Major 2.0.27 hassles #3 -- process accounting knackered

Chris Evans (chris@ferret.lmh.ox.ac.uk)
Tue, 11 Feb 1997 14:32:42 +0000 (GMT)


Hi,

After the OOPS I just posted, the swapper task went nuts and obviously
decided it was trying to exit (it claimed it was killing interrupt handler
but the machine is still up).

Whilst trying to exit, it decided to log thousands of exiting swapper
processes per second to /var/log/pacct. It filled up /var/log (what joy),
although it should be noted that process accounting handled the out of
space case far more gracefully than it used to.

Here's an example of the crap it was trying to log:

swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10
swapper S root ?? 1168531.93 secs Mon Jan 27 15:10

<repeat a lot>

2.0.30 time...?

Please cc: any replies to me.

Good luck!
Chris.