Bug#644550: linux-image-2.6.32-5-amd64: indefinite soft lockup on rm
> > > Package: linux-2.6
> > > Version: 2.6.32-38
> > > Severity: important
> > > Tags: upstream
> > >
> > > the lockup happened on unlinking a large file (several GB) on XFS on fibre
> > > channel storage (qla2xxx module). We experience very simiral trouble on
> > > this machine though, using other kernels (from stable, stable-backports and
> > > sid as well) and other fs/storage combinations (ext4 on local Dell PERC
> > > controller).
> > How long did the soft lockup state persist before you rebooted?
> 14 minutes
...after switching back to the stable kernel (where this bug report started),
the same soft lockup happened again last night after 36h of uptime. It
lasted 9h (until reboot), so it's really indefinite I guess.
The "rm" was again stuck on the same file, being produced and steadily grew
up a few hours from zero size collaboratively by these "sky-db" processes,
see the atop sample above.
To summarize: the stable kernel locks up in matter of hours/days, and the
unstable kernel crashes in days/weeks, without link to "rm" command. The
other similar machine crashed twice too (running CentOS) and we are reporting
this to the HW vendor to not waste time of kernel developers/maintainers