Resource Informer missing ingress resource on race condition
- Dominant language
- C#
- Stars
- 9.6k
- Forks
- 933
- Avg merge
- 12d 18h
- Merged PRs (30d)
- 2
Description
### Describe the bug
ingress resource not loaded in list on controller startup, and event for `ADDED` seems to be missed by the controller => ingress stuck with no controller taking ownership of this.
manually restarting controller / creating an edit event on the ingress resource fixes this behaviour
### To Reproduce
I'm not really sure how i caught this race condition, it happens in my yarp chart and only recently. There seems to be a few ticks between the controller loading and listing the ingresses, and the ingress resource `ADDED` event that the controller is missing
Got Exceptions? Include both the message and the stack trace
-->
this is the log from the startup, the correct ingress is not present in the list
[15:16:51 INF] Watching V1Ingress starting from resource version 175782309.
[15:16:51 INF] Ignoring null/squadron-67rbjo as the spec.controller is not the same as this ingress
[15:16:51 INF] Ignoring null/yarp as the spec.controller is not the same as this ingress
[15:16:51 INF] Completed synchronizing V1IngressClass resources from API server.
### Further technical details
as shown in images, for 19m no controller has taken ownership of the specified ingress resource. As stated, restarting the controller pod or triggering an ingress event will fix this issue.
- Include the version of the packages you are using : 2.1.0
- The platform (Linux/macOS/Windows) linux - k8s
Contributor guide
Assessment
This issue has not been assessed yet.