Professional Documents
Culture Documents
HCP S10 Node v2 2 0 Maintenance FE-HCPS002-17
HCP S10 Node v2 2 0 Maintenance FE-HCPS002-17
2.2.0
FE-HCPS002-17
July 2021
© 2014, 2021 Hitachi Vantara LLC. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including copying and recording, or
stored in a database or retrieval system for commercial purposes without the express written permission of Hitachi, Ltd., or Hitachi Vantara LLC
(collectively, “Hitachi”). Licensee may make copies of the Materials, provided that any such copy is: (i) created as an essential step in utilization of the
Software as licensed and is used in no other manner; or (ii) used for archival purposes. Licensee may not make any other copies of the Materials.
"Materials" mean text, data, photographs, graphics, audio, video, and documents.
Hitachi reserves the right to make changes to this Material at any time without notice and assumes no responsibility for its use. The Materials contain the
most current information available at the time of publication.
Some of the features described in the Materials might not be currently available. Refer to the most recent product announcement for information about
feature and product availability, or contact Hitachi Vantara LLC at https://support.hitachivantara.com/en_us/contact-us.html.
Notice: Hitachi products and services can be ordered only under the terms and conditions of the applicable Hitachi agreements. The use of Hitachi
products is governed by the terms of your agreements with Hitachi Vantara LLC.
By using this software, you agree that you are responsible for:
1. Acquiring the relevant consents as may be required under local privacy laws or otherwise from authorized employees and other individuals; and
2. Verifying that data continues to be held, retrieved, deleted, or otherwise processed in accordance with relevant laws.
Notice on Export Controls. The technical data and technology inherent in this Document may be subject to U.S. export control laws, including the U.S.
Export Administration Act and its associated regulations, and may be subject to export or import regulations in other countries. Reader agrees to comply
strictly with all such regulations and acknowledges that Reader has the responsibility to obtain licenses to export, re-export, or import the Document and
any Compliant Products.
Hitachi and Lumada are trademarks or registered trademarks of Hitachi, Ltd., in the United States and other countries.
AIX, AS/400e, DB2, Domino, DS6000, DS8000, Enterprise Storage Server, eServer, FICON, FlashCopy, GDPS, HyperSwap, IBM, Lotus, MVS, OS/390,
PowerHA, PowerPC, RS6000, S/390, System z9, System z10, Tivoli, z/OS, z9, z10, z13, z14, z/VM, and z/VSE are registered trademarks or trademarks
of International Business Machines Corporation.
Active Directory, ActiveX, Bing, Excel, Hyper-V, Internet Explorer, the Internet Explorer logo, Microsoft, the Microsoft Corporate Logo, MS-DOS, Outlook,
PowerPoint, SharePoint, Silverlight, SmartScreen, SQL Server, Visual Basic, Visual C++, Visual Studio, Windows, the Windows logo, Windows Azure,
Windows PowerShell, Windows Server, the Windows start button, and Windows Vista are registered trademarks or trademarks of Microsoft Corporation.
Microsoft product screen shots are reprinted with permission from Microsoft Corporation.
All other trademarks, service marks, and company names in this document or web site are properties of their respective owners.
Copyright and license information for third-party and open source software used in Hitachi Vantara products can be found at
https://www.hitachivantara.com/en-us/company/legal.html.
Preface 9
Intended audience 9
Product version 9
Release notes 9
Syntax notation 9
Related documents 10
Accessing product documentation 11
Getting help 11
Comments 11
Change list 12
Chapter 1: HCP S10 Node overview 14
HCP S Series Nodes 14
S10 Node hardware components 15
S10 Node product offerings 16
S10 Node rack options 16
S10 Node power distribution options 17
S10 Node connectivity options 18
Intended audience
This book is intended for people who are responsible for maintaining S10 Nodes that are
deployed at customer sites. The book assumes that you have experience working with both
the hardware and software of computer systems.
Product version
This book applies to release 2.2.0 or later of the HCP S Series Node.
Release notes
Read the release notes before installing and using this product. They may contain
requirements or restrictions that are not fully described in this document or updates or
corrections to this document. Release notes are available on Hitachi Vantara Support
Connect:
https://knowledge.hitachivantara.com/Documents
Syntax notation
The table below describes the conventions used for the syntax of commands in this book.
Preface
HCP S10 Node Maintenance 9
Related documents
(Continued)
Notation Meaning Example
Related documents
The following documents contain additional information about HCP S Series Nodes:
• HCP S Series Node Help (MK-HCPS000) — This Help system contains information
about configuring, monitoring, and managing an HCP S10 or S30 Node.The Help
includes information you need to effectively use the HCP S Series Management
Console. The Help also describes the physical specifications of and environmental
requirements for S10 and S30 Nodes. Additionally, the Help contains a complete
reference for the HCP S Series management API.
• HCP S10 and S30 Node API Reference (MK-HCPS004) — This book contains all the
information you need for using the HCP S Series management API with an HCP S10 or
S30 Node. This RESTful HTTP-based API enables you to configure, monitor, and
manage an S10 or S30 Node programmatically. The book explains how to use the
management API to retrieve information about and manipulate S10 and S30 Node
resources. The book also includes an introduction to the S Series Node concepts that
underlie the management API resources.
• HCP S30 Node Maintenance (FE-HCPS010) — This book contains instructions for
maintaining the physical components of an HCP S30 Node and for upgrading or
reinstalling the HCP S Series operating system and software. The book includes
instructions for adding drives and enclosures to increase capacity and for replacing faulty
components. The book also contains information about monitoring S30 Nodes.
Additionally, the book describes the physical specifications of and environmental
requirements for S30 Nodes.
• Cabling Enclosures in an HCP S30 Node (FE-HCPS012) — This Help system contains
diagrams showing the SAS cable connections for each possible enclosure in an HCP
S30 Node. The Help steps through the diagrams in the order in which the enclosures
should be cabled.
Preface
HCP S10 Node Maintenance 10
Accessing product documentation
components to a condition in which they can be used either in new S Series Nodes or as
replacement parts for existing S Series Nodes.
Getting help
Hitachi Vantara Support Portal is the destination for technical support of products and
solutions sold by Hitachi Vantara. To contact technical support, log on to Hitachi Vantara
Support Connect for contact information: https://support.hitachivantara.com/en_us/contact-
us.html.
Hitachi Vantara Community is a global online community for Hitachi Vantara customers,
partners, independent software vendors, employees, and prospects. It is the destination to
get answers, discover insights, and make connections. Join the conversation today! Go to
community.hitachivantara.com/s, register, and complete your profile.
Note: If the customer purchased the HCP S Series Node from a third party, please contact
the applicable HCP support center.
Comments
Please send us your comments on this document to doc.comments@hitachivantara.com.
Include the document title and part number, including the revision (for example, -01), and
refer to specific sections and paragraphs whenever possible. All comments become the
property of Hitachi Vantara LLC.
Thank you!
Preface
HCP S10 Node Maintenance 11
Change list
The table below outlines the significant differences between this version of HCP S10 Node
Maintenance and the previous version.
Chapter Differences
Chapter 1: HCP S10 Node overview Adds the Hitachi Universal V2B rack as a racking option
Chapter 3: Preparing for HCP S10 Node • Updates PuTTY to version 0.73
maintenance
• Updates Rawrite32 to version 1.0.6.0
Chapter 4: Configuring an HCP S Series New chapter: explains how to configure an S Series Node in DNS
Node in DNS
Chapter 5: Monitoring an HCP S10 Node Adds information about configuring an S Series Node monitoring
with the Hitachi Remote Ops monitor agent
Chapter 6: Maintaining the HCP S Series • Explains that upgrading an S Series Node to release 2.2.0
operating system and software entails approximately 20 minutes of downtime
Change list
HCP S10 Node Maintenance 12
(Continued)
Chapter Differences
Chapter 7: Maintaining data and Explains that the S Series Node can continue to process drives for
database drives reuse after an add drives or replace drives operation is complete
Chapter 9: Maintaining server modules • Updates the lsscsi command used to identify OS SSDs
Appendix: HCP S Series Node New appendix: contains a form for recording the information needed
Configuration Information form for performing a custom reinstallation of the HCP S Series OS and
software, configuring the S Series Node for the customer
environment, or recovering a server module
Change list
HCP S10 Node Maintenance 13
Chapter 1: HCP S10 Node overview
The Hitachi Content Platform (HCP) S10 Node, which is a type of HCP S Series
Node, is one of the storage products offered by Hitachi Vantara. This chapter contains a
general description of S Series Nodes and provides information about these aspects of the
S10 Node:
• Hardware components
• Product offerings
• Rack options
• Connectivity options
Important: HCP and HCP for cloud scale are the only supported clients for the S Series
Node.
HCP systems and HCP for cloud scale systems use the S Series Node implementation of
the Hitachi API for Amazon S3® to write, retrieve, and otherwise manage objects in an S
Series Node. This RESTful, HTTP-based API is compatible with Amazon S3.
• NDS-4600 enclosure — The enclosure is a container for the power and cooling
modules, server modules, and hard disk drives associated with an S10 Node.
• Rail kit — The rail kit, with left and right extension brackets, is used to mount the
enclosure in a standard nineteen-inch rack.
• Two cable management arms — The cable management arms, with left and right
extension brackets, are used to neatly arrange the power and network cables for the S10
Node and to keep the cables securely connected to the enclosure. These cables connect
the server modules and power and cooling modules in the enclosure to the power
distribution units (PDUs) and customer network.
• Front panel and light pipe assembly — The front panel and light pipe assembly
attach to the front of the enclosure.
• Two power and cooling modules — The power and cooling modules, installed in the
rear of the enclosure, provide power and cooling for the enclosure and its components.
• Two server modules — The active-active server modules, installed in the rear of the
enclosure, run the software that manages the S10 Node, provide data access, and
ensure data protection.
In the HCP S Series Management Console, in some cases, server module 1 is
represented as IOM A, and server module 2 is represented as IOM B. This convention
reflects the low-level hardware implementation.
• SATA or SAS hard disk drives (HDDs) for storing data — These SATA or SAS
drives, with drive carriers, store the data written to the S10 Node. A new S10 Node has
either 28 or 56 SATA or SAS data drives.
• Four SAS hard disk drives for storing the internal database — These SAS drives,
with drive carriers, store the internal database used by the S10 Node to hold information
such as user account and bucket definitions and various configuration settings. Like the
SATA or SAS data drives, the SAS database drives can also store data written to the
S10 Node.
• One-foot CAT6 Ethernet cable — The purple one-foot CAT6 Ethernet cable, which is
called the server interconnect cable, is used to connect the two server modules to each
other for synchronization and resource sharing purposes.
• Two IEC 320 C19 to IEC 320 C20 power cables — The two 2-meter C19-to-C20
power cables are used to connect the power and cooling modules in the enclosure to two
different PDUs.
An S10 Node comes with a 4GB USB flash drive and a USB-to-mini-USB cable for
authorized service personnel to use at the customer site for maintenance procedures.
Important: Be sure to keep the USB flash drive in a safe place in case the drive is needed
for a service procedure.
• Half-populated with twenty-eight 6TB SATA or SAS data HDDs and four 6TB SAS
database HDDs
• Fully populated with fifty-six 6TB SATA or SAS data HDDs and four 6TB SAS database
HDDs
• Half-populated with twenty-eight 10TB SAS data HDDs and four 10TB SAS database
HDDs
• Fully populated with fifty-six 10TB SAS data HDDs and four 10TB SAS database HDDs
The table below shows the total storage capacity for the possible configurations of a new
S10 Node.
Note: Due to internal storage requirements, the capacity available for storing data is less
than the total capacity provided by the SATA and SAS HDDs.
To ensure the successful installation of an S10 Node, a customer-supplied rack must meet
the requirements specified in "Customer-supplied racks" on page 22.
Hitachi Vantara
replacement part C13 C19
number Geography Phase Description Outlets Outlets
The table below describes the Hitachi Vantara-supplied horizontal PDUs used with older S10
Nodes.
Hitachi Vantara
replacement C13 C19
part number Geography Phase Description outlets outlets
Under certain circumstances, PDUs not listed in the tables above may be supported. If you
have questions about specific PDUs, check with your HCP support center.
• For the access network, each server module has two bonded 10Gb Ethernet SFP+
ports. These ports can be configured as active-active (802.3ad) or active-backup. To
create a highly available infrastructure, the ports can be connected to different physical
switches.
• For the management network, each server module has one 1Gb Ethernet 1000Base-T
port.
Both the access network and the management network can be used for management
functions. Only the access network can be used for data access.
To meet customer network connectivity requirements for the S10 Node access network,
these additional components are available from Hitachi Vantara:
• 1Gb SFP+ module — If a 1Gb Ethernet network, 1Gb SFP+ modules are required for
the access network connections. This module has an RJ45 connector. Hitachi Vantara
offers 1Gb SFP+ modules from both Brocade and Cisco.
• 10Gb SFP+ optical transceiver — If using fiber optic cables for the access network
connections, a 10Gb SFP+ optical transceiver is required on both ends of each fiber
optic cable. The 10Gb SFP+ optical transceiver has an LC type connector. Hitachi
Vantara offers a 10Gb SFP+ optical transceiver from Intel®.
• Multimode fiber optic cable — If a 10Gb Ethernet network and the distance between
the enclosure and the Ethernet switches is greater than 10 meters, fiber optic cables
must be used for the access network connections. Hitachi Vantara offers multimode
fiber optic cables in various lengths in both plenum and LSZH jacketing.
• 10Gb Twinax cable — If a 10Gb Ethernet network and the distance between the
enclosure and the Ethernet switches is less than 10 meters, Twinax cables can be used
instead of fiber optic cables for the access network connections. Hitachi Vantara offers
10Gb Twinax cables in various lengths from both Brocade and Cisco.
Mechanical details
The following sections describe the mechanical specifications and requirements for an S10
Node.
Dimensions
The labels in the figure below identify the faces of an S10 Node enclosure. Use this figure as
a reference for the table of dimensions that follows.
The Hitachi Universal V2B and V2 racks have the same physical dimensions as each other.
The table below shows these dimensions.
Parameter Inches mm
Weights
The table below shows the weights of various configurations of an S10 Node. Each weight
includes the enclosure, all the components inside the enclosure, the inner and outer
enclosure rails, and the enclosure power cables.
The weights shown do not include the PDUs or the rack required for mounting an S10 Node.
The weights of these components are shown at the end of this section.
S10 Node fully populated with thirty-two 4TB drives 219.12 99.41
and twenty-eight 6TB drives
(Continued)
S10 Node configuration Weight in lbs. Weight in kg
S10 Node fully populated with thirty-two 6TB drives 222.36 100.88
and twenty-eight 10TB drives
Caution: Due to the weight of an enclosure, you should always use a lift when installing or
uninstalling an enclosure in a rack.
The table below shows the weight of an empty Hitachi Universal V2B rack, an empty Hitachi
Universal V2 rack, a Hitachi Universal V2 PDU, and a Hitachi Vantara-supplied horizontal
PDU.
Cables
Each power cable and Ethernet cable (except the server interconnect cable) for an S10 Node
must be long enough such that 36 inches (914 mm) can be allocated to the portion of the
cable that goes from the point at which the cable is secured to the rack, through the
applicable cable management arm, to the point at which the cable connects to the back of the
enclosure. This ensures that the enclosure can be pulled partially out of the rack for servicing
without the need to disconnect the cables.
Note: The cable management arms are a required component of an S10 Node.
Customer-supplied racks
Customers can mount an S10 Node in a rack that they supply. A customer-supplied rack
must meet these requirements:
• The rack must have square holes in the vertical EIA rails.
• The rack depth must be at least 43.3 inches (1,100 mm). The use of cable management
arms, which are required with the enclosure, forces this depth requirement.
• To support the enclosure rail kit, the distance between the front and rear vertical EIA rails
in the rack must be a minimum of 20 inches (508 mm) and a maximum of 30.5 inches
(775 mm). For maximum stability, this distance should not be less than 24 inches (610
mm).
If the distance is less than or equal to 28.5 inches (724 mm), the rails can be used
without rail extenders and cable management arm extenders. If the distance is greater
than 28.5 inches, rail extenders and cable management arm extenders are required.
• The distance between the left and right vertical EIA rails must be a minimum of 18.5
inches (470 mm) wide for the entire depth of the rack, as shown below.
The figure below shows the dimensions and clearances required for mounting an S10 Node in
a customer-supplied rack. In the figure, the rails are used without the two-inch rail extenders
and two-inch cable management arm extenders, which allow the distance between the front
and rear vertical EIA rails to be two-inches greater than what the diagram shows.
Electrical details
The following sections describe the power requirements and electrical specifications for an
S10 Node.
Power system
For redundancy, an S10 Node has two power and cooling modules power supplies. These
modules should be connected to two different PDUs, which should be plugged into two
separate power sources. This setup ensures that the entire power system has no single point
of failure.
If only one power source is available, the two PDUs should be plugged into different circuits.
If only one circuit is available, the two power and cooling modules can be connected to the
same PDU as a last resort, assuming the PDU has two available outlets and enough power
capacity.
The power system input can be either single-phase or three-phase with single phase on the
outlets.
The table below shows the power system requirements for the Hitachi Universal V2 PDU.
The table below shows the power system requirements for the Hitachi Vantara-supplied
horizontal PDU.
Hitachi Vantara
replacement
part number Geography Phase Voltage Amperage
Power connections
The power connections required by the Hitachi Universal V2 PDUs differ by geography and
input phase:
¡ The three-phase, 208V, 30A PDU has a NEMA L15-30 four-wire plug, as shown
below.
• For EMEA/APAC:
¡ The single-phase, 230V, 32A PDU has an IEC 309 three-wire plug, as shown below.
¡ The three-phase, 400V, 32A PDU has an IEC309 five-wire power plug, as shown
below.
Electrical specifications
When an S10 Node is operating normally, each power and cooling module supplies half the
power that the S10 Node requires. However, if one power and cooling module is unavailable,
the S10 Node can run on the other power and cooling module alone.
The table below outlines the electrical specifications for a power and cooling module in an
S10 Node. The power system at the customer site should be sized appropriately for the
number of S10 Nodes installed at the site.
Parameter Value
(Continued)
Parameter Value
Immunity EN55024
EN61000-4-2 Electrostatic discharge
EN61000-4-3 Radiated electromagnetic field
EN61000-4-4 Burst, electrical fast transients
EN61000-4-5 Surge transients
EN61000-4-6 Conducted disturbances
EN61000-4-11 Power line interruption
Environmental details
The following sections describe the environmental specifications and requirements for an
S10 Node.
RoHS compliance
An HCP S10 Node, including all its components, is compliant with the European Union
Restriction of Hazardous Substances (RoHS) Directive (Directive 2002/95/EC), with no
exceptions or exemptions.
BNST compliance
Benzenamine, N-phenyl-, reaction products with styrene and 2,4,4-trimethylpentene (BNST)
is an antioxidant used as an additive in many industrial lubricants. Its use has been restricted
in Canada under the Prohibition of Certain Toxic Substances Regulations, 2012, which went
into effect on March 14, 2013. The Regulations include a two-year exemption for BNST used
in small-application lubricants in the electrical and electronics engineering industry.
In an HCP S10 Node, the components that use small-application lubricants are the fans in
the power and cooling modules, the fans in the server modules, and the hard disk drives. If
such a component is labeled with a manufacturing date of March 14, 2015 or later, the
component does not contain BNST and is in compliance with the Regulations. If such a
component is labeled with a manufacturing date earlier than March 14, 2015, the component
is not subject to the Regulations.
• Normal operation
• In transit
• In storage
*The maximum operating temperature value is specified at sea level and is derated 2.0% per 1,000 feet of
increased altitude.
The figure below shows the airflow from the perspective of the top of the enclosure.
When all four fans are at full speed, the total airflow is approximately 200 cubic feet per
minute.
When an S10 Node is powered on, the fans run at full speed for a short time to ensure that
they are fully operational. After that, under normal conditions, the fans run at half speed. The
cooling they provide is sufficient to prevent the hard disk drives and other components from
exceeding the manufacturer’s rated specifications throughout the range of the S10 Node
operating conditions.
If a single fan fails, the S10 Node continues to operate. However, this fault condition forces
the remaining fans to run at full speed until the condition is corrected.
The normal heat dissipation rate for a fully populated S10 Node enclosure is 212 BTU/hour.
The maximum heat dissipation rate is 307 BTU/hour.
Acoustics
At normal operation tested to ISO 7779, the acoustic level of a fully populated S10 Node
enclosure does not exceed an A-weighted sound power level of 77 decibels. Normal
operation is defined as an ambient temperature of 23°C, 60 I/Os per second per hard disk
drive, and all fans operational with no fault conditions. Under fault conditions, the acoustic
sound level can increase to 84.3 decibels.
Additionally, the S10 Node enclosure has an audible alarm that sounds under some fault
conditions and beeps three times when the enclosure cover is closed. This alarm operates at
a frequency of 2300 Hz, plus or minus 300 Hz, with a sound power level of 80 decibels at 10
cm from the sound source.
• Virtual COM port (VCP) serial port driver — Used for communication between the
laptop computer and an S10 Node server module over a serial connection.
• VCP serial port driver uninstaller — Used to uninstall the VCP serial port driver. This
uninstaller also deletes the COM ports associated with the VCP driver.
• PuTTY — Used for access to the server modules either over a serial connection or with
SSH over an Ethernet connection.
• Rawrite32 — Used to create bootable USB flash drives or DVDs that contain the HCP
S Series operating system and software installation ISO.
• Rufus — Used to reformat bootable USB flash drives so that they can be reused for
different versions of the HCP S Series operating system and software ISO.
• HCP S Series Node post-install script — Used to verify the numbers of enclosures
and drives in the S Series Node, set the S Series Node serial number, disable data-in-
flight encryption (if applicable), and disable SSH access to the S11 or S31 Node server
modules.
• CDM v2.12.00 WHQL Certified.exe — Installation file for the VCP serial port driver.
• S2B_3A19-orig.bin — Image used to flash the server module BIOS in an S10 or S30
Node.
For instructions on obtaining this file, see "HCP S Series Node tools file" on page 31.
2. In the window that opens, click Extract.
3. If the User Account Control window appears, click Yes in the window to allow the
installation program to make changes to your computer.
4. In the Device Driver Installation Wizard window, click Next.
The installation program installs the serial port driver files.
5. Click Finish.
If you are providing service on behalf of an authorized S Series Node distributor, you get the
SSH keys from the HCP S Series Node Product Management team.
You can store the SSH keys in the location of your choice on your laptop computer.
SSH keys differ by the distributor from which the customer got the S10 Node. Ensure that
you have the correct key before you use PuTTY to set up an SSH connection to an S10 Node
server module.
Note: Be sure to install PuTTY in the default location. If you don't, the hcps_post_
install.bat script will fail.
5. On the Select Start Menu Folder page, optionally select a different Start menu folder in
which to place the PuTTY and Pageant shortcuts. Then click Next.
6. On the Select Additional Tasks page:
¡ Optionally, select the option to create a desktop icon for PuTTY. This option also
creates a desktop icon for Pageant.
¡ Optionally, select the option to create a Quick Launch icon for PuTTY. This option
also creates a Quick Launch icon for Pageant.
¡ Leave the option to associate .PPK files selected.
Then click Next.
7. On the Ready to Install page, click Install.
8. In the User Account Control window, take one of these actions:
¡ If you logged in to the laptop computer with an admin user account, click Yes.
¡ If you did not log in to the laptop computer with an admin user account:
1. In the applicable fields, type the username and password for an admin user
account.
2. Click Yes.
The installation program installs PuTTY and Pageant.
9. On the Completing the PuTTY Setup Wizard page, optionally deselect the View
README.txt option. Then click Finish.
Note: SSH keys are not retained in your keychain when you close Pageant. If you restart
Pageant after closing it, you need to add the applicable SSH key to your keychain again
before you can use SSH to access an S10 Node server module.
Tip: The easiest way to open Device Manager is to enter Device Manager in the
Windows Start menu search field and then select Device Manager from the list of
matching items.
b. Set the fields for the serial connection to the values shown in the table below.
Field Value
Data bits 8
Stop bits 1
Parity None
Caution: Do not leave the USB-to-mini-USB cable connected to your laptop computer after
you've finished using the serial connection. Leaving the cable connected may cause the S
Series Node to behave in unexpected ways.
and the server module. For network access, you should use the S10 Node management
network to access the server module. However, if the management network is not
connected, you can use the S10 Node access network.
For both network access and a direct Ethernet connection, you need a CAT6 Ethernet cable:
• For network access, this cable connects the Ethernet port on the laptop computer to a
port on the customer network that has connectivity to the S10 Node access or
management network. Depending on the network configuration, either the laptop
computer is automatically assigned an IP address in the customer network, or you need
to temporarily reconfigure the laptop computer to be on the customer network.
• For a direct Ethernet connection, this cable connects the Ethernet port on the laptop
computer to the management port on the server module. In this case, you need to
temporarily reconfigure the laptop computer to be on the same subnet as the S10 Node
management network.
To reconfigure your laptop computer for SSH access to an S10 Node server module:
1. Connect the CAT6 Ethernet cable to the Ethernet port on the laptop computer and to a
port on the customer network or to the management port on the server module, as
applicable.
The management port is the third Ethernet port from the left on the back of the server
module.
2. On the laptop computer, open the Windows Control Panel. Then take one of these
actions:
¡ If the Control Panel shows categories, select Network and Internet ► Network
and Sharing Center ► Change adapter settings.
¡ If the Control Panel shows large or small icons, select Network and Sharing
Center ► Change adapter settings.
3. On the Network Connections page, double-click the local area connection for the port
to which you connected the CAT6 Ethernet cable. The connection you want shows the
model of the network controller in the laptop computer (for example, Intel(R) 82579LM).
4. In the Local Area Connection Status window, click Properties.
5. On the Networking page in the Local Area Connection Properties window, select
Internet Protocol Version 4 (TCP/IPv4). Then click Properties.
6. Make a note of the settings on the General page in the Internet Protocol Version 4
(TCP/IPv4) Properties window.
7. On the General page, select Use the following IP address. Then set the fields in that
section to the values shown in the table below.
Field Value
Important: When you no longer need SSH access to the server module, change the network
connection properties on your laptop computer back to their original settings, as noted in step
6.
To use PuTTY for SSH access to an S10 Node server module over an Ethernet connection:
1. If you have not already done so, use Pageant to add the applicable S Series Node
SSH key to the SSH keychain on your laptop computer, as described in "Adding an S
Series Node SSH key to your SSH keychain" on page 35.
2. Reconfigure your laptop computer, as described in "Reconfiguring your laptop computer
for SSH access" on the previous page.
3. Start PuTTY.
4. In the PuTTY Configuration window, configure the PuTTY session:
a. In the Category field, select Session (if it's not the current category).
Note: If a message is displayed before the PuTTY window opens, click OK or Yes to
close the message window.
7. In response to the login prompt in the PuTTY window, enter the applicable username for
the procedure you want to perform.
Tip: If you're prompted for a password at this point, your SSH keychain does not include
the necessary SSH key.
Burning the HCP S Series installation ISO image to a USB flash drive or
DVD
You use Rawrite32 to burn the HCP S Series operating system and software installation ISO
image to a USB flash drive or DVD.
Important: Burning an ISO image to a USB flash drive or DVD overwrites any existing data
on that drive. Before burning the ISO image, ensure that the USB flash drive or DVD you're
using is the one you want to use.
To run Rawrite32, you need a Windows user account that has admin privileges. You can
either log in to your laptop computer with this account before you start Rawrite32 or enter the
credentials for the account when prompted after you start Rawrite32.
To burn the HCP S Series operating system and software installation ISO image to a USB
flash drive or DVD:
1. If you are burning the ISO image to a DVD, connect the external DVDdrive to a USB port
on your laptop computer.
2. Optionally, log in to your laptop computer with a Windows admin user account.
3. Take one of these actions, as applicable:
¡ Insert the USB flash drive into a USB port on your laptop computer.
¡ Insert the DVD into the external DVD drive that's connected to your laptop
computer.
4. If you are prompted to format the drive, click Cancel.
5. Start Rawrite32.
6. In the User Account Control window, take one of these actions:
¡ If you logged in to the laptop computer with an admin user account, click Yes.
¡ If you did not log in to the laptop computer with an admin user account:
1. In the applicable fields, type the username and password for an admin user
account.
2. Click Yes.
7. In the NetBSD Disk Image Tool window, in the Target field, select the letter for the
drive into which you inserted the USB flash drive or DVD.
Important: Rawrite32 can overwrite any removable storage device connected to the
laptop computer. Be sure to select the correct drive.
13. In the Windows system tray, right-click the Safely Remove Hardware and Eject
Media icon ( ). Then, in the popup menu, select the USB flash drive or DVD, as
applicable.
14. Take one of these actions, as applicable:
¡ If you burned the ISO image to a USB flash drive, remove the USB flash drive from
the laptop computer.
¡ If you burned the ISO image to a DVD:
1. Remove the DVD from the external DVD drive.
2. Disconnect the external DVD drive from the laptop computer.
Important: Do not use Rufus to burn the installation ISO image or Hardware Setup Tool ISO
image to a USB flash drive. Rufus cannot correctly handle these image files.
To run Rufus, you need a Windows user account that has admin privileges. You can either
log in to your laptop computer with this account before you start Rufus or enter the
credentials for the account when prompted after you start Rufus.
For instructions on getting Rufus, see "HCP S Series Node tools file" on page 31.
To reformat a bootable USB flash drive:
1. Optionally, log in to your laptop computer with a Windows admin user account.
2. Insert the USB flash drive you want to reformat into a USB port on the laptop computer.
3. Run the rufus-3.5.exe file.
10. Remove the USB flash drive from the laptop computer.
• Contains instructions for configuring forward lookup zones for an S Series Node in both
Windows and Unix
• Create a forward lookup zone for the S Series Node and add host entries to that zone. In
this case, the name of the zone looks something like s-node-1.example.com.
• Add host entries for the S Series Node to an existing forward lookup zone. In this case,
the name of the zone looks something like example.com.
In either case, the zone must be configured as a primary zone.
Host entries
Each host entry for an S Series Node associates the IP address of a server module on one of
the S Series Node networks with one of the three S Series Node interfaces or with a wildcard
(*) that represents any of the interfaces.
The hostnames that correspond to the S Series Node interfaces are:
• For the Hitachi API for Amazon S3 (the S3 compatible API): hs3
If the zone for the S Series Node has entries for both the wildcard hostname and one or more
of the specific interface hostnames:
• The interface-specific host entries are used for access to the applicable interfaces.
• The wildcard host entry is used for access to the interfaces for which no interface-
specific entry exists.
Normally, a wildcard host is associated with the virtual IP addresses for the server modules
on the access network. If that network is using IPv6, the host can be associated with only
the primary virtual IPv6 addresses or with both the primary and secondary virtual IPv6
addresses.
Note: HCP and S Series Nodes always communicate over the access network. If the
access network is disabled for the management API, HCP systems cannot use the S Series
Node.
9. On the Dynamic Update page, select Do not allow dynamic updates. Then click
Next.
10. On the Completing the New Zone Wizard page, click Finish.
To add host entries to a forward lookup zone:
1. In the left side of the DNS Manager window, select the applicable forward lookup zone.
Then right-click the zone and select New Host (A or AAAA).
2. In the New Host window:
¡ In the Name field, type the hostname for the entry. For example:
– If you're using a forward lookup zone created specifically for the S Series Node,
type * or admin.
– If you're using a different forward lookup zone, type *.s-node-1 or admin.s-node-
1.
¡ In the IP address field, type the IP address for the entry.
¡ If you want to add the IP address to a reverse lookup zone that already exists, select
Create associated pointer (PTR) record.
Then click Add Host.
3. In response to the confirming message, click OK.
If you selected Create associated pointer (PTR) record and the reverse lookup zone
does not exist, the host entry is added to the forward lookup zone but not to a reverse
lookup zone.
4. Take either of these actions:
¡ To add another host entry, repeat steps 2 and 3.
¡ If you're done adding host entries, click Done.
• A domain name:
¡ If the zone is specifically for an S Series Node, the domain name looks something
like s-node-1.example.com.
¡ If you're using an existing domain, the domain name looks something like
example.com.
• The name of the file containing the A records for the zone. The A records specify the
host entries for the zone.
Each A record for an S Series Node associates an S Series Node interface or the
wildcard (*) with the IP address of one of the S Series Node server modules on one of the
S Series Node networks.
Each fully qualified domain name that can be used for access to an S Series Node is the
concatenation of the hostname in an A record with the specified domain name. For
example:
¡ If the domain name is s-node-1.example.com, the hostname for the Management
Console is admin by itself.
¡ If the domain name is example.com, the hostname for the Management Console is
admin.s-node-1.
admin IN A 10.0.0.3
admin IN A 10.0.0.4
mapi IN A 10.0.0.3
mapi IN A 10.0.0.4
hs3 IN A 10.0.0.3
hs3 IN A 10.0.0.4
In the sample file above, the admin, mapi, and hs3 hosts are all associated with the virtual
IP addresses of the server modules on the access network.
If the forward lookup zone is not specifically for the S Series Node, the A records in the file
look something like this:
admin.s-node-1 IN A 10.0.0.3
admin.s-node-1 IN A 10.0.0.4
mapi.s-node-1 IN A 10.0.0.3
mapi.s-node-1 IN A 10.0.0.4
hs3.s-node-1 IN A 10.0.0.3
hs3.s-node-1 IN A 10.0.0.4
Name: admin.s-node-1.example.com
Addresses: 10.0.0.3
10.0.0.4
If you don't see the expected IP addresses, the zone is not defined correctly.
• You can configure the S Series Node to send event log messages, log messages for
data access requests, and log messages for management API requests to one or more
syslog servers.
• Through the management API, you can get information about the current load on S
Series Node resources.
• Many hardware components have LEDs that light up to indicate certain conditions.
• Some LEDs serve as beacons. Beacons enable you to easily identify components in
your data center.
You can use the HCP S Series Management Console or management API to turn
beaconing on or off for individual components.
• The S Series Node internal logs contain detailed records of the status and activity of
various components of the software running on the HCP S10 Node.
You can use the HCP S Series Management Console or management API to insert
comments into or download the internal logs.
• If in use by an HCP system, the HCP S10 Node notifies that system about certain
abnormal conditions as they occur.
• If the Hitachi Remote Ops monitor agent is configured to monitor the HCP S10 Node,
the HCP S10 Node sends status information to the monitor agent in response to periodic
requests from the agent.
• To see detailed information about the enclosure and its components, click the gear icon (
) for the enclosure on the Hardware ► Overview page. Then, in the Detailed
Enclosure Information list, click the type of component you want more information
about.
• To see detailed information about the data or database drive in a slot or about an empty
drive slot, click the slot in the top view in the enclosure diagram on the enclosure details
page.
• To see detailed information about a server module and its components, click the gear
icon ( ) for the server module on the Hardware ► Overview page. Then, in the
Detailed Server Module Information list, click the type of component you want more
information about.
Enclosure LEDs
The front of an enclosure has eight color-coded LEDs that indicate the status of the S10
Node. The LEDs are arranged in two angled columns of four LEDs each, as shown in the
figure below.
(Continued)
LED Color State Description
Enclosure fault Amber Solid The enclosure itself or one or more components in the
enclosure are not functioning properly.
Blinking N/A
Drive fault Amber Solid One or more data or database drives in the enclosure are
not functioning properly.
Blinking N/A
Software OK Green Solid The HCP S Series software is running on the S10 Node.
Software fault Amber Solid One or more components of the S10 Node are not
functioning properly.
Off N/A
PCM fault Amber Solid The power and cooling module is not operating normally.
Blinking N/A
DC OK Green Solid The power supply DC output voltages are within the normal
operating range.
Blinking N/A
Off The power supply DC output voltages are not within the
normal operating range.
AC OK Green Solid The power and cooling module is powered on, and the
AC input power is within normal operating range.
Blinking N/A
• While a power and cooling module has a problem (such as a failed power supply or fan).
• While the enclosure contains only one power and cooling module. In this case, the alarm
is intermittent for the first five minutes and then becomes continuous.
Beaconing
The enclosure, power and cooling modules, and server modules have LEDs that can serve
as beacons. When blinking, a beaconing LED lets you easily identify the applicable
component in the customer data center. For information about the LEDs used for beaconing,
see "Physical component status indicators" on page 50.
If you have the administrator or service role, you can use the HCP S Series Management
Console or management API to turn beaconing on or off for a component.
In the Management Console, you manage beaconing from the enclosure diagram. This
diagram provides top, front, and back views of the enclosure.
To see:
• All three views of the enclosure diagram, on the Hardware ► Overview page, click the
gear icon ( ) for the enclosure.
• Only the back view of the enclosure diagram, on the Hardware ► Overview page, click
the gear icon ( ) for either one of the server modules.
4. In the back view of the enclosure diagram, click the module for which you want to turn
beaconing on or off.
5. In the Power & Cooling Module or Server Module window that opens, set the
Beacon slider to On to turn beaconing on or to Off to turn beaconing off.
The blue identity LED on the applicable module immediately starts to blink or stops
blinking, as applicable. The identity LED on a power and cooling module is the second
LED from the left on the front of the module. The identity LED on an server module is the
middle LED on the front of the module.
6. Click Close.
Internal logs
HCP S Series Nodes maintain internal logs that record the status and activity of various
components of the HCP S Series software. If a problem occurs with the S Series Node, the
internal logs can assist support personnel in diagnosing and resolving the problem.
If you have the administrator, monitor, security, or service role, you can use the HCP S
Series Management Console or management API to insert comments into the S Series Node
internal logs. You can use this capability, for example, to note unusual events that occur on
the S Series Node. Comments can later assist support personnel in understanding the
symptoms that indicate a possible problem. Comments can also assist support personnel in
determining when a problem started.
To help with troubleshooting, if you have the administrator or service role, you can download
the internal logs and send them to your HCP support center. You can use the HCP S Series
Management Console or management API to download the logs. For ease of handling, the S
Series Node downloads the logs into a single packed file. Neither this file nor the logs
themselves are encrypted.
An S Series Node generally keeps internal logs for at least 120 days. However, it keeps them
for a shorter time period if not enough space is available for them. You can download the logs
for any length of time within the period for which logs exist. When downloading the logs, be
sure to include all the days on which you observed problems with the S Series Node.
Note: Downloading the internal logs puts a heavy load on the S Series Node. Do not take
this action unless explicitly told to do so by support personnel.
To use the HCP S Series Management Console to download the internal logs:
1. Log in to the HCP S Series Management Console using a user account with the
administrator or service role.
Tip: To ensure that you consistently see correct status information during a log
download operation, use a physical IP address to access the Management Console.
Note: While the logs are being prepared for download, do not restart a server module.
Doing so causes the logs to remain in the prepare state indefinitely. If this happens,
contact your HCP support center for help.
5. When the status of both server modules is "Ready for log download," click Download.
6. When prompted, save the .zip file containing the downloaded logs in the location of your
choice.
By default, the name of the .zip file is HCPSLogs-YYYYMMDD-hhmm.zip.
You do not need to stay on the Internal Logs page while the S Series Node is packing
and downloading the logs.
7. When the status of both server modules is "Download complete," click Reset.
To configure the Remote Ops monitor agent to monitor an S Series Node, you specify this
information in the monitor agent:
• The username and password for an S Series Node user account that has the monitor
role.
Tips:
• Create an S Series Node user account specifically for use by the monitor agent.
When you create the account, deselect the option to force a password change on
the next login. Also deselect the option to have the password expire automatically.
• If you change the password for the S Series Node user account used by the monitor
agent, be sure to change the user account password in the monitor agent at the
same time.
• The IP addresses of both S Series Node server modules on the access or management
network. If you use the access network, you can specify either the physical IP
addresses or the virtual IP addresses.
Note: S Series Nodes support both IPv4 and IPv6 network connections to the monitor
agent. However, monitor-agent support for IPv6 connections varies based on the
operating system on the server where the monitor agent is running. For information about
requirements for monitor-agent servers that support IPv6 networks, see the applicable
Hitachi Remote Ops monitor agent documentation.
• One or more email addresses to which the monitor agent should send notifications of
error conditions.
• Upgrading the HCP S Series OS and software when a new HCP S Series Node release
becomes available and the customer is ready for the upgrade. Upgrades can also include
updates to S Series Node component firmware.
• Recovering the HCP S Series OS and software on a single server module after replacing
both OS SSDs in the module.
• Recovering the HCP S Series OS and software on both server modules in the rare case
that an S Series Node becomes inoperable. To perform this procedure, you need to
contact your HCP support center for instructions.
• Before you start the procedure to upgrade the HCP S Series software or apply a hotfix,
both S Series Node server modules must be running and healthy.
• When you upload an update file, the file overwrites any previously uploaded update file.
• After uploading an update file, you cannot apply the update while the internal logs are
being downloaded or while a maintenance procedure is in progress.
• While the software is being updated, you can make changes to the S Series Node
configuration. However, most configuration changes don't take effect until the software
update is complete.
• Software updates occur on one server module at a time. While the software is being
updated on one server module, all S Series Node processing occurs on the other server
module.
• When a software update finishes on the first server module, that server module
automatically reboots. When the reboot is complete, the update automatically starts on
the second server module, and processing fails over from the second server module to
the first server module. While this failover is in progress, the HCP S Series Management
Console may be briefly unavailable.
When the software update is complete on the second server module, that server module
automatically reboots. When the reboot is complete, processing is again distributed
across both server modules.
• While the software on a server module is being updated, you cannot use the physical IP
address of that module to access the HCP S Series Management Console, make
management API requests, or perform data access operations.
• If you accessed the HCP S Series Management Console by using the physical
IP address of the second server module while the software on the first server module
was being updated, when failover occurs, you lose your connection to the Management
Console. At that point, you need to log in again, this time using the S Series Node
domain name, a virtual IP address, or the physical IP address of the first server module
to access the Management Console.
• At certain points during a software update that includes a firmware update, the HCP S
Series Management Console displays an alert indicating that a firmware version
mismatch exists. Depending on which server module issued the alert, the alert may
show up as informational or as indicating an error condition. In either case, no action is
required.
• If an error occurs during the apply step of an update, you can try restarting the update. If
an error occurs again, do not try to restart the update a second time. Instead, contact
your HCP support center for help.
The VPD tool reports its progress as it checks and, if necessary, tries to fix the VPDs.
The tool then reports the outcome of its processing. Three outcomes are possible:
¡ The tool does not find any errors. The report looks like this:
To continue:
1. Enter exit to close the PuTTY window.
2. From the HCP S Series Management Console, restart server module 2:
a. On the Hardware ► Overview or Hardware ► Maintenance page, click
Power Options.
b. In the Server Module field, select Server Module 2.
c. In the Reason field, type the reason why you're restarting the server
module.
d. Click Power On.
To continue, , follow the procedure above for when the VPD tool doesn't find any
errors.
¡ The tool found one or more errors it could not fix. The report looks something like
this:
To continue:
1. Using a file transfer tool such as WinSCP, copy the .log file and all the .txt files
from the /tmp/VPDTool directory to the location of your choice on your laptop
computer. HCP support personnel will need these files to resolve the VPD
errors.
2. From the HCP S Series Management Console, restart server module 2:
a. On the Hardware ► Overview or Hardware ► Maintenance page, click
Power Options.
b. In the Server Module field, select Server Module 2.
c. In the Reason field, type the reason why you're restarting the server
module.
d. Click Restart.
3. Either try running the VPD tool again, or contact your HCP support center for
help. Do not try to run the tool again more than twice.
Important: Do not perform the software upgrade after the VPD tool discovers errors
that the tool cannot fix.
Tip: If you accidentally upload the wrong file, click Cancel to reactivate the Select and
Upload section on the Update page.
The prechecks for an update are mostly the same as the prechecks for a software
reinstallation. For a list of prechecks, see "Software reinstallation prechecks" on page 91.
If an update fails, the Restart Update button on the Update page becomes active. You can
try applying the update again by clicking Restart Update.
If an update fails more than once, contact your HCP support center for help.
Tip: Before starting the reinstallation, use the HCP S Series Node Configuration Information
form on page 215 to record the customer-specific configuration settings for the S10 Node on
which you're reinstalling the OS and software. Then keep the form with you when you
perform the reinstallation so you have easy access to the information you need.
Caution: Reinstalling the OS and software on an S Series Node destroys all data stored on
the S Series Node. Reinstallation is not the way to recover the OS and software on an S
Series Node that is experiencing problems.
What you need for reinstalling the HCP S Series OS and software
For an HCP S Series software reinstallation, you need:
• If the access ports on the server modules are not connected to switches in the customer
network, two 10Gb Twinax cables.
• If the management ports on the server modules are not connected to switches in the
customer network, one CAT6 Ethernet cable.
• One bootable USB flash drive or DVD that contains the HCP S Series OS and software
installation files. For information about preparing this drive, see "Preparing an installation
USB flash drive or DVD" on page 40.
• If the HCP S Series OS and software installation files are on a DVD, an external DVD
drive with a USB connector.
• A USB-to-mini-USB cable.
If you are providing service on behalf of an authorized S Series Node distributor, you get
the BIOS password from the HCP S Series Node Product Management team.
2. Ensure that the two server modules are firmly seated in the enclosure. If either module is
not firmly seated, using the #1 Phillips screwdriver, gently tighten the large black screw
in the top center of the module. Do not overtighten the screw.
3. If the access ports are not connected to the customer network and have SFP+ modules
installed in them, remove the modules from those ports.
4. Connect cables to the Ethernet ports:
¡ If the access ports on the server modules are not connected to switches in the
customer network:
– Use one Twinax cable to connect access port 1 on the left server module to
access port 1 on the right server module.
When connecting the cable to access port 1 on a server module, ensure that the
card in the connector at the end of the cable is facing access port 2 on that
server module.
– Use the other Twinax cable to connect access port 2 on the left server module
to access port 2 on the right server module.
When connecting the cable to access port 2 on a server module, ensure that the
card in the connector at the end of the cable is facing access port 1 on that
server module.
¡ If the management ports on the server modules are not connected to switches in the
customer network, using one CAT6 Ethernet cable, connect the management port
on the left server module to the management port on the right server module.
When inserting the cable into each port, push the cable in until the end clicks into
place.
The VPD tool reports its progress as it checks and, if necessary, tries to fix the VPDs.
The tool then reports the outcome of its processing. Three outcomes are possible:
¡ The tool does not find any errors. The report looks like this:
To continue:
1. Enter exit to close the PuTTY window.
2. From the HCP S Series Management Console, restart server module 2:
a. On the Hardware ► Overview or Hardware ► Maintenance page, click
Power Options.
b. In the Server Module field, select Server Module 2.
c. In the Reason field, type the reason why you're restarting the server
module.
d. Click Power On.
3. Wait for server module 2 to become available.
4. Proceed with the software reinstallation.
¡ The tool found one or more errors and fixed them all. The report looks something like
this:
To continue, , follow the procedure above for when the VPD tool doesn't find any
errors.
¡ The tool found one or more errors it could not fix. The report looks something like
this:
To continue:
1. Using a file transfer tool such as WinSCP, copy the .log file and all the .txt files
from the /tmp/VPDTool directory to the location of your choice on your laptop
computer. HCP support personnel will need these files to resolve the VPD
errors.
2. From the HCP S Series Management Console, restart server module 2:
a. On the Hardware ► Overview or Hardware ► Maintenance page, click
Power Options.
b. In the Server Module field, select Server Module 2.
c. In the Reason field, type the reason why you're restarting the server
module.
d. Click Restart.
3. Either try running the VPD tool again, or contact your HCP support center for
help. Do not try to run the tool again more than twice.
Important: Do not perform the software reinstallation after the VPD tool discovers
errors that the tool cannot fix.
• A default reinstallation reinstalls the OS and software with all default initial S Series
Node configuration settings, including the default distributor key (Atlantic) and server
module IP addresses. With this type of installation, you don't provide any configuration
information.
• A custom reinstallation lets you specify the initial configuration settings for the S
Series Node, including the distributor key and server module IP addresses. If your
distributor key is other than Atlantic, a custom reinstallation is required.
Tip: Before starting a custom reinstallation, use the "Appendix: HCP S Series Node
Configuration Information form" on page 214 to record the customer-specific
configuration settings for the S Series Node. Then keep the form with you when you
perform the reinstallation so you have easy access to the information you need.
Caution: Reinstalling the OS and software on an S Series Node destroys all data stored on
the S Series Node. Reinstallation is not the way to recover the OS and software on an S
Series Node that is experiencing problems.
For a default software reinstallation, you first reconfigure the left server module and reinstall
the OS on that module. Then you reconfigure the right server module and reinstall the OS on
that module. After the OS is reinstalled on both server modules, the software reinstallation
starts automatically.
Caution: Reinstalling the OS and software on an S Series Node destroys all data stored on
the S Series Node. Reinstallation is not the way to recover the OS and software on an S
Series Node that is experiencing problems.
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module again.
5. When prompted for a password, enter the BIOS password for S10 Nodes.
7. On the Advanced menu page, arrow down to USB Configuration. Then press Enter.
8. On the USB Configuration page, arrow down to USB Mass Storage Driver Support.
Then press Enter.
9. In the USB Mass Storage Driver Support popup, arrow down to Enabled. Then press
Enter.
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module again.
15. When prompted for a password, enter the BIOS password for S10 Nodes.
16. In the Please select boot device menu, arrow down to the entry for the installation
USB flash drive or DVD. Then press Enter. (The name of the drive may differ from the
name shown in the figure below.)
Caution: Option I destroys all data stored on the S Series Node. If you do not want to
destroy the data, enter e to exit the reinstallation.
The OS reinstallation starts on the left server module. When the reinstallation finishes,
the server module automatically reboots. The left server module then waits for the OS
reinstallation to finish on the right server module.
3. After the left server module starts to reboot, take one of these actions, as applicable:
¡ If the HCP S Series OS and software installation files are on a USB flash drive,
remove that drive from the USB port on the left server module.
¡ If the HCP S Series OS and software installation files are on a DVD, disconnect the
external DVD drive from the USB port on the left server module, but do not remove
the DVD from the external drive.
Reconfigure the right server module
To reconfigure the right server module:
1. Use PuTTY to establish a serial connection to the right server module, as described in
"Using a PuTTY serial connection" on page 35.
2. Take one of these actions, as applicable:
¡ If the HCP S Series OS and software installation files are on a USB flash drive,
insert that drive into a USB port on the right server module.
¡ If the HCP S Series OS and software installation files are on a DVD, connect the
external DVD drive to a USB port on the right server module.
3. Reboot the right server module in either of these ways:
¡ In the PuTTY window:
1. Right-click in the window title bar. Then select Special Command ► Break.
2. Enter b.
¡ Insert the end of a paper clip into the hole for the recessed power button in the
bottom right corner of the module and then gently pressing the paper clip against the
power button for five seconds.
4. When you see this prompt, press Delete:
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module again.
5. When prompted for a password, enter the BIOS password for S10 Nodes.
The BIOS Main page appears.
6. Arrow right to Advanced.
7. On the Advanced menu page, arrow down to USB Configuration. Then press Enter.
8. On the USB Configuration page, arrow down to USB Mass Storage Driver Support.
Then press Enter.
9. In the USB Mass Storage Driver Support popup, arrow down to Enabled. Then press
Enter.
10. Press Escape to return to the Advanced menu page.
11. Arrow right to Save & Exit.
12. On the Save & Exit page, arrow down to Save Changes and Reset. Then press Enter.
13. In response to the confirming prompt, press Enter.
The server module reboots.
14. When you see this prompt, press F7:
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module.
15. When prompted for a password, enter the Boot Manager password for S10 and S30
Nodes.
16. In the Please select boot device menu, arrow down to the entry for the installation USB
flash drive or DVD. Then press Enter.
Reinstall the OS on the right server module and the software on both server modules
To reinstall the OS on the right server module and the software on both server modules:
1. When the HCPSOS Operating System installation window appears in the PuTTY
window, arrow up to HCPSOS Operating System under SERIAL. Then press Enter.
2. In response to the prompt for the OS installation mode, enter i.
Caution: Option I destroys all data stored on the S Series Node. If you do not want to
destroy the data, enter e to exit the reinstallation.
Tip: You can close the PuTTY session and disconnect the laptop computer from the
server module at any time after you select the OS installation option. However, if you
want to monitor the reinstallation process, you should leave the PuTTY session open on
the right server module.
The portion of the reinstallation process that goes from the time the OS reinstallation starts
on the right server module to the time the database drive synchronization starts takes
approximately 25 minutes.
For a custom software reinstallation, you first reconfigure the left server module and reinstall
the OS on that module. Then you reconfigure the right server module and reinstall the OS on
that module. After the OS is reinstalled on both server modules, the software reinstallation
starts automatically.
Caution: Reinstalling the OS and software on an S Series Node destroys all data stored on
the S Series Node. Reinstallation is not the way to recover the OS and software on an S
Series Node that is experiencing problems.
¡ Insert the end of a paper clip into the hole for the recessed power button in the
bottom right corner of the module and then gently press the paper clip against the
power button for five seconds.
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module again.
5. When prompted for a password, enter the BIOS password for S10 Nodes.
8. On the USB Configuration page, arrow down to USB Mass Storage Driver Support.
Then press Enter.
9. In the USB Mass Storage Driver Support popup, arrow down to Enabled. Then press
Enter.
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module again.
15. When prompted for a password, enter the BIOS password for S10 Nodes.
16. In the Please select boot device menu, arrow down to the entry for the installation
USB flash drive or DVD. Then press Enter. (The name of the drive may differ from the
name shown in the figure below.)
Caution: Option I destroys all data stored on the S Series Node. If you do not want to
destroy the data, enter e to exit the reinstallation.
4. Respond to the next series of prompts with the applicable values for the customer-
specific S Series Node configuration. The table below lists the settings you are prompted
for, along with their default values. For prompts that have a default value in square
brackets, you can press Enter to accept that value.
Prompt Value
Distributor SSH key Name of the distributor key for the S Series
Node
Access network IPv4 address Access network IPv4 address for the server
module on the left
The default is 10.0.0.1
Access network IPv4 gateway address Access network IPv4 gateway address
Access network IPv6 address Access network primary IPv6 address for the
server module on the left
No default
Access network IPv6 prefix length Access network primary IPv6 prefix length
The default is 64.
Access network IPv6 gateway address Access network primary IPv6 gateway
address
No default
(Continued)
Prompt Value
Access network secondary IPv6 address Access network secondary IPv6 address for
the server module on the left
No default
Access network secondary IPv6 prefix Access network secondary IPv6 prefix
length length
The default is 64.
Access network secondary IPv6 gateway Access network secondary IPv6 gateway
address address
No default
Management network IPv4 address Management network IPv4 address for the
server module on the left
The default is
10.2.2.1.
Management network IPv6 prefix length Management network primary IPv6 prefix
length
The default is 64.
(Continued)
Prompt Value
No default
Server interconnect network IPv4 subnet Server interconnect network IPv4 subnet
The default is 10.1.1.0.
The OS reinstallation starts on the left server module. When the reinstallation finishes,
the server module automatically reboots. The left server module then waits for the
OS reinstallation to finish on the right server module.
5. After the left server module starts to reboot, take one of these actions, as applicable:
¡ If the HCP S Series OS and software installation files are on a USB flash drive,
remove that drive from the USB port on the left server module.
¡ If the HCP S Series OS and software installation files are on a DVD, disconnect the
external DVD drive from the USB port on the left server module, but do not remove
the DVD from the external drive.
Reconfigure the right server module
To reconfigure the right server module:
1. Use PuTTY to establish a serial connection to the right server module, as described in
"Using a PuTTY serial connection" on page 35.
2. Take one of these actions, as applicable:
¡ If the HCP S Series OS and software installation files are on a USB flash drive,
insert that drive into a USB port on the right server module.
¡ If the HCP S Series OS and software installation files are on a DVD, connect the
external DVD drive to a USB port on the right server module.
3. Reboot the right server module in either of these ways:
¡ In the PuTTY window:
1. Right-click in the window title bar. Then select Special Command ► Break.
2. Enter b.
¡ Insert the end of a paper clip into the hole for the recessed power button in the
bottom right corner of the module and then gently pressing the paper clip against the
power button for five seconds.
4. When you see this prompt, press Delete:
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module again.
5. When prompted for a password, enter the BIOS password for S10 Nodes.
The BIOS Main page appears.
6. Arrow right to Advanced.
7. On the Advanced menu page, arrow down to USB Configuration. Then press Enter.
8. On the USB Configuration page, arrow down to USB Mass Storage Driver Support.
Then press Enter.
9. In the USB Mass Storage Driver Support popup, arrow down to Enabled. Then press
Enter.
10. Press Escape to return to the Advanced menu page.
11. Arrow right to Save & Exit.
12. On the Save & Exit page, arrow down to Save Changes and Reset. Then press Enter.
13. In response to the confirming prompt, press Enter.
The server module reboots.
14. When you see this prompt, press F7:
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to the reboot the server module
again.
15. When prompted for a password, enter the Boot Manager password for S10 and S30
Nodes.
16. In the Please select boot device menu, arrow down to the entry for the installation USB
flash drive or DVD. Then press Enter.
Reinstall the OS on the right server module and the software on both server modules
To reinstall the OS on the right server module and the software on both server modules:
1. When the HCPSOS Operating System installation window appears in the PuTTY
window, with HCPSOS Operating System custom under SERIAL highlighted, press
Enter.
2. In response to the prompt for the OS installation mode, enter i.
Caution: Option I destroys all data stored on the S Series Node. If you do not want to
destroy the data, enter e to exit the reinstallation.
Prompt Value
Access network IPv4 address Access network IPv4 address for the server
module on the right
The default is 10.0.0.2.
Access network IPv4 gateway address Access network IPv4 gateway address
The default is 10.0.0.254.
Access network IPv6 address Access network primary IPv6 address for the
server module on the right
No default
Access network IPv6 prefix length Access network primary IPv6 prefix length
The default is 64.
(Continued)
Prompt Value
Access network secondary IPv6 address Access network secondary IPv6 address for
the server module on the right
No default
Access network secondary IPv6 prefix Access network secondary IPv6 prefix
length length
Management network IPv4 address Management network IPv4 address for the
server module on the right
The default is 10.2.2.2.
Management network IPv6 prefix length Management network primary IPv6 prefix
length
The default is 64.
(Continued)
Prompt Value
64
The OS reinstallation starts on the right server module. When the OS reinstallation
finishes, the right server module automatically reboots.
5. As soon as the right server module starts to reboot, take one of these actions, as
applicable:
¡ If the HCP S Series OS and software installation files are on a USB flash drive,
remove that drive from the USB port on the right server module.
¡ If the HCP S Series OS and software installation files are on a DVD, remove the
DVD from the external DVD drive. Then disconnect the DVD drive from the USB
port on the right server module.
After the right server module finishes rebooting, the HCP S Series software installation
program first performs a series of prechecks. If all the prechecks are successful, the
program reinstalls the software on both server modules.
When the software reinstallation finishes, both server modules automatically reboot, and
the S Series Node starts synchronizing the database drives.
For information about the installation prechecks, see "Software reinstallation prechecks"
on the next page.
Tip: To monitor the reinstallation process, leave the PuTTY session open on the right
server module.
The portion of the reinstallation process that goes from the time the OS reinstallation starts
on the right server module to the time the database drive synchronization starts takes
approximately 25 minutes.
Before reinstalling the HCP S Series software, the installation program performs a series of
prechecks. If any of these prechecks fail, the program does not reinstall the software.
These prechecks are also performed before an update of the OS and software on an S Series
Node.
The table below lists the prechecks that the installation program performs. For each
precheck, the table explains how the precheck can fail and what you should do in response to
a failure.
Some precheck failures identify the server modules by server interconnect network
IP address. These IP addresses are:
all network links An Ethernet interface is unavailable. Check that all the Ethernet cables are
correctly connected to both sever
modules and that the cables are
properly seated in their respective
ports.
disk space One or more OS partitions have Contact your HCP support center.
insufficient space available for the
installation.
DiskTypePrecheck A database drive is missing or is not a If the drive is missing, insert a drive of
supported hardware model. the correct type. If the drive is not the
correct type, replace the drive with the
correct type of drive.
firmware versions The firmware on one or more Contact your HCP support center.
components of the S10 Node is not the
correct version for the version of the
HCP S Series software you're trying to
reiinstall.
free space for upgrade of One or both server modules have Make a note of the failed precheck.
internal database insufficient free space for upgrading Then contact your HCP support center.
(checked only during an the internal database on the database
HCP S Series OS and drives the module uses.
software update)
(Continued)
Precheck Failure description What to do
node connections The installation program cannot Check that the server
communicate with one or both server interconnect cable is correctly
modules. connected to both sever modules and
that the cable is properly seated in both
ports.
If the cable is correctly connected but
the link LEDs for the applicable ports
are not lit, replace the cable.
RAID synchronization The partitions on the OS drives in a Reinstall the OS on the server module.
server module are not synchronized If that doesn't resolve the issue, replace
with each other. the SSDs in the server module.
repair backlog The S Series Node has storage that If the storage is not repaired within one
(checked only during an needs repair. The upgrade cannot hour, contact your HCP support center.
HCP S Series OS and proceed until the S Series Node
software update) repairs are complete.
run location The installation program cannot Reinstall the OS on the applicable
determine which version of the OS is server module.
installed on the server module on
which the program is running.
running as install The installation program is using an Contact your HCP support center.
incorrect system-internal user account.
SAS topology A server module cannot see the For an enclosure issue, check that the
(checked only during an enclosure or one or more drives in the server module is correctly seated in its
HCP S Series OS and enclosure. slot and that the pins on the back of the
software reinstallation) server module are not damaged. If the
pins are damaged, replace the server
module. If the server module is
correctly seated, reboot it. If rebooting
the server module doesn't resolve the
issue, power the enclosure off and
back on.
For a drive issue, check that all drives
are present and properly seated in their
respective slots.
services One or more services started by the OS Start the OS and software reinstallation
on either server module is still running procedure again. If the services
after 15 minutes. precheck fails again, make a note of
the failed precheck. Then contact your
HCP support center.
(Continued)
Precheck Failure description What to do
SSH connectivity A server module is not responding to Check that both server modules are
an SSH request. powered on and that the OS is installed
on each module. Check that the server
interconnect cable is correctly
connected to both sever modules and
that the cable is properly seated in both
ports.
SSH keys A problem exists with the SSH keys Contact your HCP support center.
used by the S10 Node.
systemwide SSH The server modules cannot Check that both server modules are
communicate with each other through powered on and that the OS is installed
SSH. on each module. Check that the server
interconnect cable is correctly
connected to both sever modules and
that the cable is properly seated in both
ports.
versions The version of the OS installed on a Install the correct version of the OS on
server module does support the the server module.
version of the HCP S Series software
you're trying to reiinstall.
waiting for all nodes to One or both server modules are not Check that both server modules are
become available ready for the software to be reiinstalled. powered on and that the OS is installed
on each module. Check that the server
The installation program waits two
hours for both server modules to be interconnect cable is correctly
ready. After two hours, if either server connected to both sever modules and
module is not ready, the precheck fails. that the cable is properly seated in both
ports.
While the HCP S Series OS and software are being reinstalled, various LEDs on the
enclosure and server modules either blink or are lit solid:
• While the OS is being reinstalled, the blue identity LED on the applicable server module
blinks.
• During the software installation prechecks, the blue software install LED on the
enclosure is lit solid.
• During the software reinstallation, the blue software install LED on the enclosure blinks.
When the software reinstallation is complete, this LED turns off.
• Near the end of the software reinstallation, the green software OK LED and amber drive
fault LED on the enclosure start blinking to indicate that the S10 Node has started
synchronizing the database drives.
• On the front of the enclosure, the green power LED is lit solid.
• On the front of the enclosure, the green software OK LED and the amber drive fault LED
blink while the S10 Node synchronizes the database drives. This process takes
approximately 20 minutes. When the synchronization is compete, the software OK LED
is lit solid and the drive fault LED turns off.
• On each power and cooling module, the green DC OK LED and the green AC OK LED
are all both solid. No other LEDs on the power and cooling modules are lit.
• On each server module, the green power LED is lit solid. No other LEDs on the server
modules are lit.
A software reinstallation can succeed even if a problem exists with a power and cooling
module (PCM) . In this case, the enclosure fault LED is solid, and the applicable power and
cooling LEDs are either off or blinking, as described in the table below.
Enclosure fault solid The power and Ensure that the power and cooling
and cooling module module is plugged in. If the module is
PCM no lights either is not plugged in, replace the module.
plugged in or is
faulty.
Enclosure fault solid The power and Replace the power and cooling module.
and cooling module is
PCM fault solid faulty.
For instructions on replacing a power and cooling module, see "Replacing a power and
cooling module" on page 132.
If, after you fix the problem with the power and cooling module, all of the following are true,
the software reinstallation was successful:
• Remove the installation USB flash drives from the server modules
• Disconnect the access and management network cables from the server modules
Important: Do not disconnect the purple server interconnect cable while the database
drive synchronization is in progress. Doing so causes the software reinstallation to be
unsuccessful.
• Run the HCP S Series Node post-install script to set the S Series Node serial number
and, if applicable, enable data-in-flight encryption
However, if a database drive error occurs after you remove the USB flash drives or
disconnect the cables, you need to reconnect the cables and reinsert the flash drives in order
to reinstall the software. If a database drive error occurs after you run the HCP S Series
Node post-install script, you need to run the script again after you reinstall the software.
For instructions on running the post-install script, see Step 4: "Run the post-install script" on
page 97.
A reinstallation of the HCP S Series OS and software can fail for a variety of reasons,
including faulty hardware and corrupt software. If a reinstallation fails, you can use the LEDs
on the enclosure and server modules (SMs) to help determine the cause of the failure.
The table below lists LED states that indicate a failed OS or software reinstallation and
describes the actions to take in response to them. If a reinstallation fails and the LEDs do not
match any of states listed in the table, contact your HCP support center for help.
Enclosure software fault solid The server Use the rocker switches on the power and cooling
and interconnect cable is modules to power both power and cooling
Enclosure software install solid faulty. modules off. Replace the server interconnect
and cable. Then perform a default software
Both SM faults blinking reinstallation or custom software reinstallation, as
applicable.
Enclosure software fault solid The OS reinstallation Try these in the order shown:
and failed.
SM fault solid in one SM 1. Prepare a new USB flash drive or DVD with
and the same software version as before. Use the
SM identity blinking in the same rocker switches on the power and cooling
SM modules to power both power and cooling
modules off. Then perform a default software
reinstallation or custom software
reinstallation, as applicable, using the new
USB flash drive or DVD.
(Continued)
LEDs Meaning Action
Enclosure software fault solid Either the server Use the rocker switches on the power and cooling
and module has modules to power both power and cooling
Enclosure software install solid insufficient memory or modules off. Replace the problem server module
and a faulty Ethernet port, with a new server module. Move the SSDs from
SM fault solid in one SM or one or more the problem server module to the new server
and components of the module. Then perform a default software
SM identity blinking in the same server module have
reinstallation or custom software reinstallation, as
SM incorrect firmware
applicable.
versions.
Enclosure drive fault solid or Either the database Check the database drives inside the enclosure:
blinking drive configuration is
and invalid, the enclosure • If any database drives have a blinking blue
Enclosure software fault solid cannot read from the LED, replace those drives. Then, use the
and database drives, or rocker switches on the power and cooling
Enclosure software install solid one or more database modules to power both power and cooling
and drives are faulty. modules off. Perform a default software
Either SM fault solid installation or custom software reinstallation,
and as applicable. If the reinstallation fails again
Either SM identity blinking with the same symptoms, replace the
enclosure. Then perform a default software
reinstallation or custom software reinstallation
again, as applicable.
(Continued)
LEDs Meaning Action
Enclosure drive fault solid Either one or more Check the data and database drives inside the
data or database enclosure. Replace any drives that have a blinking
drives are faulty, or an blue LED. Then, use the rocker switches on the
error occurred during power and cooling modules to power both power
the database drive and cooling modules off. Perform a default
synchronization software reinstallation or custom software
process. reinstallation, as applicable.
• Sets the S Series Node serial number for the S10 Node
• Verifies that the S10 Node has the correct number of drives
• The total number of data and database drives in the S10 Node enclosure.
A half-populated enclosure has a total of 32 data and database drives. A fully populated
enclosure has a total of 60 data and database drives.
For instructions on getting the post-install script, see.
To run the post-install script, you use your laptop computer to connect directly to the S Series
Node over the management network. This means that your laptop computer must be on the
subnet used for the management network. After a default reinstallation of the HCP S Series
OS and software, this subnet is 10.2.2.0/24.
To run the post-install script:
1. If you have not already done so, add the S Series Node root SSH key to the SSH
keychain on your laptop computer, as described in "Adding an S Series Node SSH key
11. In response to the prompt for whether the S Series Node should allow data-in-flight
encryption:
¡ To allow data-in-flight encryption, enter y.
y is the standard response to this prompt.
¡ To prevent data-in-flight encryption, enter n.
Important: Be sure to respond correctly to this prompt. If you enter y when the S10
Node is not supposed to support data-in-flight encryption, the S Series Node will be in
violation of export control regulations.
13. If the post-install script did not finish successfully, take the applicable actions to resolve
any reported issues. Then run the script again.
Important: After a successful run of the post-install script, you cannot run the script
again because SSH access is now disabled on the S10 Node. If you incorrectly enabled
or disabled support for data-in-flight encryption when you ran the script, you need to
reinstall the HCP S Series OS and software in order to run the script again.
After a default reinstallation of the HCP S Series OS and software, this IP address is
10.2.2.1.
2. When prompted, accept the self-signed HCP S Series Node SSL server certificate.
The HCP S Series Management Console opens and displays the Getting Started page
of the first-time setup wizard.
3. Click Next.
4. On the Identification page, specify the domain name and, if missing, the serial number
for the S Series Node. Then click Next.
5. On the User page, specify the password for the admin user account. Then click Next.
Note: If the customer wants to keep the password for the admin user account private,
have the customer type the password on the User page.
6. On the Access Network page, specify the access network settings. Then click Next.
7. On the Management Network page, specify the management network settings. Then
click Next.
8. On the Server Interconnect Network page, specify the subnet for the server
interconnect network. Then click Next.
9. On the DNS & Time Servers page, specify the DNS and time servers to be used by the
S Series Node. Then click Next.
10. On the Review page, review the initial S Series Node configuration settings:
¡ If all the settings are correct, click Finish.
The S Series Node reboots. When the S Series Node becomes available again, you
are automatically redirected to the Management Console.
¡ If you need to change any settings, use the Previous button to return to the
applicable page and make the corrections you want. Then return to the Review page
and, if all the settings are correct, click Finish.
11. Disconnect your laptop computer from the server module. Then change the network
connection properties on the laptop computer back to their original settings.
12. If the management port on the left server module was previously connected to a switch
in the customer network, reconnect the cable to that management port.
Step 6 (conditional): Have the customer enable SSH access to the server
modules
The post-install script disables SSH access to the S Series Node server modules. After you
configure the S Series Node, the customer can choose to enable SSH access. SSH access
facilitates troubleshooting when the customer requests support.
Alternatively, the customer can choose to leave SSH access disabled. Keeping SSH access
disabled enhances the security of the S Series Node but can increase the amount of time
required to diagnose and resolve issues.
While SSH access is disabled, this banner appears at the top of each page in the HCP S
Series Management Console:
SSH is disabled.
For users with the security role, this banner is a link to the Configuration ► Security page.
Customers who want to enable SSH access should perform the procedure themselves using
the instructions below.
Tip: Before starting the server-module recovery, use the "Appendix: HCP S Series Node
Configuration Information form" on page 214 to record the customer-specific configuration
settings for the server module you're recovering. Then keep the form with you when you
perform the recovery so you have easy access to the information you need.
• If the management ports on the server modules are not connected to switches in the
customer network, one CAT6 Ethernet cable to connect the ports to each other.
• A laptop computer with the HCP S Series Node root SSH key in the SSH keychain on
the computer. For instructions on adding this key to the keychain on your laptop
computer, see "Adding an S Series Node SSH key to your SSH keychain" on page 35.
• A USB-to-mini-USB cable.
• One bootable USB flash drive or DVD that contains the HCP S Series OS and software
installation files for the software release that's currently installed on the S10 Node.
If you don't already have an OS and software installation USB flash drive or DVD, you
need to create one, as described in "Preparing an installation USB flash drive or DVD".
• If the HCP S Series OS and software installation files are on a DVD, an external DVD
drive with a USB connector.
• A paper clip. You need a paper clip only if the HCP S Series Management Console is
currently unavailable.
If you are providing service on behalf of an authorized S Series Node distributor, you get
the BIOS password from the HCP S Series Node Product Management team.
• This information:
¡ The name of the distributor key for the S Series Node if the key is not Atlantic.
¡ The current network configuration settings for the server module you're recovering.
You need the settings for the access network, management network, and server
interconnect network.
You can use the HCP S Series Management Console or the HCP S Series
management API to get the current network configuration settings.
¡ The domain name and serial number for the S Series Node.
Tip: Before starting the server-module recovery, use the HCP S Series Node
Configuration Information form on page 215 to record the customer-specific
configuration settings for the server module you're recovering. Then keep the form with
you when you perform the reinstallation so you have easy access to the information you
need.
2. If the server module you want to recover is powered off, power on the module in either of
these ways:
¡ From the HCP S Series Management Console:
1. On the Hardware►Overview or Hardware►Maintenance page, click Power
Options.
2. In the Server Module field, select the server module you want to recover.
3. In the Reason field, type the reason why you're powering on the server module.
This text can be up to 1,024 characters long and can contain any valid UTF-8
characters, including white space.
4. Click Power On.
¡ Insert the end of a paper clip into the hole for the recessed power button in the
bottom right corner of the module and then gently press the paper clip against the
power button for two seconds.
3. When you see this prompt, press Delete:
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module again.
4. When prompted for a password, enter the BIOS password for S10 Nodes.
7. On the USB Configuration page, arrow down to USB Mass Storage Driver Support.
Then press Enter.
8. In the USB Mass Storage Driver Support popup, arrow down to Enabled. Then press
Enter.
11. On the Save & Exit page, arrow down to Save Changes and Reset. Then press Enter.
You have approximately ten seconds to do this. If you wait too long, the server module
boots using the installed OS. In this case, you need to reboot the server module again.
14. When prompted for a password, enter the BIOS password for S10 Nodes.
15. In the Please select boot device menu, arrow down to the entry for the installation
USB flash drive or DVD. Then press Enter. (The name of the drive may differ from the
name shown in the figure below.)
Caution: If the R option is not present, enter e to exit the installation. Then contact your
HCP support center for help.
Do not enter i. Option I destroys all data stored on the S Series Node.
4. Respond to the next series of prompts with the applicable values for the customer-
specific S Series Node configuration. The table below lists the settings you are prompted
for, along with their default values. For prompts that have a default value in square
brackets, you can press Enter to accept that value.
Prompt Value
Distributor SSH key Name of the distributor key for the S Series
Node
The default is Atlantic.
(Continued)
Prompt Value
Access network IPv4 address Access network IPv4 address for the server
module on the left
For the server module on the left, the
default is 10.0.0.1.
For the server module on the right, the
default is 10.0.0.1.
Access network IPv4 gateway address Access network IPv4 gateway address
The default is 10.0.0.254
Access network IPv6 address Access network primary IPv6 address for the
server module
No default
Access network IPv6 prefix length Access network primary IPv6 prefix length
The default is 64.
Access network IPv6 gateway address Access network primary IPv6 gateway
address
No default
Access network secondary IPv6 address Access network secondary IPv6 address for
the server module
No default
Access network secondary IPv6 prefix Access network secondary IPv6 prefix
length length
Access network secondary IPv6 gateway Access network secondary IPv6 gateway
address address
No default
(Continued)
Prompt Value
Management network IPv4 address Management network IPv4 address for the
server module
For the server module on the left, the
default is 10.2.2.1.
Management network IPv6 prefix length Management network primary IPv6 prefix
length
The default is 64.
Server interconnect network IPv4 subnet Server interconnect network IPv4 subnet
The default is 10.1.1.0.
(Continued)
Prompt Value
The OS reinstallation starts on the server module. When the reinstallation finishes, the
server module automatically reboots.
5. After the server module starts to reboot, take one of these actions, as applicable:
¡ If the HCP S Series OS and software installation files are on a USB flash drive,
remove that drive from the USB port on the server module.
¡ If the HCP S Series OS and software installation files are on a DVD, remove the
DVD from the external DVD drive. Then disconnect the DVD drive from the USB
port on the server module.
6. Disconnect the USB-to-mini-USB cable from your laptop computer and from the the
server module.
6. Enter this command to reinstall the HCP S Series software on the server module you're
recovering:
./genesis.py server-module-recovery --target ip-address
In this command, ip-address is the server interconnect network IP address of the server
module you're recovering. For server module 1, the fourth octet in this IP address is 1.
For server module 2, the fourth octet is 2.
The HCP S Series software installation program starts on the server module. The
program first performs a series of prechecks. If all the prechecks are successful, the
program reinstalls the software on the server module.
When the software reinstallation finishes, the server module you're recovering
automatically reboots.
For information about the reinstallation prechecks and what to do in case of a software
reinstallation failure, see "Software reinstallation prechecks" on page 91.
Similarly, while a procedure is in progress, you can use the management API to retrieve
information about it. You can also use management API to retrieve the maintenance
procedure history list.
To add, remove, or replace data or database drives, you need to open the enclosure cover.
When you open the cover, an alarm sounds intermittently. If the cover remains open for more
than five minutes, the alarm becomes continuous. The alarm stops when the cover is closed.
Important: Do not perform multiple hardware maintenance procedures at the same time (for
example, replacing the enclosure while adding data drives). Doing so can have unpredictable
results.
• Instructions for removing data and database drives from an S10 Node
• You cannot remove or replace more than one database drive in a single operation.
However, in an operation in which you remove or replace a database drive, you can also
remove or replace one or more data drives.
• While the database drive in slot 29,30, or 31 is unavailable, you cannot remove or
replace the database drive in another one of those slots. However, you can remove or
replace the database drive in slot 32.
• While the database drive in slot 32 is unavailable, you can remove or replace the
database drive in only one of slots 29,30, and 31.
• If two or more database drives have failed, one or more of these drives may not appear in
the list of drives that are eligible to be removed or replaced. If this happens:
¡ For a remove drives procedure, first remove the database drives that appear in the
list. Then add new database drives to replace the drives you removed. Wait for the
S10 Node to finish synchronizing the new database drives with the existing
database drives. Then repeat the remove drives procedure to remove the remaining
failed database drives.
¡ For a replace drives procedure, first replace the database drives that appear in the
list. Wait for the S10 Node to finish synchronizing the new database drives with the
existing database drives. Then repeat the replace drives procedure to replace the
remaining failed database drives.
• If, while a new or replacement database drive is being synchronized, the database drive
with which it is being synchronized fails, the new or replacement drive also fails, and the
status of the new or replacement drive changes to "Database drive not synchronized."
To recover from this situation, use maintenance procedures to remove the replacement
drive and then add it back again.
DANGER: Do not leave more than one enclosure extended out of the rack at the
same time. With more than one enclosure extended, the rack can tip over, causing
personal injury to people in the area.
Tip: At any time during the maintenance procedure, you can enter new notes or edit
existing notes in the Notes field on the procedure page. To do this:
1. Click in the Notes field.
2. Type or edit the text you want. This text can be up to 1,024 characters long and can
contain any valid UTF-8 characters, including white space.
3. Click Save Notes.
7. In the Select Slots list, select the slots into which you want to insert new drives. The list
includes only empty slots. It does not include slots identified as having unavailable
drives.
Selecting the box at the left end of the heading row for the slot list selects all slots in the
list, not just the slots that are currently visible. Similarly, deselecting the box deselects
all slots in the list.
8. Click Continue.
The blue identity LED on the front of the enclosure starts blinking. Additionally, the blue
LED at the bottom of each selected slot starts blinking.
9. Optionally, click Enclosure Diagram to display a diagram that shows the layout of the
slots in the enclosure. In this diagram, the slots you selected have a dark gray outline.
Tip: To ensure that you act on the correct slots, print the enclosure diagram and bring it
with you to the S10 Node.
¡ If the enclosure is secured to the rack with two small silver locking screws on each
side of the front of the enclosure, using the #1 Phillips screwdriver, unscrew those
screws until they are very loose. The screws do not come out.
2. Push in the tabs on both sides of the front of the enclosure behind the front panel and pull
the enclosure forward. Then slide the enclosure out of the rack as far as the enclosure
goes.
3. Lift both latches on the top of the enclosure and pull up to open the right side of the
enclosure cover.
4. Push down on the two release latches for the left side of the enclosure cover. Then pull
up that side to open it.
The slots you selected for the drive addition procedure have a blinking blue LED at the
bottom. Be sure to insert drives into only those slots.
If a selected slot in a half-populated enclosure contains an empty drive placeholder, the
blinking blue LED is under the placeholder. Placeholders would be in slots 27, 28, 33,
and 34.
If the LEDs for one or more selected slots are not blinking, the cause may be that those
LEDs are not functioning properly.
5. For each drive you want to insert into a slot:
a. If the slot contains an empty drive placeholder, with your finger hooked under the rim
of the hole in the top of the placeholder, pull the placeholder straight up out of the
slot.
b. Push back on the tab on the top of the drive.
The latch on the top of the drive pops open.
Caution: Do not pull the latch open past a 45-degree angle. Doing so will damage
the latch.
c. Slide the drive all the way down into the slot.
d. Close the latch, pushing down until it clicks into place.
After a few seconds, the blue LED on the top of the drive starts blinking, but at a slower
rate than that at which the LED at the bottom of the slot was blinking before the drive
was inserted into the slot.
If you remove or replace the drive in a slot that was not selected, the S10 Node
immediately starts the repair process to ensure that no data is lost and that all data is still
protected.
6. Close the left side of the enclosure cover.
7. Close the right side of the enclosure cover, pushing down until it clicks into place.
8. While pushing back the green plastic sliders at the front end of the rails on both sides of
the enclosure, push back on the enclosure to release it. Then slide the enclosure all the
way back into the rack until the enclosure clicks into place.
9. Take either of these actions:
¡ If the enclosure was secured to the rack with four medium black screws, using the
#2 Phillips screwdriver, screw those screws back into the four screw holes on the
front of the enclosure (two on each side).
¡ If the enclosure was secured to the rack with two small silver locking screws on
each side of the front of the enclosure, using the #1 Phillips screwdriver, screw
those screws back in.
2. To allow the S10 Node time to recognize the new drives, ensure that at least 30 seconds
have elapsed since the last new drive was inserted into a slot.
3. On the Add Drives page, click Done.
4. On the Add Drives page, click Verify.
The S10 Node checks that all the selected slots have acceptable drives in them. It also
checks that no drives have been inserted into slots that were not selected for the
procedure.
Note: If you click Cancel instead of or after clicking Verify, the drive addition procedure
ends, and the S10 Node treats all the newly inserted drives as failed drives. To recover
from this, use the drive removal procedure to remove the failed drives. Then use the
drive addition procedure to insert drives into the slots from which you removed the failed
drives.
For each selected slot, each unselected slot into which a drive was inserted, and each
unselected slot in which the drive was removed or replaced, the Management Console
displays a success, warning, or error message, as applicable. These messages appear
below the Verify button.
If error messages identify any slots as containing a drive with a duplicate SAS address
or a drive that doesn't have a valid SAS address, the drives in those slots are not added
to the S10 Node. To recover from this situation, follow the instructions in "SAS address
errors" on the next page.
If any drives inserted into any of the selected slots were previously used in an S Series
Node, an Actions section appears below the messages. This section lists each such
slot with the actions you can choose to take for that slot:
¡ If the drive is a native drive (that is, it was previously used in the current S10 Node,
and the HCP S Series software has not been reinstalled since the drive was
removed), the possible actions are:
– Reuse — Use the drive as is. This action is useful, for example, if the drive was
previously unintentionally removed from the enclosure. In this case, choosing
Reuse may facilitate repairs.
– Format — Format the drive and then use it. All existing data on the drive will be
lost.
¡ If the drive is a foreign drive (that is, it was previously used in a different S Series
Node or in the current S10 Node before the HCP S Series software was reinstalled),
the possible actions are:
– Format — Format the drive and then use it. All existing data on the drive will be
lost.
– Discard — Treat the drive as a failed drive. The drive can then be removed in a
new maintenance procedure and possibly reused in the S Series Node from
which it was originally removed.
If any slots that were selected for the procedure do not have drives in them, the slots
return to their previous empty state.
If any drives were removed from unselected slots, the S10 Node treats those slots as
having unavailable drives.
The S10 Node treats the following as failed drives:
¡ These types of drives inserted into selected slots:
– Previously used drives on which the S10 Node had detected errors
– Drives that are bad to start with
– Drives of an unsupported type
¡ Drives inserted into unselected slots
5. If the Actions section is present:
a. For each listed slot, select the action you want to take on the drive in that slot. To
select all the actions under the Format or Other heading, select the applicable
heading.
b. Click Confirm.
c. In the window with the confirming message, click Confirm.
The S10 Node takes the applicable action for each drive.
6. On the Add Drives page, click End Procedure.
When you add database drives, the S10 Node immediately begins synchronizing the new
drives with the existing database drives. This process takes approximately 20 minutes,
during which time, the software OK LED on the front of the enclosure blinks green and the
drive fault LED on the front of the enclosure blinks amber. When synchronization is
complete, the LED turns off.
The synchronization process has no effect on use of the S10 Node.
Note: Processing drives for reuse can take some time. After an add drives procedure is
complete, the S10 Node may continue to report drive additions or drive addition failures until
all the drives selected for reuse have been processed. This behavior is expected. While the
drives are being processed, you can continue to perform maintenance procedures, including
removing and adding back drives that failed to be added.
DANGER: Do not leave more than one enclosure extended out of the rack at the
same time. With more than one enclosure extended, the rack can tip over, causing
personal injury to people in the area.
Tip: At any time during the maintenance procedure, you can enter new notes or edit
existing notes in the Notes field on the procedure page. To do this:
1. Click in the Notes field.
2. Type or edit the text you want. This text can be up to 1,024 characters long and can
contain any valid UTF-8 characters, including white space.
3. Click Save Notes.
6. In the Select Slots list, select the slots from which you want to remove drives. The list
includes only slots containing drives that are eligible to be removed. All the slots in the
list are selected by default, except for those containing database drives associated with
an unavailable server module.
Selecting the box at the left end of the heading row for the slot list selects all slots in the
list, not just the slots that are currently visible. Similarly, deselecting the box deselects
all slots in the list.
7. Click Continue.
The blue identity LED on the front of the enclosure starts blinking. Additionally, the blue
LED on the top of the hard disk drive in each selected slot starts blinking.
8. Optionally, click Enclosure Diagram to display a diagram that shows the layout of the
slots in the enclosure. In this diagram, the slots you selected have a dark gray outline.
Tip: To ensure that you act on the correct slots, print the enclosure diagram and bring it
with you to the S10 Node.
¡ If the enclosure is secured to the rack with two small silver locking screws on each
side of the front of the enclosure, using the #1 Phillips screwdriver, unscrew those
screws until they are very loose. The screws do not come out.
2. Push in the tabs on both sides of the front of the enclosure behind the front panel and pull
the enclosure forward. Then slide the enclosure out of the rack as far as the enclosure
goes.
3. Lift both latches on the top of the enclosure and pull up to open the right side of the
enclosure cover.
4. Push down on the two release latches for the left side of the enclosure cover. Then pull
up that side to open it.
The drives that you selected for removal have a blinking blue LED on the top. Drives that
have failed but that have not been selected for removal have a solid blue LED on the top.
Be sure to remove only the drives that have a blinking blue LED.
If the LEDs for one or more selected drives are not blinking, the cause may be that those
LEDs are not functioning properly.
5. For each drive you want to remove:
a. Push back on the tab on the top of the drive.
The latch on the top of the drive pops open.
Caution: Do not pull the latch open past a 45-degree angle. Doing so will damage
the latch.
2. To allow the S10 Node time to recognize the drive removal, ensure that at least 30
seconds have elapsed since the last drive was removed from a slot.
3. On the Remove Drives page, click Done.
4. On the Remove Drives page, click Verify.
The S10 Node checks that the drives have been removed from all the selected slots. It
also checks that no drives have been removed from slots that were not selected for the
procedure.
For each selected slot, each unselected slot into which a drive was inserted, and each
unselected slot in which the drive was removed or replaced, the Management Console
displays a success, warning, or error message, as applicable. These messages appear
below the Verify button.
The S10 Node continues to treat drives that were not removed from selected slots as
failed or unavailable, as applicable. It also treats the following as failed drives:
¡ Drives inserted into selected slots
¡ Drives inserted into unselected slots
If any drives were removed from unselected slots, the S10 Node treats those slots as
having unavailable drives.
5. On the Remove Drives page, click End Procedure.
An S10 Node can operate correctly with multiple failed data drives, so you don't need to
replace failed data drives as soon as they fail. The S10 Node issues an alert when the
number of failed data drives reaches the threshold at which the drives must be replaced. This
threshold depends on the total number of data drives in the S10 Node.
You should always immediately replace failed database drives.
Before replacing a data or database drive, ensure that the replacement drive is of a supported
type for the drive you're replacing. For details, see HCP S Series Node 2.2.0 Release Notes.
DANGER: Do not leave more than one enclosure extended out of the rack at the
same time. With more than one enclosure extended, the rack can tip over, causing
personal injury to people in the area.
Tip: At any time during the maintenance procedure, you can enter new notes or edit
existing notes in the Notes field on the procedure page. To do this:
1. Click in the Notes field.
2. Type or edit the text you want. This text can be up to 1,024 characters long and can
contain any valid UTF-8 characters, including white space.
3. Click Save Notes.
7. In the Select Slots list, select the slots in which you want to replace the drives. The list
includes only slots containing drives that are eligible to be replaced. All the slots in the
list are selected by default, except for those containing database drives associated with
an unavailable server module.
Selecting the box at the left end of the heading row for the slot list selects all slots in the
list, not just the slots that are currently visible. Similarly, deselecting the box deselects
all slots in the list.
8. Click Continue.
The blue identity LED on the front of the enclosure starts blinking. Additionally, the blue
LED on the top of each selected slot starts blinking.
9. Optionally, click Enclosure Diagram to display a diagram that shows the layout of the
slots in the enclosure. In this diagram, the slots you selected have a dark gray outline.
Tip: To ensure that you act on the correct slots, print the enclosure diagram and bring it
with you to the S10 Node.
¡ If the enclosure is secured to the rack with two small silver locking screws on each
side of the front of the enclosure, using the #1 Phillips screwdriver, unscrew those
screws until they are very loose. The screws do not come out.
2. Push in the tabs on both sides of the front of the enclosure behind the front panel and pull
the enclosure forward. Then slide the enclosure out of the rack as far as the enclosure
goes.
3. Lift both latches on the top of the enclosure and pull up to open the right side of the
enclosure cover.
4. Push down on the two release latches for the left side of the enclosure cover. Then pull
up that side to open it.
The drives that you selected for replacement have a blinking blue LED on the top. Drives
that have failed but that have not been selected for replacement have a solid blue LED on
the top. Be sure to replace only the drives that have a blinking blue LED.
If the LEDs for one or more selected drives are not blinking, the cause may be that those
LEDs are not functioning properly.
5. For each drive you want to replace:
a. Remove the old drive:
1. Push back on the tab on the top of the drive.
The latch on the top of the drive pops open.
Caution: Do not pull the latch open past a 45-degree angle. Doing so will
damage the latch.
After a few seconds, the blue LED at the bottom of the slot starts blinking at a slower
rate than the rate at which the LED on the top of the drive was blinking before the
drive was removed from the slot.
b. Insert the new drive:
1. Push back on the tab on the top of the drive.
The latch on the top of the drive pops open.
2. Slide the drive all the way down into the slot.
3. Close the latch, pushing down until it clicks into place.
After a few seconds, the blue LED on the top of the drive starts blinking.
If you remove or replace the drive in a slot that was not selected, the S10 Node
immediately starts the repair process to ensure that no data is lost and that all data is still
protected.
6. Close the left side of the enclosure cover.
7. Close the right side of the enclosure cover, pushing down until it clicks into place.
8. While pushing back the green plastic sliders at the front end of the rails on both sides of
the enclosure, push back on the enclosure to release it. Then slide the enclosure all the
way back into the rack until the enclosure clicks into place.
9. Take either of these actions:
¡ If the enclosure was secured to the rack with four medium black screws, using the
#2 Phillips screwdriver, screw those screws back into the four screw holes on the
front of the enclosure (two on each side).
¡ If the enclosure was secured to the rack with two small silver locking screws on
each side of the front of the enclosure, using the #1 Phillips screwdriver, screw
those screws back in.
2. To allow the S10 Node time to recognize the replacement drives, ensure that at least 30
seconds have elapsed since the last replacement drive was inserted into a slot.
3. On the Replace Drives page, click Done.
Note: If you click Cancel instead of or after clicking Verify, the drive replacement
procedure ends, and the S10 Node treats all the replaced drives as failed drives. To
recover from this, use the drive removal procedure to remove the failed drives. Then use
the drive addition procedure to insert drives into the slots from which you removed the
failed drives.
For each selected slot, for each unselected slot into which a drive was inserted, and for
each unselected slot in which the drive was removed or replaced, the Management
Console displays a success, warning, or error message, as applicable. These messages
appear below the Verify button.
If error messages identify any slots as containing a drive with a duplicate SAS address
or a drive that doesn't have a valid SAS address, the drives in those slots are not added
to theS10 Node. To recover from this situation, follow the instructions in "SAS address
errors" on page 121.
If any drives inserted into any of the selected slots were previously used in an S Series
Node, an Actions section appears below the messages. This section lists each such
slot with the actions you can choose to take for that slot:
¡ If the drive is a native drive (that is, it was previously used in the current S10 Node,
and the HCP S Series software has not been reinstalled since the drive was
removed), the possible actions are:
– Reuse — Use the drive as is. This action is useful, for example, if the drive was
previously unintentionally removed from the enclosure. In this case, choosing
Reuse may facilitate repairs.
– Format — Format the drive and then use it. All existing data on the drive will be
lost.
¡ If the drive is a foreign drive (that is, it was previously used in a different S Series
Node or in the current S10 Node before the HCP S Series software was reinstalled),
the possible actions are:
– Format — Format the drive and then use it. All existing data on the drive will be
lost.
– Discard — Treat the drive as a failed drive. The drive can then be removed in a
new maintenance procedure and possibly reused in the S Series Node from
which it was originally removed.
The S10 Node continues to treat drives that were not replaced in selected slots as failed
or unavailable, as applicable. It also treats these as failed drives:
¡ These types of drives inserted into selected slots:
– Previously used drives on which the S10 Node had detected errors
– Drives that are bad to start with
– Drives of an unsupported type
¡ Drives inserted into unselected slots
If any drives were removed from selected or unselected slots and not replaced, those
slots are treated as having unavailable drives.
5. If the Actions section is present:
a. For each listed slot, select the action you want to take on the drive in that slot. To
select all the actions under the Format or Other heading, select the applicable
heading.
b. Click Confirm.
c. In the window with the confirming message, click Confirm.
The S10 Node takes the applicable action for each drive.
6. On the Replace Drives page, click End Procedure.
When you replace database drives, the S10 Node immediately begins synchronizing the new
drives with the existing database drives. This process takes approximately 20 minutes,
during which time, the software OK LED on the front of the enclosure blinks green and the
drive fault LED on the front of the enclosure blinks amber. When synchronization is
complete, the LED turns off.
The synchronization process has no effect on use of the S10 Node.
Note: Processing drives for reuse can take some time. After a replace drives procedure is
complete, the S10 Node may continue to report drive additions or drive addition failures until
all the drives selected for reuse have been processed. This behavior is expected. While the
drives are being processed, you can continue to perform maintenance procedures, including
removing and adding back drives that failed to be added.
Important: Do not perform multiple hardware maintenance procedures at the same time (for
example, replacing a server module while replacing the enclosure). Doing so can have
unpredictable results.
• A #1 Phillips screwdriver
Step 1: Remove the old power and cooling module from the enclosure
To remove the old power and cooling module from the enclosure:
1. Use the rocker switch on the power and cooling module to power the module off.
2. While pushing back the small tab on the bottom of the clasp on the power cable lock, pull
the clasp away from the enclosure.
3. Press down on the tab on the side of the clasp to open the clasp.
4. Unplug the power cable from the power and cooling module, and move the cable out of
the clasp.
5. Using the #1 Phillips screwdriver, unscrew the large black screw in the top center of the
power and cooling module until the screw is very loose. The screw does not come out.
6. Pull down the handle that runs across the top of the power and cooling module. The
handle stops midway down the module.
7. Use the handle to pull the power and cooling module all the way out of the enclosure.
Step 2: Insert the new power and cooling module into the enclosure
To insert the new power and cooling module into the enclosure:
1. Ensure that the handle on the new power and cooling module is lowered as far as the
handle goes.
2. Slide the power and cooling module all the way back into the power and cooling module
slot in the enclosure. When the module reaches the back of the enclosure, the handle
pops up slightly.
3. Raise the handle all the way to the top of the power and cooling module.
4. Using the #1 Phillips screwdriver, screw in the large black screw in the top center of the
power and cooling module. Do not overtighten the screw.
5. Plug the power cable into the power and cooling module.
6. Fit the power cable into the clasp on the power cable lock.
7. Push down on the top of the clasp to close it.
8. Push the clasp back toward the enclosure as far as the clasp goes to lock the power
cable plug in place.
9. Use the rocker switch on the power and cooling module to power the module on.
What you need for replacing the front panel and light pipe assembly
For replacing the front panel and light pipe assembly on an enclosure, you need:
• A #1 Phillips screwdriver. You need this item only if the enclosure is secured to the rack
with small silver locking screws.
• A #2 Phillips screwdriver.
Step 1: Remove the old front panel and light pipe assembly from the
enclosure
To remove the old front panel and light pipe assembly on the enclosure:
1. Take either of these actions:
¡ If the enclosure is secured to the rack with two medium black screws on each side
of the front of the enclosure, using the #2 Phillips screwdriver, unscrew and remove
those screws.
¡ If the enclosure is secured to the rack with two small silver locking screws on each
side of the front of the enclosure, using the #1 Phillips screwdriver, unscrew those
screws until they are very loose. The screws do not come out.
2. Push in the tabs on both sides of the front of the enclosure behind the front panel and pull
the enclosure forward. Then slide the enclosure out of the rack as far as the enclosure
goes.
3. Using the #2 Phillips screwdriver, unscrew and remove the silver self-tapping screw in
each of the four corners on the back of the front of the enclosure.
4. Pull the front panel off the front of the enclosure.
5. Gently push the top of the light pipe assembly to the left and rock it to the left to free the
right side. Then push the light pipe assembly to the right side and pull the left side out of
the enclosure.
Step 2: Attach the new front panel and light pipe assembly to the
enclosure
To attach the new new front panel and light pipe assembly to the enclosure:
1. On the front of the enclosure, insert the left side of the top of the new light pipe assembly
into the bottom rectangular hole above the left column of LEDs . Then push the light pipe
assembly to the left side and insert the right side of the top of the assembly into the
bottom rectangular hole above the right column of LEDs.
2. Align the eight pipes in the light pipe assembly with the eight LEDs on the front of the
enclosure.
3. Fit the new front panel onto the front of the enclosure so that:
¡ The pins in the four corners of the front of the enclosure fit into the holes in the four
corners of the back of the front panel
¡ The eight pipes in light pipe assembly fit into the eight round holes in the bottom
center of the front of the front panel
4. Using the #2 Phillips screwdriver, screw the four silver self-tapping screws that come
with the new front panel and light pipe assembly into the holes in the four corners in the
back of the front of the enclosure. Do not overtighten the screws.
5. While pushing back the green plastic sliders at the front end of the rails on both sides of
the enclosure, push back on the enclosure to release it. Then slide the enclosure all the
way back into the rack until the enclosure clicks into place.
6. Take either of these actions:
¡ If the enclosure was secured to the rack with four medium black screws, using the
#2 Phillips screwdriver, screw those screws back into the four screw holes on the
front of the enclosure (two on each side).
¡ If the enclosure was secured to the rack with two small silver locking screws on
each side of the front of the enclosure, using the #1 Phillips screwdriver, screw
those screws back in.
Tip: To keep the S10 Node available when you need to replace both cable management
arms, replace the arms one at a time.
• A new cable management arm kit, which contains both the left and right cable
management arms.
Cable management arm part number: ASM-01659-01-A.X
• A #1 Phillips screwdriver.
• A #2 Phillips screwdriver. You need this item only if the enclosure is secured to the rack
using medium black screws.
• A paper clip.
Step 2: Disconnect the power and Ethernet cables and detach the old
cable management arm
To disconnect the power and Ethernet cables from the enclosure and detach the cable
management arm you want to replace:
1. Disconnect the power cable that's connected to the power and cooling module on the
side of the enclosure on which you're replacing the cable management arm. To do this:
a. While pushing back the small tab on the bottom of the clasp on the power cable lock,
pull the clasp away from the enclosure.
b. Press down on the tab on the side of the clasp to open the clasp.
c. Unplug the power cable from the power and cooling module, and move the cable out
of the clasp.
2. Disconnect the Ethernet cables from the two access ports and from the management
port on the server module on the side of the enclosure on which you're replacing the
cable management arm.
3. Using the #1 Phillips screwdriver, loosen the large black screw that attaches the cable
management arm to the middle of the enclosure until the screw is very loose. The screw
does not come out.
4. Pull the cable management arm away from the enclosure, allowing the small stud on the
left arm to slide out of the slot on the right arm.
5. Take either of these actions:
¡ If the enclosure is secured to the rack with two medium black screws on each side
of the front of the enclosure, using the #2 Phillips screwdriver, unscrew and remove
those screws.
¡ If the enclosure is secured to the rack with two small silver locking screws on each
side of the front of the enclosure, using the #1 Phillips screwdriver, unscrew those
screws until they are very loose. The screws do not come out.
6. While pushing in the tabs on both sides of the front of the enclosure behind the front
panel, pull the enclosure forward a few inches out of the rack.
7. Using the #1 Phillips screwdriver, unscrew and remove the pin that attaches the cable
management arm to the outer enclosure rail or cable management arm extender, as
applicable, on the applicable side of the rack.
8. Pull the cable management arm away from the outer rail or cable management arm
extender, as applicable.
Step 3: Move the cables from the old cable management arm to the new
cable management arm
To move the cables from the old cable management arm to the new cable management arm:
1. Unwrap and remove the Velcro® strips that are wrapped around the cables in the old
cable management arm. Do not discard these strips. You will reuse them with the new
cable management arm.
2. For each Ethernet cable in the old cable management arm, move the cable out of the
arm, passing the cable out though the opening at the rack end of the arm.
3. Move the power cable out of the old cable management arm, passing the cable out
through the opening at the rack end of the arm.
4. Align the pin holes in the rail end of the new cable management arm with the pin holes at
the rear of the outer enclosure rail or cable management arm extender, as applicable.
5. Using the #1 Phillips screwdriver, screw in the pin that's used to attach the cable
management arm to the outer rail or to the cable management arm extender, as
applicable.
6. Pass the power cable through the opening at the rack end of the new cable management
arm, going from the outer side of the arm to the inner side of the arm.
7. Press the power cable into the inner side of cable management arm, starting from the
rack end of the arm.
8. For each Ethernet cable:
a. Pass the cable through the opening at the rack end of the cable management arm,
going from the outer side of the arm to the inner side of the arm.
b. Press the cable into the inner side of the cable management arm, starting from the
rack end of the arm.
9. Slide the enclosure all the way back into the rack until the enclosure clicks into place.
10. Take either of these actions:
¡ If the enclosure was secured to the rack with four medium black screws, using the
#2 Phillips screwdriver, screw those screws back into the four screw holes on the
front of the enclosure (two on each side).
¡ If the enclosure was secured to the rack with two small silver locking screws on
each side of the front of the enclosure, using the #1 Phillips screwdriver, screw
those screws back in.
11. Wrap the Velcro strips from the old cable management arm around the cables in the new
cable management arm. To wrap a Velcro strip:
a. Pass the narrow end of the strip through the bottom slot on the outside of one link of
the cable management arm, letting the strip pass under the cables in the arm and out
the other side.
b. Wrap the strip around the arm once, passing the narrow end of the strip through the
slit in the wide end of the strip and pulling the strip tightly around the link.
c. Wrap the strip tightly around the arm again, passing it through the slot on the bottom
of the link.
d. Press the narrow end of the strip against the wrapped portion of the strip to hold the
strip in place.
Step 4: Attach the new cable management arm and reconnect the
Ethernet and power cables
To attach the new cable management arm and reconnect the Ethernet and power cables:
1. Position the cable management arm so that the small stud on the outside of the
enclosure end of the left arm fits into the corresponding slot in the outside of the
enclosure end of the right arm. Then align the large black screw at the end of the arm with
the applicable hole in the middle of the enclosure.
2. Using the #1 Phillips screwdriver, screw in and tighten the screw. Do not overtighten the
screw.
3. Connect the management network cable to the management port on the applicable
server module.
When inserting the cable into the port, push the cable in until the end clicks into place.
4. Connect each access network cable to the applicable access port on the applicable
server module.
When inserting the cable into the port, ensure that the card in the nonmetal portion of the
connector at the end of the cable is facing the other access port on the same server
module.
5. Plug the power cable into the power and cooling module.
6. Fit the power cable into the clasp on the power cable lock.
7. Push down on the top of the clasp to close it.
8. Push the clasp back toward the enclosure as far as the clasp goes to lock the power
cable plug in place.
9. Use the rocker switch on the applicable power and cooling module to power the module
on.
10. Insert the end of a paper clip into the hole for the recessed power button in the bottom
right corner of the applicable server module. Then gently press the paper clip against the
power button for two seconds.
If the server module does not become available again after you press the power button,
you can power the module back on by sliding it partially out of the enclosure and then
pushing it back in. For instructions on doing this, see "Replacing a server module".
DANGER: Do not leave more than one enclosure extended out of the rack at the
same time. With more than one enclosure extended, the rack can tip over, causing
personal injury to people in the area.
• The new enclosure rail kit (for the contents of this kit, see "Enclosure rail kit" below).
Enclosure rail kit part number: ASM-01658-01-A.X
• A lift.
• A #1 Phillips screwdriver.
• A #2 Phillips screwdriver.
• A ¼-inch flat-head screwdriver. You need this screwdriver only if you're replacing a rail
that attaches to the rack.
Each outer rail has a rail extender attached to it at the rear. The two-inch-wide rail
extenders are used to add length to the outer rails when necessary to accommodate the
depth of the rack.
Each outer rail also has a cable management arm extender attached to it at the rear. The
cable management arm extenders are used to add length to the parts of the outer rails
that the cable management arms attach to. The cable management arm extenders must
be used when the rail extenders are used.
The rail extenders and cable management arm extenders are required for mounting the
enclosure in a Hitachi Universal V2 rack.
B. Two inner rails. These are the rails that attach to the left and right sides of the enclosure.
The green plastic slider in each rail has an arrowhead at the front that points toward the
rear.
The right inner rail comes set into the right outer rail. The left inner rail comes set into the
left outer rail. The rails need to be separated before they can be installed.
C. Two chassis screws. These small silver screws are used to attach the inner rails to the
enclosure.
The package containing the chassis screws also contains four medium black screws.
These screws are used for securing the enclosure to the rack.
D. Four screw plates, with 16 medium silver screws. The screw plates are used to
strengthen the vertical mounting rails in the rack where the outer rails are attached.
E. Two shelf locking brackets, with four medium silver screws. The shelf locking brackets,
which attach to the front vertical mounting rails, have the screw holes for the top locking
screws on the enclosure.
Step 2: Disconnect the power and Ethernet cables and detach the cable
management arms
To disconnect the power and Ethernet cables from the enclosure and detach the cable
management arms:
1. Disconnect the two power cables from the power and cooling modules. To disconnect a
power cable:
a. While pushing back the small tab on the bottom of the clasp on the power cable lock,
pull the clasp away from the enclosure.
b. Press down on the tab on the side of the clasp to open the clasp.
c. Unplug the power cable from the power and cooling module, and move the cable out
of the clasp.
2. Disconnect the Ethernet cables from the two access ports and from the management
port on each server module.
3. Using a #1 Phillips screwdriver, loosen the two large black screws that attach the cable
management arms to the middle of the enclosure until the screws are very loose. The
screws do not come out.
¡ If the enclosure is secured to the rack with two small silver locking screws on each
side of the front of the enclosure, using the #1 Phillips screwdriver, unscrew those
screws until they are very loose. The screws do not come out.
6. Push in the tabs on both sides of the front of the enclosure behind the front panel and pull
the enclosure forward. Then slide the enclosure out of the rack as far as the enclosure
goes.
7. Detach the cable management arms from the outer enclosure rails or cable management
arm extenders on both sides of the rack. To detach a cable management arm:
a. Using the #1 Phillips screwdriver, unscrew and remove the pin that attaches the
cable management arm to the outer rail or to the cable management arm extender,
as applicable.
b. Pull the cable management arm away from the outer rail or cable management arm
extender, as applicable.
¡ For the right outer rail, press up on the latch on the inner rear of the extended part to
release the part. Then push the extended part back into the rack as far as the part
goes.
Step 4: Separate the inner and outer rails in the new enclosure rail kit
In a new enclosure rail kit, the left and right inner rails are set inside the left and right outer
rails, respectively. ou need to separate the inner and outer rails from each other so that you
can attach the inner rails to the enclosure and the outer rails to the rack.
For more information about the enclosure rail kit, see "Enclosure rail kit" on page 143.
To separate an inner rail from its outer rail:
1. Slide the inner rail out of the front end of the outer rail as far as the inner rail goes.
2. While pressing the green tab on extended part of the outer rail, pull the inner rail to
release it from the outer rail. Then slide the inner rail all the way out of the outer rail.
• If only an outer rail is damaged, you need to replace both the inner and outer rails on the
applicable side of the enclosure and rack. (Alternatively, you can remove the old inner rail
from the enclosure and check whether it slides smoothly into the new outer rail. If it does,
you can replace only the outer rail.)
• If only an inner rail is damaged, check whether the new inner rail slides smoothly in the
existing outer rail on the applicable side of the rack. If it slides in smoothly, you need to
replace only the inner rail. If it doesn't, you need to replace both the inner and outer rails.
To check whether the new inner rail slides smoothly in the old outer rail:
1. With the rear of the inner rail aligned with the front of the outer rail, slide the inner rail
back into the outer until it comes to a stop.
2. While pressing the green tab on the inner rail, push the inner rail back to release it.
Then slide the inner rail all the way back into the rack.
3. Slide the inner rail out of the front end of the outer rail as far as the inner rail goes.
4. While pressing the green tab on the inner rail, pull the inner rail to release it from the
outer rail. Then slide the inner rail all the way out of the rack.
• If both the inner rail and outer rail are damaged on a given side, you need to replace both
the inner rail and rails on that side.
If the new enclosure rail kit is a different model from the old enclosure rail kit, you need to
replace both the inner and outer rails on both sides of the enclosure and rack.
2. While lifting the silver tab on the old rail, slide the rail toward the front of the enclosure as
far as the rail goes. Then pull the rail off the enclosure.
3. Position the new inner rail on the applicable side of the enclosure so that the green
plastic slider in the rail faces out and the studs on the enclosure fit into the holes in the
rail. Then slide the rail toward the back of the enclosure as far as the rail goes.
4. Using the #1 Phillips screwdriver, screw in and tighten one of the new silver chassis
screws that come in the enclosure rail kit.
Tip: If you have access to the sides of the rack, you can adjust the sliding panel when
you install the rail in the rack instead of adjusting it ahead of time. However, you still
need to remove the rail extender if necessary and loosen the six silver screws that hold
the sliding panel in place before you install the rail.
The sliding panel has a small circular window that shows the length of the rail from the
front of the rail to the rear of the sliding panel, in inches. The panel has a small oblong
window that shows lines indicating ⅛-inch intervals.
For a Hitachi Universal V2 rack, set the total length of the rail plus the two-inch rail
extender to just under 29 inches. At this length, some portion of the number 27 shows in
the small circular window. (A Hitachi Universal V2 rack requires the use of the rail
extenders.)
b. Adjust the position of the sliding panel to extend the rail to the length you want.
c. Using the ¼-inch flat-head screwdriver, tighten the six silver screws.
The rail extender is attached to the rear of the sliding panel. To remove the rail extender,
using the #2 Phillips screwdriver, unscrew and remove the four medium silver screws
that attach the extender to the rail.
Tip: To ensure a proper fit, measure the rail against the rack before attaching it to the
rack. You may need to adjust the length of the rail more than once to get it right.
3. If you removed the rail extender from the rail, remove the cable management arm
extender from the rail. To do this, using the #1 Phillips screwdriver, unscrew and remove
the pin that attaches the extender to the rail.
4. Attach the new rail to the front and rear of the rack. To do this:
¡ At the front of the rack:
1. On the front of the vertical mounting rail, position one of the screw plates so that
the bottom of the plate is aligned with the bottom of the bottom rack unit for the
enclosure.
2. On the back of the vertical mounting rail, position the outer enclosure rail such
that the bottom of the lip at the front of the outer rail is aligned with the bottom of
the screw plate and the PEM fastener on the lip sits in the sixth hole down from
the top of the screw plate.
3. Using the #2 Phillips screwdriver, screw four of the medium silver screws that
come with the screw plates into the screw plate. Start with the hole that's fourth
from the top of the screw plate, then do the bottom hole in the screw plate, and
finally do the holes that are sixth and seventh from the top of the screw plate.
Tighten the screws.
¡ At the rear of the rack:
1. On the front of the vertical mounting rail, position one of the screw plates such
that the bottom of the plate is aligned with the bottom of the bottom rack unit for
the enclosure.
2. On the back of the vertical mounting rail, position the outer enclosure rail such
that the bottom of the lip at the rear of the outer rail is aligned with the bottom of
the screw plate.
3. Using the #2 Phillips screwdriver, screw four of the medium silver screws that
come with the screw plates into the screw plate. Start with the hole that's fourth
from the top of the screw plate, then do the bottom hole in the screw plate, and
finally do the holes that are sixth and seventh from the top of the screw plate.
Tighten the screws.
5. Attach one shelf locking bracket to the each of the vertical mounting rails at the front of
the rack. To attach a shelf locking bracket:
a. On the back of the vertical mounting rail, position the shelf locking bracket so that:
– The top of the bracket is a little above the top of the screw plate
– The PEM® fastener on the bracket sits in the second hole down from the top of
the screw plate
– The two holes in the bracket are aligned with the top hole in the screw plate and
third hole down from the top of the screw plate
b. Using the #2 Phillips screwdriver, screw two of the medium silver screws that come
with the shelf locking brackets into the top hole of the screw plate and the third hole
down from the top of the screw plate. Tighten the screws.
Step 9: Reattach the cable management arms and reconnect the Ethernet
and power cables
To reattach the cable management arms and reconnect the Ethernet and power cables:
1. While pushing in the tabs on both sides of the front of the enclosure behind the front
panel, pull the enclosure forward a few inches out of the rack.
2. Using the #1 Phillips screwdriver, unscrew and remove the pin that's used to attach the
left cable management arm (when viewed from the rear of the rack) to the rear of the left
outer enclosure rail or cable management arm extender, as applicable.
3. Align the pin holes in the rail end of the left cable management arm with the pin holes in
the left outer rail or cable management arm extender, as applicable.
4. Using the #1 Phillips screwdriver, screw in the pin that attaches the left cable
management arm to the left outer rail or cable management arm extender, as applicable.
5. Repeat steps 2 through 4 with the right cable management arm.
6. Slide the enclosure all the way back into the rack until the enclosure clicks into place.
7. Take either of these actions:
¡ If the enclosure was secured to the rack with four medium black screws, using the
#2 Phillips screwdriver, screw those screws back into the four screw holes on the
front of the enclosure (two on each side).
¡ If the enclosure was secured to the rack with two small silver locking screws on
each side of the front of the enclosure, using the #1 Phillips screwdriver, screw
those screws back in.
8. Align the large black screw at the enclosure end of the left cable management arm with
the lefthand hole in the middle of the enclosure.
9. Using the #1 Phillips screwdriver, screw in and tighten the screw. Do not overtighten the
screw.
10. Position the right cable management arm so that the small stud on the outside of the left
arm fits into the corresponding slot on the outside of the right arm. Then align the large
black screw at the end of the right cable management arm with the righthand hole in the
middle of the enclosure.
11. Using the #1 Phillips screwdriver, screw in and tighten the screw. Do not overtighten the
screw.
12. Connect each management network cable to the management port on the applicable
server module.
When inserting the cables into the ports, push the cables in until they click into place.
13. Connect each access network cable to the applicable access port on the applicable
server module.
When inserting the cable into the port, ensure that the card in the nonmetal portion of the
connector at the end of the cable is facing the other access port on the same server
module.
14. Plug the power cable back into the applicable power and cooling module.
15. Fit the power cable into the clasp on the power cable lock.
16. Push down on the top of the clasp to close it.
17. Push the clasp back toward the enclosure as far as the clasp goes to lock the power
cable plug in place.
18. Use the rocker switches on the power and cooling modules to power both modules on.
When replacing the enclosure for an S10 Node, you need to replace the righthand cover of
the new enclosure with the righthand cover of the old enclosure. This is because the serial
number for the S10 Node is on a label attached to that cover.
After replacing the enclosure, you need to check the firmware versions on the SAS
expanders in the new enclosure. If any expanders have the wrong firmware version, you
need to perform a firmware update.
• If the old enclosure is secured to the rack using four small silver locking screws, either of
these:
¡ If the carton containing the new enclosure also contains a package of four medium
black screws, those screws
¡ If the carton containing the new enclosure does not also contain four medium black
screws, four 8-32 x ½" machine screws
• A lift
• A #1 Phillips screwdriver
• A #2 Phillips screwdriver
Tip: At any time during the maintenance procedure, you can enter new notes or edit
existing notes in the Notes field on the procedure page. To do this:
1. Click in the Notes field.
2. Type or edit the text you want. This text can be up to 1,024 characters long and can
contain any valid UTF-8 characters, including white space.
3. Click Save Notes.
Caution: Do not try to move the enclosure by yourself. Doing so can result in personal
injury. Always use at least two people to move an enclosure.
8. If any of the foam packing pieces remain on the enclosure, remove them.
9. Use the lift to move the enclosure to a work surface that has room for both the old and
new enclosures.
10. Remove the two straps from the enclosure.
Step 3: Disconnect the power and Ethernet cables and detach the cable
management arms
To disconnect the power and Ethernet cables from the enclosure and detach the cable
management arms:
1. Use the rocker switches on the power and cooling modules to power both modules off.
2. Disconnect the two power cables from the power and cooling modules. To disconnect a
power cable:
a. While pushing back the small tab on the bottom of the clasp on the power cable lock,
pull the clasp away from the enclosure.
b. Press down on the tab on the side of the clasp to open the clasp.
c. Unplug the power cable from the power and cooling module, and move the cable out
of the clasp.
3. Disconnect the Ethernet cables from the two access ports, the management port, and
the server interconnect port on each server module.
4. Using a #1 Phillips screwdriver, loosen the two large black screws that attach the cable
management arms to the middle of the enclosure until the screws are very loose. The
screws do not come out.
¡ If the enclosure is secured to the rack with two small silver locking screws on each
side of the front of the enclosure, using the #1 Phillips screwdriver, unscrew those
screws until they are very loose. The screws do not come out.
2. Push in the tabs on both sides of the front of the enclosure behind the front panel and pull
the enclosure forward. Then slide the enclosure out of the rack as far as the enclosure
goes.
3. Remove the front panel and light pipe assembly from the enclosure. To do this:
a. Using the #2 Phillips screwdriver, unscrew and remove the silver self-tapping screw
in each of the four corners on the back of the front of the enclosure.
b. Pull the front panel off the front of the enclosure.
c. Gently push the top of the light pipe assembly to the left and rock it to the left to free
the right side. Then push the light pipe assembly to the right side and pull the left side
out of the enclosure.
Set the front panel and light pipe assembly aside. You will need them for the new
enclosure.
4. Position the lift so that the lift platform is below the enclosure.
5. Raise the platform to just below the enclosure, aligning the platform so that the bulk of
the enclosure will rest on it when the enclosure is pulled out of the rack.
6. While pressing in the green tabs on the rails on both sides of the enclosure, pull the
enclosure to release it from the rack. Then pull the enclosure completely out of the rack.
7. Using the lift, move the enclosure to the work surface on which you put the new
enclosure.
8. Slide the extended part of each outer enclosure rail back into the rack:
¡ For the left outer rail, press down on the latch on the inner rear of the extended part to
release the part. Then push the extended part back into the rack as far as the part
goes.
¡ For the right outer rail, press up on the latch on the inner rear of the extended part to
release the part. Then push the extended part back into the rack as far as the part
goes.
Step 5: Move the inner rails from the old enclosure to the new enclosure
Each inner enclosure rail must be on the same side on the new enclosure as it was on the old
enclosure. To ensure that the rails are on the correct sides on the new enclosure, move the
rails one at a time.
To move an inner rail from the old enclosure to the new enclosure:
1. Using the #1 Phillips screwdriver, unscrew and remove the silver chassis screw that
attaches the rail to the enclosure.
2. While lifting the silver tab on the rail, slide the rail toward the front of the enclosure as far
as the rail goes. Then remove the rail from the enclosure.
3. Position the rail on the applicable side of the new enclosure so that the green plastic
slider in the rail faces out and the studs on the enclosure fit into the holes in the rail. Then
slide the rail toward the back of the enclosure until it clicks into place.
4. Using the #1 Phillips screwdriver, screw in and tighten the small silver chassis screw.
Step 6: Move the power and cooling modules from the old enclosure to
the new enclosure
The power and cooling modules do not need to be in the same positions in the new enclosure
as they were in the old enclosure.
To move a power and cooling module from the old enclosure to the new enclosure, with the
rear of each enclosure facing you:
1. Using the #1 Phillips screwdriver, unscrew the large black screw in the top center of the
power and cooling module until the screw is very loose. The screw does not come out.
2. Pull down the handle that runs across the top of the power and cooling module. The
handle stops midway down the module.
3. Use the handle to pull the power and cooling module all the way out of the enclosure.
4. Ensure that the handle on the power and cooling module is lowered as far as the handle
goes.
5. Slide the power and cooling module all the way back into one of the power and cooling
module slots in the new enclosure. When the module reaches the back of the enclosure,
the handle pops up slightly.
6. Raise the handle all the way to the top of the power and cooling module.
7. Using the #1 Phillips screwdriver, screw in the large black screw in the top center of the
power and cooling module. Do not overtighten the screw.
Step 7: Move the server modules from the old enclosure to the new
enclosure
Each server module must be in the same position in the new enclosure as it was in the old
enclosure. To ensure that the modules are in the correct positions in the new enclosure,
move the modules one at a time.
Important: If the server modules are not in the same positions in the new enclosure as they
were in in the old enclosure and you later need to recover the S10 Node, the recovery
operation will cause the S10 Node not to function properly.
To move a server module from the old enclosure to the new enclosure, with the rear of each
enclosure facing you:
1. Using the #1 Phillips screwdriver, unscrew the large black screw in the top center of the
server module until the screw is very loose. The screw does not come out.
2. Pull down the handle that runs across the top of the server module. The handle stops
midway down the module.
3. Use the handle to pull the server module all the way out of the enclosure.
4. Ensure that the handle on the server module is lowered as far as the handle goes.
5. Slide the server module all the way back into the applicable server module slot in the new
enclosure. When the module reaches the back of the enclosure, the handle pops up
slightly.
6. Raise the handle all the way to the top of the server module.
7. Using the #1 Phillips screwdriver, screw in the large black screw in the top center of the
server module. Do not overtighten the screw.
Step 8: Move the data and database drives from the old enclosure to the
new enclosure
Each database drive must be in the same slot in the new enclosure as it was in the old
enclosure. These are the drives in slots 29, 30, 31, and 32 (the two immediately to the left of
the center divider and the two immediately to the right of the center divider, all in the middle
row).
To ensure that the database drives are in the correct slots in the new enclosure, move them
one at a time.
Important: If the database drives are not in the same slots in the new enclosure as they
were in in the old enclosure, the HCP S Series software may not run when you power the
S10 Node back on. If this happens, you need to recover the S10 Node. For more information
about S10 Node recovery, contact your HCP support center.
The data drives also should be in the same positions in the new enclosure as they were in in
the old enclosure.
To move the additional data drives from the old enclosure to the new enclosure:
1. Open the covers on the old and new enclosures. To open an enclosure cover:
a. Lift both latches on the top of the enclosure and pull up to open the right side of the
enclosure cover (when viewed from the front of the enclosure).
b. Push down on the two release latches for the left side of the enclosure cover. Then
pull up that side to open it.
2. Move each data or database drive from the old enclosure to the new enclosure. To move
a data or database drive:
a. Push back on the tab on the top of the drive.
The latch on the top of the drive pops open.
b. Pull up the latch until it clicks into place.
c. Pull up on the latch to slide the drive out of the slot.
d. Slide the drive all the way down into the applicable slot in the new enclosure.
e. Close the latch, pushing down until it clicks into place.
3. Close the left side of the enclosure cover.
Step 9: Swap the right sides of the enclosure covers between the two
enclosures
The serial number label for an S10 Node is on the top right side of the enclosure cover. The
right side of the enclosure cover is the side on the right when the front of the enclosure is
facing you.
To retain the serial number label for the S10 Node, you need to swap the right sides of the
enclosure covers between the old and new enclosures.
To swap the right sides of the enclosure covers between the old enclosure and the new
enclosure:
1. On each enclosure, if the right side of the enclosure cover is not open, lift both latches on
the top of the enclosure and pull up to open the cover.
2. Remove the right side of the enclosure cover from each of the enclosures. To remove the
right side of a cover:
a. Use the ¼-inch nut driver or ¼-inch socket and socket wrench to unscrew and
remove the two small black screws in each of the three hinges on the right side of
the cover (when viewed from the front of the enclosure).
b. Lift the right side of the cover off the enclosure.
3. Swap the removed covers between the old and new enclosures.
4. Attach the swapped covers to the enclosures. To attach the right side of a cover:
a. Place the cover on the enclosure, aligning the holes in the cover hinges with the
holes on the right side of the enclosure.
b. Use the ¼-inch nut driver or ¼-inch socket and socket wrench to screw the small
black screws into the six hinge holes. Tighten the screws.
5. Close the covers on the old and new enclosures. To close an enclosure cover:
a. Close the left side of the enclosure cover.
b. Close the right side of the enclosure cover, pushing down until the cover clicks into
place.
assembly to the left and insert the right side of the top of the assembly into the bottom
rectangular hole above the right column of LEDs.
9. Align the eight pipes in the light pipe assembly with the eight LEDs on the front of the
enclosure.
10. Fit the front panel onto the front of the enclosure so that:
¡ The pins in the four corners of the front of the enclosure fit into the holes in the four
corners of the back of the front panel
¡ The eight pipes in light pipe assembly fit into the eight round holes in the bottom
center of the front of the front panel
11. Using the #2 Phillips screwdriver, screw the four silver self-tapping screws into the holes
in the four corners in the back of the front of the enclosure. Do not overtighten the
screws.
12. While pushing back the green plastic sliders at the front end of the rails on both sides of
the enclosure, push back on the enclosure to release it. Then slide the enclosure all the
way back into the rack until the enclosure clicks into place.
13. Take one of these actions:
¡ If the old enclosure was secured to the rack with four medium black screws, using
the #2 Phillips screwdriver, screw those screws into the four screw holes on the
front of the new enclosure.
¡ If the old enclosure was secured to the rack with four small silver locking screws
and the new enclosure comes with four medium black screws, using the #2 Phillips
screwdriver, screw the medium black screws into the four screw holes on the front
of the new enclosure.
¡ If the old enclosure was secured to the rack with four small silver locking screws
and the new enclosure does not come with medium black screws, screw four 8-32 x
½" machine screws into the screw holes on the front of the new enclosure.
Step 11: Reattach the cable management arms and reconnect the
Ethernet and power cables
To reattach the cable management arms and reconnect the Ethernet and power cables:
1. Align the large black screw at the end of the left cable management arm (when viewed
from the rear of the rack) with the lefthand hole in the middle of the enclosure.
2. Using the #1 Phillips screwdriver, screw in and tighten the screw. Do not overtighten the
screw.
3. Position the right cable management arm so that the small stud on the outside of the left
arm fits into the corresponding slot on the outside of the right arm. Then align the large
black screw at the end of the right cable management arm with the righthand hole in the
middle of the enclosure.
4. Using the #1 Phillips screwdriver, screw in and tighten the screw. Do not overtighten the
screw.
5. Connect the purple server interconnect cable to the server interconnect port on each
server module.
6. Connect each management network cable to the management port on the applicable
server module.
When inserting the cables into the ports, push the cables in until they click into place.
7. Connect each access network cable to the applicable access port in the applicable
server module.
When inserting the cable into the port, ensure that the card in the nonmetal portion of the
connector at the end of the cable is facing the other access port on the same server
module.
8. Plug the power cable into the power and cooling module.
9. Fit the power cable into the clasp on the power cable lock.
10. Push down on the top of the clasp to close it.
11. Push the clasp back toward the enclosure as far as the clasp goes to lock the power
cable plug in place.
Note: If you replace the enclosure and then click Cancel instead of or after clicking
Verify, the enclosure replacement procedure ends unsuccessfully, and the new
enclosure is marked failed.
The VPD tool reports its progress as it checks and, if necessary, tries to fix the VPDs.
The VPD tool then reports the outcome of its processing. Three outcomes are possible:
¡ The tool does not find any errors. The report looks like this:
To continue:
1. Enter exit to close the PuTTY window.
2. From the HCP S Series Management Console, restart server module 2:
a. On the Hardware ► Overview or Hardware ► Maintenance page, click
Power Options.
b. In the Server Module field, select Server Module 2.
c. In the Reason field, type the reason why you're restarting the server
module.
d. Click Restart.
3. Wait for server module 2 to become available.
4. Proceed to Step 14: "Check and, if necessary, update the SAS expander
firmware on the new enclosure" on the next page.
¡ The tool found one or more errors and fixed them all. The report looks something like
this:
To continue, follow the procedure above for when the VPD tool doesn't find any
errors.
¡ The tool found one or more errors it could not fix. The report looks something like
this:
To continue:
1. Using a file transfer tool such as WinSCP, copy the .log file and all the .txt files
from the /opt/rhino/sil/bin/VPDTool directory to the location of your
choice on your laptop computer. HCP support personnel will need these files to
resolve the VPD errors.
Important: Do not perform the SAS expander firmware step after the VPD tool
discovers errors the tool cannot fix.
Step 14: Check and, if necessary, update the SAS expander firmware on
the new enclosure
After replacing the enclosure, you need to check the firmware versions on the SAS
expanders in the new enclosure. If any of the SAS expanders have the wrong firmware
version, you need to perform a firmware update.
The procedure for checking and updating the SAS expander firmware also checks and, if
necessary, updates the firmware on all other components of the S10 Node.
You can perform a firmware update in either of these ways:
• One server module at a time. In this case, while a server module is running the update,
the module is unavailable. The whole process takes approximately ten minutes per
server module. However, during that time, the S10 Node remains available.
• Both server modules at the same time. In this case, the whole process takes
approximately ten minutes. However, during that time, the entire S10 Node is
unavailable.
Checking and updating firmware while the S10 Node remains available
With the procedure in this section, the firmware check and update run on one server module
at a time. The S10 Node remains available during the entire process.
To check and, if necessary, update S10 Node component firmware:
1. If you have not already done so, add the S Series Node service SSH key to your
SSH keychain, as described in "Adding an S Series Node SSH key to your SSH
keychain" on page 35.
2. Use PuTTY to establish an SSH connection to either server module, as described in
"Using a PuTTY SSH connection" on page 36.
3. At the login prompt, enter service.
4. Enter this command to check whether any components have firmware that needs to be
updated:
sudo firmware.py show --check
This command displays a list of all currently installed component firmware versions. The
firmware versions seen by the server module you're logged in to are shown on the left.
The firmware versions seen by the other server module are shown on the right.
The list of the currently installed component firmware versions is followed by a list of the
component firmware that needs to be updated (that is, a list of the components that failed
the firmware version check). A failure statement has this format:
component failed firmware version check- Found: current-version Expected: expected-version
This command shuts down HCP S Series software on the server module you're
logged in to. The firmware update then starts on that module. After the firmware
update is complete, the software restarts on that module.
b. Enter this command to verify that all the necessary updates have been made to the
firmware seen by the server module you're logged in to:
sudo firmware.py show --check --local
If the output from this command includes any failure statements, contact your HCP
support center for help.
If the output from this command does not include any failure statements:
1. Enter this command to determine when startup processing on the server module
is complete:
admin -t 1200 status waitUntilStarted
• While the firmware check is in progress, the S10 Node remains available
4. Enter this command to check whether any components have firmware that needs to be
updated:
sudo firmware.py show --check
This command displays a list of all currently installed component firmware versions. The
firmware versions seen by the server module you're logged in to are shown on the left.
The firmware versions seen by the other server module are shown on the right.
The list of the currently installed component firmware versions is followed by a list of the
component firmware that needs to be updated (that is, a list of the components that failed
the firmware version check). A failure statement has this format:
component failed firmware version check- Found: current-version Expected: expected-version
This command shuts down HCP S Series software on both server modules. The
firmware update then starts on each module. After the firmware update is complete
on a module, the software restarts on that module.
b. Enter this command again to verify that all the necessary firmware updates have
been made:
sudo firmware.py show --check
If the output from this command includes any failure statements, contact your HCP
support center for help.
An S10 Node can take up to ten minutes to become available after a firmware update. If
any components are not running and healthy and at least ten minutes have passed since
the firmware update finished, contact your HCP support center for help.
6. Enter b to return to the HCP S Series Node Service Menu.
7. Enter q to return to the command-line prompt.
8. Enter y to confirm that you want to leave the HCP S Series Node Service Menu.
Important: Do not perform multiple hardware maintenance procedures at the same time (for
example, replacing the enclosure while replacing a server module). Doing so can have
unpredictable results.
This chapter explains how to determine the cause of the problem when a server module is
unavailable. The chapter also contains instructions for:
Degraded OS SSD
The OS SSDs in an S10 Node server module are identified as SSD0 and SSD1. SSD0 is
closer to the right side of the server module. SSD1 is closer to the left side. On the
motherboard, the card edge connector for SSD0 is labeled J11. The card edge connector for
SSD1 is labeled J10.
This alert specifies which server module contains the problem SSD but not which SSD has
the problem. As a result, you cannot tell from the alert whether you need to replace SSD0 or
SSD1.
To determine which SSD you need to replace:
1. If you have not already done so, add the S Series Node service SSH key to your
SSH keychain, as described in "Adding an S Series Node SSH key to your SSH
keychain" on page 35.
2. Use PuTTY to establish an SSH connection to the server module specified in the alert,
as described in "Using a PuTTY SSH connection" on page 36.
3. At the login prompt, enter service.
4. At the command-line prompt, enter this command:
cat /proc/mdstat
The response to this command is a list of pairs of partitions on the OS SSDs and
database drives.
Each pair of partitions is identified by the letters md followed by a partition number (for
example, md5). The pairs of partitions on the SSDs are numbered 1, 2, 5, 6, 7, 8, and 9.
The pair of partitions on the database drives is numbered 11 or 12.
Each partition in a given pair of SSD partitions is the partition with the applicable number
on one of the SSDs. Partitions are identified by the sd device name of the SSD they're
on, followed by the applicable partition number. An sd device name has the form sdletter
(for example, sda). So, for example, partition 5 on the SSD named sda is sda5.
If both partitions in a pair of partitions are accessible, the entry for that pair ends with a
string consisting of one occurrence of the letter U for each partition in the pair, enclosed
in square brackets (like this: [UU]).
If one or both partitions in a pair of partitions are inaccessible, the string in square
brackets at the end of the entry for that pair is missing one occurrence of the letter U for
each inaccessible partition.
Also, if a partition is inaccessible, either of these is true:
¡ The partition name is followed by the letter F in parentheses, as shown below for
partition 2 on an SSD with device name sdb.
¡ The partition name is missing, as shown below, where partition 6 on an SSD with
device name sdb is missing.
The SSD to replace is the one with one or more inaccessible partitions.
5. To see which OS SSD an sd device name identifies, enter this command:
lsscsi | grep -e StorFly -e VikingDisk -e VPFEM
The response to this command consists of one line for each accessible SSD.
The sd device name for each listed SSD is in the rightmost column. The value in the first
column indicates which SSD the sd device name identifies:
¡ 0:0:0:0 is SSD0.
¡ 1:0:0:0 is SSD1.
2. Reboot the server module by inserting the end of a paper clip into the hole for the
recessed power button in the bottom right corner of the module and then gently pressing
the paper clip against the power button for five seconds if the module is still powered on
or two seconds if the module is powered off.
3. Monitor the server module startup process in the PuTTY window on the laptop computer:
¡ If you see a message indicating either that a boot device cannot be found or that the
server module cannot boot from a given device, both SSDs in the module are bad
and need to be replaced.
¡ If the startup process stops before you see such a message, the server module
itself has a problem and needs to be replaced.
Notes:
• The server module replacement procedure described in this document can be used with
S Series Nodes at release 2.1.1 or later. During this procedure, the S Series Node
remains available.
If the S Series Node is at a release below 2.1.1, you need to use the server replacement
procedure described in HCP S10 Node Maintenance for the applicable release. In this
case, the procedure requires that the S Series Node be shut down.
• Typically, during a firmware update, the server module being updated is unavailable for
both management purposes and data access for approximately five minutes.
Note: Before using the new server module, check the pins on the back of the module. If
any pins are bent, do not use the module. Instead, return it to the manufacturer and
request a new one.
• If the SSDs in the old server module cannot be reused, two new SSDs
SSD part number: ASM-01667-01-A.X
• A #0 Phillips screwdriver
• A #1 Phillips screwdriver
• If you move the SSDs from the old server module to the new server module, a laptop
computer with:
¡ PuTTY and Pageant installed on it. For information about installing PuTTY and
Pageant on a laptop computer, see "Installing PuTTY and Pageant" on page 34.
¡ The S Series Node service SSH key in the SSH keychain on the computer. For
instructions on adding this key to the keychain on your laptop computer, see "Adding
an S Series Node SSH key to your SSH keychain" on page 35.
• If the operating system and software need to be recovered on the new server module:
¡ A USB-to-mini-USB cable.
¡ One bootable USB flash drive or DVD that contains the installation files for the
version of the HCP S Series software that's currently installed on the S10 Node.
If you don't already have an OS and software installation USB flash drive or DVD,
you need to create one, as described in "Preparing an installation USB flash drive or
DVD".
• If the HCP S Series OS and software installation files are on a DVD, an external DVD
drive with a USB connector.
Step 1 (conditional): Shut down the server module you want to replace
If the server module you want to replace (that is, the old server module) is still running, you
need to shut it down. You can shut down a server module from the HCP S Series
Management Console or by using the management API.
To shut down the server module from the Management Console:
1. Log in to the HCP S Series Management Console using a user account with the
administrator or service role.
2. Go to the Hardware ► Overview or Hardware ► Maintenance page.
3. Click Power Options.
4. In the Server Module field, select the server module you want to shut down.
Server module 1 is the server module on the left in the enclosure when viewed from the
back. Server module 2 is the server module on the right.
5. In the Reason field, type the reason why you're shutting down the server module. This
text can be up to 1,024 characters long and can contain any valid UTF-8 characters,
including white space.
6. Click Shut Down.
2. Using the #1 Phillips screwdriver, unscrew the large black screw in the top center of the
server module until the screw is very loose. The screw does not come out.
3. Pull down the handle that runs across the top of the server module. The handle stops
midway down the module.
4. Use the handle to pull the server module all the way out of the enclosure.
Moving the SSDs from the old server module to the new server module
To move the SSDs from the old server module to the new server module:
1. Remove the panel from the right side of each of the old and new server modules. To
remove a panel:
a. Using the #0 Phillips screwdriver, remove the small silver screw from the top of the
panel.
b. Lift the panel out of the server module.
2. In the old server module, remove the foam drive stabilizer that's holding the SSDs in
place.
3. Pull one SSD out of the old server module. When pulling the SSD, you may need to
wiggle it from side to side to loosen it.
4. Insert the SSD from the old server module into the new server module in the same
position as the SSD was in in the old server module. To insert the SSD:
a. Align the SSD with the black card edge connector on the bottom of the server
module.
b. Push the SSD all the way down into the connector.
5. Repeat steps 3 and 4 with the second SSD.
6. In the new server module, place the foam drive stabilizer from the old server module on
the SSDs, fitting the tops of the SSDs into the slots in the stabilizer.
7. Replace the side panel on each server module. To replace a side panel:
a. Insert the bottom tab on the panel into the slot at the bottom of the panel opening.
b. Push the panel into position.
c. Using the #0 Phillips screwdriver, screw the small silver screw back into the hole in
the top of the panel. Tighten the screw.
2. In the old server module, remove the foam drive stabilizer that's holding the SSDs in
place.
3. Insert the new SSDs into the new server module. To insert an SSD:
a. Align the SSD with the black card edge connector on the bottom of the server
module.
b. Push the SSD all the way down into the connector.
4. In the new server module, place the foam drive stabilizer from the old server module on
the SSDs, fitting the tops of the SSDs into the slots in the stabilizer.
5. Replace the side panel on each server module. To replace a side panel:
a. Insert the bottom tab on the panel into the slot at the bottom of the panel opening.
b. Push the panel into position.
c. Using the #0 Phillips screwdriver, screw the small silver screw back into the hole in
the top of the panel. Tighten the screw.
Note: If you moved the SSDs from the old server module to the new server module,
while the HCP S Series software is starting on the new server module, the module
automatically reboots one time.
3. Raise the handle all the way to the top of the server module.
4. Using the #1 Phillips screwdriver, screw in the large black screw in the top center of the
server module. Do not overtighten the screw.
5. Reconnect the access network, management, and server interconnect cables to the
Ethernet ports on the new server module. Be sure to connect each cable to the same port
as the one from which you removed the cable.
When inserting each access network cable into the applicable access port, ensure that
the card in the nonmetal portion of the connector at the end of the cable is facing the
other access port on the same server module.
When inserting the management and server interconnect cables into the applicable
ports, push each cable in until the end of the cable clicks into place.
Step 5: Check and, if necessary, update the firmware on the new server
module
To check and, if necessary, update the firmware on the new server module and, if applicable,
the new OS SSDs:
1. If you have not already done so, add the S Series Node service SSH key to your
SSH keychain, as described in "Adding an S Series Node SSH key to your SSH
keychain" on page 35.
2. Use PuTTY to establish an SSH connection to the new server module, as described in
"Using a PuTTY SSH connection" on page 36.
3. At the login prompt, enter service.
4. Enter this command:
sudo firmware.py show --auto --local --tryLocalUpdate
This command:
a. Checks the currently installed firmware versions on the new server module.
b. If any of the firmware needs to be updated, updates that firmware.
As part of the firmware update process, the HCP S Series software shuts down on
the new server module. Typically, the update process takes less than five minutes,
after which the software restarts. However, if the update includes the BMC
firmware, the update process can take up to 45 minutes. Also, if the update includes
the BMC firmware, at the end of the process, the new server module automatically
reboots.
c. Rechecks the currently installed firmware versions on the new server module.
If all the firmware is up to date, this message appears:
All hardware components are set to the proper versions
If any of the firmware is still not up to date, this failure message appears for each
piece of outdated firmware:
component failed firmware version check- Found: current-version Expected: expected-version
If any failure messages appear, contact your HCP support center for help.
5. Reboot the new server module. To do this, insert the end of a paper clip into the hole for
the recessed power button in the bottom right corner of the module and then gently press
the paper clip against the power button for five seconds.
6. Use PuTTY to a establish a new SSH connection to the new server module.
7. At the login prompt, enter service.
8. Enter this command to update the BIOS on the new server module, regardless of
whether any firmware was updated:
sudo firmware.py bios prod
During the BIOS update process, the new server module automatically reboots.
What you need for replacing one or both SSDs in a server module
For replacing OS SSDs, you need:
• A #0 Phillips screwdriver
• A #1 Phillips screwdriver
• If the operating system and software need to be recovered on the server module, which
is the case only when you replace both OS SSDs in the module:
¡ A USB-to-mini-USB cable.
¡ One bootable USB flash drive or DVD that contains the installation files for the
version of the HCP S Series software that's currently installed on the S10 Node.
If you don't already have an OS and software installation USB flash drive or DVD,
you need to create one, as described in "Preparing an installation USB flash drive or
DVD".
• If the HCP S Series OS and software installation files are on a DVD, an external DVD
drive with a USB connector (only if replacing both OS SSDs).
Important: Before you shut down the server module, if you're replacing only one SSD, you
need to know the location of that SSD. For information about locating an SSD, see
"Degraded OS SSD" on page 174.
To shut down the server module that contains the SSD or SSDs you want to replace:
1. Log in to the HCP S Series Management Console using a user account with the
administrator or service role.
2. Go to the Hardware ► Overview or Hardware ► Maintenance page.
3. Click Power Options.
4. In the Server Module field, select the server module you want to shut down.
Server module 1 is the server module on the left in the enclosure when viewed from the
back. Server module 2 is the server module on the right.
5. In the Reason field, type the reason why you're shutting down the server module. This
text can be up to 1,024 characters long and can contain any valid UTF-8 characters,
including white space.
6. Click Shut Down.
2. Using the #1 Phillips screwdriver, unscrew the large black screw in the top center of the
server module until the screw is very loose. The screw does not come out.
3. Pull down the handle that runs across the top of the server module. The handle stops
midway down the module.
4. Use the handle to pull the server module all the way out of the enclosure.
3. Pull the SSD you're replacing or both SSDs, as applicable, up out of the server module.
When pulling an SSD, you may need to wiggle it from side to side to loosen it.
4. Insert the new SSD or SSDs, as applicable, into the server module. To insert an SSD:
a. Align the SSD with the black card edge connector on the bottom of the server
module.
b. Push the SSD all the way down into the connector.
5. Replace the foam drive stabilizer on the SSDs, fitting the tops of the SSDs into the slots
in the stabilizer.
6. Replace the side panel on the server module. To do this:
a. Insert the bottom tab on the panel into the slot at the bottom of the panel opening.
b. Push the panel into position.
c. Using the #0 Phillips screwdriver, screw the small silver screw back into the hole in
the top of the panel. Tighten the screw.
4. Using the #1 Phillips screwdriver, screw in the large black screw in the top center of the
server module. Do not overtighten the screw.
5. Reconnect the access network, management, and server interconnect cables to the
Ethernet ports on the server module. Be sure to connect each cable to the port from
which you removed the cable.
When inserting each access network cable into the applicable access port, ensure that
the card in the nonmetal portion of the connector at the end of the cable is facing the
other access port on the same server module.
When inserting the management and server interconnect cables into the applicable
ports, push each cable in until the end of the cable clicks into place.
Step 5: Check and, if necessary, update the firmware on the new SSD or
SSDs
To check and, if necessary, update the firmware on the new OS SSD or SSDs:
1. If you have not already done so, add the S Series Node service SSH key to your
SSH keychain, as described in "Adding an S Series Node SSH key to your SSH
keychain" on page 35.
2. Use PuTTY to establish an SSH connection to the server module in which you replaced
the OS SSD or SSDs, as described in "Using a PuTTY SSH connection" on page 36.
3. At the login prompt, enter service.
4. If you replaced a single SSD in the server module, wait for the rebuild of that SSD to
finish.
To check whether the SSD rebuild has finished, enter this command:
cat /proc/mdstat
If the output for each md device looks something like this, the rebuild is complete:
md1 : active raid1 sda1[0] sdb1[1]
1045504 blocks super 1.2 [2/2] [UU]
If the output for any md device contains a line that includes something like this, the
rebuild is not yet complete:
[==>..................] recovery = 12.6%
This command:
a. Checks the currently installed firmware versions on the new SSD or SSDs and on
other components in the server module that has the new SSD or SSDs.
b. If any of the firmware needs to be updated, updates that firmware.
As part of the firmware update process, the HCP S Series software shuts down on
the server module that has the new SSD or SSDs. Typically, the update process
takes less than five minutes, after which the software restarts. However, if the
update includes the BMC firmware, the update process can take up to 45 minutes.
Also, if the update includes the BMC firmware, at the end of the process, the new
server module automatically reboots.
c. Rechecks the currently installed firmware versions on the new SSD or SSDs and on
other components in the server module that has the new SSD or SSDs.
If all the firmware is up to date, this message appears:
All hardware components are set to the proper versions
If any of the firmware is still not up to date, this failure message appears for each
piece of outdated firmware:
component failed firmware version check- Found: current-version Expected: expected-version
If any failure messages appear, contact your HCP support center for help.
For example:
sudo os_drive_maint.sh add /dev/sda
• A paper clip
2. Connect the new server interconnect cable to the server interconnect port on each server
module.
3. Insert the end of a paper clip into the hole for the recessed power button in the bottom
right corner of the unavailable server module. Then gently press the paper clip against
the power button for two seconds.
If the server module does not become available again after you press the power button,
you can power the module back on by sliding it partially out of the enclosure and then
pushing it back in. For instructions on doing this, see"Replacing a server module".
• Hitachi Universal V2 PDUs, which are installed vertically on the sides of a rack
• The blue and red sections each have three C13 outlets
• On the Americas models, the yellow section has two C13 outlets
• On the EMEA/APAC models, the yellow section has three C13 outlets
The C13 outlets are counted separately in each section:
• Outlets in the blue section are counted starting from the end of the PDU.
• Outlets in the red section are counted starting from the end of the blue section.
• Outlets in the yellow section are counted starting from the end of the red section.
Each outlet is identified by these properties, in order:
• The number of the outlet within its section, preceded by a hyphen (-)
For example, the second C13 outlet in the red section is R1C13-2.
The power inlet cable is attached to the blue end of the PDU.
The figure below shows the layout of the PDU models used with S10 Nodes.
• Left and right U# columns that show the rack units in which each enclosure is installed
• Left and right PDU outlet columns that show only the used outlets
• Left and right PDU configuration columns that show the number of PDUs installed on
each side of the rack and the orientation of each PDU
Each enclosure connects to the outlets that, in the diagram, are aligned with the bottom rack
unit occupied by that enclosure.
For each enclosure, the power and cooling module on the left connects to a PDU on the left
side of the rack. The power and cooling module on the right connects to a PDU on the right
side of the rack.
Diagram legend
In the power-cable connection diagram:
• Rack units in italics with a gray background in the U# column show the PDU bracket
locations.
• Outlet background colors (blue, red, and yellow) correspond to the circuits on the PDUs.
• The "PWR" label on each PDU shows which end of the PDU has the power inlet cable.
The PDUs in the diagrams are not drawn to scale.
For an explanation of the outlet identifiers, see "Hitachi Universal V2 PDUs" on page 190.
This PDU model has two sets of three outlets. The set on the left is labeled M1. The set on
the right is labeled M2. In each set, the outlet on the left is labeled 1 (one), and the outlet on
the right is labeled 3. The middle outlet, outlet 2, is not labeled.
The figure below shows a pair of PDUs oriented as they would be when installed in a rack.
The table below shows the PDU outlet connections for the power and cooling modules in S10
Node enclosures. The specified locations for the power and cooling modules are as viewed
from the rear of the rack.
Power and
Enclosure cooling
rack unit module PDU pair PDU position Outlet
U37 Top pair, top PDU: M1-2 Top pair, left PDU: Y1C19-1
Top pair, bottom PDU: M1-2 Top pair, right PDU: Y1C19-1
U33 Top pair, top PDU: M2-1 Top pair, left PDU: R1C19-1
Top pair, bottom PDU: M2-1 Top pair, right PDU: R1C19-1
U29 Top pair, top PDU: M2-3 Top pair, left PDU: B1C19-1
Top pair, bottom PDU: M2-3 Top pair, right PDU: B1C19-1
U23 Middle pair, top PDU: M1-2 Middle pair, left PDU: Y1C19-1
Middle pair, bottom PDU: M1-2 Middle pair, right PDU: Y1C19-1
U19 Middle pair, top PDU: M2-1 Middle pair, left PDU: R1C19-1
Middle pair, bottom PDU: M2-1 Middle pair, right PDU: R1C19-1
U15 Middle pair, top PDU: M2-3 Middle pair, left PDU: B1C19-1
Middle pair, bottom PDU: M2-3 Middle pair, right PDU: B1C19-1
U11 Bottom pair, top PDU: M1-2 Bottom pair, left PDU: B1C19-1
Bottom pair, bottom PDU: M1-2 Bottom pair, right PDU: B1C19-1
U07 Bottom pair, top PDU: M2-1 Bottom pair, left PDU: R1C19-1
Bottom pair, bottom PDU: M2-1 Bottom pair, right PDU: R1C19-1
U03 Bottom pair, top PDU: M2-3 Bottom pair, left PDU: Y1C19-1
Bottom pair, bottom PDU: M2-3 Bottom pair, right PDU: Y1C19-1
Replacing a PDU
If a PDU fails or is damaged, you can replace it. The table below shows the replacement
options.
If you use a vertical PDU to replace one PDU in a pair of horizontal PDUs, the resulting pair
has one PDU of each type. This configuration is allowed.
If the old PDU is a Hitachi Vantara-supplied horizontal PDU and the new PDU is a
Hitachi Universal V2 PDU, be sure to plug each cable into the outlet that corresponds to
the outlet on the cable label. For the outlet correspondences, see "PDU outlet
correspondences" on page 195.
b. Using the #2 Phillips screwdriver and two of the silver screws that come with the
PDU, attach the bracket to the PDU.
2. At the rear of the rack, use the caged nut tool to attach the four caged nuts to the top and
bottom holes (two on the left and two on the right) in the vertical mounting rails in the rack
unit in which you're mounting the PDU. Orient the caged nuts so that the brackets on the
nut grip the left and right sides of the holes.
3. At the rear of the rack, position the PDU in the rack so that:
¡ The sides of the PDU brackets that have the oblong holes are in front of the vertical
mounting rails (when viewed from the rear of the rack)
¡ The oblong holes in the PDU brackets are aligned with the holes to which you
attached the caged nuts
4. Using the #2 Phillips screwdriver and the four large black screws, attach the PDU to the
rack.
Note: Depending on the action you want to perform, you may also need the root SSH
key in your keychain.
cluster_ssh
The cluster_ssh tool does any of these:
Parameter Description
-h|--help Displays and describes the syntax for the cluster_ssh tool; then exits.
--version Displays the version number of the currently installed cluster_ssh tool; then
exits.
(-u|--user) username Tells cluster_ssh to log into each server module with the specified username. If
you omit this parameter, cluster_ssh logs in with the username you used to log
into the server module on which you're running the tool.
(-t|--timeout_min) Specifies the number of minutes cluster_ssh should wait for any given
timeout-minutes command to finish before canceling the comand. The default is 2,880.
(-T|--max_threads) Specifies the maximum number of threads cluster_ssh can use to run any given
max-threads command on each server module. The default is 25.
-v|--verbose Reports cluster_ssh activity in greater detail. If an error occurs while cluster_
ssh is running, rerunning the same cluster_ssh command with this parameter
can help identify the problem.
(Continued)
Parameter Description
(-l|--level) message- Retrieves all the log messages with the specified severity level or a more severe
severity-level level from the log files in the /var/rhino/logs directory on each server module.
Valid values for message-severity-level are:
FINEST
FINER
FINE
INFO
WARNING
SEVERE
-r|--reboot Tells cluster_ssh to shut down the HCP S Series software on each server
module and then reboot the module.
This parameter is mutually exclusive with the -l, --level, -H, --halt, -i, --
inputScript, and command-string parameters.
-H|--halt Tells cluster_ssh to shut down the HCP S Series software on each server
module and then power off the module.
This parameter is mutually exclusive with the -l, --level, -r, --reboot, -i, --
inputScript, and command-string parameters.
(Continued)
Parameter Description
"command-string" Specifies one or more commands for cluster_ssh to execute on each server
module. cluster_ssh executes multiple commands on each module in the order
in which you specify them.
Use a semicolon (;) or a double ampersand (&&) to separate each command
from the next:
• The semicolon means proceed with the next command even if the current
command fails.
• The double ampersand means do not proceed if the current command fails.
The double quotation marks (") are required when you specify more than one
command or when you specify a single command with one or more options. You
can omit the quotation marks around a single command with no options.
This parameter is mutually exclusive with the -l, --level, -r, --reboot, -H, --
halt, -i, and --inputScript parameters.
Note: To reboot server modules, use the -r or --reboot parameter instead of a
reboot command. With a reboot command, the server module on which you ran
the tool may shut down before it can communicate the request to the other server
module. Similarly, use the -H or --halt parameter instead of a shutdown
command.
cluster_ssh examples
Here are three sample cluster_ssh commands:
• The cluster_ssh command below echoes the ping command on each server module.
You can use a command like this on one server module to check whether the other
server module is running.
cluster_ssh.py "echo ping"
• The cluster_ssh command below displays the current date and time followed by
information about the current run of the HCP S Series software on each server module.
cluster_ssh.py "date;uptime"
• The cluster_ssh command below retrieves all the messages with a severity level of
SEVERE from the log files in the /var/rhino/logs directory on each server module.
The messages appear in the cluster_ssh output.
cluster_ssh.py --level SEVERE
cluster_get
The cluster_get tool retrieves one or more files from each server module and stores them on
the server module on which you ran the tool. By default, the files are stored in subdirectories,
one per server module, of a particular directory, but you can choose to have them stored
directly in that directory. The name of each subdirectory is the server interconnect network
IP address of the applicable server module.
Syntax for the cluster_get tool is:
cluster_get.py
(-h|--help)
|--version
|([(-u|--user) username]
[(-t|--timeout_min) timeout-minutes]
[(-T|--max_threads) max-threads]
[-d|--disable-unique-ify]
[-v|--verbose]
((-l|--logs)
|((-i|--image) [--days number-of-days])
|((source-file|source-directory) destination-directory)))
Parameter Description
-h|--help Displays and describes the syntax for the cluster_get tool; then exits.
--version Displays the version number of the currently installed cluster_get tool; then
exits.
(-u|--user) username Tells cluster_get to log into each server module with the specified username. If
you omit this parameter, cluster_get logs in with the username you used to log
into the server module on which you're running the tool.
(-t|--timeout_min) Specifies the number of minutes cluster_get should wait for any given
timeout-minutes command to finish before canceling the command. The default is 2,880.
(-T|--max_threads) Specifies the maximum number of threads cluster_get can use to run any given
max-threads command on each server module. The default is 25.
-d|--disable- Puts all the returned files directly in the directory identified by the destination-
unique-ify directory parameter. With the -d or --disable-unique-ify parameter, files
with the same name that come from different server modules overwrite each
other.
Without the -d or --disable-unique-ify parameter, cluster_get puts each
returned file in a directory named with the server interconnect network IP
address for the server module from which the files was retrieved. These
directories are located in the directory specified by the destination-directory
parameter.
The -d and --disable-unique-ifyparameters are not valid with the -l, --logs,
-i, and --image parameters.
-v|--verbose Reports cluster_get activity in greater detail. If an error occurs while cluster_get
is running, rerunning the same cluster_get command with this parameter can
help identify the problem.
-l|--logs Retrieves and compresses the S Series Node internal log files in the
/var/rhino/logs directory on each server module. The log files from each
server module are compressed into a single file named access-network-ip-
address_log.tar.bz2, which is then stored in a subdirectory of the current
directory. The name of each subdirectory is the server interconnect network IP
address for the server module from which the log files were retrieved.
The -l and --logs parameters are not valid with the -i, --image, source-file,
source-directory, and destination-directory parameters.
For information about the internal log files, see "Internal logs" on page 55.
(Continued)
Parameter Description
-i|--image Retrieves and compresses the S Series Node internal log files in the
/var/rhino/logs directory on each server module, along with other files
containing server module information that’s useful for identifying problems. The
log files and other information from each server module are compressed into a
single file named fully-qualified-server-module-hostname-image.tar.bz2,
which is then stored in a subdirectory of the current directory. The name of each
subdirectory is the server interconnect network IP address for the server module
from which the log files were retrieved.
The -i and --image parameters are not valid with the -l, --logs, source-file,
source-directory, and destination-directory parameters.
For information about the internal log files, see "Internal logs" on page 55.
Important: This operation is CPU intensive. It also makes reviewing the logs in
real time more difficult for support personnel. Do not request this operation
unless explicitly told to do so by HCP support personnel.
--days number-of-days Specifies the number of days worth of logs to retrieve from the /var/rhino/logs
directory when the -i or --image parameter is specified in the cluster_get
command. Days are counted backward starting from the current day.
The --days parameter is valid only with the -i and --image parameters
source-file Specifies the file cluster_get should retrieve from each server module. The file
specification must include the absolute path to the file.
The source-file parameter is not valid with the -l, --logs, -i, --image, and
source-directory parameters.
source-directory Specifies a directory from which cluster_get should retrieve all files on each
server module, including the files in any subdirectories of that directory,
recursively. The directory specification must be an absolute path.
The source-directory parameter is not valid with the -l, --logs, -i, --image,
and source-file parameters.
destination-directory Specifies the directory in which cluster_get should put the returned files. The
directory specification can be an absolute path or a path relative to the current
directory. If the specified directory doesn’t exist, cluster_get creates it.
cluster_get examples
Here are three sample cluster_get commands:
• The cluster_get command below retrieves the build.version file from each server
module and stores it in a directory named with the server interconnect network IP
address of that server module. cluster_get creates these directories in the
/service/temp directory on the server module on which the command was run. You
can use this command to verify that the same version of the HCP S Series software is
installed on both server modules.
cluster_get.py /opt/rhino/build.version /service/temp
• The cluster_get command below retrieves a copy of the S Series Node internal logs
from the current day and the previous day from each server module, along with other files
containing useful information. For each server module, the retrieved files are
compressed into a single file named using the fully qualified hostname of that module
and stored in a subdirectory of the current directory. The subdirectory is named with the
server interconnect network IP address of the applicable server module.
cluster_get.py -i --days 2
cluster_put
The cluster_put tool stores one or more files on each server module. Syntax for this tool is:
cluster_put.py
(-h|--help)
|--version
|([(-u|--user) username]
[(-t|--timeout_min) timeout-minutes]
[(-T|--max_threads) max-threads]
[-v|--verbose]
(source-file|source-directory)
[destination-directory])
Parameter Description
-h|--help Displays and describes the syntax for the cluster_put tool; then exits.
--version Displays the version number of the currently installed cluster_put tool; then
exits.
(-u|--user) username Tells cluster_put to log into each server module with the specified username. If
you omit this parameter, cluster_put logs in with the username you used to log
into the server module on which you're running the tool.
(-t|--timeout_min) Specifies the number of minutes cluster_put should wait for any given
timeout-minutes command to finish before canceling the command. The default is 2,880.
(-T|--max_threads) Specifies the maximum number of threads cluster_put can use to run any given
max-threads command on each server module. The default is 25.
-v|--verbose Reports cluster_put activity in greater detail. If an error occurs while cluster_put
is running, rerunning the same cluster_put command with this parameter can
help identify the problem.
source-file Specifies the file cluster_put should store. cluster_put copies this file only from
the server module on which you're running the tool.
The file specification can include an absolute path or a path relative to the
current directory.
source-directory Specifies a directory containing the files cluster_put should store. cluster_put
copies all files in the specified directory, including the files in any subdirectories
of that directory, recursively. cluster_put copies the files only from the server
module on which you're running the tool.
The directory specification must be an absolute path.
(Continued)
Parameter Description
destination-directory Specifies the directory in which cluster_put stores the files on each server
module. The specified directory must already exist.
The directory specification must be an absolute path.
If you don't specify a destination directory, cluster_put uses the soruce directory
as the destination directory. If the source directory is specified by a relative path,
cluster_put creates an absolute destination directory path by appending that
path to the path to the current directory.
cluster_put examples
Here are two sample cluster_put commands:
• The cluster_put command below copies the fix-3.sh file from the current directory on
the server module on which you're running the tool to the existing
/service/temp/fixes directory on each server module.
cluster_put.py fix-3.sh /service/temp/fixes
Field Value
Identification
Domain name
Serial number
Password
Access network
IP mode o IPv4
o IPv6
Access network: IPv4 settings
(Continued)
Field Value
Duplex o Auto
o 10GbE Full
Bonding mode o active-backup
o 801.3ad
MTU o 1500
o 9000
VLAN ID (optional)
Management network
IP mode o IPv4
o IPv6
Management network: IPv4 settings
(Continued)
Field Value
Duplex o Auto
o 10MbE Half
o 10MbE Full
o 100MbE Half
o 100MbE Full
o 1GbE Full
MTU o 1500
o 9000
VLAN ID (optional)
Subnet