Failed deploy model due to InternalFailure (500)
#3.773 aberto em 17 de jul. de 2024
Métricas do repositório
- Stars
- (4.317 estrelas)
- Métricas de merge de PR
- (Métricas PR pendentes)
Description
Describe the bug ALB managed by the controller is completely unresponsive, every action on the listeners return a 500 (even on AWS console).
Only errors in the log are: {"level":"error","ts":"2024-07-16T19:51:59Z","msg":"Reconciler error","controller":"ingress","object":{"name":"alb-ext"},"namespace":"","name":"alb-ext","reconcileID":"81af4f98-71d3-4a2a-870a-b431053e6738","error":"InternalFailure: \n\tstatus code: 500, request id: 617ba366-572a-4da0-8f7f-0ef2ac3d110b"}
Steps to reproduce
No idea, just tried to deploy a new ingress. The LB worked fined, I deployed an ingress which had an issue on with an invalid secretName in alb.ingress.kubernetes.io/auth-idp-oidc, but I quickly fixed it. The LB went to this degraded state just after that. So I guess the issue is closely related to https://github.com/kubernetes-sigs/aws-load-balancer-controller/issues/2819, but the related secret is now fixed. And even if it wasn't fix, the whole ALB shouldn't be down because of a single misconfigured Ingress, right ?
I tried to delete the ALB and let the controller recreate it, but it recreated it in the same state.
Environment
- AWS Load Balancer controller version:
v2.8.1 - Kubernetes version:
1.30 - Using EKS (yes/no), if so version?
eks.2
Additional Context: I also opened a ticket in AWS support, because even if the ALB is misconfigured, it shouldn't raise 500. It should give more context about the misconfiguration, right now I'm just in the dark.