[AG-TECH] RAT crash

[Piers O'Hanlon]

Hi,

2009/7/6 Philippe d'Anfray <Philippe.d-Anfray at cea.fr>:
> Bonjour,
>
> I launched rat manually with the same parameter "as in the venue".  It
> works normally then
> the messages  are issued  all together just before the rat window
> disappears.
> The first time a segfault was reported in the logs  but not the second
> time.
>
The 'Bad address' error (according to 'man errno') corresponds to the
"EFAULT" error which for recvfrom() indicates: "EFAULT The receive
buffer pointer(s) point outside the process'€™s address space". This
points to a memory bug in RAT/common - I've not see this particular
error before - we'll need to try to work out what is causing it.... Do
you see any thing happen before RAT crashes - e.g. do you click on a
particular button, or does a particular participant join the session
or does this always happen after a set amount of time. Can you see if
it correlates to any other event or is totally random?

Thanks,

Piers.

> Cordialement
>
> Philippe
>
> aristote at aristotest4:~$ rat -4.4.01 -C MARCEL -S 8168306a1e5b045610539184ffd9c0d5 -f L16-16K-Mono accessgrid-bridge.renater.fr/50064
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> aristote at aristotest4:~$ rat -4.4.01 -C MARCEL -S 8168306a1e5b045610539184ffd9c0d5 -f L16-16K-Mono accessgrid-bridge.renater.fr/50064
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> recvfrom: Bad address
> Not enough memory
> aristote at aristotest4:
>
>
> Message in the log for the first crash:
>
> Jul 6 13:29:43 aristotest4 kernel: [ 1393.943534] rat-4.4.01-medi[4973]: segfault at 0 ip 0809735a sp bfa365e0 error 6 in rat-4.4.01-media[8048000+6f000]
>
>
>
>
> Christoph Willing a écrit :
>>
>> On 02/07/2009, at 1:22 AM, D'ANFRAY Philippe wrote:
>>
>>> Bonjour,
>>>
>>> We're running AG 3.2 beta 1  with Ubuntu 9.04. The machine has a
>>> creative sound card and we're running the linux driver
>>> (/XFiDrv_Linux_Public_US_1.00). After a few minutes rat always
>>> crashes i.e. the rat window just disappears and there's no message if
>>> we run in debug mode.
>>>
>>> I eventually found   messages like the following(s)  in the log
>>> files  (but not everytime):
>>>
>>> Jun 29 17:38:02 aristotest4 kernel: [16153.505141]
>>> rat-4.4.01-medi[14426]: segfault at 0 ip 080910c7 sp bfe2bcc0 error 6
>>> in rat-4.4.01-media[8048000+6f000]
>>>
>>> Jun 29 13:21:53 aristotest4 kernel: [  784.482769]
>>> rat-4.4.01-medi[4790]: segfault at 0 ip 0809355c sp bfdb7f40 error 4
>>> in rat-4.4.01-media[8048000+6f000]
>>
>>
>> Philippe,
>>
>> As a test, could you run rat from a terminal command line, with
>> something like:
>>     rat 233.45.67.89/45678
>>
>> and let it run till it crashes and then report any output that appears
>> in the terminal please?
>>
>>
>> chris
>>
>>
>> Christoph Willing                       +61 7 3365 8316
>> QCIF Access Grid Manager
>> University of Queensland
>>
>>
>
>