Put daemon.log under faster rotation

[justinsb]

We need to make sure that all our log files rotate sufficiently fast, e.g. daemon.log is currently defaulting to weekly rotation.

(Likely a dup of some others, but this is a work-tracking issue rather than a general issue)

[krogon-dp]

#2816 #3047

[cmaloney]

Would it be feasible to make it so that someone could disable the rsyslogd/classic log handling, and just rely on journald/journalctl (Possibly opt in). journald is much better at not allowing disk space to fill up (Works when messages come in for rotation / fixed size log, rather than trying to "guess" when files will get so big they need to be rotated).

Definitely journald is a sysadmin consideration, and has other behaviors which people may/may not like, but would give people who can/want to move to it (esp. since things already log to it) an option to fairly simply solve this.

[chrislovecnm]

Move to 1.9?

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[blakebarnett]

doesn't seem like a release blocker

[chrislovecnm]

I agree ...

/lifecycle frozen
/remove-lifecycle stale

[mikesplain]

I was taking a look at this, @justinsb what's the best way to do this? I don't see a existing task to modify a config file. Logrotate doesn't support duplicates so, should we modify the overrides in /var/log/rsyslog?