SMP: Background fsck deadlocks CURRENT and RELENG_5 solid

[Doug White]

On Thu, 19 Aug 2004, Martin Blapp wrote:

> > > I'm currently not able to reboot a paniced or unsincedy disk without
> > > doing a normal fsck. Just when the background fsck starts, the box
> > > locks solid. The load at this time is not significant.
> >
> > You might have a corrupted fs from earlier panics.  Do a full fsck in single
> > user mode and keep doing it until you have no errors (i.e., if it fixes
> > anything run fsck again in a loop until you get no fixups at all).
>
> Nope. This happens on two different SMP boxes. If background fsck comes
> into play (whatever happened before is not important), the boxes lock.

Can't say I 've had problems with this while debugging the IPI hangs.  And
those are rock solid.

Perhaps you're short on disk space on / and the deltas from the snapshot
take you over the edge?  Or have some sort of large I/O operation run just
after boot?

-- 
Doug White                    |  FreeBSD: The Power to Serve
dwhite at gumbysoft.com          |  www.FreeBSD.org