Kong proxy goes unresponsive with normal traffic - kubelet restarts the container since liveness probe fails #11709
Unanswered
kranthirachakonda
asked this question in
Help
Replies: 1 comment
-
Hi @kranthirachakonda , I want to know how you solved it later. I seem to have encountered this error too. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Looks like we are running into Kong#8152 issue with 2.8.3 and also 3.0.1. We had new route/service added recently and there is increase in traffic but not huge. May ~15% increase. Only proxy container inside Kong pods gets restart due to /status timing out. We tried changing the timeout/period intervals but that didnt help much. We dont see our containers going above the cpu/memory requests.
Behavior is identical, where kong proxy container goes into unresponsive mode for about 4-5 mins and processes again. We also observed that while I gave memory requests of 4G, proxy container never goes above 512Mi. Not sure if there are default limits within LuaVM or threads etc. Can you please help me in debugging further.
db-less
kong:2.8.3
kubernetes-ingress-controller:2.8.2
Events:
These are few errors I see while proxy containers goes into a condition where status page fails to respond within 5s, causing kubelet to restart when liveness probe fails.
[lua] targets.lua:248: could not reschedule DNS resolver timer: process exiting, context: ngx.timer
[lua] init.lua:409: [cluster_events] failed to start recurring polling timer: process exiting, context: ngx.timer
Admin page
Status page
Beta Was this translation helpful? Give feedback.
All reactions