GitLab SSH Handshake failed: EOF #2536
-
|
After changing the a Github repo to Gitlab repo, i've run into the following error: The error occurs at random times, sometimes at night and others mid day. When it happens it sends around 2-10 messages on slack and then nothing for hours/days. Flux is installed trough Terraform Flux check output: ❯ flux check
► checking prerequisites
✗ flux 0.17.2 <0.27.3 (new version is available, please upgrade)
✔ kubectl 1.22.1 >=1.18.0-0
✔ Kubernetes 1.20.12-gke.1500 >=1.16.0-0
► checking controllers
✔ helm-controller: deployment ready
► ghcr.io/fluxcd/helm-controller:v0.17.0
✔ image-automation-controller: deployment ready
► ghcr.io/fluxcd/image-automation-controller:v0.20.0
✔ image-reflector-controller: deployment ready
► ghcr.io/fluxcd/image-reflector-controller:v0.16.0
✔ kustomize-controller: deployment ready
► ghcr.io/fluxcd/kustomize-controller:v0.21.0
✔ notification-controller: deployment ready
► ghcr.io/fluxcd/notification-controller:v0.22.0
✔ source-controller: deployment ready
► ghcr.io/fluxcd/source-controller:v0.21.2
✔ all checks passed |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 3 replies
-
|
Hi, I've seen this exact issue when using GitLab. It seems like an issue on the gitlab server itself. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the answer, i will up the If you get a solution merged, i would be glad to hear from you! |
Beta Was this translation helpful? Give feedback.
-
|
Do you happen to know what the root cause on the GitLab server might be ? We're seeing the same issue and would like to fix it on the GitLab server. |
Beta Was this translation helpful? Give feedback.
-
|
Seeing the same thing as well, maybe 2 or 3 times a day |
Beta Was this translation helpful? Give feedback.
Hi, I've seen this exact issue when using GitLab. It seems like an issue on the gitlab server itself.
It has helped to increase the
spec.intervalfor the git repositories to reduce the frequency of this failure. But it still happens once in a while.I've been working on sending a positive notification after such failures to signal that everything is alright in fluxcd/source-controller#604 . I hope that'll help in knowing about the failure better.