Azure Scale Set VM monitoring: How to deal with host or monitoring unavailable?

[rsmsdk]

Hi

I apologize if this is a generic questions.

We have configured Dynatrace monitoring of an Azure Scale Set where we get alerts for "Host or monitoring unavailable" everytime a VM is shut down automatically.

The problems stick and have to be removed manually.

This alerting add confusion for our Ops team, since the VM naming is dynamic and different everytime a new VM is added /removed by the defined schedule.

How does other deal with alerting and monitoring of Scale Sets?

I'm not part of the Azure team internally, so if configuration should be different on that side, any valid questions is valued.

Thank you.

[Mizső]

Hi @rsmsdk,

I am a little bit confused regarding your post. You metioned both: problems and alerts. I do not have solution for problem view. Regarding alerting you can define an alerting profile for AWS with an event filter, so if your DT is integrated with an ITSM tool eg SNOW or Opsgenie you will receive tickets / alets from AWS except "Host or monitoring unavailable":

I hope it helps.

Best regards,

Mizső

[rsmsdk]

Sorry about the confusion, we receive alerts for these problems, but I obviously need to be clear here.

The thing is that we want to receive alerts for host or monitoring unavailable, but the nature of scale sets from my experience, seem to be random host names appearing and disappearing with new hostnames each times. The nature of this causes alot of host unavailable by design, and I was wondering how other people deal with it.

[AntonPineiro · Accepted solution]

Hi,

You can adjust the sensitivity of anomaly detection for infrastructure base on your needs.

Best regards