[PLUG] Help diagnosing system restarts

Aaron Ten Clay aaron at madebyai.com
Thu Oct 11 18:10:45 UTC 2007


alan wrote:
> On Thu, 11 Oct 2007, Aaron Ten Clay wrote:
>
>   
>> All,
>>
>> I've got a box setup for my mother who is 300 miles away from me. I've
>> got logsentry emailing me unusual system activity once an hour and I can
>> ssh into the box from where I am.
>>
>> Here's the problem:
>>
>> I started noticing a couple days ago kernel boot messages in some of the
>> hourly reports. I thought it was a bit odd but not too serious. This
>> morning I see that every hour for the past 24 hours her system has
>> rebooted itself. There are no obvious messages pertaining to the reboot
>> in the log; presumably something catastrophic happens and the disk isn't
>> sync'd before the reboot.
>>
>> The only pattern I've noticed thus far, by looking at the timing of the
>> kernel boot messages, is that it's slowly getting later in the hour. For
>> example:
>>
>> 00:32:28
>> 01:32:48
>> 02:33:05
>> 03:33:26
>> 04:33:44
>> 05:34:03
>> 06:34:24
>> 07:34:41
>> 08:35:05
>> 09:35:20
>> 10:35:41
>>
>> and so on. Any ideas what might be causing this? I spoke to her and she has noticed some unusual fan activity, but I'm not sure how that could explain the reboots.
>>     
>
> Check the hourly cron jobs.  Could be something silly like "makewhatis" or 
> it could be something else worth worrying about.
>
> Should be in /etc/cron.hourly.
>
>   
Thanks for the suggestion. I've disabled cron, we'll see if it still 
restarts. The only things cron should be running, aside from emailing 
hourly reports, is a daily rdiff-backup run that happens at 2PM.



More information about the PLUG mailing list