Am 29.08.20 um 01:56 schrieb Jonathan Billings:
On Aug 28, 2020, at 17:53, Leon Fauster via CentOS centos@centos.org wrote:
Is cron running in EL8 with stripped CAPs of? Does some one have an idea to address this?
In general, we no longer use tmpwatch at all. In CentOS 7 and 8, use systemd-tmpfiles. Here is a blog post that describes it pretty well:
https://developers.redhat.com/blog/2016/09/20/managing-temporary-files-with-...
Thanks, it seems that this migration will take more time :-).
Okay, systemd-tmpfiles is a reasonable solution for one scenario here but some questions still persists for a second one:
As you see - the next clean up will be in 23h
# systemctl status systemd-tmpfiles-clean.timer
● systemd-tmpfiles-clean.timer - Daily Cleanup of Temporary Directories Loaded: loaded (/usr/lib/systemd/system/systemd-tmpfiles-clean.timer; static; vendor preset: disabled) Active: active (waiting) since Sat 2020-08-29 21:53:11 CEST; 53min ago Trigger: Sun 2020-08-30 22:07:52 CEST; 23h left Docs: man:tmpfiles.d(5) man:systemd-tmpfiles(8)
# systemctl list-timers
NEXT LEFT LAST PASSED UNIT ACTIVATES Sun 2020-08-30 22:07:52 CEST 23h left Sat 2020-08-29 22:07:52 CEST 39min ago systemd-tmpfiles-clean.timer systemd-tmpfiles-clean.service
What triggers cleanups that must be done every hour for example?
# cat /usr/lib/tmpfiles.d/app-tmp.conf e /srv/app/*/tmp - - - 1h
# man tmpfiles.d
is mentioning (m)minutes, (s)econds and even us (microseconds).
Do I need to override the systemd-tmpfiles-clean.timer unit?
# systemctl cat systemd-tmpfiles-clean.timer ... [Timer] OnBootSec=15min OnUnitActiveSec=1d
It seems that this is more prepared for daily clean ups?
-- Leon