Resolved /tmp not automatically cleaned (after backup)

[Jelle_Timmer]

One of my Plesk-servers (the only one with CentOS 7.3 and Onyx as initial config, currently installed version 17.0.17#11) has weekly issues regarding an increasing /tmp folder.

At least twice a week i get messages from the Server Health Manager warning me about
"Disk > Partition '/tmp' utilization". The first message tells me the status changed from green to yellow, and the next morning is changed to red. These messages are always delivered right after it's scheduled backup-time, so that's why i think it's backup-related.

It is the ONLY Plesk server out of 6 that behaves this way.
No issues:
1x CentOS 6.8 with 17.0.17 (updated from 12.5.30)
2x CentOS 6.8 with 12.5.30
1x CentOS 7.3 with 12.5.30
1x CentOS 7.3 with 12.5.30

All of these servers have the same backup-schedule, backup to the same FTP-repository, albeit to different paths, but none of these have any issues.

I have cleaned the /tmp folder manually for a few weeks, but it would be nice if it was cleaned the way it should!

Someone with similar problems or a solution?

Thank you!

 

[danami]

I would suggest that you install tmpwatch:

yum install tmpwatch

Then you can create a cron to clean out your tmp directory . tmpwatch allows you to exclude certain files or only delete files older than a certain date.

Example:

/usr/sbin/tmpwatch --verbose --mtime 72 -x /tmp/.font-unix -x /tmp/.ICE-unix /tmp

Type "man tmpwatch" for more information

 

[Jelle_Timmer]

Thanks for your suggestion. I'll look into it!

 

[yiannis yiannis]

Hello,
I have exactly the same issue.

So, is this considered as a known bug/issue of Onyx?
The only way (other than manually) to clean that folder is to use the tmpwatch software or a cron job?