Always on alerts suppresion if one node is down

Question

Hi,

We are facing an issue with the db monitoring during weekend patching activity. Case is like if there are two servers and if one goes down, then all the load comes to the second server. We keep the first server in maintenance but the always on nodes which are joining these servers have multiple instance and they start to throw alert. Is there any way through which we can suppress the alerts during that period.

Answer 1

Hello,

Thank you for posting in Q&A forum.

You can set a rule to disable alert processing to prevent alerts from being displayed during maintenance. For details, please refer to the following link:

https://learn.microsoft.com/en-us/azure/site-recovery/site-recovery-monitor-and-troubleshoot

I hope the information above is helpful.

Best regards

Zunhui

============================================

If the Answer is helpful, please click "Accept Answer" and upvote it.

Answer 2

Hi,

I understand the challenge you're facing with multiple instances on Always On nodes generating alerts during maintenance. Here are a few strategies to help suppress these alerts:

1. Maintenance Mode for Specific Instances: Instead of putting the entire server into maintenance mode, you can target specific instances. This way, only the instances that are affected by the maintenance will have their alerts suppressed.
2. Custom Groups and Overrides: Create custom groups in SCOM for the instances that are part of the Always On configuration. Apply overrides to these groups to disable alerting during the maintenance window.

By implementing these strategies, you can ensure that alerts are suppressed only for the relevant instances, minimizing unnecessary noise during maintenance periods.

Answer 3

Please try the following step to temporarily suppress alerts related to an unavailable node in a SQL Server Always On availability group:

1. Identify Alert Criteria:
   • Determine the criteria that indicate the unavailability of a node in the Always On availability group (e.g., node status is down, cluster state is unhealthy).
2. Create a Management Pack:
   • Open the SCOM console.
   • Navigate to Administration > Management Packs.
   • Right-click on Management Packs and select Create Management Pack.
   • Provide a name and description for the new management pack and click Create.
3. Identify Alert-Generating Rules:
   • Navigate to Authoring > Management Pack Objects > Monitors or Rules.
   • Identify the rules or monitors that generate alerts related to the unavailable node.
4. Create Overrides:
   • Right-click on the rule or monitor that generates alerts for the node.
   • Select Overrides > Override the Monitor or Override the Rule > For a specific object of class: <Select the node object>.
   • Configure the override to disable alert generation for the specific condition (e.g., node status is down).
   • Set the duration for which the alert suppression should be in effect (e.g., during maintenance window).
5. Set Alert Suppression Duration:
   • Specify the start and end time for the alert suppression to ensure alerts are only suppressed when the node is expected to be down.

By following these steps and customizing them to your specific environment and requirements, you can effectively define alert suppression rules in SCOM to manage alerts related to unavailable nodes in a SQL Server Always On availability group.