Hi,
I am experiencing daily restarts of the apigee-watcher and controller manager pods in my Apigee Hybrid setup. No other pods are affected, and these specific pods do not show any errors or events indicating a reason for the restart. I’ve checked logs, resource limits, node health, and events, but everything appears normal. What could be causing only these two pods to restart daily, and how can I troubleshoot this further or prevent it?
Thanks.
Thank you for your question, @shayan - We’re following this thread and making sure it gets the attention it deserves. We’re also inviting the community to join in and share their expertise.
Hi @shayan – sorry to hear about the troubles.
The good news is that neither of those components are on the critical path for processing API traffic, but daily restarts are not to be expected. Is it precisely once per day?
What version of Apigee hybrid are you using and what Kubernetes platform is it installed in?
I see that you’ve already checked all the obvious places for clues without much luck. For this type of behaviour, I would definitely expect something to show in the pod logs. If you’re not seeing anything in the logs, I recommend you raise a support case explaining all of the information you have already. Support can help gather more details about your cluster and installation which could lead to some pointers. Feel free to include a link to this post as well.
Thanks!
Hi @omidt ,
We are using a forward proxy instead of a firewall and have specified the proxy settings in the overrides. The rest of the components (runtime, synchronizer, etc.) pods are working fine, but we are still facing the issue.