repo-server rpc error: code = Canceled desc = context canceled

[maximemoreillon]

Checklist:

• I've searched in the docs and FAQ for my answer: https://bit.ly/argocd-faq.
• I've included steps to reproduce the bug.
• I've pasted the output of argocd version.

Describe the bug

After installing Argo-CD in my cluster, the repo-server pod ends up in CrashLoopBackOff. Logs are availab le hereunder.

This happens on Microk8s and RKE2 clusters that I have on-premise, behind an HTTP proxy.
I have configured the HTTP_PROXY, HTTPS_PROXY and NO_PROXY environment variables and can confirm they propagate to the pod.

NO_PROXY is set as follows: localhost,127.0.0.1,172.16.0.0/12,10.0.0.0/8,.svc,.cluster.local,0.0.0.0,10.

Installing in similar fashion in an EKS cluster that has direct internet access (No HTTP proxy) works without problem.

To Reproduce

1. Have an on-premise RKE2 or Microk8s cluster behind an HTTP proxy
2. Install Argo CD as per documentation or via helm.

Expected behavior

The pod should be running without crashing.

Version

argocd: v3.2.3+2b6251d
  BuildDate: 2025-12-24T12:35:36Z
  GitCommit: 2b6251dfedb54de40596272a73ed1fb19d740219
  GitTreeState: clean
  GoVersion: go1.25.0
  Compiler: gc
  Platform: windows/amd64
argocd-server: v3.2.3+2b6251d

Logs

time="2026-01-11T22:33:21Z" level=info msg="starting grpc server"
time="2026-01-11T22:33:44Z" level=error msg="&{0xc000672ab0 0xc001166a00 {} 0x4fba00 true false true {{} {0 0}} {{} 0} 0xc0009abd40 {0xc00115a2d0 map[] false false} map[] false 0 -1 503 false false false [] {{} 0} [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0] [0 0 0] {{} {0 0}} <nil> true} rpc error: code = Canceled desc = context canceled"
time="2026-01-11T22:33:50Z" level=info msg="started call" grpc.component=server grpc.method=Check grpc.method_type=unary grpc.request.deadline="2026-01-11T22:34:49Z" grpc.service=grpc.health.v1.Health grpc.start_time="2026-01-11T22:33:50Z" grpc.time_ms=0.016 peer.address="127.0.0.1:34724" protocol=grpc
time="2026-01-11T22:33:50Z" level=info msg="finished call" grpc.code=OK grpc.component=server grpc.method=Check grpc.method_type=unary grpc.request.deadline="2026-01-11T22:34:49Z" grpc.service=grpc.health.v1.Health grpc.start_time="2026-01-11T22:33:50Z" grpc.time_ms=0.159 peer.address="127.0.0.1:34724" protocol=grpc
time="2026-01-11T22:34:04Z" level=error msg="&{0xc000884bd0 0xc001310640 {} 0x4fba00 true false true {{} {0 0}} {{} 0} 0xc0001033c0 {0xc0013163c0 map[] false false} map[] false 0 -1 503 false false false [] {{} 0} [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0] [0 0 0] {{} {0 0}} <nil> true} rpc error: code = Canceled desc = context canceled"
time="2026-01-11T22:34:14Z" level=error msg="&{0xc000673320 0xc001167040 {} 0x4fba00 true false true {{} {0 0}} {{} 0} 0xc0009abd40 {0xc00115a5a0 map[] false false} map[] false 0 -1 503 false false false [] {{} 0} [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] [0 0 0 0 0 0 0 0 0 0] [0 0 0] {{} {0 0}} <nil> true} rpc error: code = Canceled desc = context canceled"
time="2026-01-11T22:34:20Z" level=info msg="got signal terminated, attempting graceful shutdown"
time="2026-01-11T22:34:20Z" level=info msg="clean shutdown"