Re: PROBLEM: System locks up after "spurious 8259A interrupt: IRQ7"

[Till Doerges]

On Mon, Jan 14, 2002 at 01:00:19AM +0100, Till Doerges wrote:

> sporadically my machine hangs (no response to keyboard, can't connect
> via network, no interaction via remote-control) and the last thing I
> see (if I see anything at all) is something like
> 
> --- snip ---
> [...]
> Jan 13 21:30:00 atlan CROND[2876]: (till) CMD (fetchmail&> /dev/null)
> Jan 13 21:30:09 atlan kernel: spurious 8259A interrupt: IRQ7.
> Jan 13 21:36:12 atlan syslogd 1.3-3: restart (remote reception).
> Jan 13 21:36:12 atlan kernel: klogd 1.3-3, log source = /proc/kmsg started.
> [...]
> --- snap ---

Sorry, for not having done this in the first mail already: I've found
another occurence of the problem in a logfile, which I had forgotten
about:

--- snip ---
[...]
Nov 20 14:30:00 atlan CROND[4109]: (root) CMD (   /sbin/rmmod -as)
Nov 20 14:30:00 atlan CROND[4110]: (till) CMD (fetchmail&> /dev/null)
Nov 20 14:34:45 atlan kernel: spurious 8259A interrupt: IRQ7.
Nov 20 14:34:55 atlan kernel: ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
Nov 20 14:34:55 atlan kernel: hdc: lost interrupt
Nov 20 14:35:05 atlan kernel: ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
Nov 20 14:35:05 atlan kernel: hdc: lost interrupt
Nov 20 14:35:15 atlan kernel: ide_dmaproc: chipset supported ide_dma_lostirq func only: 13
Nov 20 14:35:15 atlan kernel: hdc: lost interrupt
^@^@^@^@^@^@Nov 20 19:17:25 atlan syslogd 1.3-3: restart (remote reception).
Nov 20 19:17:25 atlan kernel: klogd 1.3-3, log source = /proc/kmsg started.
Nov 20 19:17:25 atlan kernel: Inspecting /boot/System.map-2.4.7-int-ac3
[...]
Nov 20 19:17:26 atlan kernel: Linux version 2.4.7-int-ac3 ([e-mail]) (gcc version 2.96 20000731 (Red Hat Linux 7.1 2.96-85)) #1 Wed Aug 1 01:54:15 CEST 2001
[...]
--- snap ---

This was 2.4.7-ac3 compiled w/ an earlier version of gcc
(2.96-85). And in deed, my problem was sometimes accompanied by a
'kernel: hdc: lost interrupt'. But this seems to have gone, because I
removed one disk from the system and moved hdc -> hda.

That there's such a long period between these two occurences, doesn't
mean, that my machine did not hang in the meantime, but I did not find
anything in the logs. Perhaps the messages didn't make it through?

Bye -- Till