Applicable to:
- Plesk for Linux
Symptoms
/var/lib/plesk/task-manager/db/db.sqlite3-wal
was not rotated and it grew to 20gb:
# du -hsx /var/lib/plesk/task-manager/db/
300K /var/lib/plesk/task-manager/db/db.sqlite3
32K /var/lib/plesk/task-manager/db/db.sqlite3-shm
20G /var/lib/plesk/task-manager/db/db.sqlite3-wal
Cause
The issue was caused by loops due to dbus outages(which are most likely environment-related since it’s a containerized server).:
# zcat /var/log/plesk/task-manager.log-* |grep "event loop" |tail -n5
WARNING 2024/06/18 13:04:35 task=67727 process=199615 Process exit is being handled 32.802810521s after it has actually exited. Unless this happens at Task Manager startup, this indicates either a lost dbus event, or an overloaded event loop.
WARNING 2024/07/22 13:05:04 task=73354 process=205242 Process exit is being handled 48.073929002s after it has actually exited. Unless this happens at Task Manager startup, this indicates either a lost dbus event, or an overloaded event loop.
WARNING 2024/07/25 01:58:03 task=73922 process=205778 Process exit is being handled 47.61608075s after it has actually exited. Unless this happens at Task Manager startup, this indicates either a lost dbus event, or an overloaded event loop.
WARNING 2024/09/09 13:04:32 task=82693 process=214581 Process exit is being handled 20.978413216s after it has actually exited. Unless this happens at Task Manager startup, this indicates either a lost dbus event, or an overloaded event loop.
WARNING 2024/09/09 13:04:32 task=82717 process=214604 Process exit is being handled 40.628116044s after it has actually exited. Unless this happens at Task Manager startup, this indicates either a lost dbus event, or an overloaded event loop.
Resolution
-
Connect to the server via SSH
-
Stop the task manager:
# systemctl stop plesk-task-manager
-
Remove the database:
# rm -rf /var/lib/plesk/task-manager/db/db.sqlite3-*
-
Start task manager:
# systemctl start plesk-task-manager
Comments
0 comments
Please sign in to leave a comment.