Postgres service stuck in crash loop — catatonit:2 failed to exec pid1

naftalina

HOBBYOP

24 days ago

Hi, my Postgres service is in an infinite restart loop since 2026-05-20 04:47 UTC. It runs the standard

ghcr.io/railwayapp-templates/postgres-ssl:18 image (was working fine for months, last successful checkpoint 2026-05-19

22:11 UTC, then nothing).

Logs only show:

Mounting volume on: /var/lib/containers/railwayapp/bind-mounts/39e02866-.../vol_b4bag9aspoxnl9ou

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

Manual Restart from dashboard did not fix it (same loop continues).

Project ID: 72356486-a30d-4b64-956b-cbc0baedc631
Service ID: 5f58364e-7ac7-4805-a96e-18f73c6c574d
Environment: production (EU West)
Volume ID: vol_b4bag9aspoxnl9ou

Could you check if the latest build of postgres-ssl:18 is broken? My production app (Console Hub, on App Store and

Play Store) is down because users can't authenticate.

Thank you, urgent.

Solved

3 Replies

gadatos

PRO

24 days ago

i have same Error: read ECONNRESET

at TCP.onStreamRead (node:internal/stream_base_commons:216:20) {

errno: -104,

code: 'ECONNRESET',

syscall: 'read'

}

Error: read ECONNRESET

at TCP.onStreamRead (node:internal/stream_base_commons:216:20) {

errno: -104,

code: 'ECONNRESET',

syscall: 'read'

}

Error: read ECONNRESET

at TCP.onStreamRead (node:internal/stream_base_commons:216:20) {

errno: -104,

code: 'ECONNRESET',

syscall: 'read'

}

Railway

BOT

24 days ago

Your Postgres service crash-looping with the "failed to exec pid1" error aligns with the major service disruption that occurred between 02:25 UTC and 07:57 UTC on May 20, which affected the GCP Image Registry and EU West among other components. That incident is now resolved. Please try a fresh redeploy (not just a restart) from the three-dot menu on your latest deployment, as the image should now pull correctly from the restored registry.

Status changed to Awaiting User Response Railway • 24 days ago

naftalina

HOBBYOP

24 days ago

The "catatonit: failed to exec pid1: No such file or directory" error is consistent with a corrupted image pull during the recent platform-wide service disruption that affected our Image Registry, now resolved. Redeploying your Postgres service (three-dot menu on the crashed deployment, then "Redeploy") should pull a fresh image and resolve the crash loop - your data on the volume is unaffected. Risolto...

Status changed to Awaiting Railway Response Railway • 24 days ago

Status changed to Solved naftalina • 24 days ago

Welcome!