This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Database Cluster Failover Alert

Hi All,

I've tried to configure an alert to trigger whenever the system name of a node changes to Node2 (the node 2 of the clusters ends with 2, in my case). PFB the screenshot of the same.

pastedImage_0.png

After this configuration I've failed over the cluster to Node 2, the system name got updated in the node details page but the alert didn't triggered.

Can anyone please help point out what's wrong over here? Or, could help with a better way of doing this.

Many Thanks!

  • Hi vinay.by​,

    Any suggestions here, please ?

    Thanks !

  • I will have to give it a try myself, but why is the alert configured so, its just a basic question, you would know your system better than me for sure emoticons_happy.png, why aren't you actually monitoring the process or service or event logs for cluster failover? Do you have SAM in your environment ? If you don't then it makes sense.

    As well please paste your complete trigger condition and scrub the information which is client specific, let me look into the complete trigger condition.

  • HI Vinay,

    Yes, I do have SAM. Have seen the SAM template available for this but the requirement is to get an alert only when the cluster is failed over to Node2. We shouldn't get alerts when it fail overs to Node1.

    I've attached the screenshot of the trigger condition in the question description.

  • Perfect, in that case just create an alert like the one mentioned below:

    1. Start monitoring Cluster Service on both the nodes, create a template -> create a cluster service monitor and place this template on both the nodes.

    2. Create an alert only when the below condition is met

    a. 'Cluster Service' up on Node2 AND 'Cluster Service' down on Node1

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

    Second approach

    1. Enable Event log monitoring on Node 2 and capture/monitor specific Event ID (Event Log Monitor on Node 2) 'Cluster Service failover' Event Id or 'Cluster Service coming up' Event ID.

    2. Trigger an alert based on the Event ID

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

    Third approach

    Why don't you change that trigger condition of yours and give it a try, rather than ends with , try 'contains' Node2 / 'is-equals' Node2

  • Hi vinay.by​,

    Unfortunately, none of these worked! emoticons_sad.png