ArgoCD apps in degraded state since upgrade to 2.12 #20716
Replies: 3 comments
-
The issue is related to missing pods due to auto scaling. I'll keep investigation and post how I fixed it. Seb. |
Beta Was this translation helpful? Give feedback.
-
Can you share more details, like screenshots, logs, events, please? |
Beta Was this translation helpful? Give feedback.
-
The issue is solved. It seems that ArgoCD's healthcheck method has evolved between version and supports more resource types. In my case it was a KafkaConnector kind that was with a bad status because of missing pods due to auto scaling process. In previous version, ArgoCD didn't care about the KafkaConnector state, but from version 2.12, it does. I've added below resource customizations (which is not perfect but do the job)
Hope it could help. Seb. |
Beta Was this translation helpful? Give feedback.
-
Hi,
I'm facing an issue upgrading ArgoCD 2.12.
I'm currently using the argo-cd helm chart 7.3.11 (argocd version v2.11.7).
I tried to upgrade to helm chart 7.7.0 (v2.13.0) and noticed that apps related to the same helm chart switched to "degraded" state after upgrade.
I rollback to 7.3.11 and everything was well again.
I've notice that the error appears as soon as we switched to v2.12 version (7.4.0 helm chart).
Is there any changes in the 2.12 version which could explain this behaviour?
Thanks for your help.
Seb.
Beta Was this translation helpful? Give feedback.
All reactions