Couldn't delete HttpChaos because request too large #4076
Comments
|
Update. I repaired the large HttpChaos resource and delete it by the following commands. kubectl delete validatingwebhookconfigurations.admissionregistration.k8s.io chaos-mesh-validation-auth
kubectl patch --type=merge httpchaos delaytest2 -p '{"status":null}’But I think there should be a limitation for the event recorder to avoid the resource being too large. |
dearboll23
changed the title
|
Hi @dearboll23 , could you help us to reproduce this issue? Could you share the yaml definition of the "large HTTPChaos" resource? |
|
The resource is here. Seems like there's no limitation or rotation of the field |
|
I try to reproduce and create the httpchaos locally and failed to delete. It is hung over and there isn't any log about |
|
I think the author of the issue has a good point in saying that the event in status cannot grow indefinitely. I can't even apply the yaml provided by |
Bug Report
What version of Kubernetes are you using?
Client Version: v1.24.2
Server Version: v1.24.11-gke.1000
What version of Chaos Mesh are you using?
ghcr.io/chaos-mesh/chaos-mesh:v2.5.1
What did you do? / Minimal Reproducible Example
I want to delete a HttpChaos CR, but the deletion is hanging over, and the chaos-controller's log shows the target HttpFault resource is very large, with an error message
request too large, seems like an error message from etcd log. So I guess the issue is the size of HttpChaos is too large to delete it in etcd server. When I show the HttpChaos as yaml format, thestatusfield is really very long recording all the history of itself.What did you expect to see?
What did you see instead?
Output of chaosctl
The text was updated successfully, but these errors were encountered: