Can't find what you need?


• Ask the Community
• Create a Case
Reset Search
 

 

Article

How to detect and recover from HA state not synchronized for Management Modules?

« Go Back

Information

 
TitleHow to detect and recover from HA state not synchronized for Management Modules?
Question
In order to achieve hitless failover for management modules in case of failure, the HA state needs to be checked and ensure the HA state is synchronized with both of the redundant management modules.

When the MMs are out of synchronization, failover will not work as expected. It will lead to unexpected downtime and outage of the network.
Environment
Answer
I. Checking for HA State in synchronization:
Use CLI to check HA State:

SHOW HA

Sample output:

When MMs have HA-state in sync onVDX 8770 (NOS 6.0.2b):

VDX8770# sh slot
Rbridge-id 1:

Slot Type Description ID Status
----------------------------------------------------------------------
M1 MM Management Module 112 ENABLED
M2 MM Management Module 112 ENABLED
S1 SFM Switch Fabric Module 113 ENABLED#
S2 SFM Switch Fabric Module 113 ENABLED#
S3 SFM Switch Fabric Module 113 ENABLED
L1 LC48X10G 48-port 10GE card 114 ENABLED
L2 VACANT
L3 VACANT
L4 VACANT
# = At least one enabled SFM in these slots is required.


VDX8770# sh ha
M2: Active, Cold Recovered, Dual MM, Redundant, State synchronized
M1: Standby, Dual MM, Redundant, State synchronized
L1/0: Active, Cold Recovered, Dual Partitions, Redundant, State in sync
1: Standby, Dual Partitions, Redundant, State in sync


When MMs have HA-state in sync onVDX 6740 (NOS 6.0.1):

VDX6740# sh ha
SW/0: Active, Cold Recovered, Dual Partitions, Redundant, State synchronized
SW/1: Standby, Dual Partitions, Redundant, State synchronized


When MMs have HA-state Out of sync onVDX 8770 (NOS 6.0.2b):

VDX8770# sh ha
M2: Active, Cold Recovered, Dual MM, Redundant,
M1: Standby, Dual MM, Redundant
L1/0: Active, Cold Recovered, Dual Partitions, Redundant
1: Standby, Dual Partitions, Redundant


When MMs have HA-state Out of sync onVDX 6740 (NOS 6.0.1):

VDX6740# sh ha
SW/0: Active, Cold Recovered, Dual Partitions, Redundant
SW/1: Standby, Dual Partitions, Redundant


It is evident that "State synchronized" is not printed in the output of SHOW HA when the MMs are not in sync.

II. Use CLI, HA SYNC Start, to initiate HA synchronization of the MM. It takes, in normal case, about 3-4 minute to get synchronization process complete.

Example:

VDX6740# sh ha
SW/1: Active, Cold Recovered, Dual Partitions, Redundant, State synchronized
SW/0: Standby, Dual Partitions, Redundant, State synchronized


VDX6740# ha sync start
Are you sure you want to start HA state synchronization [y/n]? y

VDX6740# sh ha
SW/1: Active, Cold Recovered, Dual Partitions, Redundant, State synchronized
SW/0: Standby, Dual Partitions, Redundant, State synchronized


You can see the HA State sync progress in RASLOG:

VDX6740# sh log ras rev count 10
Rbridge-id: 13
NOS: 6.0.2a
2016/09/06-18:29:45, [HASM-1100], 47678, SW/1 | Active, INFO, VDX6740T-1G, HA State is in sync.

2016/09/06-18:29:25, [SEC-1204], 47677, SW/0 | Standby, INFO, VDX6740T-1G, Root access mode is configured to default(SSH/Telnet/Console).

2016/09/06-18:29:25, [SEC-1197], 47676, SW/0 | Standby, INFO, VDX6740T-1G, Changed account root.

2016/09/06-18:27:03, [SEC-3051], 47675, SW/0 | Standby, INFO, VDX6740T-1G, The license key *B dI0F8CzpNI094T:1RVMyfgn1uopB6j0RdWK8XZU0z0VLC0,ikib9g1,wl:NXANFmV,1DOinmcrUjUglJkhNq3pLSuI:YOC4dVO5NJsh:VvypVMSj7s1X,F3hBzmGv7VT3vfX# is Added.

2016/09/06-18:27:02, [SEC-3051], 47674, SW/0 | Standby, INFO, VDX6740T-1G, The license key PORT_40G_UPGRADE
is Removed.

2016/09/06-18:27:02, [HASM-1012], 47673, SW/1 | Active, INFO, VDX6740T-1G, HA State starts to sync.

2016/09/06-18:27:02, [HASM-1012], 47672, SW/1 | Active, INFO, VDX6740T-1G, HA State starts to sync.
Additional notes

Feedback

 

Was this article helpful?


   

Feedback

Please tell us how we can make this article more useful.

Characters Remaining: 255