2 hours ago
My production service whatsapp-ams-audit is currently crashed because a stale process lock file remains on the persistent volume.
Project: noble-tenderness
Service: whatsapp-ams-audit
Volume ID: fd61bd0c-85e2-4ba9-852f-ab24a0195f7b
Volume mount: /data
The application fails during startup with:
Writer lock exists or is unavailable. Confirm the previous writer stopped and reconcile uncertain saves before clearing the lock.
Error location: server.mjs:119
The stale file that needs to be removed is:
/data/teacher-auth.sqlite.writer.lock
I have already tried Railway CLI volume browse/delete, but both fail with an SFTP initialization timeout. The service is crashed, so the Railway Console also reports “No running instances” and cannot access the mounted filesystem.
Please remove ONLY this stale file:
/data/teacher-auth.sqlite.writer.lock
Please DO NOT delete, modify, replace, detach, or recreate the volume.
Please DO NOT touch /data/teacher-auth.sqlite or any other file.
Please DO NOT modify environment variables, application code, or other production data.
Once this single stale lock file has been removed, please let me know so I can restart the existing service.
1 Replies
2 hours ago
This thread has been opened as a bounty so the community can help solve it.
Status changed to Open Railway • about 2 hours ago
an hour ago
You can temporarily set the start command to sleep infinity, then you can access and remove the files from service console or CLI, remove the start command once you done and the application should start normally