2 Commits

Author SHA1 Message Date
Hartmut 5fd650460e docs(gitea): bump postgres stop_grace_period to 120s
CI / Lint (push) Waiting to run
CI / Unit Tests (push) Waiting to run
Docker Deploy Test / Fresh-Linux Docker Deploy (push) Waiting to run
CI / Architecture Guardrails (push) Has started running
CI / Typecheck (push) Has started running
CI / Assistant Split Regression (push) Has started running
CI / Build (push) Has been cancelled
CI / E2E Tests (push) Has been cancelled
Release Image / Build And Push Images (push) Has been cancelled
60s was not enough when the DB has active WAL writes from recent CI
runs. 120s gives postgres the headroom for a clean shutdown and avoids
the slow crash-recovery fsync on the next start.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-12 14:35:14 +02:00
Hartmut 6a37abb8c1 docs(gitea): swap runner base image to catthehacker/ubuntu:act-latest
node:20-bookworm has no docker CLI, which caused release-image.yml and
any workflow using docker login/buildx to fail with "docker: command
not found" despite the socket mount being in place.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-12 14:17:05 +02:00
+10 -4
View File
@@ -64,11 +64,13 @@ services:
image: postgres:16-alpine
container_name: gitea-db
restart: unless-stopped
# Geben wir Postgres Zeit für sauberen Shutdown beim Stop/Replace.
# Geben wir Postgres großzügig Zeit für sauberen Shutdown beim Stop/Replace.
# Ohne diesen Grace muss beim nächsten Start Crash-Recovery laufen
# (fsync über alle Files) — auf HDD-backed QNAP-Storage dauert das
# schnell 5-10 Minuten und blockt Gitea beim Start.
stop_grace_period: 60s
# 120s ist bewusst großzügig: bei viel WAL-Write (CI-Läufe mit Artefakten)
# kann auch ein sauberer Shutdown 30-60s dauern.
stop_grace_period: 120s
environment:
- POSTGRES_USER=gitea
- POSTGRES_PASSWORD=UGi2VZA7SgYGov
@@ -86,9 +88,12 @@ services:
- gitea
environment:
- GITEA_INSTANCE_URL=http://gitea:3000
- GITEA_RUNNER_REGISTRATION_TOKEN=${GITEA_RUNNER_REGISTRATION_TOKEN}
- GITEA_RUNNER_REGISTRATION_TOKEN=218iFl8s3a6uJxntyoobzu24pQJBGGVIWmdtJbXh
- GITEA_RUNNER_NAME=qnap-runner-1
- GITEA_RUNNER_LABELS=ubuntu-latest:docker://node:20-bookworm,ubuntu-22.04:docker://node:20-bookworm
# catthehacker/ubuntu:act-latest statt node:20-bookworm, weil sonst
# `docker`-CLI in Job-Containern fehlt und Workflows wie release-image.yml
# (docker login/buildx) mit "docker: command not found" scheitern.
- GITEA_RUNNER_LABELS=ubuntu-latest:docker://catthehacker/ubuntu:act-latest,ubuntu-22.04:docker://catthehacker/ubuntu:act-22.04
- CONFIG_FILE=/config.yaml
networks:
- gitea
@@ -223,6 +228,7 @@ Zusätzlich: QNAP **Storage & Snapshots** → Volume-Snapshots für `/share/Cont
- Socket-Mount prüfen
- `container.network` in der inline-generierten Runner-Config muss zum echten Docker-Netzwerknamen passen (`docker network ls`)
- Fehler `docker: command not found` → Job-Container hat kein Docker-CLI. Runner-Label muss ein Image verwenden, das `docker` mitbringt (z.B. `catthehacker/ubuntu:act-latest`). `node:*`-Images reichen nicht, weil dort nur Node installiert ist
**`uses: actions/checkout@v4` schlägt fehl:**