circagraph
Sign inSign up

openchronology/timeline-app on GitHub

The OpenChronology repository on GitHub, mirrored by och-import-github: issues and pull requests as durations (ongoing while open), what happened to them as moments, plus releases, milestones, commits and branches. The Branches view shows the most recently active branches each in a track of its own.

Source: https://github.com/openchronology/timeline-app

3,131 moments · 2026-10-05 CE 20:21:49 UTC – 2026-11-16 CE 12:10:45 UTC · by @athan · @athan

#373 Package upgrades start the enabled mail worker again (#371)

–

  • github
  • github/openchronology/timeline-app
  • github/openchronology/timeline-app#373
  • pull request
  • state:merged

Fixes #371 An upgrade stops the server and the mail worker; `openchronology-migrate` started only the server, so an enabled mail worker stayed down and account email waited in `oc_mail_outbox`. **Debian and RPM** (`deploy/debian-server/openchronology-migrate`, which the RPM also ships): after restarting the server, `systemctl is-enabled --quiet openchronology-mail.service && systemctl restart openchronology-mail.service`, printing "OpenChronology mail worker restarted." A disabled worker stays stopped; `--no-restart` still restarts nothing. `postinst` and the RPM's `openchronology-setup` do the same where they restart the server. **FreeBSD**: the upgrade stops `openchronology_mail` too; `openchronology-migrate` now restarts it when it is enabled. **Homebrew**: upgrades don't stop the agent, but it keeps running the old version; `openchronology-migrate` restarts the mail agent when it is loaded. **Windows**: the installer already restarted an Automatic `OpenChronologyMail`; `openchronology-migrate.cmd` now does too (and still fails if the server's restart fails). **Compose**: `docs/docker.md` says to recreate `mail-worker` with the app. **Tests**: `test/server-mail-upgrade.mjs`, used by `server-deb.mjs` and `server-rpm.mjs`. The containers have no systemd, so a stand-in `systemctl` (with `/run/systemd/system`) keeps the units' enabled state and runs the mail worker as its unit would (with a placeholder Resend key; the outbox is empty). It checks that the upgrade stops the enabled worker, that `openchronology-migrate` starts it and it stays up, that `--no-restart` starts nothing, and that a disabled worker stays stopped. The FreeBSD, Homebrew and Windows tests check that their workers, which are not enabled, are not restarted. **Docs**: README.Debian, README.RPM, README.FreeBSD, README.Homebrew, README.Windows.txt, the deploy READMEs and an "Upgrading" note in `docs/administration.md`. Local: `node test/server-deb.mjs` (debian13) and `node test/server-rpm.mjs` (fedora44) pass on freshly built packages, and `npm test` passes (361/361). 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_011fFY8eruEi1xAkkMBni13B https://github.com/openchronology/timeline-app/pull/373

Related

Open #373 Package upgrades start the enabled mail worker again (#371) in the timeline

#373 Package upgrades start the enabled mail worker again…