In such cases, you may need not only the external control protocol but also
Verint's Integration Framework or Unify server. These can interpret a wide range of CTI
feeds. This is beyond the scope of this document.
This Appendix
The remaining sections of this appendix provide:
● a brief overview of the Java interface classes. For detailed information refer to the
JavaDocs for this package.
● an Overview of the underlying TCP/IP control protocol
Introduction
● a reference guide for the message sequences exchanged between recorder and
controller
Port Allocations
The Unify/External Control recording modes are provided for backwards compatibility with
existing deployments only. New deployments should NOT allocate ports to these modes.
Instead, use the external controller with ports in the following modes:
● Station Bulk - for all service observe based recording.
● Conferenced - for single-step conference recording where the external controller needs
to provide additional tagging or is splitting long recordings into segments (as needed
with many auto-diallers). Do NOT use ports in this mode if you want to initiate
recordings from the external controller. You must allow the CSCM to establish
conferences before sending TAG, STOP (and then subsequent START) commands. Do
NOT send "START SSC:nnnn:" commands as these will conflict with the recorders
allocation of ports to calls.
● On Demand - for single-step conference recording where the external controller is
determining which calls are to be recorded. It can either instruct the recorder to
establish a single-step conference; do so itself or support manual conferencing.
! Important:
Important: For backwards compatibility with existing deployments (in which Unify does
not send the FALLBACK OFF command), the recorder automatically takes
FALLBACK off those ports that are assigned to the Unify/External control
(Service Observe) mode. In this case, these ports only revert to FALLBACK
mode when ALL external controllers have disconnected. You can override
this behaviour by setting unify.defaultrecord=true in the properties file but for
new deployments should be using Station Bulk ports instead.
Master + Slave Systems

This topology only supports Conferenced mode recording. Hence the only permitted uses
of Unify/External controllers are:
● Tagging calls
● Splitting (and tagging) long calls (e.g. autodialler sessions) with STOP and subsequent
START commands.
You cannot use the external controller to initiate recordings (e.g. by sending in START
STN:nnn: SSC:mmm: commands) as this will conflict with the Master's port allocation
algorithms.
Issue 5 May, 2008 235

Connect Unify/External controller to the Master (and Standby if present) only. Do NOT
connect Unify to the Slaves. All ports on the slaves will be configured in On Demand Mode
and the slave recorders will already be configured to consider the Master (and Standby if
present) to be their "external controllers". There should be no need for the Unify/External
controller to be aware that ports are on slave recorders as it communicates with the master
(and standby) exactly as if the ports were on the master (or standby).
The only exception to the above is that when using Quality Monitoring in "CSCM Plus"
mode with external tagging, the following limitations apply to post-call tagging:
● Calls made before the most recent restart of the active recorder cannot be post-call
tagged
● Only the most recent 1000 calls can be post-call tagged. (Change this default value with
eqconnect.guidcache=nnnn in the properties file).
● Following a switchover from master to standby or vice versa, only calls started by the
currently active recorder can be post-call tagged.
These limitations are not normally of any concern to most applications as these tag calls
very shortly after the call ends. They may affect batch mode tagging - such as overnight
processing. In such cases, consider using additional table(s) in the eWare database
instead and updating these directly rather than tagging the recording itself.
Java API Toolkit
Java API Toolkit

Rather than write directly to the socket level interface described below, you may use the
Java language API provided. This contains the following components:
Packages
com.swhh.cti.csapi.lang Provides a high-level API for controlling the recorder.
com.swhh.cti.csapi.msg Provides Recorder Control Protocol message classes

and interfaces as well as classes that implement the
java.io.Reader interface for reading these messages from
streams.
com.swhh.cti.csapi.service Provides a low-level API for communication with the
recorder.
com.swhh.cti.csapi.tools Contains two debugging tools as well as examples of
applications that use the low-level API.
com.swhh.cti.csapi.ui Contains a test harness with a graphical user interface
that also serves as an example application built on top of
the high-level API classes.
JavaDoc for these classes is available on request. To view the JavaDoc without a Java
IDE, unzip the contents and double-click index.html for a top-down view. For a
comprehensive reference index, double-click indexall.html.
Issue 5 May, 2008 237

TCP/IP Protocol Overview
Connection Method
An external controller can drive the recorder via a TCP/IP socket using the protocol
described below.
You configure the Master recorder with the IP address and socket number of the
controller(s) to which it should connect. Connection defaults to port 1414 but you can
specify an alternate port. For example, you must specify port 1415 to connect to the PDS
controller and port 1416 to connect a Slave recorder to a Master recorder.
Note:
Note: This setting was previously on the administration page for Unify/External
Control ports but has been moved to the main server settings page as it now
applies to all recording modes.
The recorder creates a socket and attempts to connect to each specified address. If it fails
to connect, it will try again every 60s.
You can use multiple, independent controllers connected to a single recorder.
The controlling application should open and bind a server socket and accept incoming
connection requests.
General Protocol Specification

Messages take the form:
<protocol version number> <command> (<command specific data>)
'\x1b'
When sending messages, follow these rules:
1. Transmit messages in UTF-8 encoding
2. All messages and responses are terminated with an escape character (27 / 0x1b).
3. Separate fields with space characters.
4. Enclose field contents in double quotation marks where the field needs to contain a
space - as, for example, with filenames. Always enclose filenames in double-quotes.
5. Delimit sub-fields with colon characters ":".
6. Apart from HELLO messages (see below) and responses, all messages must be
responded to in the form:
<protocol version number> <error code> <channel identifier>
<error qualifier> <additional info> '\x1b'
Where
<error qualifier> is typically "DESC:xxxxxxx:"
and
<additional info> may be 0 or more instances of
"PARAM:pppppp:"
The response to a successful command is therefore:
1 0 STN:nnn: \x1b
and a typical error response is
1 -1 STN:1234: DESC:"Invalid Parameter": PARAM:STN \x1b
7. Ignore unexpected or malformed responses.
8. Send commands in uppercase, but test for them case insensitively.
Channel Identification
In most messages below, the recorder's channels are identified by Station Number
STN:nnnnn:.
Where nnnnn can be 1 to 9 digits long and may include leading zeroes.
The use of this station number varies according to the type of recording port being referred
to:
● Ports using a Service Observe mode (Station Bulk and Station Executive mode) are
referred to by the station number that they are observing.
● Ports using conferencing are referred to by the station number of the softphone that the
recorder is using.
XML Tagging
All tagging information is passed straight through the recorder into the XML file associated
with the recording. The recorder only looks for an escape character to delimit each
command. The intention is that the controlling application will pass through valid XML tags
which the recorder will insert into the body of the call's XML file. These may include valid
Issue 5 May, 2008 239

expansions of control and other characters (e.g. %20 etc). The entire xml string will be
inserted between opening and closing "taglist" tags. For example, the command
1 TAG STN:1234: INUM:800369000000009: <spare1>some
data</spare1><spare4>John Doe</spare4>\x1b
will result in call reference 800369000000009, which is currently being recorded on STN
1234 being tagged with the additional fields:
<taglist><spare1>some data</spare1><spare4>John
Doe</spare4></taglist>
If you intend to use Contact Viewer as your search and replay tool (rather than the integral
search and replay within the recorder) you should only use the eight special user-defined
fields, labelled "spare1" to "spare8". These can be relabelled for end user viewing as part
of the query design using Contact Viewer. You may use other fields but the recorder only
uploads these eight "spare" fields automatically to the Viewer database.
Basic Call Tagging

The recorder uses DMCC's CallInformationServices interface to tag recordings in all
modes except Conferenced (for which it uses Avaya CT derived information). This
imposes a load on the switch. If the external controller has access to all details about the
call, there may be no need for the recorder to request these details itself. You can choose
whether or not the recorder collects these details itself by setting property file entries as
specified in the table below:
Port Type Default Override with this in prop Comment
Station Bulk stnbulk.ocpneeded=false Disable only if external

controller will provide all
necessary details (bar
date/time and duration)
Station Executive stnexec.ocpneeded=false
On Demand ondemand.ocpneeded=false
Meeting meeting.ocpneeded=false
Conferenced Not Applicable Uses Avaya CT
(TSAPI) info.
Unify (Service unify.ocpneeded=true Enable if external
Observe) Disabled controller does not
have all call info.
Unify (Conferenced) unify.ocpneeded=true
Port Type Default Override with this in prop Comment
Quality Not Applicable

Live Monitor
If the recorder is collecting call details itself, the external controller can ask to be advised
whenever these change. Use the REQUEST ACTIVITY ON command to do so.
Fallback Mode
When a recorder starts, it automatically runs in locally controlled mode - where call events
it detects cause it to start/stop recording. This is also known as "fallback" mode.
If the external controller only wishes to TAG calls with additional data, it may be happy to
leave the ports in this mode. On the other hand, if it wishes to take control of the recording
decisions then it must instruct the recorder to relinquish control of the ports it wants to use.
It does this with the FALLBACK OFF command - either for a specific STN or for all ports.
If the heartbeat to the external controller that turned FALLBACK off is subsequently lost,
the channel(s) will revert to fallback mode and recording will continue as if the controller
had never been present.
Use the station specific command to take control of specific stations if you want to allow
other stations to continue to be controlled by the recorder or if you are using multiple
controllers, each of which manages its own set of stations.
Issue 5 May, 2008 241

Examples
Typical examples of how to use an external controller are given below. You should read
these in order as later examples refer back to earlier ones.
Third-party CTI Control

A bank using Genesys to route calls through an Avaya switch and wishes to record certain
calls according to user defined data and events occurring within their call routing
application. They would consider two designs:
Service Observe Based

If most - and potentially all calls at any given time - are to be recorded, a service observe
approach would be appropriate:
1. Configure Station Bulk recording for all stations to be recorded.
2. The controller establishes contact with the recorder (exchanges HELLO messages)
3. The controller notes the stations that are ONLINE (and may alarm if any are missing)
4. The controller takes charge of all stations by sending 2 FALLBACK OFF
5. The controller sends START and STOP messages to control recording.
You should note that a STOP message disables all recording (not just the call in
progress) until a START message is sent (or the recorder falls back).
6. The recorder performs basic tagging itself but the controller sends Genesys user data
through as TAG commands.
Should the controller fail, the recorder will revert to normal Station Bulk mode.
For a system that is tolerant of recorder failure, use a Standby recorder. This will contact
the controller as and when it takes over from the primary.
Single-step Conference Approach

If only a small number of calls are to be recorded, it may be more cost effective to provide
a pool of ports and conference these in as needed. This approach also allows supervisors
to continue to use service observe.
In this case:
1. Configure a pool of On Demand ports - large enough for worst case concurrent
recording need.
Examples
3. The controller notes the pool of ports that are ONLINE (and may alarm if there are too
few)
4. When a call is to be recorded, the controller picks an available port and sends it a
START message with SSC parameter specifying the station that is already on that
call..
5. When recording is no longer required, the controller sends that port a HANGUP
message,
6. On hanging up (whether instructed to by the controller or if the call ends) the recorder
sends a HUNGUP message to the controller. Only then can the port be reused for
another call.
7. The recorder performs basic tagging itself but the controller sends Genesys user data
through as TAG commands.
Should the controller fail, no recording will happen. For fault tolerance, use a pair of
controllers and have one or the other send messages to the recorder.
For large systems or to make the system tolerant to recorder failure, use a pool of N+1
independent recorders.
Additional Call Tagging

User wants to tag each call with a customer identifier that is known to an in-house
application. This knows the station handling the call and the customer on the call in
real-time i.e. during the call.
In this case, the recorder will be configured for either Station Bulk or Conferenced mode as
appropriate for the basic recording requirement. The approach to tagging will differ slightly.
Service Observe Approach

If Station Bulk recording is used:
1. Configure Station Bulk recording for all stations to be recorded.
3. The controller notes the stations that are ONLINE (and may alarm if any are missing)
4. Any time the controller receives a STARTED message, it looks up its current map of
which stations v customers and sends a TAG command to apply the appropriate
customer identifier to the call.
Should the controller fail, the recorder continues to record as normal in Station Bulk mode
but calls will not be tagged with customer id. A fault tolerant pair of controllers can be used
if required.
Issue 5 May, 2008 243

Single-step Conference Approach

If Conferenced mode is used:
1. Configure Conferenced mode as normal.
3. The controller notes the recorder ports that are ONLINE (and may alarm if any are
missing) and requests ACTIVITY ON for each.
4. Any time the controller receives an UPDATE message, it looks in the XML data for a
station identifier that it recognises. If it finds one, it sends a TAG message with the
appropriate customer identifier. (CAUTION: this will, in turn cause another UPDATE
message to come back. Do not go recursive! If the UPDATE message already includes
a customer identifier, do not send TAG again.)
5. Should a recorder port hang up, it will send "1 HUNGUP STN:nnnn:". The external
controller should use this message to indicate that the port is once again available for
use.
Should the controller fail, the recorder continues to record as normal in Conferenced mode
but calls will not be tagged with customer id. A fault tolerant pair of controllers can be used
if required.
TCP/IP Message Sequences

The table below shows all supported message sequences:
Controlling Application Dirn Recorder Sends Comments

Sends
On recorder first establishing connection with controller’s socket (default 1415)

<<< 1 HELLO The controller can determine which
CMAPICapture recorder is which from the port that it
NNNNNN connects on, its IP address or its serial
number NNNNNN. When using Unify, the
former method is normally used and
hence each recorder should be
configured to contact a different port
number.
1 HELLO XXXXX >>> Where XXXXX is the application’s name
or other identifier. The recorder does not
parse beyond “HELLO”. The recorder
writes the full message to the log file but
does not act on the name given.
<<< 1 ONLINE For each softphone configured and
STN:nnnnn: available, the recorder sends this
message. This allows the controller to
build a table of available stations and
respond with 1 FALLBACK OFF
STN:nnnnn: if it wishes to take control of
that port.
To take control of recording on all ports
2 FALLBACK OFF >>> External controller must now
enable/disable recording on this station.
<<< 10 Confirmed
To relinquish control of recording on all ports
2 FALLBACK ON >>> Recorder now controls stop/start on all
ports in all modes
<<< 10 Confirmed
To take control of recording on a specified station
2 FALLBACK OFF >>> External controller must now
STN:nnnn: enable/disable recording on this station.
Issue 5 May, 2008 245


Sends
<<< 1 0 STN:nnnnn: Confirmed

To relinquish control of recording on a specified station
2 FALLBACK ON >>> Recorder now controls stop/start on this
STN:nnnn: station.
To activate additional notifications
1 REQUEST ACTIVITY ON >>> The controller can ask that it is advised of
STN:nnnn: call details in addition to basic
started/stopped messages.
Start Recording (optionally, set up single-step conference)
1 START STN:nnnnn: >>> Start recording on station specified
SSC:mmmm: <xml data> (nnnn) and tag with optional data
supplied.
Note: SSC parameter is If SSC parameter is provided, the
optional. recorder will use TSAPI to request that
the recorder’s softphone (nnnnn) be
single-step conferenced onto the call
currently active on station (not agent or
VDN) mmmm. Use this parameter with
OnDemand, Conferenced and Unify
(Conferenced) ports only.
Failure to attempt to establish a
connection (e.g. because the call
terminated before connection could be
established) is indicated by an error
message with “NOSSC” in the
description. Parameter 1 repeats the
target address and Parameter 2 gives the
error message.
Stop Recording
1 STOP STN:nnnnn: <xml >>> Stop recording on station specified and
data> tag with optional data supplied.
Note: This command does not cause
softphone to hang up.
External controller wishes to terminate single-step conference recording on known softphone

Sends
1 HANGUP STN:NNNN: >>> Note that the STOP command simply

<optional XML data> stops recording – it does not cause the
port to hang up. Hence an additional
command is required where a port is
conferenced into a call.:
This command instructs the specified
station to hang up – implicitly terminating
any current recording and, optionally,
tagging it with the XML data provided.
However, the recording state
(enabled/disabled) of the softphone is
NOT changed.
On recording starting
<<< 1 STARTED Unify can note the INum to ensure
STN:nnnn: subsequent tag commands go to the right
INUM:nnnnnnnnn segment even if that call has finished.
nnnnnn: Unify can also parse out the station
number if it wants to check for orphaned
records hanging around from this station
which are obviously now in error.
On recording stopping
<<< 1 STOPPED We could drop the call if told to but we
STN:nnnn: can’t be sure it won’t stop before then so
INUM:nnnnnnnnn maybe best to always report when call
nnnnnn: drops.
Don’t see any need to include tagging as
we will not normally be getting any
additional tagging beyond stn number.
On recording port hanging up
<<< 1 HUNGUP
STN:nnnn:
On recording getting new or changed info (NAILED mode and only if ACTIVITY ON previously
requested for this port)
<<< 1 UPDATE Info available for call specified on station
STN:nnnn: specified. Unify may need to interpret
INUM:nnnnnnnnn some of the xml to determine recording
nnnnnn: <xml action. Will send all call info on each
data> update so Unify can cache single string.
On Unify wanting to tag call
Issue 5 May, 2008 247


Sends
1 TAG STN:nnnn: >>> This allows tagging to be sent for the

INUM:nnnnnnnnnnnnnnn: current and previous calls on a channel.
<xml data> The recorder will use nnnnnnnnnnnnnnn
to determine whether the tagging is for
the current call or the previous one. In the
latter case, it will open the XML file and
append the details given.
Additional tagging should currently be
confined to the 8 pre-defined fields
spare1 to spare8. i.e. <xml data> should
be of the form <spare1>some
data</spare1>.
Call Break – Unify initiated
1 BREAK STN:nnnn: <xml >>> Recorder breaks the call seamlessly. The
data> optional tag data is applied to the stopped
call. Send further TAG command to tag
new call.
Noend and nostart fields are set true on
the old and new calls as per MediaStore.
<<< 1 STOPPED
STN:nnnn:
INUM:nnnnnnnnn
nnnnnn:
<<< 1 STARTED Same STN but different INum
STN:nnnn:
INUM:nnnnnnnnn
nnnnnn:
Call Break at recorder due to max segment duration exceeded
<<< 1 STOPPED
STN:nnnn:
INUM:nnnnnnnnn
nnnnnn:
<<< 1 STARTED Same STN but different INum
STN:nnnn:
INUM:nnnnnnnnn
nnnnnn:

Sends
<<< 1 UPDATE Only sent if ACTIVITY ON previously

STN:nnnn: requested. Unify can identify this as a
INUM:nnnnnnnnn broken call from the “nostart” field being
nnnnnn: <xml “true”.
data> Unify tagged data will have been rolled
over from previous call and included in
this call’s data already.
Call Masking
Note: The properties file setting rec.maskallowed must be set to true for this feature to be enabled.
3 PAUSE STN:nnnn: >>> Instructs the recorder to begin ’masking’.
The recorder continues to record, but
replaces the real audio with a fixed tone
pattern. This is useful for portions of
conversations that are confidential.
<<< 3 PAUSED
STN:nnnn:
3 RESUME STN:nnnn: >>> Instructs the recorder to revert to normal
recording of the real audio.
<<< 3 RESUMED
STN:nnnn:
On shutdown
1 BYE <reason> >>> <reason> can be
1. FALLBACK – recorder should
continue to answer calls and record
(though in most cases, no calls will
appear unless Unify is sending them to
the recorder’s stations). Without knowing
the mappings in Unify, it is not as easy to
drop into a fallback mode – unless station
numbers are also configured somewhere.
2. SHUTDOWN – recorder will stop
recording on Unify controlled ports.
Allows “busying out”.
<<< 1 BYE <extra info>
message
On channel error
Issue 5 May, 2008 249


Sends
<<< 1 ERROR Softphone unusable e.g. cannot register.

STN:nnnn: In AUTO-ANSWER mode, Unify should
mark this port faulty and not attempt to
use. Hopefully, an ONLINE message will
follow shortly as the recorder re-registers
the port.
Unify should probably be written to
recognise faulty behaviour too e.g. if it
doesn’t get STARTED from softphone it
tried to place a recording on, it needs to
report this and busy out the faulty one?
On channel being restored
<<< 1 ONLINE Softphone available
STN:nnnn:
Appendix E: Fault Tolerant Systems
In addition to using fault tolerant components within servers as described in High

Availability Systems on page 59, recording systems can be made tolerant of many server
and network failure conditions. This appendix details how such systems are designed and
configured, how they handle failures and how to upgrade them. This Appendix assumes
that the reader is already familiar with design, installation and operation of single server
recording systems as described in the body of this manual.
● Redundant SAN on page 252
● Duplicated recording on page 253
● Standby Recorder Options on page 254
● Mode of operation on page 276
● Standby Recorder Configuration on page 281
● Upgrade Process for Primary/Standby on page 283
Issue 5 May, 2008 251

Fault Tolerant Systems
Redundant SAN
If you already have a fault tolerant storage network in place, you can assign the calls
storage path to an area of this storage system. This is supported for locally connected
drives - NOT for Network Attached Storage (NAS).
You are then reliant on the mirroring or other redundancy and standby mechanisms
associated with the storage network. These mechanisms typically include a tape library
backup and might include hierarchical file storage (HFS), in which older files are replaced
by small tokens that allow the system to retrieve the original content from the tape library.
The recorder runs successfully with Tivoli Storage Manager and might also support other
similar systems. However, Verint does not proactively test against these systems. Connect
to them at your own risk. If you want to use such a system, turn the recorder's own disk
management function off. To do this, add the line disk.manager=false to the properties file.
This stops the recorder from deleting the oldest files as the available disk capacity falls to
1GB. You must then ensure that the available space on the /calls partition does not fall
below 1GB.
Duplicated recording
Duplicated recording
Where recording is controlled by an external application (such as Unify) establishing
conference calls onto the recorder's ports, a fault tolerant system can be deployed by using
two totally independent recorders and establishing 4-way, instead of 3-way conferences.
By conferencing in one port from each of the two recorders, a fully redundant recording
system can be delivered.
! WARNING:
WARNING: Although the recorders are running in parallel, you must also ensure that the
mechanism that is establishing the conferences is, itself, fault tolerant. This
may imply duplicated Unify servers and/or duplicated CTI feeds. The default
behavior of recorders is to fall back into basic recording mode (where
possible) should connectivity to Unify be broken. This often suffices as an
acceptable failure mode.
You must ensure that the mechanism that establishes the conferences can handle the
error condition that occurs when one of the recorders fails. It must establish the other
conference party to the remaining, good recorder rather than fail totally.
Note:
Note: A parallel recording system inevitably requires twice the resources on the
Communication Manager, both in terms of IP_API_A licenses and VoIP
resources. This option is, therefore, an expensive one.
! WARNING:
WARNING: The overall maximum of 6 parties on a conference might become a
constraint if two of these are recording ports. Call scenarios in which two
calls merge (transfer and conferencing) - and where each of these is already
a 4 party call due to duplicated recording can exceed these limits and are
not supported. To avoid this, recording ports are removed from a call as
soon as it goes on hold. In this way there are less parties present on the call
when it later merges with a consultation call.
Issue 5 May, 2008 253

Standby Recorder Options

You can configure additional Verint Contact Recorders to act as standby to a primary
recorder. These would normally be in "standby" mode and only become "active" when they
detect the failure of the primary they are shadowing. Where you have multiple sites, you
can use multiple standby recorders. Each standby can back up a specified primary or a
specific subset of a primary recorder's configuration.
This section describes:
● Prerequisites for high availability
● Standby recorder licensing
● Known limitations
● Supported fall back scenarios
● Supported failure modes
● Mode of operation
● The impact of a switch-over
● Restoration of a primary recorder once it has been repaired
● Upgrade procedures specific to primary / standby topologies
● A summary comparison with hardware switch-over units
● "Supported Master/Standby Topologies
Prerequisites for high availability

To get maximum benefit from a standby recorder topology, you must adopt the appropriate
elements of an overall strategy for high availability:
Centralized database and replay

Although the standby unit(s) can work without a centralized database and replay tools like
Viewer, you need these elements to provide a seamless, single point of access to all
recordings. Without them, those recordings made on a standby recorder are only
accessible by accessing the url of that recorder. With the central database, all recordings
made on any of the recorders, including the standby(s), are consolidated into a single,
searchable pool.
Fault tolerant storage

Both primary and standby recorders must be configured with RAID1 or (preferably) RAID 5
arrays or be using a fault tolerant SAN on which recordings are kept. Using standby
recording provides fault tolerant recording. However, it does not itself provide for fault
tolerant STORAGE of recordings.
Off-site archive
It is also assumed that, where DVD+RW disks are written, that these will be stored away
from the recorders, ideally in a separate building for maximum protection. Should these be
needed for regular access, a copy should be made and kept off-site. Unless the fault
tolerant disk system is spread across multiple buildings, off-site storage of archive media is
essential.
Fault tolerant network connectivity

Where centrally located primary and standby recorders are co-located or are in disaster
recovery sites with fault tolerant connections between them, it is critical that the primary
and standby recorders are connected by way of a fully fault tolerant network topology.
Failure of any one networking component (NIC, cable, Ethernet switch or router) between
them must not result in connectivity being lost between them. Each standby recorder
assumes that failure to communicate with the primary indicates that the primary has failed.
It will attempt to take over from it. If this is not the case, then the two recorders will clash as
they attempt to use the same softphone numbers on the switch. You must provide multiple
independent paths or teamed NICs with independent network paths between primary and
standby as described in Fault Tolerant Paths on page 84.
Where remotely sited standby recorders are provided so as to record should the remote
site lose connectivity with the center and fall back to ESS mode, you should ensure that
the remote standby is connected to the primary recorder over the same network path that
the Avaya components use. This ensures the it will lose contact with the center - and
hence become active - should the site fall into ESS mode.
Location of recorders
Where possible, you should locate the standby recorder(s) in a different building from the
primary recorders. You should also provide diverse network routing between the recorders
and the Avaya switch components that they interact with (AE Services, C-LAN and
MedPros).
One additional Media Processing Resource per Media Gateway

You must provide one additional media processor per Media Gateway to ensure that failure
of a single card does not impact the ability of a recorder to record at its configured capacity.
Issue 5 May, 2008 255

Standby recorder licensing

The standby recorder is given a special server license that:
● modifies its Administration User Interface to support the entry of the addresses of the
recorder it is to back up.
● restricts it from being used as a primary recorder.
Once the Standby Server is licensed and its primary identified, the standby copies the
channel licensing details from the Primary. However, when the standby recorder is
manually or locally configured, it does not copy channel license details from the primary. In
this circumstance, you must ensure that it is licensed appropriately.
When you enter a license key that identifies a recorder as a standby unit, you must
subsequently stop and restart the recorder in order for it to take notice of this setting.
In laboratory conditions, when switching between primary and standby licenses, a restart is
required.
Automated Configuration Shadowing

If you are using service observe based recording (Station Bulk, Station Executive) you will
probably want the standby recorder on each site to start recording those stations that are
on that site when failover occurs. Remote standby recorders copy the entire configuration
from the central primary recorder. In the event of network failure, they start recording those
stations that are marked with that standby recorder's serial number as their standby
recorder. They will do so using softphones that are marked with their serial number as their
Standby recorder. These settings are under the "Advanced" link for the range of
softphones or stations.
Manual Configuration of Standby

If you are using conferenced mode recording, it is unlikely that your fallback scenarios will
include full CTI links. In such cases you can instruct the Standby recorder NOT to copy
configuration from the primary. You are then free to configure it manually such that it
operates as you require under fallback conditions. This may be a simpler approach such
as recording all stations on its site.
See Standby Server on page 182 for instructions on how to specify this option.
Known limitations
Deploying standby recorders:
● Does not guarantee access to recordings made by the failed server. See Centralized
database and replay above.
● Does not guarantee "no loss" of recording on failure. See Mode of operation below.
● Requires that the standby recorder can contact the primary recorder when it is first
configured so that its configuration can be copied.
● Does not allow configuration changes to configuration while a standby recorder is active
(unless the standby is configured for local configuration). You must restore the primary
recorder before making any changes to the configuration.
● Does not support the use of Telephone Replay ports on the primary and standby
recorder (unless you force local configuration). Place these on a central replay server
(or pair of servers if fault tolerant replay is required).
Distributed Systems
The primary and standby operations are designed to work well in distributed systems
where loss of connectivity between sites would otherwise case recording to fail. The
following Avaya failover modes are supported:
● Locally Survivable Processors (LSP)
● Enterprise Survivable Servers (ESS)
AE Server to Recorder Mapping

Each group of primary recorders and each group of standby recorders should have their
own AE Services box. This provides an extra layer of redundancy. If both the primary and
standby share the same AE Server then a failure of this server will affect both. If they
connect to different AE Servers then the primary recorder can detect this and ask the
standby to take over. As it uses a different AE server, it should be able to do so
successfully.
Where recording is taking place across the enterprise, always place an AE server on the
same site as the recorder that will use it so that the remote recorders can continue to
operate should the WAN fail.
Locally Survivable Processors (LSP)

LSP is where a remote site has an S8300 chassis that normally acts as a slave to a central
S8700. When the connection to the central site fails, the S8300 takes over in LSP mode
Issue 5 May, 2008 257

and uses a copy of the configuration from the central switch server. Because the copied
switch configuration (from the central site) cannot show the S8300 itself offering a DAPI
connection, the switch will not allow the recorder to establish a DAPI link to it under
fallback conditions. Therefore, recordings that are made under these conditions have
limited call details - that is, they do not have data in the agent, VDN, or UCID fields.
You must:
● specify the S8300 on the LSP site as the backup C-LAN for the recorder ports that you
wish to use when in LSP failover mode
● specify the appropriate remote Verint Contact Recorder as the Standby recorder for the
stations to be recorded there in LSP mode.
Three options are supported for LSP:
a. Standalone Recorder(s) on LSP Site(s) - In this enterprise configuration, the
recorders and AE servers are based on the satellite site(s) only. There are no
standby recorders. Each Verint Contact Recorder is configured to register its
softphones through a C-LAN at the central site. The backup C-LAN address for the
softphones is the S8300 on the same site as the recorder. In normal operation, the
recorder uses the C-LAN on the S8700. However, when LSP mode is invoked, the
recorder will reregister softphones on the backup C-LAN - in this case, the S8300 on
the remote site.
All calls which are taking place at the time will be lost but all new calls will be
recorded by the standby.
b. Automatic Standby Recorder(s) on LSP Site(s) - In this enterprise configuration,
the primary recorders are based at a central site and the standby recorders are
based at the satellite sites. The recording ports on the primary recorder are split into
groups. Each group of ports reflects the ports on a standby recorder that will take
over in failure mode.
A group of ports is configured with:
- The serial number of the standby Recorder that is to back these up in failover
mode
- Their C-LAN setting configured to point first to a C-LAN at the central site
(primary) and secondly, at a C-LAN on the satellite site (backup).
The port allocations made on the primary recorder are similarly marked with the
serial number of the standby recorder. The standby takes over this part of the total
recording load in failover mode.
The configuration of the primary recorder is copied periodically to all the standby
recorders on the remote sites as normal. When a failure occurs, the standby
recorder connects only those ports marked with its serial number to the standby
C-LAN and allocate them to the recording tasks that are marked for this recorder to
perform in failover mode.
c. Manually Configured Standby Recorder(s) on LSP site(s) - The system topology
is the same as in b) above, with the primary recorder(s) based at a central site and
the standby recorders based at the local sites. However, the standby recorders are
setup for manual configuration (by means of a property file entry) and are typically
configured to perform station bulk recording of the stations at the LSP site. To set
this up, the primary recorder is configured without the backup or standby settings.
Then, each standby recorder is configured manually to operate as required when
LSP mode is invoked.
Enterprise Survivable Servers (ESS)

Enterprise Survivable Servers are where one or more S8500-based Communication
Manager servers can take control of a portion of the overall system if the main server fails.
If an S8500 is present on a remote site then it can take over the running of any gateways
on that site including the C-LANs on the gateway.
This is better than LSP mode from a recording perspective as this can offer a DAPI link to
the standby recorder even when operating in failover mode i.e. disconnected from the
central site. Two options are supported:
1. Standalone Recorder(s) and AE Server(s) on ESS Site(s) - In this configuration, the
recorders are based on the S8500 site(s) only. There are no standby recorders. Here
the recorders are configured to use a C-LAN on their site and, after ESS is invoked,
this same C-LAN is still present and usable. The recorder may encounter a short
interruption in service but it continues to function after the local S8500 has switched in.
You do not need to setup the recorder for this to take place, as, apart from the brief
interruption of service, the recorder is not aware of any changes.
2. Primary Recorder centrally + Standby Recorder(s) on ESS Site(s): - If the WAN is
lost, the standby recorders determine that the primary is down. When they no longer
detect the heartbeat signal over the WAN, the standby recorders take over. They
operate in an identical manner to the standby recorders in LSP mode as described
above for both the manual and local standby LSP mode. However, unlike LSP mode,
their AE Server establishes a DAPI link and provides full call tagging.
Supported failure modes

The Standby recorder is designed to become active within a few seconds of a complete
system failure of the primary recorder it is configured to "shadow." These failure modes
include:
● Catastrophic server failure
● Total network isolation
● Accidental power down or power failure
● Sustained failure of the link to AE Services or (where used) Avaya CT
● Disk full
Issue 5 May, 2008 259

Standby recorders and Unify/External Control

Unify/External Controller connected to Master/Standby pair
When using Unify or an equivalent external controller with a primary/standby recorder pair,
it should send any START/STOP commands to the recorder that sent it port ONLINE
messages.
Only the active recorder will send STARTED/STOPPED messages. Unify should only send
TAG commands to this active recorder.
Unify/External Controller connected to Master only

If the external controller is essential to the operation of the Master (for example, if it is
setting up single step conferences) then the master needs to recognize that it is not viable
if it has no links to external controllers. Set the property unify.required in the
properties file to indicate this. You can then provide either:
● a standby recorder, locally configured for Station Bulk mode
● a second Unify/External controller connected to the standby recorder
With this property set, the Unify/External controller can force a switchover to the standby
by sending a 1 BYE SHUTDOWN message and then not sending the HELLO response
when the recorder re-establishes connection. When it is ready to take control again, it can
send the HELLO message and the recorder will take over from the standby again.
Supported Master/Standby Topologies

The following pages describe the supported Master/Standby topologies. Use the table
below to determine the appropriate configuration for your system then refer to that table for
details of sizing, configuration, how faults are handled and how to restore the system
following a failure.
For further discussion of these options, read the remainder of this Appendix.
Recording System Size Avaya Topology See Table

Mode
Station Bulk Single Single Site A

Recording Recorder
One or more remote sites with B
ESS support.
No backup for stations recorded at
central site.
One or more remote sites with C
ESS support.
Backup for stations recorded at
central site.
Multiple Single Site D
Independent
Recorders
Conferenced Single Single Site A
Recording Recorder
One or more remote sites with E
ESS support.
No backup for stations recorded at
central site.
Multiple Single Site F
Recorders
One or more remote sites with G
ESS support.
Backup for stations recorded at
central site.
Master/Standby Topology A
If your System Recording Station Bulk Recording or Conferenced

Type is "A" Mode Recording
System Size Single Recorder
Avaya Topology Single Site
To protect Recorder Critical Failure
against…
AE Server Critical Failure
Network Failure of path between recorder and AE
Server
Issue 5 May, 2008 261

Master/Standby Topology A
Provide Standby CPU size as per Master recorder. Disk may

Recorder be smaller as long as you restore the master
promptly following any switchover.
Standby AE Identical to other AE Server
Server
Network Fault tolerant connection between master
Connectivity and standby recorder. (Teamed NICs and
independent switches or two completely
independent paths.)
Replay Server Provide eWare/Viewer or Central Replay
Server if you want to avoid having to search
separately across Master and Standby
recorders.
Configure Master To connect to one AES only
Recorder
Standby To connect to other AES only.
Recorder To connect to Master via fault tolerant path
as above.
Other recording configuration will be copied
automatically from the Master recorder.
In the event of a When It may take a few minutes for the switch to
failure switchover release existing softphone registrations and
occurs associated service observes.
Calls in progress will be recorded as soon as
the softphones on the standby can be
registered and service observe is allowed.
Until problem is Standby will record exactly as the Master
resolved would have.
When problem Master will allow Standby to continue
is resolved recording until you either restart the whole
system or instruct it to restore via url
/servlet/cscm?cmd=restore.
Do this out of hours as there will be a brief
interruption to recording as the system
switches back.
Master/Standby Topology B
If your System Recording Station Bulk Recording

Type is "B" Mode
Avaya Topology One or more remote sites with ESS support.
No backup for stations at central site.
against…
Server
Failure of path between an ESS site and
central recorder
Provide Standby For each ESS site you want to protect,
Recorder(s) provide a standby with CPU sized according
to the number of channels to be recorded on
that site. Disk may be smaller as long as you
restore the master promptly following any
switchover.
Standby AE On each ESS site you want to protect
Server
Network Network path between each remote standby
Connectivity recorder and the central recorder. Use same
path(s) as are used for the Avaya switch so
that if the site goes into ESS mode, the
standby will also have lost connectivity to the
(central) master recorder.
recorders.
Issue 5 May, 2008 263

Configure Master To connect to central AES only.

Recorder Split the softphone ranges on the Settings >
Communication Manager page into groups
suitable for use by each standby. Use the
Advanced Settings link to specify which
Standby recorder should use each block of
softphones.
Set appropriate Standby recorder serial
number (under Advanced settings link) for
each station or range of stations to be
recorded in Station Bulk mode (unless using
localconfig option below)
Each Remote Each to connect to AES on its site only.
Standby To connect to Master via network path as
Recorder above.
automatically from the Master recorder and
each standby recorder will record only those
stations set with its serial number as their
designated standby recorder.
ALTERNATIVELY set
standby.localconfig=true and
configure the standby itself if you want it to
record different stations when in ESS mode.
Ensure each standby has its own set of
softphones in this case.
In the event of a When If the master recorder has failed, it may take
failure switchover a few minutes for the switch to release
occurs existing softphone registrations and
associated service observes.
Until problem is Standby will record only those stations
resolved marked with its serial number as their
standby recorder (OR if using the localconfig
option above, it will record whatever it has
been configured to record).
When problem Master will attempt to take control as soon as
is resolved it recovers. This is to ensure that any stations
without a remote standby assigned are
recorded. Remote sites that are in ESS
mode will remain in ESS mode until manually
reset so will not be affected by this.
Standby will continue to record until reset.
You must therefore reset the standby at the
same time as you restore the Avaya system
from ESS mode to centralized control.
switches back.
Master/Standby Topology C

Type is "C" Mode
Avaya Topology One or more remote sites with ESS support
AND backup needed for stations at central
site being recorded.
against…
Server
central recorder
Issue 5 May, 2008 265

Provide Standby At Central Site, provide a Standby recorder

Recorders with same CPU size as the Master.
For each ESS site you want to protect,
provide a standby with CPU sized according
that site.
Disks on all Standbies may be smaller as
long as you restore the master promptly
following any switchover.
Standby AE Identical to existing AES at central site.
Server One on each ESS site you want to protect
Network Fault tolerant connection between Master
Connectivity and central Standby recorder. (Teamed NICs
and independent switches or two completely
independent paths.)
Network path between each remote standby
recorder and the central recorder. Use same
recorders.
Configure Master To connect to one central AES only.

Recorder Split the softphone ranges on the Settings >
Communication Manager page into groups
suitable for use by each standby. Use the
Advanced Settings link to specify which
Standby recorder should use each block of
softphones.
Set appropriate Standby recorder serial
number (under Advanced settings link) for
each station or range of stations to be
recorded in Station Bulk mode (unless using
localconfig option below)
Central Standby Set standby.main=true
Recorder To connect to the other central AES (not the
one the master connects to).
To connect to Master via fault tolerant path
as above.
Standby To connect to Master and Central Standby
Recorder via same network path Avaya system relies
on for central control.
automatically from the Master recorder and
each standby recorder will record only those
stations set with its serial number as their
designated standby recorder.
ALTERNATIVELY set
standby.localconfig=true and
configure the standby itself if you want it to
record different stations when in ESS mode.
Ensure each standby has its own set of
softphones in this case.
Issue 5 May, 2008 267

In the event of a If Master Central standby will take over. Remote

failure recorder fails standbys will stay inactive.
It may take a few minutes for the switch to
release existing softphone registrations and
associated service observes.
the softphones on the central standby can be
When The affected standby will register the
switchover to appropriate softphones and start to record
ESS occurs the stations assigned to it (or locally
configured if that option has been set).
Until problem is Appropriate standby will record as above.
resolved
When problem If the Master recorder failed, DO NOT
is resolved RESTART IT UNTIL OUT OF HOURS.
STOP THE CENTRAL STANDBY BEFORE
RESTARTING THE MASTER.
Failure to do so will result in the remote
standbies being told to go active.
If the problem was with network connectivity,
then the remote site will have dropped into
ESS mode and the master will not be able to
take control until this is restored as below.
Remote Standby will continue to record until
reset.
You must therefore reset the remote standby
at the same time as you restore the Avaya
system from ESS mode to centralized
control.
switches back.
Master/Standby Topology D

Type is "D" Mode
System Size Multiple Independent Recorders
Master/Standby Topology D
To protect Recorder Critical Failure of any Master Recorder

against…
Server
Provide Standby CPU size for each as per the Master
Recorders (one recorder it is backing up. Disk may be
for each master smaller as long as you restore the affected
recorder) master promptly following any switchover.
Standby AE Identical to other AE Server (provide more on
Server very large systems)
Network Fault tolerant connection between each
Connectivity master and its standby recorder. (Teamed
NICs and independent switches or two
completely independent paths.)
Replay Server Should already be in place. No need to
change
Configure Each Master To connect to one AES only. Spread masters
Recorder across the available AE Servers.
Each Standby To connect to one AES only - not the same
Recorder one as that of the Master recorder it is
shadowing.
To connect to the one Master it is shadowing
via fault tolerant path as above.
automatically from the appropriate Master
recorder.
In the event of a When It may take a few minutes for the switch to
failure switchover release existing softphone registrations and
occurs associated service observes.
Until problem is Each Standby will record exactly as the
resolved Master it is shadowing would have.
When problem Master will allow its Standby to continue
is resolved recording until you either restart both Master
and its Standby or instruct it to restore via url
interruption to recording as this pair of
recorders switches back.
Issue 5 May, 2008 269

Master/Standby Topology E
If your System Recording Conferenced Recording

Type is "E" Mode
Avaya Topology One or more remote sites with ESS support.
No backup for stations at central site.
against…
Server
central recorder
Provide Standby For each ESS site you want to protect,
Recorder(s) provide a standby with CPU sized according
restore the master promptly following any
switchover.
Standby AE On each ESS site you want to protect
Server
Network Network path between each remote standby
Connectivity recorder and the central recorder. Use same
recorders.
Configure Master To connect to central AES only.
Recorder
Standby To connect to Master via network path as
Recorder above.
Set standby.localconfig=true and
configure the standby to record the local
stations. You may want to have the standby
record in Station Bulk mode rather than rely
on local TSAPI services.
Master/Standby Topology E
In the event of a When Calls in progress will be recorded as soon as

failure switchover the softphones on the standby can be
occurs registered and service observe is allowed.
Until problem is Standby will record whatever it has been
resolved configured to record.
When problem Standby will continue to record until reset.
is resolved You must therefore reset the standby at the
same time as you restore the Avaya system
from ESS mode to centralized control.
switches back.
Master/Standby Topology F

Type is Mode
System Size Multiple Recorders
against…
Server
Provide Standby CPU size as per Master recorder. Disk of
Recorder Master and Standby do not record calls so
may be small.
Standby AE Identical to other AE Server
Server
Slave Provide one more slave recorder than is
Recorders needed to support maximum concurrent
load. These operate as a pool so N+1 are
required for fault tolerance.
independent paths.)
Replay Server Should be already provided.
Issue 5 May, 2008 271

Master/Standby Topology F
Configure Master To connect to one AES only.

Recorder ZERO concurrent recording channels in
Conferenced page (all recording is done on
slaves).
Standby To connect to other AES only.
slaves)
as above.
Slave To connect to Master and Standby recorders
Recorders as their "external controllers".
All ports in On Demand mode.
Do NOT configure Conferenced mode or any
Avaya CT settings.
In the event of a When Apart from a few seconds delay, calls in
failure switchover progress should be unaffected if the Master
occurs fails.
If a slave fails, the calls being recorded on it
will be reassigned to the remaining slaves
within a few seconds.
Until problem is Standby will control recordings exactly as the
resolved Master would have.
When problem Master will allow Standby to continue
is resolved recording until you either restart the whole
"/servlet/cscm?cmd=restore".
switches back.
A failed slave, on the other hand, can
restored without danger. Note that its ports
will be added to the back of the available port
list so it may be some time before calls are
assigned to it.
Master/Standby Topology G

Type is "G" Mode
System Size Multiple Recorders
Avaya Topology One or more remote sites with ESS support
AND backup needed for stations at central
site being recorded.
against…
Server
Provide Central Standby CPU size as per Master recorder. Disk of
Recorder Master and Standby do not record calls so
may be small.
Remote For each ESS site you want to protect,
Standby provide a standby with CPU sized according
Recorder(s) to the number of channels to be recorded on
restore the system promptly following any
switchover.
Standby AE One on central site identical to existing AE
Server Server
AND one on each ESS site you want to
protect.
Slave Provide one more slave recorder than is
Recorders needed to support maximum concurrent
load. These operate as a pool so N+1 are
required for fault tolerance.
independent paths.)
Network path between each remote standby
recorder and the central recorders. Use
same path(s) as are used for the Avaya
switch so that if the site goes into ESS mode,
the standby there will also have lost
connectivity to the central recorders.
Replay Server Should be already provided.
Issue 5 May, 2008 273

Configure Master To connect to one AES only.

slaves).
Central Standby To connect to other AES only.
slaves)
as above.
Remote Each to connect to AES on its site only.
Standby To connect to Master and Central Standby
Recorder(s) via same network path as used by Avaya
components so it will detect failure under
same conditions as ESS mode is activated.
Set standby.localconfig=true and configure
the standby to record the local stations. You
may want to have the standby record in
Station Bulk mode rather than rely on local
TSAPI services.
Slave To connect to Master and Standby recorders
Recorders as their "external controllers".
All ports in On Demand mode.
Do NOT configure Conferenced mode or any
Avaya CT settings.
In the event of a When Apart from a few seconds delay, calls in

failure switchover to progress should be unaffected if the Master
Central Standby fails.
occurs If a slave fails, the calls being recorded on it
will be reassigned to the remaining slaves
within a few seconds.
When remote Remote Standby will register phones and
loses contact start to record as per its local configuration.
with centre
Until problem is Standby(s) will continue to record as above.
resolved
When problem If Central Standby took over, Master will
is resolved allow the Central Standby to continue
recording until you either restart the whole
switches back.
A failed slave, on the other hand, can
restored without danger. Note that its ports
will be added to the back of the available port
list so it may be some time before calls are
assigned to it.
If remote standby took over due to loss of
connection to center, it will continue to record
even when connectivity is re-established
(since the Avaya system will continue to
operate in ESS mode). You must therefore
reset the standby at the same time as you
restore the Avaya system from ESS mode to
centralized control.
switches back.
Issue 5 May, 2008 275

Mode of operation
This section describes how the standby recorder is configured and how it monitors the
health of its primary recorder. It also explains how it takes over when needed and returns
to standby mode when appropriate.
Standby configuration (automatic)

If you do not make alterations to the standby machine's properties file, it automatically
copies the license details and settings from its primary.
Standby configuration (manual)

You only have to configure a standby server manually if you have forced local configuration
in the cscm.properties file. A Standby recorder is configured manually in the following
way:
● to match the primary recorder it is to shadow in most respects, for example, system
timeouts, maximum recording durations and so on.
● with the IP addresses of the primary recorder that it is to shadow.
Power-On
A common case is that all recorders are powered on within a second or two of each other.
To avoid destabilizing the system, the standby unit will always wait for a pre-defined period
after it starts trying to contact the primary recorder before it assumes that it is dead. This
period allows for slight variations in boot time and manual power-on of one unit after
another.
The time-period is determined by the standby.connecttimeout=xxx property in the
properties file. (xxx is in seconds).
Standby mode
Once configured, the standby unit attempts to establish TCP/IP socket connections to the
primary recorder(s) over the one or more IP addresses it has been configured with. If it
succeeds, it downloads the configuration details of the primary recorder that are necessary
Mode of operation
for it to take over in the event of the primary failing. It continues to refresh these details
every minute, thus keeping up to date with configuration changes.
Heartbeat messages are exchanged every second. If the standby unit detects a critical
error - such as failure to connect to AES - it will shut down its link(s) to the Master so as to
avoid any danger of the Master relying on it for fallback when in fact it is unable to function.
Failure Detection
The standby recorder will attempt to take over from the primary recorder in the following
circumstances:
● The primary requests that the standby unit take over because it has detected a fatal
error such as hard disk full, switch connectivity lost completely.
● On startup, if connection cannot be established with primary after 120 seconds (default -
and accurate only to within one minute).
● All TCP/IP sockets to the master(s) failing and attempts to re-establish them every
second are still failing after 30 seconds.
● Either or both TCP/IP sockets to the Master are still active but the primary does not
respond to repeated heartbeat polls for 30 seconds.
The three timeout settings above may be overridden by setting the property values
standby.connecttimeout, standby.reconnecttimeout and
standby.inactivitytimeout respectively in the properties file. Times are specified in
seconds.
These same parameters are also used by the primary recorder to report corresponding
failures of the standby although it will not do anything other than raise an alarm if this
occurs. The default values for these timeouts can be overridden on the primary using the
same property settings.
A primary recorder is not aware of standby recorder(s) until it has established contact with
a standby for the first time. Thereafter it will expect to establish contact with that standby
recorder always. (This can only be reversed by editing the underlying configuration
database).
Disk Space Monitoring

Once a minute, each recorder will monitor the available disk space on the partitions used
for the operating system, call details storage and recording storage. If any of these drops
below 500MB a warning will be raised.
To override this default; set disk.warnAtMB=nnn in the properties file.
Issue 5 May, 2008 277

Active mode
On inferring failure of or being instructed to by the primary, the Standby unit will adopt the
appropriate port allocations and attempt to register softphones as per the automatically
copied configuration or its local configuration.
It will configure the softphones in the same way that the failed recorder had been using
them prior to failure (unless you have forced local configuration). Where softphones are
marked with specific standby recorder numbers, only those marked for a given standby will
be registered.
Recordings will be made and uploaded to the eWare database or Central Replay Server (if
present).
Return to Standby mode

A standby recorder will return to Standby mode:
● If the standby unit is shutdown before the failed primary is restored. When the standby
unit is rebooted, the primary responds within the timeout mentioned above and the
standby recorder remains in standby mode.
● If the failed unit recovers it may instruct the standby recorder to return to idle
immediately - or not until the administrator forces this (see Restoring the Primary on
page 279). At this time, the standby unit will truncate all current recordings and
un-register all softphones allowing the primary to take over again. Note that this
switch-back has the same impact on recordings as the initial switch-over. Hence you
should only bring a failed unit back on-line out of hours.
Switchover Implications
It takes a few seconds to detect most of the failure modes. Although configurable using the
properties file, this interval is a compromise between rapid detection of true failure versus
risk of false alarms and/or "yo-yoing," a condition where the system goes unstable. The
system defaults aim to detect failures within 10 seconds and should not normally be
altered.
When a failure occurs and the standby recorder becomes active:
● Recordings in progress are interrupted. The partially completed .wav files for the
recordings in progress might be manually recoverable (professional services
chargeable). In G.729A mode, up to 1 minute of audio is buffered in memory and hence
will not have been appended to the file. In G.711 mode, files are appended to every 30
seconds.
Mode of operation
For Station Bulk and Station Executive recording modes:

● The standby recorder switches in within seconds, registers its softphones and starts
recording new calls and calls still in progress from the point at which the softphones are
registered and service observe is re-established.
For On Demand and Meeting recording modes:
● Recordings in progress are dropped. The standby recorder switches in within seconds,
registers its softphones and awaits new calls arriving on these ports.
The worst case loss of recording on a port is therefore determined by the speed with which
the switch can re-register the failed softphones and establish service observe again. The
more channels on the failed recorder, the longer this process takes.
Restoring the Primary

The fault condition that caused the standby to become active can be removed in a number
of ways, not all of which are ideal. These methods are:
● Power restored to primary
● Network connectivity restored to primary
● Connection to AE Services is restored
● Primary repaired and reinstated.
If there is only one standby recorder, the master recorder will allow it to continue so as not
to interrupt recordings unnecessarily. The administrator may force the master to take
control again by logging into the administration web pages on the master and then entering
the url http://servername:8080/servlet/cscm?cmd=restore
This will cause the master to take over. It should be performed out of hours as there will be
a brief interruption in recording as the system switches over.
In cases where multiple standby recorders are, between them, providing backup for a
master, this is usually less desirable than having the (single) master regain control. It is
also more difficult for the Master recorder to be confident that all recordings are being
handled by the standbys. In this case, a recovering Master will retake control automatically
- with a brief interruption to service as it does so.
The one exception to this is a remote standby configured as part of an ESS. A remote
standby will remain recording until it is shutdown. It should be restarted as part of the
procedure to unpartition the switch. The master will take control of the stations that were
partitioned once the switch is rejoined.
Issue 5 May, 2008 279

Comparison with hardware switch-over units

Traditionally, high availability digital recording has been provided by inserting an "N+1
Switchover Unit" between the recorders and their audio sources. Although this provides a
slightly faster switchover, it has several negatives in comparison to the software only
approach used here:
● The cost of the switchover unit makes N+1 systems uneconomical for N=1 or 2
● The additional cabling required to and from the switchover unit introduces further cost,
failure mechanisms and opportunities for mis-configuration (swapped cables) that might
not be noticed until after a failure has occurred.
● The switchover unit itself introduces a significant single-point of failure into the system.
Standby Recorder Configuration
Standby Recorder Configuration
Configuration Differences
The following Administration pages differ on a Standby recorder:
● System Settings > License - Check that the license shows as a "Standby" server type.
The number of channels is equal to that on the primary recorder that this one is to
shadow.
Note:
Note: If you do not invoke local configuration in the properties file, the standby
recorder automatically copies license details from the primary.
● System Settings > Server - If using a DVD+RW drive for archive, set the path as usual
and set the volume label root to a string that is unique to this standby recorder. This will
let you distinguish DVDs written on the standby unit from those written on the primaries.
- The timeout and purge settings should be kept identical to those on the primary
recorder.
- An additional setting shows and allows you to enter the IP addresses of the primary
recorder that this recorder is to shadow. A primary recorder should have two NIC
cards, each with its own IP address. Both addresses should be configured on the
standby to enhance the reliability of the network link between them. The default IP
port number will be 1414 but you may override this by adding the port number to the
IP address, separated by a colon, for example, 192.134.34.45:1419. If you enter
a port number, this must match the configuration of the primary recorder, which also
defaults to 1414. You can override this default by setting the primary.localport
property in the cscm.properties file. Unless you are confident that your DNS
server is fault tolerant, you should specify numeric IP addresses to ensure that
connectivity does not fail because of address translation problems.
- You are strongly advised to create a separate email account, using a separate
SMTP server from that used by the primary. This way, failure of an SMTP server will
be noticed as you continue to receive e-mails from the standby unit but not the daily
heartbeats from the primary - or vice versa. It is critical that you ensure that the
standby unit is ready to take over at any time.
- The IP Station Security Code should be entered the same as for the primary.
! Important:
Important: Do not use the Advanced settings in the port entry form for the standby
recorder.
Issue 5 May, 2008 281

● Security > Users - This page is identical to that on the primary recorder. However,
since a central database and replay are nearly always used in systems with standby
units, there is little point in entering "replay only" users. Simply use this page to maintain
the administrators' account(s).
● Port Allocations - These pages are provided but their contents are not editable as
these are copied from the primary recorder (once contact has been established with it).
● Status > Alarms and Events - This page is identical to that on the primary recorder
though some different alarms will be generated - when the standby establishes contact
with the primary and should it ever try to take over from the failed primary. You should
use the email settings on the System Settings > Server page to ensure that you are
advised of these alarms within 10 minutes of one occurring.
● Status > System Overview - In addition to the normal fields that show on primary
recorders, a line will be added to the top table showing the status of the two links to the
primary recorder that the standby is configured to shadow.
● Status > Port States - When the recorder is in standby mode, no ports are visible.
When it takes over from the failed primary, this page should be the same as it would
have been on the primary had it still been running.
● Status > Peak Activity - This page is identical to that on the primary recorder but has
an added use on the standby recorder. Click on the Restart Peak Activity Count link at
the bottom left of this page to zero the counters. When you next view this page, if any of
these counts are non-zero you can infer that the standby unit has gone active since you
reset the count. This is a quicker check than looking through many pages of alarms and
events.
Upgrade Process for Primary/Standby
Upgrade Process for Primary/Standby

A standby recorder can only act as standby to a primary that is running the same version of
recorder software. Please follow these steps when upgrading fault tolerant systems:
1. Take the standby unit OFF the network.
2. Upgrade the primary to the most current version.
3. Confirm correct operation. If any serious problems occur, shut the primary down and
reconnect the standby.
4. When confident that the primary is running correctly, upgrade the standby recorder
5. Reconnect the standby to the network.
Issue 5 May, 2008 283

Appendix F: AIC Connector
This appendix describes the AIC connector. This is a ready-to-use client of the External
Control Interface for Avaya Interaction Center. It covers the following topics:
● Overview on page 286
● Installation on page 289
● Programmer’s Guide on page 290
Issue 5 May, 2008 285

AIC Connector
Overview
The Verint AIC connector is an AIC server process that allows AIC clients and servers to
control the the recorder. It provides the following features to client developers and
scriptwriters:
● Start recording a station
● Stop recording a station
● Tag a particular recording with user defined data
● Enable recording of all calls at a station with automatic tagging
● Disable recording of that station
All these features are made available as simple AIC calls, without the need for complex
integration projects.
“Tagging” is a feature of the connector and the recorder that allows clients to associate
arbitrary information with each recording. These User-Defined Fields (UDFs), or ‘tags’, are
stored in the calls details database and can be included in searches and viewed in the
search and replay web applications. UDFs are tag-value pairs. (All calls will automatically
have the standard set of tag fields.)
The connector seamlessly supports:
● Multiple recorders (if more than one is needed to provide the total number of recording
channels)
● The recorder’s primary/standby high-availability mode
When there are multiple AIC servers for resilience, the connector can be installed on
multiple AIC servers and it will allow recording control of all recorders from either AIC
server.
The connector itself requires no configuration; it determines its configuration from the
recorder’s configuration. The recorder is basically configured in the same way that it would
be for Station Bulk recording, but with a few minor changes.
The AIC connector requires Version 7.7 (or higher) of the recorder. It works in Service
Observe mode only.
Modes of Operation
There are two basic modes of operations:
● Finer, event driven control
● Coarser, connector-managed control
Overview
In the finer mode, clients of the connector are sent events whenever a call segment starts,
and are able to send tagging information via the connector to the recorder for each specific
call segment. To operate in this mode a client must register with the connector server for
events. It calls the start recording method of the connector. For each call segment that is
recorded, the client is sent an event containing the unique recording file serial number. The
client can tag each call segment differently if required. The client must stop recording when
it is no longer required.
In the coarser mode, clients do not need to register for, or handle, events. The connector
deals with event handling. In this mode the client enables recording for a particular
station and supplies a list of user-defined fields. The connector automatically tags each
subsequent call segment for that station with the tags supplied when recording was
enabled. The client need take no action, but loses the ability to tag each call segment
differently. The client should disable recording when it is no longer required.
Principles of Operation
The connector is an AIC server process, and runs on the AIC server. It is available for
Solaris and Windows. The connector is started and managed by AIC like any server
process.
The recorders are configured to contact the connector, which listens on a well-known
TCP/IP port number. If there is more than one connector, each recorder is configured with
all addresses and connects to all of them.
The recorders are administered in the usual way for Station Bulk mode, except that they
accept external control. The recorder will establish service observe on each target station
using one of its recording channels. The target stations can be split across multiple
recorders as required.
As the recorders connect to the switch and establish Service Observe on the target
stations, they announce their ability to record the target stations to the connector. When a
client requests recording of a particular station the connector forwards the request to the
relevant recorder. At the start of each call segment the recorder sends an event to the
connector with the unique serial number of the recording file. The connector sends events
to any registered clients and sends tagging information to the recorder as required.
High Availability
The connector integrates seamlessly with the recorder’s primary / standby high-availability
arrangement. If the primary recorder fails, it disconnects from the connector, and the
secondary connects instead. The connector proceeds to send recording commands to the
secondary recorder.
Issue 5 May, 2008 287

AIC Connector
The recorder can connect to more than one external controller. This allows two connectors
to be installed on separate AIC servers. Either connector can be used to route messages
to the recorder.
Installation
Installation
Before the connector can be installed its IDL must be appended to the AIC server’s master
IDL file, which must then be re-compiled. The resulting ‘pk’ file must be distributed. The
connector’s IDL is in the file cscm.idl on the distribution CD.
The connector consists of a single executable file that must be copied to the AIC server’s
directory. It is typically installed in the same directory as the rest of the AIC server
executables.
Use the AIC admin screens to add an entry for the Verint connector. There are no
parameters required. Start the connector on the AIC server(s) where it will run.
Configuration
Use the ContactStore administration screens to configure the recordable stations. If there
is more than one ContactStore, spread the recorded stations evenly across the recorders.
Specify the stations or ranges of stations on the Unify page (rather than the Station Bulk
page); this allows the connector to stop and start recording.
The recorder needs to know that it should continue to request the parties, VDN and UCID
information from the switch. Edit the properties file and add the following line:
unify.ocpneeded=true
Configure the address of the AIC server(s) in the URL field of the Unify page. If there are
two AIC servers specify both separated by a semicolon.
Check the status of the ‘Unify’ link(s) on the Status page and correct any alarms.
Issue 5 May, 2008 289

AIC Connector
Programmer’s Guide
The interface to the server is described by its IDL, which is shown below.
interface CSCM : GeneralS {
ORBStatus Enable(in string station, in SeqCouple tags);
ORBStatus Disable(in string station);
ORBStatus Start(in string station);
ORBStatus Stop(in string station);
ORBStatus Tag(in string station, in string inum, in SeqCouple
tags);
ORBStatus Assign();
};
The first two methods (Enable and Disable) are used in the coarser-grained mode of
operation; the other four (Start, Stop, Tag and Assign) are used in the finer-grained mode.
Note:
Note: It is entirely possible to use any names for the tags sent to the recorder, but
you should stick to certain conventions. The normal choices for the tag
names are ‘spare1’ through ‘spare8’. If you use these tag names the rest of
the Verint infrastructure will make them available throughout the suite of
applications. If you use other tag names these will only be accessible
through the recorder’s own search and replay application. For example, you
may wish to tag calls with customer account number and zip code. You
should pass a Sequence containing two Couples: spare1=12345,
spare2=10010. Even though zip codes can be numbers they should be
passed as a string. The Viewer search and replay application allows you to
change the names of these fields to more convenient, human-readable
names when you design page layouts later.
Enable(in string station, in SeqCouple tags)

Instructs the connector to enable recording of the specified station, and to tag all
subsequent call segments with the tag-value pairs given. If a call is already in progress,
recording starts immediately.
Programmer’s Guide
Disable(in string station)

Cancels recording started by an earlier Enable. If a call is in progress, recording is disabled
but what was recorded so far is retained.
Start(in string station)

Instructs the connector to enable recording of the specified station. If a call is in progress,
recording starts immediately.
Stop(in string station)

Cancels recording started by an earlier Start. If a call is in progress, recording is disabled
but what was recorded so far is retained.
Tag(in string station, in string inum, SeqCouple tags)

Instructs the recorder to associate the tags with the recording already in progress (with
serial number ‘inum’).
Assign()
Registers with the connector to receive recording start and stop events. It is these events
that contain the inum required for the call-specific tagging.
! WARNING:
WARNING: Note that the two modes of operation must not be mixed. Clients must either
choose the Enable/Disable method, or the Start/Stop/Tag method. Clients
may assign a session in either mode in order to receive events. Creating a
session is optional with the first method and required with the second.
Issue 5 May, 2008 291

AIC Connector
Appendix G: PDS/PCS Connector
This appendix describes the PDS/PCS connector. This is a ready-to-use client of the
External Control Interface for Avaya Predictive Dialler / Proactive Contact. It covers the
following topics:
● Installation on page 295
Issue 5 May, 2008 293

PDS/PCS Connector
Overview
The Verint PDS/PCS connector is an optional server process that runs on the same server
as the recorder. It uses events from the dialler to:
● Split the otherwise long, single recording of a dialler session into separate recordings,
one for each interaction.
● Tag each of the separate recordings with the dialler Job Name.
The PDS/PCS connector requires Version 7.7 (or higher) of the recorder. It works in Station
Bulk and Conferenced modes. The connector works with PDS v12, PCS 3.0 and PCS 4.0.
When working with PCS 4.0, security can be enabled with a property file entry.
Principles of Operation
The connector is an optional process that runs on the same server as the recorder..
The recorder is configured to contact the connector, which listens on a well-known TCP/IP
port number. A recorder always connects to the connector running on the same server.
The recorders are administered in the usual way for Station Bulk or Conferenced mode,
except that they accept external control.
The connector is automatically informed by the recorder which stations are being recorded.
The connector monitors recordings and makes a note of any Agent information. It is able to
build a map of which agents are logged on at which station (it does not have access to CTI
to determine this in more conventional ways).
The connector monitors dialler events for Logon events. If the ’terminal’ ID matches either
a recorded station or an agent ID that has been seen at a recorded station, the connector
notes the recorder port associated with this dialler session.
The connector disables recording on that port. When the dialler agent is on a call,
recording is re-enabled, and the dialler call information is sent to the recorder. When the
dialler agent is between calls recording is again disabled. When the dialler agent logs off,
recording control is returned to the recorder.
The dialler job name is shown in the Service column of the search and replay screen.
Installation
Installation
The PDS/PCS connector is part of the recorder installation kit. No additional installation
steps are needed.
Configuration
Use the recorder administration screens to configure the recordable stations. Specify the
stations or ranges of stations on the Station Bulk page; this allows the connector to stop
and start recording.
Add the address of the PDS connector in the URL(s) of Unify/External control port(s) to
connect to setting on the System Settings > Server administration page. Specify
localhost as the address.
The PDS/PCS connector has its own properties file (pdscon.properties). A sample
properties file is supplied but must be edited before the connector will work.
Edit it and insert site specific data.
The following keys are mandatory:
● PDS.UserName - the PDS client username
● PDS.Password - the PDS client password
● PDS.DialerHost - the name of the PDS dialer
● OAIAddr - the IP address of the recorder’s NIC card to be used for communication with
the PDS
● license.key - the license key
The following setting is optional:
● PDS.NSIOR - the Name Service IOR
Note:
Note: If the IOR is not provided the connector attempts to look up the Name
Service using corbaloc (and must be able to resolve the IP address of the
dialler by name - make sure it is in DNS)
● PDS.Secure - true to enable PCS4 security.
Issue 5 May, 2008 295

PDS/PCS Connector
Choosing the dialler version

There are different personality files for the three different versions of PDS/PCS. You must
create a symbolic link to enable the correct version. In the /opt/witness/lib directory
there are three separate files: pds12_idl.jar, pcs30_idl.jar and pcs40_idl.jar.
Create a symbolic link called idl.jar in the same directory to the appropriate jar. For
example to select PCS 3.0 as the version enter the command:
ln -s pcs30_idl.jar idl.jar
Changing the log level

Change the log4j.rootLogger key.
There are two valid values between the "=" and the "," - DEBUG or INFO.
Note that at DEBUG level the log file is likely to fill the disk completely within a few days.
Use DEBUG level with caution.
Controlling the service

You should add the connector as a Linux service. As root, enter chkconfig --add
pdscon. If you want the service to start automatically, enter chkconfig --levels 345
pdscon on.
As root, you can control the service.
● service pdscon start
● service pdscon stop
● service pdscon restart
● service pdscon status
Automatic restart
If the PDS shuts down nightly, consider adding cron entries to stop and start the connector.
Shut down the connector after all agents have left and restart the next day before agents
begin to log on.
Appendix H: Non-standard Hardware
This appendix discusses considerations for non-standard hardware such as blade servers.
It covers the following topics:
Issue 5 May, 2008 297

Non-standard Hardware
Overview
This manual and the automated installation processes assume a "typical" rack-mounted
server as the server hardware. However, ContactStore has been installed on other
hardware sucessfully, including blade servers. This appendix covers the considerations for
such non-standard hardware.
Disks
SAN
Blade servers are normally equipped with locally attached SAN which is ideal for
ContactStore. See Storage at Each Recorder on page 43 for further details.
Partitioning
The standard installation assumes either one or two physical disks, or one RAID array. In a
more complex environment there could be separate arrays for different partitions. For
example, a RAID 0 mirror for everything except call storage and a RAID 5 array for call
storage. Edit the kickstart script to specify the appropriate disk device for each partition by
changing the "on" parameter. See Expert kickstart options on page 84.
NICS
Blade servers often have redundant network configurations. It is therefore more likely that
ContactStore should be installed with just one IP address (eth0) instead of two (eth0 and
eth1).
DVD
Local archive will typically be unavailable since a blade will not have dedicated access to a
DVD recorder. Contact Archive is the prefered solution in this environment.
Overview
Soundcard
The optional local soundcard will likely be unavailable. This is not a problem as a desktop
machine can be used to test replay.
Kickstart
If no floppy disk is available for kickstart installation, use the network-based kickstart
instructions in Performing a kickstart install without a floppy on page 83.
Issue 5 May, 2008 299

Non-standard Hardware
Appendix I: Advanced Security Settings
This appendix discusses some features and prerequisites for advanced security. It
includes:
● Installing Unlimited Strength Encryption on page 302
● Installing a Signed SSL Certificate on page 303
● Changing Tomcat Port Numbers on page 307
● Encrypting Properties File entries on page 308
● Configuring Viewer and Archive to use HTTPS on page 309
● Configuring Archive for a Key Management Server on page 313
Issue 5 May, 2008 301

Advanced Security Settings
Installing Unlimited Strength Encryption

To support key sizes larger than 128 bits, you will need to replace the standard Java
Jurisdiction Policy Files using the following instructions:
1. Using your browser, go to the following link:
http://java.sun.com/javase/downloads/index_jdk5.jsp
2. Follow the link entitled:
Java Cryptography Extension (JCE) Unlimited Strength Jurisdiction Policy Files 5.0 at
the foot of the page.
3. Download and unzip the files.
4. Follow the instructions in the readme.txt. (The path for installation is
/usr/java/jre1.5.0_13/lib/security).
5. You should note the instruction to back up the existing jars that will be replaced
Installing a Signed SSL Certificate

If you want to install your own SSL certificate, you must replace the certificate distributed
with the application. Your replacement certificate must be specific to your installed server.
Selecting a Certificate Authority (CA)

If you do not already use a certificate authority, you can use:
● http://www.freessl.com/starterssl/starterssl.html - FreeSSL requires that the web server
has a fully qualified domain name (e.g. contactrecorder.bigcorp.com or
contactrecorder.division.bigcorp.com) and needs to be able to send an email to an
address like ssladmin@bigcorp.com or administrator@division.bigcorp.com. The list of
addresses can be found on their website, and it includes admin, ssladmin, root, and
administrator.
● http://www.instantssl.com - InstantSSL is more flexible and allows intranet addresses
(such as WINS names and IP addresses) as well as fully qualified domain names.
Backing up the Keystore file

In the instructions which follow, replace <installdir> with the location into which you
installed Verint ContactStore for Communication Manager.
The certificates and keys are stored beneath your installation folder in the file:
/opt/witness/keystore/keystore.jks
Because this file contains the original, distributed certificate, it is important to make a
backup of it. You will delete this file during the remaining steps. Should it be necessary to
restore the original certificate, you can copy the backup to the original filename.
Creating the new Certificate

If you would like to test this implementation, you can practice this procedure with a
certificate authority's 30-day trial certificate. Then, to implement real certificates, you can
start over from this point.
To create a certificate:
1. Create a new certificate with the real URL of the Verint ContactStore for
Communication Manager.
Issue 5 May, 2008 303

2. Log onto the server and change directory as follows:

cd /opt/witness/keystore
3. Remove the original keystore file
rm keystore.jks
4. Run the java keytool utility with
/javadirectory/bin/keytool -genkey -keystore keystore.jks
-alias tomcat -keyalg RSA
5. Fill in the Keytool prompts with the following:
Password: Contact5tor3
Note:
Note: You must type this password, exactly as shown. It is case sensitive.
a. First & Last Name: enter the FQDN, IP address or intranet name
b. Organizational Unit: enter your division
c. Organization: enter your company name
d. City/Location: enter your location
e. State/Province: enter your state
f. Country Code: enter the ISO 2 letter code for your country (for example, GB is the
code for United Kingdom)
6. Enter yes if the information is correct.
7. Hit enter when prompted for the second password.
8. Restart the Verint ContactStore for Communication Manager service.
9. Access the Administration pages via https.
10. Check that the certificate matches the information entered.
11. Double click the padlock icon. Internet Explorer should warn you that the certificate is
unsigned. However, it should no longer display a message that indicates the certificate
does not match the web server name.
Tip:
Tip: If you do get a warning that the certificate does not match, check that the
Common Name matches the URL. Double click the padlock, select the
details tab, and click the Subject line. This displays the Common Name.
Generating a Certificate Signing Request

You need a Certificate Signing Request (CSR) as the first step of the signing process.
When you have it, paste it into the Certificate Authority's web page. To generate a CSR:
1. Re-run the keytool command
/javadirectory/bin/keytool -certreq -keystore keystore.jks
-alias tomcat
2. Enter the password - which is Contact5tor3.
3. Copy and paste the output into the CA's web page. (Include the BEGIN and END
lines.)
4. Complete the verification process
5. Reply to the verification emails and other verification steps until you obtain a signed
certificate back from the CA.
Importing the CA's certificates

Before you can import your certificate reply, you need to import the certificate authority's
root certificate and any intermediate certificates between their root and your certificate.
To acquire these certificates:
1. Download these certificates from the certificate authority's website.
2. Save the root as rcert.crt and any intermediate as icert.crt.
If you have more than one intermediate certificate, give them separate filenames.
To import all your certificates:
1. Import the root certificate by running keytool:
/javadirectory/bin/keytool -import -keystore keystore.jks
-alias root -file rcert.crt
2. Enter the password - which is Contact5tor3.
3. Import the intermediate (if required).
-alias inter -file icert.crt
If you have more than one intermediate certificate, import them as inter1, etc.
4. Import your signed certificate.
5. Save the file the CA sent as cert.crt.
Issue 5 May, 2008 305

6. Import with the keytool.

-alias tomcat -file cert.crt
7. Restart the Verint ContactStore for Communication Manager service.
8. Access the administration pages using https.
9. Double click the padlock icon and ensure that Internet Explorer no longer displays a
message that the certificate is unsigned.
Backing up the keystore file

The keystore file now contains:
● the random private key that is unique to this web server
● the signed certificate you just paid for
These two are linked and cannot be regenerated, so it is important to back up the keystore
file. If either one of these components is lost, you must regenerate the certificate and pay
again to get it signed.
Changing Tomcat Port Numbers
Changing Tomcat Port Numbers

You can change the default http and https ports (8080 and 8443 respectively) by editing
/opt/witness/tomcat5525/config/server.xml.
Locate the two Connector elements on roughly lines 20 and 30 and change 8080 to the
chosen plain port number and 8443 to the chosen secure port number.
You must also set the following property in the properties file:
cscm.localport=nnnn
specifying the replacement for 8080.
Note:
Note: If you are using Central Replay Server you must change all the recorders
and the Central Replay Server. The port numbers must be consistent across
all of these servers for upload to the Central Replay Server to work.
If you use Viewer you should alter the URL entered in the Viewer administration
screens to your chosen port number.
Note:
Note: If you change the https port from the default of 8443 and you do not allow
http access, you must also set the following line in the properties file so that
attempts to use http are redirected to the correct https socket.
https.socket=nnnn
The server.xml may be overwritten on subsequent upgrades. You should keep a copy of
the file after editing it so that this can be compared with any changed version and the
appropriate set of merged changes determined after the upgrade.
Issue 5 May, 2008 307

Encrypting Properties File entries

Verint provides a tool to encrypt your passwords so that they can safely be placed in
properties files. You will need this tool if you want to, for example, change the postgresql
password.
Obtain the WitsBSUserCredentials tool from Verint WAS support and install it following the
instructions provided with it.
To encrypt a password for use in the properties file:
1. Launch the tool.
2. Select Other for the application type.
3. Enter a dummy username (it is not used) together with the password to be encrypted.
4. Use the encrypted password in the properties file.
Configuring Viewer and Archive to use HTTPS

To allow Viewer and Archive to connect to CSCM using HTTPS, you must first create and
install an customised certificate for each recorder in your environment. See Installing a
Signed SSL Certificate on page 303. You must obtain certificates with the exact recorder
names, as entered in the Capture Platform setting in Viewer.
Keep the intermediate and root certificates that you received with the signed certificate and
imported into the keystore. You will also need to import these into Viewer and Archive.
Importing the public certificates into Viewer and Archive

The following procedure describes how to import the public certificates into a windows
server. This process must be followed on both the Viewer and the Archive server. This is
necessary so that Viewer and Archive can validate the cscm server.
1. Choose Start > Run, type mmc, and then click OK. The Microsoft Management
Console (MMC) window opens
2. Click File > Add Remove Snap-in.
3. In the Add/Remove Snap-in window, click Add.
4. In the Add Standalone Snap-in window, click Certificates, and then click Add.
5. Choose Computer Account (to manage the certificate store on the local computer),
and then click Next. The Select Computer window opens.
6. Choose Local Computer, and then click Finish.
Now that the snap-in is available you should import any intermediate certificate(s).
1. In the Console window under Certificates, right-click Intermediate Certificate
Authorities, and then select All Tasks > Import. The Certificate Import Wizard window
opens.
2. Click Next. The File to Import window displays.
3. Click Browse
4. Browse for the intermediate certificate file, select the Files of type drop-down list X.509
Certificate (*.cer, *.crt), and select it.
5. Click Next.
6. The Certificate Store dialog box opens.
7. Select Place all certificates in the following store, and then select the Trusted Root
Certificate Authorities store click Next.
8. Check the displayed information, and then click Finish.
Issue 5 May, 2008 309

Finally import the root certificate.

1. In the Console window under Certificates, right-click Intermediate Certificate
Authorities, and then select All Tasks > Import. The Certificate Import Wizard window
opens.
2. Click Next. The File to Import window displays.
3. Click Browse
4. Browse for the root.cer file, select the Files of type drop-down list X.509 Certificate
(*.cer, *.crt), and select it.
5. Click Next.
6. The Certificate Store dialog box opens.
7. Select Place all certificates in the following store, and then select the Trusted Root
Certificate Authorities store click Next.
8. Check the displayed information, and then click Finish.
Configure Viewer to use https and a secure port

By default Viewer will not use https or the secure port to connect to the cscm recorder, both
these features need to be enabled on Viewer. This is done using the Viewer utility
EyrBSConfigurationViewer which is located in the following directory on the Viewer server,
note by default Viewer uses http and once https is enabled will default to use port 50150.
C:\Program Files\ComPlus Applications\{A639A196-338F-48E0-B76B-E223C00B5617}
Configure Viewer to use https

1. Navigate to the EyrBSConfigurationViewer application and right-click in explorer,
select run as…
2. The Run As dialog appears, select The following user option and specify the account
which Viewer is running under.
3. The Configuration Viewer application is launched.
4. Navigate to Central Configuration > Systems >Main Eyretel System > General
5. In the right-hand pane right-click the SSLOnly value and select Modify
6. Change the value from False to True.
7. Select OK.
Configure Viewer to use the capture platforms https port

8. Still using the Configuration Viewer, navigate to Central Configuration > Systems >
CapturePlatformName > General, where CapturePlatformName is the capture platform
name of the cscm recorder as configured in Viewer.
9. In the right-hand pane right-click and select New > Value a new value is created, set
the name to SSLHostPort.
10. Now right-click the new value and select Modify
11. Set the value to the https port in use by the cscm recorder, the recorders default https
port is 8443.
12. Select OK.
13. Now close the Configuration Viewer.
14. Now you should launch Component Services and navigate to Com+ Applications,
select Application Portal, right-click and select shutdown.
15. Viewer should now use https when communicating with the cscm recorder.
Configure Archive to use https

By default Archive will not use https to connect to the cscm recorder, this is done by editing
a file located on the Archive server.
Configure Archive to use https

1. Open the following file in a text editor
C:\Program Files\Verint Systems\conf\SecurityConfig.xml
2. Locate the following xml element in the file <SSLOnly>false</SSLOnly>
3. Change the word false to true and save the file.
4. Now use the Archive Management UI to stop and start the archive service.
5. Archive will now use https to contact to the cscm recorder.
As a result of changing this file the archive system will raise two Major alarms indicating
that the security configuration file has been tampered and that there is an SSL
Configuration Error, both of these can be ignored – examples of these are as follows.
File Tampered, 04/24/2008 8:44 AM, On Archive781, file "C:\Program Files\Verint
Systems\conf\securityconfig.xml" may have been tampered with. Please refer to the
System Administration Guide for details on how to correct a tampered file.
SSL Configuration Error, 04/24/2008 8:46 AM, SSL is configured for Content Server on
host ARCHIVE781, but a required file is missing. The file is C:\Program Files\Verint
Issue 5 May, 2008 311

Systems\conf\security\cert.pem. Please deploy the file or remove SSLConfiguration by

modifying RecorderGeneral.xml ssllistenport.
Note:
Note: Archive will use the capture platforms https port as configured using the
Viewer Configuration tool as described in the previous section, the error
message above is related to a core recorder content server configuration
which is not part of the cscm recorder deployment and hence can be
ignored. The certificates referred to in the alarm are not necessary for cscm
as the public certificates are available via the windows certificate store as
described earlier.
Configuring Archive for a Key Management Server
Configuring Archive for a Key Management Server
Configuring Archive
1. Go to C:\Program Files\Witness Systems\conf\Securityconfig.xml

2. Change the AtRestDataEncryption value to True
3. Go to C:\Program Files\Witness Systems\conf\RSAProvider.cfg
4. Enter the fully qualified domain name of the KMS for kms.address, and change the
kms.port to 7443.
Signing and Delivering the certificate
1. Navigate to the \CAUtility folder on the CA's PC and generate a certificate for the
Archive server by typing the following command line at a DOS prompt: genCert
<ServerName> <CA's Organization Name> where ServerName is the full server name
or just the host name of the server.
You will be prompted for:
● The password of the CA's key file (cakey.pem)
● An export password to protect the output svr_cert_key.p12 file
Successful execution of the above command generates a svr_cert_key.p12 file
containing the signed server certificate, its matching private key, and the CA's
certificate in PKCS#12 format in the \CAUtility\signed_certs\ServerName folder of the
CA's PC. NOTE: The signed server certificate is valid for two years.
2. Deliver the generated file to the C:\Program Files\Witness Systems\conf\Security
folder on the Archive server. To ensure security, the export password should not be
transported along with the PKCS#12 file.
Installing the Certificate

To enable SSL, after successful execution of the installation script, you must first execute
the following procedure. The procedure converts and stores the necessary certificate files
Issue 5 May, 2008 313

for all recorder and JAVA components and must be executed before any other certificate
installation procedures. This procedure is required for encryption of data at rest and/or
encryption of data in transit.
1. After delivering the svr_cert_key.p12 certificate file to the server, navigate to the
C:\Program Files\Witness Systems \conf\security folder and run the following
command at a DOS prompt:
installCerts export_password
where export_password is the entered password when the certificate was signed by
the CA. Running the script does the following:
● Extracts the certificate chain (cert.pem), the server private key (key.pem), and a
single server certificate (rsa_app_cert.pem) from the svr_cert_key.p12 file. The
rsa_app_cert.pem file needs to be loaded into RSA KMS to identify the key
management client (KMC) used by recorder applications on this server if At Rest
data encryption with RSA Key Management needs to be enabled.
● Creates a cacert.pem in the INSTALL_DIR\conf\security\trustedCA directory.
● Adds the CA certificate (cacert.pem) to the trusted CA certificate keystore
(JAVA_HOME\jre\lib\security\cacerts) of the running JRE with Impact360 as the
alias.
● Creates a svr_cert_key_dp.p12 file with the same content as svr_cert_key.p12
except with a new password changeit, which is the default password for the JAVA
keystore.
Note:
Note: Run the above script as often as necessary in case errors occur, such as an
incorrect password.
2. Load the file svr_cert_key.p12 into the Windows Personal certificate store of the
Computer Account of the server with the export password using the Certificate snap-in
of MMC (Start > Run > MMC). See Importing Certificates and Keys on page 65. Make
sure that the option Automatically select the certificate store based on the type of
certificate is selected. This option adds the server certificate to the Personal certificate
store and the CA's certificate to the Trusted Root Certificate Authority certificate store.
You may need to refresh the view to see the certificates.
3. Restart Tomcat if applicable
Glossary
ACD Automatic Call Distribution. This is a feature offered by the Avaya

Communication Manager that queues and distributes incoming calls to
waiting agents. Calls are queued until an agent is available. If multiple agents
are available, calls are distributed on an equitable basis.
AE Services Avaya's Application Enablement Services are APIs to services such as

telecoms, database, and so on. They include DMCC and Avaya CT.
AGENT Logon ID An unique identification number assigned to a particular agent. The agent
uses this number when logging on. The agent ID is not associated with any
particular phoneset
ANI Automatic Number Identification (Service). The provision of calling party

information (typically, telephone number or billing/account number) to the
called party.
Avaya CT Avaya Computer Telephony is the CTI link running on the AE Services
platform that the recorder uses to observe activity on the Communication
Manager and to establish single-step conferences in Conferenced recording
mode. It is also used by eQuality Balance (V5). Avaya CT implements the
TSAPI standard and is often referred to as such.
BHCA Busy Hour Call Attempts. The number of calls that are attempted during the
switches busy hour. Typically slightly higher than, but often used
interchangeably with BHCC
BHCC Busy Hour Call Completions. The number of calls that are completed during
the switches busy hour. Typically slightly lower than, but often used
interchangeably with BHCA
Communication Now known as Device, Media and Call Control (DMCC)

Manager API (CMAPI)
Codec An abbreviation of COder/DECoder. A device or program that converts

signals from one form to another. In this context, between different digital
audio compression standards.
Issue 5 May, 2008 315

CTI
CTI Computer Telephony Integration - typically an interface through which a

computer system can be advised of events occurring within a telephony
system and/or control the telephony system. Avaya CT is an example of
such a link.
Device, Media and Call A software platform (part of AE Services, previously known as
Control (DMCC) Communciation Manager API or CMAPI) that applications such as the
recorder use to create softphones that can participate in calls made on
Avaya Communication Manager-based systems.
DID Direct Inward Dialing. An attribute of a trunk. The CO passes the extension
number of the called party over a DID trunk to the PBX when offering a call to
the PBX. The PBX is then able to automatically route the call to that
extension without requiring operator/attendant assistance. In this way, a
single trunk can terminate calls for many different extensions (but not
simultaneously).
DVD+RW Digital Versatile Disk + Read/Write. One of several competing optical storage
standards, supported by a wide range of manufacturers.
HFS Hierarchical File Storage system. Typically a combination of hard disk and
tape drives plus controlling software that automatically migrates little used
files to cheaper storage media.
IP Internet Protocol. IP specifies the format of packets and the addressing

scheme for internet data. The IP, like the postal system, allows you to
address a package and drop it in the system. The packet may traverse
multiple networks on the way to its ultimate destination.
IVR Interactive Voice Response. A system/facility that plays voice menus to

callers, and acts upon user input (typically, DTMF digits from a touch tone
phone). It is sometimes called VRU (Voice Response Unit).
JTAPI Java Telephony Application Program Interface. A CTI standard which, in

Avaya's case is provided via TSAPI (Avaya CT).
NAS Network Attached Storage. A term used for RAID, tape and other mass
storage systems which have an integral network connection such as
Ethernet or fibre-channel.
NIC Network Interface Card. An expansion board that you insert into a computer
so the computer can connect to a network. Most NICs are designed for a
particular type of network, protocol, and media, although some can serve
multiple networks.
TSAPI
Primary Controller A Verint Contact Recorder that is being shadowed by one or more Standby
Controllers. The Primary will default to "Active" on startup but will ask a
Standby to take over its role if it determines that it cannot record e.g. a disk
fills.
QoS Quality of Service
RAID Redundant Array of Inexpensive Disks. RAID controllers use two or more
hard disk drives together for fault tolerance and enhanced performance.
RAID disk drives are used frequently on servers.
SAN Storage Attached Network. A high-speed network that is typically part of an

overall network of computing resources for an enterprise, in which the
software knows the characteristics of storage devices and the quantity and
value of the data stored in those devices.
Skill Hunt Group A telephone number that is used to route calls to agents on the basis of the
skills needed to handle the call. Agents are assigned to one or more such
skill groups according to their expertise and appropriate calls are routed to
them when they are available.
Slave Recorder A Verint Contact Recorder being controlled by the active Controlling
Recorder. i.e. recording what it is told to record.
Softphone (SOFTware PHONE) In this context, a software-based emulation of an Avaya

VoIP phone. Multiple such emulations run on the Verint Contact Recorder
and each can participate in a telephone call in order to record it.
Standby Controller A Verint Contact Recorder that shadows a Primary Controller and takes over
its role ("goes active") should the primary fail or appear to fail (e.g. the
standby loses contact with it).
Tagging Adding details to the database of call recordings so that recordings can later
be retrieved by searching through the available data fields.
Trunk A communications link between a PBX and the public central office (CO), or
between PBXs, or between COs.
TSAPI Telephony Services Application Program Interface. A CTI interface standard

to which Avaya CT conforms.
Issue 5 May, 2008 317

UTC
UTC Universal Time Coordinated. A time scale that couples Greenwich Mean
Time, which is based solely on the Earth's inconsistent rotation rate, with
highly accurate atomic time. When atomic time and Earth time approach a
one second difference, a leap second is calculated into UTC. UTC, like
Greenwich Mean Time, is set at 0 degrees longitude on the prime meridian.
VDN Vector Directory Number. An extension number used in Avaya's ACD

software to connect calls to a vector for processing. The VDN by itself can be
dialed to access the vector from any extension connected to the switch. (See
also Vector.)
Vector A list of steps that processes calls in a user-defined manner. The steps in a
vector can send calls to splits, play announcements and music, disconnect
calls, give calls a busy signal, or route calls to other destinations. (See also
VDN.)
VoIP Voice over Internet Protocol. A means of transmitting telephone calls over
the packet-switched IP network - as distinct from TDM.
VPN Virtual Private Network. Private, or restricted, communications networks

which use encryption and other security measures to transmit information
through a public network such as the Internet and avoid unauthorized use.
VRU Voice Response Unit. A device that plays voice menus to a caller and
responds to caller instructions entered on a touch tone phone. Also known as
Interactive Voice Response (IVR).

Acr Avaya PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Acr Avaya PDF

Uploaded by

Copyright:

Available Formats

Verint ContactStore for

Planning, Installation and Administration Guide

About This Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Issue 5 May, 2008 3

Issue 5 May, 2008 5

Search and Replay . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 136

Search and Display UDFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

Issue 5 May, 2008 7

This Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234

Issue 5 May, 2008 9

The Verint ContactStore for Communication Manager Planning, Installation and

Issue 5 May, 2008 11

Summary of information included in this guide

Chapter Title Description

Chapter 1: This chapter provides an overview of the design options

Chapter 2: This chapter gives details of the prerequisites for Verint

Chapter 3: This chapter gives details of the steps to install Verint

Chapter 4: This chapter gives details of the steps to configure Verint

Chapter 5: This chapter provides detais of regular maintenance

Chapter 6: This chapter sets out the available and recommended

Chapter 7: This chapter details the rarely used advanced settings of

Appendix A: This appendix provides technical details about Verint

Glossary The glossary defines the terms you need to understand

Conventions used in this guide

To show... This style is used For example...

Issue 5 May, 2008 13

To show... This style is used For example...

Situations that can result

Issue 5 May, 2008 15

What’s New in Version 7.8

Issue 5 May, 2008 17

Verint ContactStore for Communication Manager Server

Optional Server Applications

Issue 5 May, 2008 19

Additional Verint Systems applications

Issue 5 May, 2008 21

● eQuality Now (agent training tools)

Bulk Search and Replay

Issue 5 May, 2008 23

Central Search and Replay

External Search and Replay

Sampled Recording for Quality Assessment

Quality Recording Channels (eQuality Balance V5 only)

Quality Replay Channels

Issue 5 May, 2008 25

Bulk Recording Modes

Station Bulk Recording

Externally Controlled Recording

Bulk plus Quality

Ad-hoc or Occasional Recording Modes

Issue 5 May, 2008 27

Station Executive Recording

Issue 5 May, 2008 29

Live Monitor Ports

Recording Mode Uses

Issue 5 May, 2008 31

Limited Number of Service Observers per Station

Limited Number of Service Observers per Call

Assign Ports to Stations not Agents/VDNs

Bulk Recording requires dedicated port

Issue 5 May, 2008 33

Personal Station Access (PSA)

Transfer via Coverage