Cloudron Forum

@james said in PostgreSQL Service Stuck in a Restart Loop - Health Check Failing:

Since the last 3 lines state:

Oct 13 12:20:45 2025-10-13 10:20:45.712 UTC [15] LOG: database system is ready to accept connections
Oct 13 12:20:45 done
Oct 13 12:20:45 server started

And it still restarts after that? Just to make sure, maybe increase the memory limit even more.

Yes, it starts again over and over. Increasing memory to 4GB did not change anything.

Thank you for your feedback.

I have increased the memory to 2GB and rebootet several times. Here's the looping log:

Oct 13 12:19:48 server started
Oct 13 12:19:48 ==> Waiting for Postgres
Oct 13 12:19:48 pg_ctl: server is running (PID: 15)
Oct 13 12:19:48 /usr/lib/postgresql/16/bin/postgres "-D" "/var/lib/postgresql/16/main"
Oct 13 12:19:48 ==> Postgres is running
Oct 13 12:19:48 ==> Resetting root password
Oct 13 12:19:48 ALTER ROLE
Oct 13 12:19:48 ==> Upgrading pgvectors
Oct 13 12:19:48 2025-10-13 10:19:48.537 UTC [30] root@postgres ERROR: database "cloudronpgvectorupdate" already exists
Oct 13 12:19:48 2025-10-13 10:19:48.537 UTC [30] root@postgres STATEMENT: CREATE DATABASE cloudronpgvectorupdate
Oct 13 12:19:48 ERROR: database "cloudronpgvectorupdate" already exists
Oct 13 12:20:40 ==> Detected existing installation
Oct 13 12:20:40 ==> Copying updated config
Oct 13 12:20:40 ==> Updating existing postgresql
Oct 13 12:20:40 waiting for server to start....2025-10-13 10:20:40.763 UTC [15] LOG: starting PostgreSQL 16.8 (Ubuntu 16.8-0ubuntu0.24.04.1) on x86_64-pc-linux-gnu, compiled by gcc (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0, 64-bit
Oct 13 12:20:40 2025-10-13 10:20:40.765 UTC [15] LOG: listening on IPv4 address "0.0.0.0", port 5432
Oct 13 12:20:40 2025-10-13 10:20:40.766 UTC [15] LOG: listening on IPv6 address "::", port 5432
Oct 13 12:20:40 2025-10-13 10:20:40.777 UTC [15] LOG: listening on Unix socket "/var/run/postgresql/.s.PGSQL.5432"
Oct 13 12:20:40 2025-10-13 10:20:40.813 UTC [18] LOG: database system was interrupted; last known up at 2025-10-13 10:19:48 UTC
Oct 13 12:20:45 ....2025-10-13 10:20:45.582 UTC [18] LOG: database system was not properly shut down; automatic recovery in progress
Oct 13 12:20:45 2025-10-13 10:20:45.599 UTC [18] LOG: redo starts at 6/546D6180
Oct 13 12:20:45 2025-10-13 10:20:45.599 UTC [18] LOG: invalid record length at 6/546D7CF8: expected at least 24, got 0
Oct 13 12:20:45 2025-10-13 10:20:45.599 UTC [18] LOG: redo done at 6/546D7CA0 system usage: CPU: user: 0.00 s, system: 0.00 s, elapsed: 0.00 s
Oct 13 12:20:45 2025-10-13 10:20:45.640 UTC [16] LOG: checkpoint starting: end-of-recovery immediate wait
Oct 13 12:20:45 .2025-10-13 10:20:45.682 UTC [16] LOG: checkpoint complete: wrote 3 buffers (0.0%); 0 WAL file(s) added, 0 removed, 0 recycled; write=0.010 s, sync=0.006 s, total=0.047 s; sync files=4, longest=0.003 s, average=0.002 s; distance=6 kB, estimate=6 kB; lsn=6/546D7CF8, redo lsn=6/546D7CF8
Oct 13 12:20:45 2025-10-13 10:20:45.712 UTC [15] LOG: database system is ready to accept connections
Oct 13 12:20:45 done
Oct 13 12:20:45 server started```

Hello,

My server recently experienced a severe CPU overload. I have resolved the overload, but my PostgreSQL service is now stuck in a constant restart loop, which prevents apps like n8n from starting.

I have done troubleshooting and have confirmed the following:

• docker ps | grep postgres shows the container is permanently in a Restarting state.
• A full server reboot, even with all other apps stopped, did not resolve the issue.
• The docker logs postgresql command does not show any fatal crash errors. Instead, it shows a repeating loop where the database successfully performs automatic recovery, reports database system is ready to accept connections, and then is terminated a few seconds later.
• This suggests the PostgreSQL service is running correctly, but is failing Cloudron's internal health check and being killed.
• Because the database is never online for more than a few seconds, apps that depend on it (like n8n) cannot start and fail with a DNS lookup error (getaddrinfo EAI_AGAIN postgresql).

The core server health is now good (htop is low, disk space is fine), but this PostgreSQL loop is preventing me from using my apps.

Could you please advise on how to diagnose the failing health check or how to force the PostgreSQL service into a stable state?