[BUG]

[ahaghgoo]

Describe the bug:

• Postgres crashes when importing large OPML file. Now postgres keeps crashing on startup. Here is what I see on the backend logs:

Postgres:

2023-01-01 11:34:42.286 UTC [1] LOG:  starting PostgreSQL 13.9 on x86_64-pc-linux-musl, compiled by gcc (Alpine 12.2.1_git20220924-r4) 12.2.1 20220924, 64-bit
2023-01-01 11:34:42.289 UTC [1] LOG:  listening on IPv4 address "0.0.0.0", port 5432
2023-01-01 11:34:42.289 UTC [1] LOG:  listening on IPv6 address "::", port 5432
2023-01-01 11:34:42.509 UTC [1] LOG:  listening on Unix socket "/var/run/postgresql/.s.PGSQL.5432"
2023-01-01 11:34:42.524 UTC [22] LOG:  database system was interrupted while in recovery at 2023-01-01 11:34:33 UTC
2023-01-01 11:34:42.524 UTC [22] HINT:  This probably means that some data is corrupted and you will have to use the last backup for recovery.
2023-01-01 11:34:42.541 UTC [22] LOG:  database system was not properly shut down; automatic recovery in progress
2023-01-01 11:34:42.544 UTC [22] LOG:  redo starts at 0/2B440D0
2023-01-01 11:34:42.780 UTC [22] LOG:  redo done at 0/4FFFC08
2023-01-01 11:34:42.781 UTC [22] PANIC:  could not write to file "pg_wal/xlogtemp.22": No space left on device
2023-01-01 11:34:42.782 UTC [1] LOG:  startup process (PID 22) was terminated by signal 6: Aborted
2023-01-01 11:34:42.782 UTC [1] LOG:  aborting startup due to startup process failure
2023-01-01 11:34:42.785 UTC [1] LOG:  database system is shut down`

TTRSS:

Exception while creating PDO object:SQLSTATE[08006] [7] connection to server at "database.postgres" (172.29.8.2), port 5432 failed: Connection refused`
    Is the server running on that host and accepting TCP/IP connections?

Have you googled?

• Yes, but I couldn't find anything on what to do in order to clean up postgres database within docker and to prevent this from happening again.

Deployment method

• docker-compose
• ttrss image: wangqiru/ttrss:latest
• postgres image: postgres:13-alpine

Deployment environment

• OS: QNAP NAS QuTS hero FW h4.5.4.2217
• Browser: Safari 16.1 (18614.2.9.1.12)

To Reproduce Steps to reproduce the behavior:

• I've linked to my OPML file here: https://drive.google.com/open?id=1kprkvWyC6EekArOPPhX-9RZxF6PcjTc-&authuser=ahaghgoo%40g.ucla.edu&usp=drive_fs

• All I did is import the file and TTRSS would populate articles but midway through fetching articles it would crash and it would no longer be accessible.

Expected behavior

• I would be able to import my OPML file with no issues.

Screenshots If applicable, add screenshots to help explain your problem.

Error logs

[21:18:53/31] Received SIGCHLD, 0 active tasks left.
[21:19:52/31] 0 active tasks, next spawn at 60 sec.
[21:20:52/31] 0 active tasks, next spawn at 0 sec.
[21:20:53/31] Spawned child process with PID 1817 for task 0.
[21:20:53/31] Spawned child process with PID 1819 for task 1.
<pre>Exception while creating PDO object:SQLSTATE[08006] [7] could not translate host name "database.postgres" to address: Name does not resolve</pre><pre>Exception while creating PDO object:SQLSTATE[08006] [7] could not translate host name "database.postgres" to address: Name does not resolve</pre>[21:20:54/31] Child process with PID 1817 reaped.
[21:20:54/31] Child process with PID 1819 seems active but lockfile is unlocked.
[21:20:54/31] Received SIGCHLD, 0 active tasks left.
[21:20:54/31] Received SIGCHLD, 0 active tasks left.
[21:21:53/31] 0 active tasks, next spawn at 60 sec.
[21:22:53/31] 0 active tasks, next spawn at 0 sec.
[21:22:54/31] Spawned child process with PID 1827 for task 0.
[21:22:54/31] Spawned child process with PID 1829 for task 1.
<pre>Exception while creating PDO object:SQLSTATE[08006] [7] could not translate host name "database.postgres" to address: Name does not resolve</pre><pre>Exception while creating PDO object:SQLSTATE[08006] [7] could not translate host name "database.postgres" to address: Name does not resolve</pre>[21:22:55/31] Child process with PID 1827 seems active but lockfile is unlocked.
[21:22:55/31] Child process with PID 1829 reaped.
[21:22:55/31] Received SIGCHLD, 0 active tasks left.
[21:22:55/31] Received SIGCHLD, 0 active tasks left.
[21:23:54/31] 0 active tasks, next spawn at 60 sec.
[21:24:54/31] 0 active tasks, next spawn at 0 sec.
[21:24:55/31] Spawned child process with PID 1838 for task 0.
[21:24:55/31] Spawned child process with PID 1840 for task 1.
<pre>Exception while creating PDO object:SQLSTATE[08006] [7] could not translate host name "database.postgres" to address: Name does not resolve</pre><pre>Exception while creating PDO object:SQLSTATE[08006] [7] could not translate host name "database.postgres" to address: Name does not resolve</pre>[21:24:56/31] Child process with PID 1838 seems active but lockfile is unlocked.

Additional context Add any other context about the problem here.

[TonyRL]

2023-01-01 11:34:42.781 UTC [22] PANIC: could not write to file "pg_wal/xlogtemp.22": No space left on device

Either your storage is full or you don't enough permission to write on the mounted directory