[tmpfs] /run full

General support questions
Post Reply
DBE
Posts: 6
Joined: 2018/01/07 21:45:17

[tmpfs] /run full

Post by DBE » 2018/01/07 22:00:42

Hi,

I have a problem with /run, its size doesn't decrease so when it's full, I have to reboot.
[root@SRV ~]# df -h
tmpfs 5,9G 2,4G 3,6G 41% /run
df_run.png
df_run.png (28.73 KiB) Viewed 2306 times

The size of /run/log/journal decrease but doesn't decrease /run
:shock:

[root@SRV ~]# du -k /run | sort -nr
606984 /run
606208 /run/log/journal/d5b1a7595c414aef867926a90584852b
606208 /run/log/journal
606208 /run/log
du_run.png
du_run.png (48.67 KiB) Viewed 2306 times
How to decrease the size of /run ?
Thanks for your help.

User avatar
TrevorH
Site Admin
Posts: 33202
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: [tmpfs] /run full

Post by TrevorH » 2018/01/08 12:08:01

It's more likely that you've run across a bug that occurs when systemd packages are updated and you do not reboot afterwards. It loses its dbus connection and stops cleaning up old data in /run. For now the safe way to proceed is to reboot every time the systemd packages are updated, in the same way that you would if glibc were.
The future appears to be RHEL or Debian. I think I'm going Debian.
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 and 6 are deadest, do not use them.
Use the FAQ Luke

DBE
Posts: 6
Joined: 2018/01/07 21:45:17

Re: [tmpfs] /run full

Post by DBE » 2018/01/08 21:43:12

Thanks for your reply.

I don't have old log in /run/log/
(du_run.png)

I reboot the server each time /run is full

DBE
Posts: 6
Joined: 2018/01/07 21:45:17

Re: [tmpfs] /run full

Post by DBE » 2018/02/11 22:59:03

How can I resolve this problem?

Why the size of /run increase (90%) with df while the size seems to be stable with du?
I have to reboot the server to decrease the size of /run

Thanks.

DBE
Posts: 6
Joined: 2018/01/07 21:45:17

Re: [tmpfs] /run full

Post by DBE » 2018/02/13 00:25:20

99%, I have to reboot again to free tmpfs /run
Screenshot_2.png
Screenshot_2.png (43.26 KiB) Viewed 2145 times
Is there an another way instead rebooting server to free space in /run ?

du of /run is stable.
Screenshot_1.png
Screenshot_1.png (65.51 KiB) Viewed 2145 times
du -h /run/log/journal/
408M /run/log/journal/d5b1a7595c414aef867926a90584852b
408M /run/log/journal/

journalctl --disk-usage
Archived and active journals take up 408.0M on disk.

[root@PWCFEM31 d5b1a7595c414aef867926a90584852b]# df -h
tmpfs 5,9G 5,8G 112M 99% /run

What I expect is 408M instead of 5.8G :o

DBE
Posts: 6
Joined: 2018/01/07 21:45:17

Re: [tmpfs] /run full

Post by DBE » 2018/02/13 00:29:14

what I try:
In /etc/systemd/journald.conf, I set these parameters:
RuntimeMaxUse=400M
RuntimeKeepFree=2G

systemctl kill --signal=SIGUSR2 systemd-journald.service
journalctl --vacuum-size=400M
journalctl --vacuum-time=1d

DBE
Posts: 6
Joined: 2018/01/07 21:45:17

Re: [tmpfs] /run full

Post by DBE » 2018/03/25 22:05:09

It doesn't resolve the problem
Please help... :roll:
1803260903.png
1803260903.png (44.46 KiB) Viewed 2042 times

hunter86_bg
Posts: 2019
Joined: 2015/02/17 15:14:33
Location: Bulgaria
Contact:

Re: [tmpfs] /run full

Post by hunter86_bg » 2018/03/26 04:05:50

Run a full update and reboot.

Post Reply