cause of reboot

[Julian Elischer]

On 9/30/13 11:08 PM, Jason Hellenthal wrote:
> Also  . . . grep -ri  . . .
>
>
>
> On Sep 30, 2013, at 11:06, Jason Hellenthal <jhellenthal at dataix.net> wrote:
>
> And /etc/ ?
> And /var/cron ?
>
>
>
> On Sep 30, 2013, at 11:00, Emre Çamalan <mailist at yandex.com> wrote:
>
> Yes I checked also it , such as ;
> "grep -i 'reboot' /root/scripts/* -R"
> "grep -i 'halt' /root/scripts/* -R"
> "grep -i 'shutdown' /root/scripts/* -R"
>
>
> 30.09.2013, 17:33, "Jason Hellenthal" <jhellenthal at dataix.net>:
>> Did you check for the command reboot in all of the periodic scripts and cron jobs as well as the command shutdown ?
>>
>> On Sep 30, 2013, at 10:24, Emre Çamalan <mailist at yandex.com> wrote:
>>
>> Hi,
>> my server reboots every night and same o'clock last 10 days. But this machine's uptime was 96days.
>>
>> Suddenly reboot this machine and now this continue every night again and again.
>>
>> I didn't find any reason and I didn't change anything else.
>>
>> I looked "last" command,
>> reboot ~ ~   AM 03.15  ~
>>
>> I checked "dmesg -a" , it has only generic things no other things.
>>
>> I checked "/var/log/messages"
>>
>> I wrote script and checked swapinfo , cpu and memory but all of them was normal before reboot.
>>
>> I checked crontab and scripts.
>>
>> I checked "crashinfo"  but no dump from kernel.
>>
>> I didn't find any useful info from this commands.
>>
>> I checked "praudit /var/audir/20130930..CrashRecovery"
>> and I didn't understand very well from this file but I think this means my machine crash and reboot isn't it??
>>
>> And How can I understand what is the reason of rebooting my FreeBSD8.3 server.
>>
>> Please help I need to find cause of reboot..
>> _______________________________________________
>> freebsd-hackers at freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
>> To unsubscribe, send any mail to "freebsd-hackers-unsubscribe at freebsd.org"

Firstly, does this machine have the kernel debugger installed?
if not then a kernel page fault will look like a reboot.

if you know when this is going to occur, then I suggest that you
set up some logging of every new process run, to a second machine, or 
possibly
you might just ssh int the machine and leave 'top' running
when it reboots you shoud get a snapshot of what's going on.

you could just rename 'reboot' and see if it still happens.
If not, then replace reboot (and friends) with a script that reports 
who called it.