Date: 2026-02-07
Cluster context: kubernetes-admin@kubernetes (API: https://192.168.1.9:6443)
As of 2026-02-07, Gitea is reachable and healthy and Gitea Actions is ready to execute jobs (runner is registered and online).
The repository storage location was moved to a path writable by the Gitea container UID/GID while keeping the data on the normal filesystem for emergency access.
Ready (k8s v1.35.0).gitea exists.deployment/gitea is Available with pod/gitea Running.ingress.networking.k8s.io/gitea) with hosts:
gitea.app.langitea-private.app.lanNodePort:
:30331:30402http://192.168.1.9:30331/ with Host: gitea.app.lan returns 200 OK.service/gitea-http responds on :3000.Giteas SQLite DB contains Actions tables:
action_runner count is 1 (runner registered).action_run and action_task counts are 0 (no jobs have executed).action_runner_token count is 1 (a registration token exists; do not print the token).A Kubernetes deployment runner (gitea/act_runner) is deployed and is declaring successfully to Gitea.
The Gitea pod is configured with:
gitea/gitea:1.24.6USER_UID=1240000012, USER_GID=1240000009/mnt/fastdata/k8s-pvcs/gitea/git -> /data/gitNote: previously repos were mounted from /mnt/bigdata/personal/projects/git and Gitea could not write there due to permission mismatch; that is no longer the active repo mount.
Recent logs show a webhook delivery attempt failing:
http://192.168.1.227:8080/webhookno route to hostIf your publish pipeline relies on that webhook target, it needs network/routing fixes.
kube-proxy was observed still pointing at the old API server IP (192.168.1.216:6443), which broke in-cluster Service routing (pod -> ClusterIP timeouts). It was updated to 192.168.1.9:6443 and restarted.