After enabling tests with fixed JBEAP-17170 (KUBE_PING) test failure popped up in case of DNS_PING. Current ratio is 1:2 in lab.

Thats the first time I see it failing. DNS_PING variant was created during investigation of JBEAP-17170 and was ignored with KUBE_PING variant because of JBEAP-17170. So it wasn't run many times so far in lab. This one failure occured on OCP3.

I am attaching test log, matching events and log of pods which are available at the end of test. I dont have available logs of pods with old version of application before rolling update. Could you spot from these resources what the problem could be?

As far as I can tell, I don't see split brain occurred as in case of JBEAP-17170. And I don't see any exception in pods. That makes me think 500 error occurred on "old" pods.

I will continue with investigation, but I would be glad for any hints.