postgres db fail on restart
solvedev-art
HOBBYOP

a month ago

Deployment ID of the crashed Postgres-prod service:

fee4cde6-5cd7-49f6-9326-0cf50928c8d2

Error message:

ERROR (catatonit:2): failed to exec pid1: No such file or directory.

I have this error on my postgresdb. what can i do to fix it?

Solved$10 Bounty

Pinned Solution

ve-jo
HOBBY

a month ago

catatonit: failed to exec pid1 error has been reported in multiple Postgres/Railway services after the May 19/20 platform incident

(https://blog.railway.com/p/incident-report-may-19-2026-gcp-account-outage)

Try to do a redeploy of a database service.

5 Replies

Status changed to Open Railway about 1 month ago


ve-jo
HOBBY

a month ago

catatonit: failed to exec pid1 error has been reported in multiple Postgres/Railway services after the May 19/20 platform incident

(https://blog.railway.com/p/incident-report-may-19-2026-gcp-account-outage)

Try to do a redeploy of a database service.


solvedev-art
HOBBYOP

a month ago

if i deploy new postgres, Will I lose all data from it?


solvedev-art

if i deploy new postgres, Will I lose all data from it?

ve-jo
HOBBY

a month ago

No. You're deploying the postgres service itself. Volume that connected to it are still being the same.

To actually "lose the data" you will need to wipe the volume separately. Thats how volumes works.


ve-jo
HOBBY

a month ago

So in your current postgres just find your deployment, click three dots and "Redeploy".


85ed
HOBBY

a month ago

What was suggested above is correct redeploying the Postgres service itself does not automatically delete your database volume/data.

In Railway, the database data lives on the attached volume separately from the running container.

The:

“catatonit: failed to exec pid1”

error usually points more toward a broken container/runtime startup state than actual Postgres data corruption.

So the safest first recovery step is:

  • redeploy the existing Postgres service
  • keep the same attached volume
  • avoid deleting/wiping the volume unless absolutely necessary

Only removing the volume itself would risk permanent data loss.

Given the timing, this also aligns with several container/runtime issues reported after the May platform incident.


Status changed to Solved brody about 1 month ago


Welcome!

Sign in to your Railway account to join the conversation.

Loading...