Bonding Does Not Work Using Mode 4 802.3ad (LACP) - Red Hat Customer Portal

5/20/2020 Bonding does not work using mode 4 802.
3ad(LACP) - Red Hat Customer Portal
(https://access.redhat.com/)

 Read Red Hat’s business continuity plan (https://www.redhat.com/en/about/here-to-

help/business-continuity) for COVID-19. Have feedback? Let us know 
(https://redhatvoc.co1.qualtrics.com/jfe/form/SV_6rsJsRslhOYK8VD).
Bonding does not work using mode 4 802.3ad(LACP)

$ SOLUTION VERIFIED - Updated August 23 2019 at 11:52 PM - English ()
Red Hat Insights can detect this issue
Proactively detect and remediate issues impacting your systems.
View matching systems and remediation

(http://cloud.redhat.com/insights/advisor/recommendations/bonding_partner_mac|CHECK_
Environment
Red Hat Enterprise Linux - all versions
Cisco Switch that supports lacp bonding
Issue
Four network devices setup in bonding mode 4 link aggregation control protocol(lacp,802.ad)
show 4 aggregation ids with the number of ports in the aggregation is 1. Additionally, the
Partner Mac Address: value registers a null-value.
https://access.redhat.com/solutions/160283 1/11
5/20/2020 Bonding does not work using mode 4 802.3ad(LACP) - Red Hat Customer Portal
Ethernet Channel Bonding Driver: v3.6.0 (September 26, 2009)
Bonding Mode: IEEE 802.3ad Dynamic link aggregation

Transmit Hash Policy: layer2 (0)
MII Status: up
MII Polling Interval (ms): 200
Up Delay (ms): 0
Down Delay (ms): 0
802.3ad info
LACP rate: slow
Aggregator selection policy (ad_select): stable
Active Aggregator Info:
Aggregator ID: 2
Number of ports: 1
Actor Key: 17
Partner Key: 1
Partner Mac Address: 00:00:00:00:00:00
Slave Interface: eth1

MII Status: up
Speed: 1000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: xx:xx:xx:xx:xx:7a
Aggregator ID: 1
Slave queue ID: 0

MII Status: up
Speed: 1000 Mbps
Duplex: full
Permanent HW addr: "xx:xx:xx:xx:xx:xx"
Aggregator ID: 2 <------- ID = 2
Slave queue ID: 0

MII Status: up
Speed: 1000 Mbps
Duplex: full
Permanent HW addr: xx:xx:xx:xx:xx:7c
Aggregator ID: 3 <------- ID = 3
Slave queue ID: 0

MII Status: up
Speed: 1000 Mbps
Duplex: full
Permanent HW addr: xx:xx:xx:xx:xx:20
Aggregator ID: 4
Slave queue ID: 0

Resolution
If LACP is using one switch, (single mode), then the aggregator ID will be the same number.
If LACP is using two switches, then the aggregator ID will be different numbers.
Disclaimer: Steps contained herein from third-party vendor(s) are provided for convenience only.
Red Hat has not reviewed the steps and is not responsible for the content or its availability. The
inclusion of any instructions from a third-party vendor does not imply endorsement by Red Hat of
the website or their entities, products or services. You agree that Red Hat is not responsible or liable
for any loss or expenses that may result due to your use of (or reliance on) these steps.
STEP 1 - Remove previous related switch config

Remove all switch configuration that will be associated with your lacp bonding, old
configurations may cause issues with the new configuration
Enter the config mode of the switch then for each type:
Etherchannel: no int poX , with X being the previously configured channel group.
Vlan Configurations 'no interface vlanx` with X being the vlan number.
Interfaces default interface gi0/1 with GI representing GigabitEthernet, replace this
with your type.
STEP 2 - Configure switchports

Configure the individual switchports as listed below substituting the appropriate allowed and
native VLANs for the desired topology.
interface GigabitEthernet1/0/1
description XenServer1 - eth0
switchport trunk encapsulation dot1q
switchport trunk native vlan 2
switchport trunk allowed vlan 2-11
switchport mode trunk # if lacp doesnt work, you may want to try `switchport mode access`
speed 1000
duplex full
channel-protocol lacp # This step is requirement if you require lacp
channel-group 2 mode active # This step will create the port channel 2
no shutdown
STEP 3 - Configure(https://access.redhat.com/)
EtherChannel 
Configure the EtherChannel listed below. Verify that the VLAN configuration that is entered to
the overlying Port-channel layer matches the underlying interface configuration.
interface Port-channel1
description Etherchannel Team for XenServer1
switchport mode trunk
spanning-tree portfast
spanning-tree bpduguard enable
no shutdown
Make sure things are configured properly by running the follow commands.
STEP 4 - Check Configuration

Where X is the switchport number, interface or PortChannel
show running-config interface Gi0/X
show running-config interface Gi0/X
show running-config interface port-channelX
show interface status
show ip int brief
show int gi0/X
show int port-channelX switchport
STEP 5 - Check Etherchannel Status

This will show (SU) for the Port-channel that is functional, and (P) for any interface that is
participating in the link.
show etherchannel 2 summary
Number of channel-groups in use: 4

Number of aggregators: 4
Group Port-channel Protocol Ports

------+-------------+-----------+-----------------------------------------------
2 Po2(SU) LACP Gi1/0/1(P) Gi2/0/1(P)
STEP 6 - Check O/S status


cat /proc/net/bonding/bond0
Root Cause
Whilst it can be a valid configuration to have "slaves to active aggregator" less than "total
slaves in bond", or even one slave per aggregator ID, in this instance it was desired to have all 4
slaves in the one aggregator.
Improper switch configuration was causing the bond to show different link Aggregation IDs,
only 1 port in each aggregator, and a Partner MAC of all zeros.
Partner MAC address that contain all zeros can also be caused by a duplicate mac address.
After removing the duplicate mac address, the lacp was able to resume operating normally.
Diagnostic Steps
Verify the bonding and interface scripts in the etc/sysconfig/network-scripts directory are
configured correctly.
DEVICE="eth1"
HWADDR="xx:xx:xx:xx:xx:7A"

ONBOOT="yes"
BOOTPROTO="none"
MASTER="bond1"
SLAVE="yes"
USERCTL="no"
DEVICE="eth10"
HWADDR="xx:xx:xx:xx:xx:xx"
ONBOOT="yes"
BOOTPROTO="none"
MASTER="bond1"
SLAVE="yes"
USERCTL="no"
DEVICE="eth2"
ONBOOT="yes"
BOOTPROTO="none"
MASTER="bond1"
SLAVE="yes"
USERCTL="no"
DEVICE="eth9"
ONBOOT="yes"
BOOTPROTO="none"
MASTER="bond1"
SLAVE="yes"
USERCTL="no"
DEVICE="bond1"
ONBOOT="yes"
BOOTPROTO="none"
IPADDR="192.168.0.2"
NETMASK="255.255.255.0"
USERCTL="no"
BONDING_OPTS="mode=4 miimon=100"
Look at the switch configuration and the etherchannel summary for the bond. It only shows 1
active port for the etherchannel set in place. Also, the switch configuration below
demonstrates that the individual interface configuration does not match the EtherChannel
show command output.
$show etherchannel summary


#sh etherchannel summary
Group Port-channel Protocol Ports
------+-------------+-----------+-----------------------------------------------
2 Po2(SU) - Gi9/18(D) Gi9/19(D) Gi9/20(P) Gi9/21(D)
interface GigabitEthernet2/0/1
description XenServer1 - eth1
switchport mode trunk
speed 1000
duplex full
channel-protocol lacp
channel-group 2 mode active
An example of a system with lacp
[root@MYOPTPDN19 Desktop]# cat /proc/net/bonding/bond0
Ethernet Channel Bonding Driver: v3.7.1 (April 27, 2011)

Bonding Mode: IEEE 802.3ad Dynamic link aggregation

Transmit Hash Policy: layer2 (0)
MII Status: up
MII Polling Interval (ms): 100
Up Delay (ms): 0
Down Delay (ms): 0
802.3ad info
LACP rate: slow
Min links: 0
Aggregator selection policy (ad_select): stable
Active Aggregator Info:
Aggregator ID: 3
Number of ports: 1 <---- only one member
Actor Key: 13
Partner Key: 1

MII Status: up
Speed: 10000 Mbps
Duplex: full
Permanent HW addr: xx:xx:xx:xx:xx:e0
Slave queue ID: 0
Aggregator ID: 3 <---- different ID then for the other port
Actor Churn State: none
Partner Churn State: churned
Actor Churned Count: 0
Partner Churned Count: 1
details actor lacp pdu:
system priority: 65535
port key: 13
port priority: 255
port number: 1
port state: 77
details partner lacp pdu:
oper key: 1
port priority: 255
port number: 1
port state: 1

MII Status: up
Speed: 10000 Mbps
Duplex: full
Permanent HW addr: xx:xx:xx:xx:xx:e8
Slave queue ID: 0
Aggregator ID: 4 <---- different ID then for the other port

Actor Churn State: churned

Partner Churn State: churned
Actor Churned Count: 1
Partner Churned Count: 1
details actor lacp pdu:
port key: 13
port priority: 255
port number: 2
port state: 69
details partner lacp pdu:
oper key: 1
port priority: 255
port number: 1
port state: 1
This means that the bond is not aggregating. In the above example two aggregator's are
created with one individual port in each interface.
Check the switch configuration if it is LACP enabled for both ports.
Product(s) Red Hat Enterprise Linux (/taxonomy/products/red-hat-enterprise-linux)
Category Troubleshoot (/category/troubleshoot)
Tags bonding (/tags/bonding) network (/tags/network)
This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions
that Red Hat engineers have created while supporting our customers. To give you the knowledge
you need the instant it becomes available, these articles may be presented in a raw and unedited
form.
People who viewed this solution also viewed
LACP bond does not change the Aggregator ID of a slave without detected link.
Solution - May 31, 2017
LACP bond mode 4 slaves get different aggregation id. switch and servers configs look
g gg g g
correct 
Solution - Sep 13, 2017
Multiple LACP bonds have the same Aggregator ID
Solution - Feb 9, 2017
1 Comments
24 September 2014 3:51 PM (https://access.redhat.com/solutions/160283#comment-821573)
MR Manoj Rana Rana (/user/8293993)
(/user/8293993)
----total-cpu-usage---- -dsk/total- -net/bond0----net/eth0----net/eth1----
NEWBIE net/eth2----net/eth3- ---paging-- ---system--
5 Points usr sys idl wai hiq siq| read writ| recv send: recv send: recv send: recv send: recv send| in
out | int csw
0 1 97 2 0 0| 20M 3423k| 0 0 : 0 0 : 0 0 : 0 0 : 0 0 | 0 0 | 12k 12k
0 0 100 0 0 0| 0 16k|1532B 2432B:1160B 1394B: 124B 790B: 0 0 : 248B 248B| 0 0 | 160
138
0 0 100 0 0 0| 0 388k|1530B 1536B: 796B 642B: 124B 646B: 610B 248B: 0 0 | 0 0 |
306 330
0 0 100 0 0 0| 0 392k|4949B 1536B:4577B 642B: 124B 646B: 0 0 : 248B 248B| 0 0 |
314 288
0 0 100 0 0 0| 0 28k|7573B 1536B:6188B 642B: 124B 646B:1261B 248B: 0 0 | 0 0 | 316
368
0 0 100 0 0 0| 0 0 | 984B 2733B: 612B 642B: 124B 646B: 0 0 : 248B 1445B| 0 0 | 228
142
0 0 100 0 0 0| 0 0 |1515B 1536B: 612B 642B: 124B 646B: 779B 248B: 0 0 | 0 0 | 266
322
0 0 100 0 0 0| 0 40k|5197B 2423B:4825B 1529B: 124B 646B: 0 0 : 248B 248B| 0 0 |
622 561
0 0 100 0 0 0| 0 28k|3040B 11k:1096B 642B: 124B 646B:1820B 10k: 0 0 | 0 0 | 311 410
0 0 100 0 0 0| 708k 452k| 13k 2301B: 13k 642B: 124B 662B: 0 0 : 248B 997B| 0 0 | 452
479 
0 0 100 0 0 0| 0 0 |1640B 1568B: 677B 658B: 124B 662B: 839B 248B: 0 0 | 0 0 | 294
342
0 0 100 0 0 0| 0 124k| 892B 3586B: 520B 642B: 124B 1366B: 0 0 : 248B 1578B| 0 0 |
292 202 ^C
≤ Reply (/Ajax_comments/Reply/160283/821573)
All systems operational (https://status.redhat.com)
Privacy Statement (http://www.redhat.com/en/about/privacy-

policy)
Customer Portal Terms of Use
(https://access.redhat.com/help/terms/)
All Policies and Guidelines
(http://www.redhat.com/en/about/all-policies-guidelines)
Copyright © 2020 Red Hat, Inc.

Bonding Does Not Work Using Mode 4 802.3ad (LACP) - Red Hat Customer Portal

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bonding Does Not Work Using Mode 4 802.3ad (LACP) - Red Hat Customer Portal

Uploaded by

Copyright:

Available Formats

5/20/2020 Bonding does not work using mode 4 802.

3ad(LACP) - Red Hat Customer Portal

 Read Red Hat’s business continuity plan (https://www.redhat.com/en/about/here-to-

Bonding does not work using mode 4 802.3ad(LACP)

Red Hat Insights can detect this issue

Proactively detect and remediate issues impacting your systems.

View matching systems and remediation

Slave Interface: eth1

Slave Interface: eth10

Slave Interface: eth2

Slave Interface: eth9

STEP 1 - Remove previous related switch config

STEP 2 - Configure switchports

STEP 4 - Check Configuration

STEP 5 - Check Etherchannel Status

show etherchannel 2 summary

Number of channel-groups in use: 4

Group Port-channel Protocol Ports

STEP 6 - Check O/S status

Partner Mac Address: 00:00:00:00:00:00

$show etherchannel summary

An example of a system with lacp

Bonding Mode: IEEE 802.3ad Dynamic link aggregation

Slave Interface: eth0

Slave Interface: eth1

Aggregator ID: 4 <---- different ID then for the other port

Product(s) Red Hat Enterprise Linux (/taxonomy/products/red-hat-enterprise-linux)

Category Troubleshoot (/category/troubleshoot)

Tags bonding (/tags/bonding) network (/tags/network)

People who viewed this solution also viewed

Solution - May 31, 2017

Multiple LACP bonds have the same Aggregator ID

Solution - Feb 9, 2017

All systems operational (https://status.redhat.com)

Privacy Statement (http://www.redhat.com/en/about/privacy-

You might also like