systemd questions

Lennart Poettering mzerqung at 0pointer.de
Sun May 15 19:19:40 UTC 2011


On Sun, 15.05.11 19:28, Miloslav Trmač (mitr at volny.cz) wrote:

> 
> On Sun, May 15, 2011 at 6:57 PM, Lennart Poettering
> <mzerqung at 0pointer.de> wrote:
> > Well, adding 30s-sleep()s here and there and everywhere is a hack, not a
> > clean fix. But if you have a lot of sync to disk, and/or a slow disk,
> > then 30s might not be enough and you are fucked.
> I can't see where would that "a lot of sync to disk" come from: the
> filesystems are synced, RAID has been marked clean. What else needs to
> be written back?

If the root dir is on RAID the MD device cannot be stopped on shutdown
currently. Look for the messages like "md: stopping all md devices" you
can see when you shut down.

> > "We always did it this way" is not a fix for a problem. The race is
> > real, and should be fixed properly.
> Sure, when/if somebody feels this is important enough to rewrite all
> of the UPS drivers.  In the mean time, future Fedoras should work at
> least as well as F14 does.

I think "work as badly" would be better wording here.

Lennart

-- 
Lennart Poettering - Red Hat, Inc.


More information about the devel mailing list