INFO: task snmpd:398 blocked for more than 120 seconds.

[Joakim Tjernlund]

Norbert van Bolhuis <nvbolhuis at aimvalley.nl> wrote on 2010/11/09 10:47:46:
>
> Joakim Tjernlund wrote:
> >>> I can't make out what is causing this hang every now an then:
> >>>
> >>> INFO: task snmpd:398 blocked for more than 120 seconds.
> >> My problem with that 'error' message is that there is no way
> >> for a driver to disable it on a per-process basis.
> >> We have some processes whose 'normal state' is to sleep
> >> uninterruptibly in the kernel. Shutdown is handled by
> >> an explicit request (not by sending a signal).
> >> The processes could be kernel worker threads (except that
> >> is is ~impossible to handle them exiting from a loadble
> >> kernel module) so are actually children of a daemon sat
> >> inside an ioctl() request that never terminates!
> >>
> >> However, on the face of it, your case does look as though
> >> the mutex is fubar'ed.
> >>
> >> Might be worth (somehow) dumping the mutex state.
> >
> > ehh, it is locked, isn't it? How to find who locked it
> > and forgot to release it?
> >
>
> You could add show_state to check_hung_task
>
> It worked for me while solving some (application) task hangups.

Will have a look at that too, thanks.

Meanwhile i have tried to turn on various LOCK debug and
that results in an unbootable kernel.
Somewhere the dev. tree. gets mangled and it happens when
DEBUG_LOCK_ALLOC is set.
Don't know why and I must move on.
Someone else might want to try if DEBUG_LOCK_ALLOC works
on embedded ppc boards, especially 83xx.

    Jocke