Is there currently any way how to extract or enrich the Dynatrace alerts with the host/instance information? Our use case would be that we can monitor and see certain events from our health checks trough Dynatrace via exception messages and we would want to react on those events automatically.
We are collecting the instance information trough request parameter but as far I know we don't have any way to enrich the generated alerts with request parameter or any other information which could reveal the problematic instance since all of the data is bundled on the same service currently. So currently we could get generate the custom alert from those events but without the actual service instance information which is causing little headache since our goal is to feed some automation system to make the needed recovery processes automatically.
We would like to utilize the Dynatrace as a event source in our case since most of our users have access there. But don't have any good idea currently how to pull out the information automatically on alerts which service instance is actually having the issue.
Basically the case is little bit similar like on this RFE or any other one where we would need pull out service level data per instance basis.
we had a really good blog about how to make alerts go from generic to meaningful and insightful. Its a really good read: https://www.dynatrace.com/news/blog/more-effective-ai-to-human-interactions-with-dynatrace-davis/
Thanks for the blog post and I got quite good ides from it to consider with other topics.
Our situation and need is little bit different since we are trying to avoid the case where every server would have it's own process group and therefore each service would be split per server basis.
So our case is that each service (and therefore each process group) contains multiple different instances which are running the same service.
The problematic part is that we would need to provide the information automatically if the Service1 is facing the problem on server3 so we could operate it automatically. Currently we are just getting the information out that the problem happens on Service1 without straight indication that the problem is happening only on one server instance.
We are currently playing around with the topic that can we pull out automatically somehow from Dynatrace that which server instance we would need to operate or should we just utilize some other event source than Dynatrace in our case.