Re: Occasional panic on 5.1.4_PATCH

[Peter Eisch <peter%boku.net@localhost>]

On 8/28/14, 1:28 AM, "Masanobu SAITOH" <msaitoh%execsw.org@localhost> wrote:

>On 2014/08/28 15:27, Masanobu SAITOH wrote:
>> On 2014/08/26 1:40, Peter Eisch wrote:
>>> fatal page fault in supervisor mode
>>> trap type 6 code 2 eip c0167da7 cs 8 eflags 10246 cr2 0 ilevel 4
>>
>> eip is the program counter. The trap was occured when executing
>> an instruction at c0167da7. Do "objdump /netbsd" and check
>
>objdump -D /netbsd
>

Yes, that's what I'm looking for — thank you!  I should probably disable
the auto-reboot and capture a backtrace.  The problem is good in that that
it is very repeatable.  It is dropping inside fr_queueappend() which makes
sense as the reboot closely follows an ipfilter or ipnat reload.

I'll keep digging.

>
>> around c0167da7. You will know which function was executed.
>>
>>> panic: trap
>>> Begin traceback...
>>> fatal page fault in supervisor mode
>>> trap type 6 code 0 eip c05b0841 cs 8 eflags 10246 cr2 4 ilevel 4
>>> panic: trap
>>> Faulted in mid-traceback; aborting...rebooting...
>>>
>>> Hi,
>>>
>>>
>>> I managed to get the above off serial console as it restarted.  Could
>>> someone please point me to how I might be able to figure out what
>>>event(s)
>>> are happening to cause this panic?  Given the age of the hardware, this
>>> could be a factor but it only happens a day or so after modifying
>>> /etc/ipf.conf or /etc/ipnat.conf.  Otherwise it runs for without issue.
>>>
>>> Hardware: Dell 1850, installed in 2009, running mostly GENERIC i386
>>>kernel.
>>>
>>> I have other systems running this same hardware w/5.1.2 kernel just
>>>fine.
>>>
>>> (Yes, if I could update it to 6.x I would, but it's more likely to be
>>> replace with an ASA-5525X.)
>>>
>>> Thanks,
>>>
>>> peter
>>>
>>>
>>
>>
>
>
>-- 
>-----------------------------------------------
>                 SAITOH Masanobu (msaitoh%execsw.org@localhost
>                                  msaitoh%netbsd.org@localhost)