17 days ago
Hi Railway team,
Our PostgreSQL service crashed and entered a crash loop yesterday (2026-05-19) that required a manual restart to
recover. We're concerned it may happen again and would like help understanding the root cause.
Service details:
- Project: hanhoggan-project
- Service: PostgreSQL
- Volume ID: vol_fyg6s18n9zht6sge
What happened (from Postgres logs):
At 22:20 UTC the database received a fast shutdown request and shut down. After that the container entered a repeated
crash loop with this error:
Mounting volume on:
/var/lib/containers/railwayapp/bind-mounts/f5dd0b40-d041-41b2-9faa-e5bdfff88f42/vol_fyg6s18n9zht6sge
ERROR (catatonit:2): failed to exec pid1: No such file or directory
This repeated dozens of times. The service eventually recovered on its own but caused ~30 minutes of downtime for our
production application.
Questions:
- What triggered the fast shutdown at 22:20 UTC?
- Why did the container fail to restart (failed to exec pid1: No such file or directory)?
- Is there a known issue with the volume mount on this service?
- How can we prevent this from happening again?
We would also appreciate advice on setting up automatic backups for this service.
Thank you.
1 Replies
17 days ago
The crash loop your PostgreSQL service experienced on May 19 was caused by a platform-wide service disruption that has since been resolved. The "failed to exec pid1" errors were a side effect of the underlying infrastructure issue, not a problem with your volume or service configuration. No action is needed on your end to prevent recurrence. For automated backups, you can configure daily, weekly, or monthly backup schedules in your service's Backups tab, and for more granular recovery you can enable Point-in-Time Recovery.
Status changed to Awaiting User Response Railway • 17 days ago
Status changed to Solved Railway • 10 days ago