[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [cobalt-users] Qube2 Crashes without obvious cause



On Fri, 08 Dec 2000 17:01:21 +1100, Malcolm McLeary <mmcleary@xxxxxxx> wrote:

:>It would be nice if the box had a hardware watchdog to do a hardware reset
:>if/when it crashed.

I guess one could be programmed into the monitor?

:>> Anything in the
:>> /var/cobalt/ monitor logs (since they get updated every 15 minutes - unless
:>> you have changed crontab like I have to do a monitor every 6 hours)? You can
:>> check memory and disk usage there.
:>
:>Hadn't looked there it might be interesting to correlate the "stop" time in
:>/var/log/messages with these logs.  It might also be interesting to disable
:>monitoring altogether to see if its a "monitor" which is causing the
:>problem, but wouldn't that cause it to crash every 15 minutes?

I doubt it is the monitor, but those logs may give you a hint as to what as
failing at the time of the lockup. Uhm, anything being logrotated around that
time?

:>
:>> What changes have been made to the system
:>> or the environment that might suggest a cause?
:>
:>Up until now none ... its as delivered.  The user had been complaining about
:>"File System Full" email so messages so I moved /usr/doc to /home/doc to
:>free up a little space on /.

You did put in a ln -s for /usr/doc right?

Those messages should not be sent unless the file space hits 80% or 90%. 

:>[root /root]# df -m
:>Filesystem         MB-blocks    Used Available Capacity Mounted on
:>/dev/hda1                290     211       79     73%   /
:>/dev/hda3                193      25      168     13%   /var
:>/dev/hda4              18172     577    17595      3%   /home
:>[root /root]#
:>
:>It was running at 78%.

Qube2s typically arrive with 78% on the root partition. A couple of other
thoughts. The /var/Cobalt logs did not all rotate correctly (I posted a
message about this several months ago). I believe one of the patches fixed
that. Look at both /var/cobalt and /var/log and make sure the logs have a
matching zipped log (if it has been around log enough to have needed zipping).