Argocd helm chart deployment cannot reach argocd redis pod

[lamenawuer]

Checklist:

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

Describe the bug

With a fresh install of the latest ArgoCD helm chart my cluster seems to be badly detected and I cannot add repositories. Looking at the logs I see that it tries to reach the Redis service to no avail (10.97.57.226). I've tried to undeploy and redeploy with helm and it still does not work

To Reproduce

Fresh install of the latest ArgoCD helm chart with no particular values excepting an ingress

Expected behavior

Cluster status displays ok and repositories can be added successfully

Screenshots

Version

argocd: v2.8.2+dbdfc71.dirty
  BuildDate: 2023-08-25T03:59:20Z
  GitCommit: dbdfc712702ce2f781910a795d2e5385a4f5a0f9
  GitTreeState: dirty
  GoVersion: go1.21.0
  Compiler: gc
  Platform: linux/amd64
argocd-server: v2.7.11+ec195ad.dirty
  BuildDate: 2023-08-07T14:40:17Z
  GitCommit: ec195adad84c61c6151d553b9fdce3c258b1325d
  GitTreeState: dirty
  GoVersion: go1.19.10
  Compiler: gc
  Platform: linux/amd64
  Kustomize Version: v5.0.1 2023-03-14T01:32:48Z
  Helm Version: v3.11.2+g912ebc1
  Kubectl Version: v0.24.2
  Jsonnet Version: v0.19.1

Logs

time="2023-09-05T14:20:31Z" level=warning msg="Failed to resync revoked tokens. retrying again in 1 minute: dial tcp 10.97.57.226:6379: i/o timeout"                                                                         
time="2023-09-05T14:21:31Z" level=warning msg="Failed to resync revoked tokens. retrying again in 1 minute: dial tcp 10.97.57.226:6379: i/o timeout"                                                                         
time="2023-09-05T14:21:54Z" level=info msg="received unary call /version.VersionService/Version" grpc.method=Version grpc.request.content= grpc.service=version.VersionService grpc.start_time="2023-09-05T14:21:54Z" span.ki
time="2023-09-05T14:21:54Z" level=info msg="finished unary call with code OK" grpc.code=OK grpc.method=Version grpc.service=version.VersionService grpc.start_time="2023-09-05T14:21:54Z" grpc.time_ms=0.496 span.kind=server time="2023-09-05T14:21:54Z" level=info msg="received unary call /version.VersionService/Version" grpc.method=Version grpc.request.content= grpc.service=version.VersionService grpc.start_time="2023-09-05T14:21:54Z" span.ki
time="2023-09-05T14:21:54Z" level=info msg="finished unary call with code OK" grpc.code=OK grpc.method=Version grpc.service=version.VersionService grpc.start_time="2023-09-05T14:21:54Z" grpc.time_ms=0.435 span.kind=server
time="2023-09-05T14:22:04Z" level=info msg="received unary call /version.VersionService/Version" grpc.method=Version grpc.request.content= grpc.service=version.VersionService grpc.start_time="2023-09-05T14:22:04Z" span.ki
time="2023-09-05T14:22:04Z" level=info msg="finished unary call with code OK" grpc.code=OK grpc.method=Version grpc.service=version.VersionService grpc.start_time="2023-09-05T14:22:04Z" grpc.time_ms=0.426 span.kind=server time="2023-09-05T14:22:31Z" level=warning msg="Failed to resync revoked tokens. retrying again in 1 minute: dial tcp 10.97.57.226:6379: i/o timeout" 

[gruberdev]

That might be related to #14725 and #13683.

Also, there seems to be a mismatch of versioning between the components:

• argocd-server: v2.7.11+ec195ad.dirty
• argocd: v2.8.2+dbdfc71.dirty

I have encountered a similar issue that forced me to downgrade to v2.6.x, in order to get it working again as intended.

[lamenawuer]

Hi thanks for the answer, so I just force the global image to be a 2.6.x?

[lamenawuer]

Hi @gruberdev, could you please explain in detail what I would need to change from my side make this work?