Skip to content

increase liveness probe and readiness probe limits for helm-operator

I did this in !388 (merged) already and it seems to make the helm-operator startup a bit less buggy. I hope it fixes this problem in this pipeline too:

root@master:~# kubectl get pods -A
NAMESPACE     NAME                              READY   STATUS             RESTARTS   AGE
kube-system   coredns-854c77959c-vv7fc          1/1     Running            0          11d
kube-system   metrics-server-86cbb8457f-q8sdm   1/1     Running            0          11d
oas           flux-memcached-5dbc947678-thhkm   1/1     Running            0          89m
oas           flux-6ccf67466b-9h55l             1/1     Running            0          89m
oas           local-flux-5d8cfdc7c6-5g2dq       1/1     Running            0          88m
oas           helm-operator-5f9cc8c4ff-nl95p    0/1     CrashLoopBackOff   31         89m
root@master:~# kubectl describe pod -n oas helm-operator-5f9cc8c4ff-nl95p 
...
Events:
  Type     Reason     Age                    From               Message
  ----     ------     ----                   ----               -------
  Normal   Scheduled  89m                    default-scheduler  Successfully assigned oas/helm-operator-5f9cc8c4ff-nl95p to master
  Normal   Pulling    89m                    kubelet            Pulling image "docker.io/fluxcd/helm-operator:1.2.0"
  Normal   Pulled     88m                    kubelet            Successfully pulled image "docker.io/fluxcd/helm-operator:1.2.0" in 8.715501395s
  Warning  Unhealthy  87m (x6 over 88m)      kubelet            Liveness probe failed: Get "http://10.42.1.2:3030/healthz": dial tcp 10.42.1.2:3030: connect: connection refused
  Normal   Killing    87m (x2 over 88m)      kubelet            Container flux-helm-operator failed liveness probe, will be restarted
  Normal   Created    87m (x3 over 88m)      kubelet            Created container flux-helm-operator
  Normal   Started    87m (x3 over 88m)      kubelet            Started container flux-helm-operator
  Normal   Pulled     48m (x16 over 88m)     kubelet            Container image "docker.io/fluxcd/helm-operator:1.2.0" already present on machine
  Warning  BackOff    9m (x302 over 85m)     kubelet            Back-off restarting failed container
  Warning  Unhealthy  3m57s (x111 over 88m)  kubelet            Readiness probe failed: Get "http://10.42.1.2:3030/healthz": dial tcp 10.42.1.2:3030: connect: connection refused
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information