Re: periodic crashes
kmself@ix.netcom.com writes:
> On Mon, Mar 13, 2000 at 09:05:13PM -0000, Pollywog wrote:
> > I just ran the "last" command, and I noticed that my machine has crashed
> > several times since March 1 and in each instance, the time was the same, 17:01
> > UTC.
> >
> > What is the best way to track down the cause of the crashes?
>
> As noted elsewhere, 17:01 is the time the system rebooted, not that it
> died.
>
> Look through /var/log for any files updated since shortly before 17:01,
> and look in them for any activity going on prior to 17:01. If you're
> getting regular crashes at the same time, you've very likely either got
> a very reliable cleaning service which is disconnecting power at the
> same time every day, or a cron job that sets your system off.
>
> The latter may not be a broken script -- it's possible that there are
> hardware or other system problems which are being aggrevated by certain
> processing.
For example, I've seen things like this when I had a bad area on a
hard drive. The system was scheduled to run a backup at, for example,
1am and was crashing at about 3am every morning. It would only crash
when it got to files associated with the bad area of the disk, and
these files were infrequently accessed by anything else and so it
didn't appear to be hardware related, but in the end it was.
I've also seen similiar things with bad memory. The particular chip
was "high" in the address range and so the system ran fine until a LOT
of memory was needed and then it accessed the bad chip and would
crash.
Gary
Reply to: