Hi,
On Thu, Feb 07, 2002 at 04:02:37AM -0500, Andres Salomon wrote:
> Linux incandescent 2.4.18-pre7-ac3 #1 Tue Feb 5 01:00:50 EST 2002 i686 unknown
>
> I came home today to a bunch of processes apparently hung in jbd's
> do_get_write_access(); apparently, something deadlocked, where no
> processes could write to one of my partition's journal. That's my
> theory, anyways. Several processes had stacks similar to:
> Trace; c011271c <sleep_on+3c/50>
> Trace; d08ecac6 <[jbd]do_get_write_access+2e6/4f0>
They are blocked waiting on the journal thread, and:
> And, kjournald had the following:
>
> Call Trace: [<c0131f0a>] [<d08ee6e6>] [<c0112380>] [<d08f08eb>]
> [<d08f07c0>]
> [<c010f4e8>]
> Warning (Oops_read): Code line not seen, dumping what data is available
>
> Trace; c0131f0a <__wait_on_buffer+6a/90>
... the journal thread is waiting for some IO to complete.
> Not sure what caused the actual deadlock; the partition uses lvm and
> ext3, and the underlying controller uses hedrick's pdc driver.
Were you doing anything particular on the LVM at the time? Is there
anything in the log which might indicate a failed IO? The trace above
just indicates that IO was initialised and never completed --- it
doesn't really give us any clue WHY that IO was lost.
Cheers,
Stephen
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Feb 07 2002 - 21:01:01 EST