How to reduce time it takes to perform a health check on dead pod

[heramb_sawant]

Hi Team,
Node/host having k8s pods , later for some reason pods were killed. Though pods were not exist(already killed) , host having oneagent , was still looking for tcp connectivity with process running on that pod. 

Can someone please tell me how we can minimize the time it takes to perform a health check on dead pod?  currently its more than 15 min  and alert remain open for long time,  it is  too long.  I would say DT should look for the pod for one minute and if doesn’t exist don’t report.

Regards,
Heramb Sawant

[heramb_sawant]

Any help will be appreciated. 

[IzabelaRokita]

Hey @heramb_sawant, did you manage to find the solution to your problem? If so, it would be amazing if you've shared it with the rest of the Community! If not, let me know, and I'll look for some further assistance 😊

[Mizső · Accepted solution]

Hi @heramb_sawant,

You can play with the time at the anomaly detection section globally or namespace level. I guess non-ready pod state could be good for you. 

Best regards,

János