15 Commits
Author SHA1 Message Date
rock e28e3cae81 fix: wait for dind to be ready before starting runner daemon 2026-09-07 13:22:09 -07:00
rock febfb055ae fix: use tcp://localhost:2375 for dind (no TLS, no socket permission issues) 2026-09-07 13:20:20 -07:00
rock 9d61c42b99 fix: run runner as root to access dind socket 2026-09-07 13:18:54 -07:00
rock 594e752d47 fix: runner uses unix socket instead of TLS TCP for dind
Job containers spawned by the runner run inside dind. With TCP+TLS
(tcp://localhost:2376), localhost inside those containers doesn't
reach the dind sidecar. Unix socket at /run/docker.sock works because
both runner and dind share the /run emptyDir.

Also disables DOCKER_TLS_CERTDIR so dind creates the socket instead
of only listening on TLS TCP.
2026-09-07 13:16:49 -07:00
rock 1a8e87131b ci: fix runner labels + CoreDNS rewrite + cleanup
- Runners use public images (code.forgejo.org/forgejo/runner:6)
- Labels pull from Docker Hub: golang:1.26, node:22, rust:1-bookworm
- Add CoreDNS api.riotpiao.com rewrite
- Fix runner re-registration to keep labels in sync
- Add unified CI pattern docs to CLAUDE.example.md
- Remove dead .forgejo/ workflow dir (Forgejo uses .gitea/)
2026-09-07 13:01:56 -07:00
rock 51c66299e9 forgejo-runner: gc every 30min instead of daily
- Fix template to use .Values.gc.schedule instead of hardcoded cron
- Change schedule from daily 03:00 UTC to every 30 minutes
- Prevents DinD PVC fill-up (was at 93% before manual prune)
2026-09-02 19:52:41 -07:00
rock 686c962ea9 fix(ci): aggressive GC for heavy Rust cargo builds
- GC CronJob runs 2x daily (02:00 & 05:00 UTC) instead of once (03:00)
- Delete incomplete actcache uploads immediately (tmp/ dirs from failed writes)
- Reduce actcache retention from 3 days to 1 day
- Prevents cargo cache backlog on rust runner under heavy commit load
- Incomplete entries were accumulating 900MB+ each, filling 1Gi PVC instantly
2026-08-30 09:23:09 -07:00
rock 91729f65e9 feat(ci): add GC CronJob for runner cleanup, expand reg PVCs to 20Gi
- Add gc-cronjob.yaml: daily prune of DinD Docker images/volumes/build-cache
  and actcache across all forgejo-runner pods. Keeps :latest tagged images,
  deletes non-latest older than 72h.
- Expand runner reg PVCs from 1Gi to 20Gi (all three runners) to prevent
  action tool cache from filling disk.
- Rust runner gets explicit 20Gi persistence override.
- GC only renders from golang (default) values to avoid duplicate resources.
2026-08-30 07:08:59 -07:00
Story Crater Bot d6fcf1016e refactor(forgejo-runner): template PVC names off Release.Name for multi-instance reuse 2026-08-21 16:30:41 -07:00
Story Crater Bot 7234eb596f fix(forgejo-runner): job containers must use host network to reach dind 2026-08-21 16:22:19 -07:00
Story Crater Bot 7cb438e25d fix(forgejo-runner): egress to ingress-nginx by namespace, not a stale LB IP 2026-08-21 16:22:19 -07:00
Story Crater Bot c46e69fd43 fix(forgejo-runner): allow job containers to mount /docker-certs/client so docker login/build/push work 2026-08-21 16:22:19 -07:00
Story Crater Bot 720181c900 feat: let the runner build and the cluster pull from the Forgejo registry
- Runner egress: allow 192.168.1.160/32:443. forgejo.riotpiao.com resolves to
  the ingress LB, inside the 192.168.1.0/24 block the NetworkPolicy denies, so
  docker push hung until timeout.
- dind CA: also mount homelab-ca at /etc/docker/certs.d/forgejo.riotpiao.com/,
  the path dockerd actually reads for per-registry trust.
- Pull secret: dockerconfigjson for the api namespace; /v2/ answers 401.
- AppProject: allow the Forgejo repo as a source for api-gw.
2026-08-19 21:48:01 -07:00
Story Crater Bot b863b6974e fix(forgejo-runner): cicd ns PSS privileged (dind needs it) + mount homelab-ca as ConfigMap not Secret — runner RS created 0 pods under baseline PSS, then FailedMount because homelab-ca is a ConfigMap trust bundle, not a Secret 2026-08-18 15:08:04 -07:00
Story Crater Bot 41f5b05395 refactor(k8s): consolidate to infra/+apps/ single-source tree, dedicated per-app CNPG (authentik-db/temporal-db), wire monitoring-config, forgejo→cicd ns, drop orphan/stale (data-schemas, ollama, story-crater, sqs/argocd, key-rotation) 2026-08-18 15:08:03 -07:00