Postgres crashed
cococunigan
HOBBYOP

2 months ago

Hi, I did what similar thread suggest. I increased the volume. But the db keeps crashing. Any help please.

2026-03-20 08:21:27.647 UTC [7] LOG: listening on Unix socket "/var/run/postgresql/.s.PGSQL.5432"

2026-03-20 08:21:27.656 UTC [33] LOG: database system was interrupted while in recovery at 2026-03-20 08:21:23 UTC

2026-03-20 08:21:27.656 UTC [33] HINT: This probably means that some data is corrupted and you will have to use the last backup for recovery.

2026-03-20 08:21:28.031 UTC [33] LOG: database system was not properly shut down; automatic recovery in progress

2026-03-20 08:21:28.035 UTC [33] LOG: redo starts at 0/3FE30788

2026-03-20 08:21:30.510 UTC [33] LOG: redo done at 0/48FFEEC8 system usage: CPU: user: 0.15 s, system: 0.08 s, elapsed: 2.47 s

2026-03-20 08:21:30.526 UTC [33] FATAL: could not write to file "pg_wal/xlogtemp.33": No space left on device

2026-03-20 08:21:30.532 UTC [7] LOG: startup process (PID 33) exited with exit code 1

2026-03-20 08:21:30.532 UTC [7] LOG: terminating any other active server processes

2026-03-20 08:21:30.533 UTC [7] LOG: shutting down due to startup process failure

2026-03-20 08:21:30.544 UTC [7] LOG: database system is shut down

Solved

4 Replies

Railway
BOT

2 months ago

Your Postgres volume is still completely full, which is why the WAL recovery process fails on every restart attempt. The Hobby plan default volume is 5GB, and the recovery process itself needs additional free space to replay WAL files. You can live resize the volume further to give Postgres enough room to complete recovery, but note that Postgres WAL replay during recovery can temporarily require more space than normal operation.


Status changed to Awaiting User Response Railway 2 months ago


Railway

Your Postgres volume is still completely full, which is why the WAL recovery process fails on every restart attempt. The Hobby plan default volume is 5GB, and the recovery process itself needs additional free space to replay WAL files. You can [live resize](https://docs.railway.com/volumes#live-resizing-the-volume) the volume further to give Postgres enough room to complete recovery, but note that Postgres WAL replay during recovery can temporarily require more space than normal operation.

cococunigan
HOBBYOP

2 months ago

I was on a free plan. Then increased the volume from 500 mb to 5 gb. You think I should increase it further ? How much ?


Status changed to Awaiting Railway Response Railway 2 months ago


2 months ago

Hi! The volume resize from 500 MB to 5 GB didn't fully apply on our end — your database was still seeing the old 500 MB of space, which is why it couldn't recover. We've fixed this and redeployed your service. You should now have the full 5 GB available. We've also shipped a fix to prevent this from happening on future volume resizes.


Status changed to Awaiting User Response Railway 2 months ago


Status changed to Solved jake 2 months ago


cococunigan
HOBBYOP

2 months ago

I waited few hours then I started a new db. Thank you


Status changed to Awaiting Railway Response Railway 2 months ago


Status changed to Solved chandrika 2 months ago


Welcome!

Sign in to your Railway account to join the conversation.

Loading...