ActiveGate failover not working for IBM MQ Extension 2.0

[AK]

Hi All,

We are facing some issue with queue manager availability metric events that is configured for IBM MQ extension 2.0.

Following basic metric selector query is used to report queue manager availability,

ibmmq.queue_manager.availability:filter(in("dt.entity.ibmmq:queue_manager",entitySelector("type(ibmmq:queue_manager)"))):splitBy("dt.entity.ibmmq:queue_manager")

ActiveGate groups having 4 servers each are configured to handle failover situation.

However, failover mechanism is not working. Whenever there patching or maintenance activity on any of ActiveGate server, then queue manager availability alerts get triggered.

We thought to adjust the violating samples to 8 samples and sliding window to 15 minutes but in this case we might miss the actual alerts. Has anyone faced the similar issue?

@Babar_Qayyum, do you have any insights on this?

Regards,

AK

[Peter_Youssef]

Hi @AK 

For the Failover situation:

• Ensure each couple of ActiveGates are configured in AG group, so you will have (2) AG Groups, so in failover scenarios if one group fail the other will take over.
• Ensure Network zone is configured for the environment.

Validate against the current configurations if the below requirements are met or something is missing:

For Full information regarding the current situation, you can also validate against your current deployment model either it's through Local Oneagents deployments or Remotely: 

• IBM MQ 

Regards,

[AK]

@Peter_Youssef., thank you for the details. We have most of the things covered from the above suggested pointer except network zone, is that really causing issue with failover mechanism?

For information, we are using ActiveGates to monitor IBM MQs.

Regards,

AK

[Peter_Youssef]

Hi @AK 

Appreciate your feedback, It's also best practice to ensure that you have 2 AGs deployed in the configured network zone to run the extensions or routing OA traffic, for ensuring the failover mechanism is achieved.

KR,

Peter.