topic Re: Health rule based on application changes in Splunk AppDynamics

Health rule based on application changes

CommunityUser — Thu, 20 Jul 2017 13:03:57 GMT

I would like to create a health rule that is based on application changes. For instance, a rule that is violated when the app server of a specific node has been restarted more than x times within a specified time range. It seems that this value is not considered as a metric and therefore cannot be referenced in the health rule violation wizard. Am i missing something here or is this simply not supported?

Re: Health rule based on application changes

CommunityUser — Fri, 21 Jul 2017 03:02:32 GMT

I am not sure what is the intent of the health rule, if it is to check at what times the app went down then you can use the following.

Overall Application Performance > Agent > App > Availability,

At the tier level, the value of the above metric is equal to total number of nodes reporting to the tier, at the node level, the value of the above metric will always be one.

Perhaps you can use the sum of the metircs over a period of time and check against your threshold?

To your original question, i dont think number of restarts is saved any where.

May be you can write a shell script to find that out on your server and then use extensions to report that value as a custom metric?

Cheers,

Gurmit.

Re: Health rule based on application changes

CommunityUser — Fri, 21 Jul 2017 09:18:39 GMT

Hi Gurmit,

The intent of the rule is to easily detect if an application is recycling frequently, as this may indicate that a private memory limit for instance, is exceeded easily. I suppose checking the availabilty of a tier could also cover this in some way. I will try this first. Thanks for the quick reply!

Regards,

Simon