Hello, I have had several instances over the last month of Asterisk freezing, sometimes after 12 hours, sometimes after 8 days. The common elements are that: - all Zap channels lock[hangups don't register and no new calls in or out] - no new in/outbound calls can be made on Zap or SIP channels - people who are still connected to calls can continue to talk - in the CLI interface, you can "show channels" and view other info like "sip show peers", but you cannot do a "stop now" only way to stop asterisk process is a kill -9 - all attempted connections to the manager interface time out - the processor load and RAM usage on the machine is low. I am experiencing these freezes on two separate identical RedHat 9 machines. On each I have 4 T1s connected and a usual concurrent call volume of 45 Zap/SIP conversations at once 14 hours a day. The processor load is never that high and the RAM usage is less than half. There are never any consistent errors or warnings in the logs or the CLI. I do have several AGI perl scripts in the dialplan and the SIP phones that connect are almost all Grandstream 102's, That's all I can think of that may be causing any problems. I have mpg123/musiconhold deactivated. Has this happened to anyone else out there? Does anyone have any suggestions? or ideas as to what may be causing it? Thanks for any help, MATT---
>I have had several instances over the last month of Asterisk freezing, > Does anyone have any suggestions? or ideas as to what may be causing it?Sounds like some type deadlock Take a look here for my tought and how to debuig dead locks http://lists.digium.com/pipermail/asterisk-dev/2003-November/002384.html Then let us know when you haved posted the info on bugs.digium.com
Thanks for the debug procedural steps. That's a lot of stuff to do. I think I will attempt to do what you suggest on one of my servers and just try some streamlining and reduction of functionality in the other server and see if that fixes it. Since it can take upto 2 weeks for a server to freeze it may be a while before I post anything on bugs(and it'll probably take me a few days to turn on thread logging and recompile the asterisk code) Thanks, MATT--- -----Original Message----- From: TC [mailto:trclark@shaw.ca] Sent: Thursday, December 04, 2003 11:58 PM To: asterisk-users@lists.digium.com Subject: Re: [Asterisk-Users] Asterisk freezing HELP>I have had several instances over the last month of Asterisk freezing, > Does anyone have any suggestions? or ideas as to what may be causing it?Sounds like some type deadlock Take a look here for my tought and how to debuig dead locks http://lists.digium.com/pipermail/asterisk-dev/2003-November/002384.html Then let us know when you haved posted the info on bugs.digium.com _______________________________________________ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users
Do you type "reload" at the cli a few times a day? If so try not reloading Asterisk and I'll bet Asterisk stop blocking. If you don't normally reload the box you will need to trouble shot normally. mattf wrote:>Hello, > >I have had several instances over the last month of Asterisk freezing, >sometimes after 12 hours, sometimes after 8 days. The common elements are >that: >- all Zap channels lock[hangups don't register and no new calls in or out] >- no new in/outbound calls can be made on Zap or SIP channels >- people who are still connected to calls can continue to talk >- in the CLI interface, you can "show channels" and view other info like >"sip show peers", but you cannot do a "stop now" only way to stop asterisk >process is a kill -9 >- all attempted connections to the manager interface time out >- the processor load and RAM usage on the machine is low. > >I am experiencing these freezes on two separate identical RedHat 9 machines. >On each I have 4 T1s connected and a usual concurrent call volume of 45 >Zap/SIP conversations at once 14 hours a day. The processor load is never >that high and the RAM usage is less than half. > >There are never any consistent errors or warnings in the logs or the CLI. > >I do have several AGI perl scripts in the dialplan and the SIP phones that >connect are almost all Grandstream 102's, That's all I can think of that may >be causing any problems. I have mpg123/musiconhold deactivated. > >Has this happened to anyone else out there? > >Does anyone have any suggestions? or ideas as to what may be causing it? > >Thanks for any help, > >MATT--- >_______________________________________________ >Asterisk-Users mailing list >Asterisk-Users@lists.digium.com >http://lists.digium.com/mailman/listinfo/asterisk-users > >
I do "extensions reload" maybe once a day, does that have the same effect? Can AGI perl scripts cause the locking as I have described it? Anyone have any experience with that? Thanks, MATT--- -----Original Message----- From: James Sizemore [mailto:james@deny.org] Sent: Friday, December 05, 2003 8:36 AM To: asterisk-users@lists.digium.com Subject: Re: [Asterisk-Users] Asterisk freezing HELP Do you type "reload" at the cli a few times a day? If so try not reloading Asterisk and I'll bet Asterisk stop blocking. If you don't normally reload the box you will need to trouble shot normally. mattf wrote:>Hello, > >I have had several instances over the last month of Asterisk freezing, >sometimes after 12 hours, sometimes after 8 days. The common elements are >that: >- all Zap channels lock[hangups don't register and no new calls in or out] >- no new in/outbound calls can be made on Zap or SIP channels >- people who are still connected to calls can continue to talk >- in the CLI interface, you can "show channels" and view other info like >"sip show peers", but you cannot do a "stop now" only way to stop asterisk >process is a kill -9 >- all attempted connections to the manager interface time out >- the processor load and RAM usage on the machine is low. > >I am experiencing these freezes on two separate identical RedHat 9machines.>On each I have 4 T1s connected and a usual concurrent call volume of 45 >Zap/SIP conversations at once 14 hours a day. The processor load is never >that high and the RAM usage is less than half. > >There are never any consistent errors or warnings in the logs or the CLI. > >I do have several AGI perl scripts in the dialplan and the SIP phones that >connect are almost all Grandstream 102's, That's all I can think of thatmay>be causing any problems. I have mpg123/musiconhold deactivated. > >Has this happened to anyone else out there? > >Does anyone have any suggestions? or ideas as to what may be causing it? > >Thanks for any help, > >MATT--- >_______________________________________________ >Asterisk-Users mailing list >Asterisk-Users@lists.digium.com >http://lists.digium.com/mailman/listinfo/asterisk-users > >_______________________________________________ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users
> -----Original Message----- > From: TC [mailto:trclark@shaw.ca] > Sent: Thursday, December 04, 2003 11:58 PM > >>I have had several instances over the last month of Asterisk freezing, >>Does anyone have any suggestions? or ideas as to what may be causing it? > > Sounds like some type deadlock > Take a look here for my tought and how to debuig dead locks > http://lists.digium.com/pipermail/asterisk-dev/2003-November/002384.html > > Then let us know when you haved posted the info on bugs.digium.com >Added to: http://www.voip-info.org/tiki-index.php?page=Asterisk+debugging Small editing applied, TC: Please check. Thank you! /Olle
Has anyone out there had the freezing problem(where they have to kill asterisk with kill -9) on any linux distro other than RedHat? What other distros do people out there use with their production Asterisk systems? MATT--- -----Original Message----- From: Philipp von Klitzing [mailto:klitzing@pool.informatik.rwth-aachen.de] Sent: Friday, December 05, 2003 1:39 PM To: asterisk-users@lists.digium.com Subject: RE: [Asterisk-Users] Asterisk freezing HELP Hi!> I'm glad other people are seeing the same problem I've been seeing and > posted about a day or so ago. My * is running on rh9 with most recent > kernel with up2date.In my case: RH 7.2, kernel 2.4.20, reload through cron job each 30 min. Reason: http://bugs.digium.com/bug_view_page.php?bug_id=0000558 Philipp _______________________________________________ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users
On Friday 05 December 2003 14:44, mattf wrote:> Has anyone out there had the freezing problem(where they have to > kill asterisk with kill -9) on any linux distro other than RedHat? > > What other distros do people out there use with their production > Asterisk systems?We have no problem with freezing, and we use Mandrake 9.1 on most production systems. -Tilghman
No problems with freezing: using Redhat 9 in production systems (2) (but IVR only traffic in these), and Fedora in house for testing -Scott Scott M. Stingel Emerging Voice Technology Inc. Email: scott@evtmedia.com <mailto:scott@evtmedia.com> URL: www.evtmedia.com <http://www.evtmedia.com>> -----Original Message----- > From: asterisk-users-admin@lists.digium.com > [mailto:asterisk-users-admin@lists.digium.com] On Behalf Of > Tilghman Lesher > Sent: Friday, December 05, 2003 8:55 PM > To: asterisk-users@lists.digium.com > Subject: Re: [Asterisk-Users] Asterisk freezing HELP > > > On Friday 05 December 2003 14:44, mattf wrote: > > Has anyone out there had the freezing problem(where they have to > > kill asterisk with kill -9) on any linux distro other than RedHat? > > > > What other distros do people out there use with their production > > Asterisk systems? > > We have no problem with freezing, and we use Mandrake 9.1 on most > production systems. > > -Tilghman > > _______________________________________________ > Asterisk-Users mailing list > Asterisk-Users@lists.digium.com > http://lists.digium.com/mailman/listinfo/asterisk-users > >
Running multiple boxes with RH7.3 and RH9.
They all freeze every few days and always when processing an inbound SIP
call. Sometimes it will freeze 5 times in a row when processng a SIP call
from the same user.
We have been unable to reproduce this so no bug has been opened.
The only suspicious thing is that they always seem to freeze when the SIP
channel is right around this line:
if (pkt->owner->owner) {
/* XXX Potential deadlocK?? XXX */
ast_queue_hangup(pkt->owner->owner, 0);
ast_mutex_unlock(&pkt->owner->owner->lock);
And as you can see somebody did put in that "deadlock" comment already
so
they must suspect something is up.
Regards,
Andres
----- Original Message -----
From: "Scott Stingel" <scott@evtmedia.com>
To: <asterisk-users@lists.digium.com>
Sent: Friday, December 05, 2003 4:11 PM
Subject: RE: [Asterisk-Users] Asterisk freezing HELP
> No problems with freezing: using Redhat 9 in production systems (2) (but
> IVR only traffic in these), and Fedora in house for testing
>
> -Scott
>
> Scott M. Stingel
> Emerging Voice Technology Inc.
>
> Email: scott@evtmedia.com <mailto:scott@evtmedia.com>
> URL: www.evtmedia.com <http://www.evtmedia.com>
>
>
>
> > -----Original Message-----
> > From: asterisk-users-admin@lists.digium.com
> > [mailto:asterisk-users-admin@lists.digium.com] On Behalf Of
> > Tilghman Lesher
> > Sent: Friday, December 05, 2003 8:55 PM
> > To: asterisk-users@lists.digium.com
> > Subject: Re: [Asterisk-Users] Asterisk freezing HELP
> >
> >
> > On Friday 05 December 2003 14:44, mattf wrote:
> > > Has anyone out there had the freezing problem(where they have to
> > > kill asterisk with kill -9) on any linux distro other than
RedHat?
> > >
> > > What other distros do people out there use with their production
> > > Asterisk systems?
> >
> > We have no problem with freezing, and we use Mandrake 9.1 on most
> > production systems.
> >
> > -Tilghman
> >
> > _______________________________________________
> > Asterisk-Users mailing list
> > Asterisk-Users@lists.digium.com
> > http://lists.digium.com/mailman/listinfo/asterisk-users
> >
> >
>
> _______________________________________________
> Asterisk-Users mailing list
> Asterisk-Users@lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-users
>
> > The only suspicious thing is that they always seem to freeze when the SIP > channel is right around this line: > if (pkt->owner->owner) { > /* XXX Potential deadlocK?? XXX */ > ast_queue_hangup(pkt->owner->owner, 0); > ast_mutex_unlock(&pkt->owner->owner->lock);Would you be able to follow the How to Deug a dead lock next time & post the info to bugs.digium.com ??
Another scary thing I noticed this week is that the newer the CVS the more
frequently it happens:
CVS from 2003-10-01 once every few days
CVS from 2003-11-03 2 times a day to once every two days
CVS from 2003-12-03 1 to 4 times a day
All on otherwise identical systems. Has anything been added to the SIP
implementation of Asterisk that may be causing this(I'm assuming that SIP is
the cause)?
MATT---
-----Original Message-----
From: TeleSIP [mailto:ricvil@telesip.net]
Sent: Friday, December 05, 2003 4:26 PM
To: asterisk-users@lists.digium.com
Subject: Re: [Asterisk-Users] Asterisk freezing HELP
Running multiple boxes with RH7.3 and RH9.
They all freeze every few days and always when processing an inbound SIP
call. Sometimes it will freeze 5 times in a row when processng a SIP call
from the same user.
We have been unable to reproduce this so no bug has been opened.
The only suspicious thing is that they always seem to freeze when the SIP
channel is right around this line:
if (pkt->owner->owner) {
/* XXX Potential deadlocK?? XXX */
ast_queue_hangup(pkt->owner->owner, 0);
ast_mutex_unlock(&pkt->owner->owner->lock);
And as you can see somebody did put in that "deadlock" comment already
so
they must suspect something is up.
Regards,
Andres
----- Original Message -----
From: "Scott Stingel" <scott@evtmedia.com>
To: <asterisk-users@lists.digium.com>
Sent: Friday, December 05, 2003 4:11 PM
Subject: RE: [Asterisk-Users] Asterisk freezing HELP
> No problems with freezing: using Redhat 9 in production systems (2) (but
> IVR only traffic in these), and Fedora in house for testing
>
> -Scott
>
> Scott M. Stingel
> Emerging Voice Technology Inc.
>
> Email: scott@evtmedia.com <mailto:scott@evtmedia.com>
> URL: www.evtmedia.com <http://www.evtmedia.com>
>
>
>
> > -----Original Message-----
> > From: asterisk-users-admin@lists.digium.com
> > [mailto:asterisk-users-admin@lists.digium.com] On Behalf Of
> > Tilghman Lesher
> > Sent: Friday, December 05, 2003 8:55 PM
> > To: asterisk-users@lists.digium.com
> > Subject: Re: [Asterisk-Users] Asterisk freezing HELP
> >
> >
> > On Friday 05 December 2003 14:44, mattf wrote:
> > > Has anyone out there had the freezing problem(where they have to
> > > kill asterisk with kill -9) on any linux distro other than
RedHat?
> > >
> > > What other distros do people out there use with their production
> > > Asterisk systems?
> >
> > We have no problem with freezing, and we use Mandrake 9.1 on most
> > production systems.
> >
> > -Tilghman
> >
> > _______________________________________________
> > Asterisk-Users mailing list
> > Asterisk-Users@lists.digium.com
> > http://lists.digium.com/mailman/listinfo/asterisk-users
> >
> >
>
> _______________________________________________
> Asterisk-Users mailing list
> Asterisk-Users@lists.digium.com
> http://lists.digium.com/mailman/listinfo/asterisk-users
>
_______________________________________________
Asterisk-Users mailing list
Asterisk-Users@lists.digium.com
http://lists.digium.com/mailman/listinfo/asterisk-users
I sure will try. (I am working on it right now) Thanks, Andres. ----- Original Message ----- From: "TC" <trclark@shaw.ca> To: <asterisk-users@lists.digium.com> Sent: Friday, December 05, 2003 4:47 PM Subject: Re: [Asterisk-Users] Asterisk freezing HELP> > > > The only suspicious thing is that they always seem to freeze when theSIP> > channel is right around this line: > > if (pkt->owner->owner) { > > /* XXX Potential deadlocK?? XXX */ > > ast_queue_hangup(pkt->owner->owner, 0); > > ast_mutex_unlock(&pkt->owner->owner->lock); > Would you be able to follow the How to Deug a dead lock next time > & post the info to bugs.digium.com ?? > > > _______________________________________________ > Asterisk-Users mailing list > Asterisk-Users@lists.digium.com > http://lists.digium.com/mailman/listinfo/asterisk-users >
Hi TC, I followed your instructions. Today I managed to lock up one of our Asterisk boxes with an incoming SIP call. It locks up right after seeing I this line: Dec 8 13:48:34 WARNING[4101]: File chan_sip.c, Line 456 (retrans_pkt): Maximum retries exceeded on call fc19b3c44627f4d6@192.168.0.160 for seqno 1058 (Response) I was able to follow down to STEP #7 of your Deadlock debug Instructions, but the file in step #8 is not present. In any case here is the relevant thread that is locked: Thread 6 (Thread 4101 (LWP 8372)): #0 0x420292e5 in sigsuspend () from /lib/i686/libc.so.6 #1 0x40039679 in __pthread_wait_for_restart_signal () from /lib/i686/libpthread.so.0 #2 0x4003b4b9 in __pthread_alt_lock () from /lib/i686/libpthread.so.0 #3 0x4003816e in pthread_mutex_lock () from /lib/i686/libpthread.so.0 #4 0x41be8235 in retrans_pkt (data=0x8138e50) at ../include/asterisk/lock.h:89 #5 0x08051e05 in ast_sched_runq (con=0x80eeac8) at sched.c:355 #6 0x41bfba86 in do_monitor (data=0x0) at chan_sip.c:5409 #7 0x40036fef in pthread_start_thread () from /lib/i686/libpthread.so.0 (gdb) thread 6 [Switching to thread 6 (Thread 4101 (LWP 8372))]#0 0x420292e5 in sigsuspend () from /lib/i686/libc.so.6 (gdb) frame 2 #2 0x4003b4b9 in __pthread_alt_lock () from /lib/i686/libpthread.so.0 (gdb) Is this enough to open a ticket in the BUG tracker? Is there anything else I should look for? I have not restared Asterisk yet. Thanks, Andres. ----- Original Message ----- From: "TC" <trclark@shaw.ca> To: <asterisk-users@lists.digium.com> Sent: Friday, December 05, 2003 4:47 PM Subject: Re: [Asterisk-Users] Asterisk freezing HELP> > > > The only suspicious thing is that they always seem to freeze when theSIP> > channel is right around this line: > > if (pkt->owner->owner) { > > /* XXX Potential deadlocK?? XXX */ > > ast_queue_hangup(pkt->owner->owner, 0); > > ast_mutex_unlock(&pkt->owner->owner->lock); > Would you be able to follow the How to Deug a dead lock next time > & post the info to bugs.digium.com ?? > > > _______________________________________________ > Asterisk-Users mailing list > Asterisk-Users@lists.digium.com > http://lists.digium.com/mailman/listinfo/asterisk-users >
Hello, I've continued to have lockups since last week and this is really beginning to drive me crazy. Could this be a problem with specific hardware or a specific setting in the configuration files that causes these lockups for some and not others? If everyone that is experiencing Asterisk lockups would email me their setup: - system hardware/software i.e.(ASUS w/P4 2.6 GHz 2GB DDR RAM - Redhat 9.0 SMP) - hardphones/softphones they are using and how many - Zap hardware they are using and what is hooked up to them - send me extensions.conf and sip.conf files I will attempt to figure out what is common amongst the unlucky and submit my findings to the list If anyone that has had no lockups for at least a month would like to send me the above information that also may help me as well. Thanks for your help, MATT--- -----Original Message----- From: TeleSIP [mailto:ricvil@telesip.net] Sent: Monday, December 08, 2003 2:30 PM To: asterisk-users@lists.digium.com Subject: Re: [Asterisk-Users] Asterisk freezing HELP Hi TC, I followed your instructions. Today I managed to lock up one of our Asterisk boxes with an incoming SIP call. It locks up right after seeing I this line: Dec 8 13:48:34 WARNING[4101]: File chan_sip.c, Line 456 (retrans_pkt): Maximum retries exceeded on call fc19b3c44627f4d6@192.168.0.160 for seqno 1058 (Response) I was able to follow down to STEP #7 of your Deadlock debug Instructions, but the file in step #8 is not present. In any case here is the relevant thread that is locked: Thread 6 (Thread 4101 (LWP 8372)): #0 0x420292e5 in sigsuspend () from /lib/i686/libc.so.6 #1 0x40039679 in __pthread_wait_for_restart_signal () from /lib/i686/libpthread.so.0 #2 0x4003b4b9 in __pthread_alt_lock () from /lib/i686/libpthread.so.0 #3 0x4003816e in pthread_mutex_lock () from /lib/i686/libpthread.so.0 #4 0x41be8235 in retrans_pkt (data=0x8138e50) at ../include/asterisk/lock.h:89 #5 0x08051e05 in ast_sched_runq (con=0x80eeac8) at sched.c:355 #6 0x41bfba86 in do_monitor (data=0x0) at chan_sip.c:5409 #7 0x40036fef in pthread_start_thread () from /lib/i686/libpthread.so.0 (gdb) thread 6 [Switching to thread 6 (Thread 4101 (LWP 8372))]#0 0x420292e5 in sigsuspend () from /lib/i686/libc.so.6 (gdb) frame 2 #2 0x4003b4b9 in __pthread_alt_lock () from /lib/i686/libpthread.so.0 (gdb) Is this enough to open a ticket in the BUG tracker? Is there anything else I should look for? I have not restared Asterisk yet. Thanks, Andres. ----- Original Message ----- From: "TC" <trclark@shaw.ca> To: <asterisk-users@lists.digium.com> Sent: Friday, December 05, 2003 4:47 PM Subject: Re: [Asterisk-Users] Asterisk freezing HELP> > > > The only suspicious thing is that they always seem to freeze when theSIP> > channel is right around this line: > > if (pkt->owner->owner) { > > /* XXX Potential deadlocK?? XXX */ > > ast_queue_hangup(pkt->owner->owner, 0); > > ast_mutex_unlock(&pkt->owner->owner->lock); > Would you be able to follow the How to Deug a dead lock next time > & post the info to bugs.digium.com ?? > > > _______________________________________________ > Asterisk-Users mailing list > Asterisk-Users@lists.digium.com > http://lists.digium.com/mailman/listinfo/asterisk-users >_______________________________________________ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users
UPDATE: We were able to consistently reproduce this problem using a Grandstream phone with buggy firmware. Mark Spencer logged into our Asterisk and identified the issue. He said it was a typo in an ast_mutex_lock. After fixing it, the problem seems to have been solved. We have now repeated about 100 calls and no lockup (with the buggy phone we were able to lock it up in under 7 calls). CVS should now reflect his fix. And by the way, do not use firmware > 1.0.4.18 on GS phones. It contains a nasty SIP Port bug. Regards, Andres. ----- Original Message ----- From: "TeleSIP" <ricvil@telesip.net> To: <asterisk-users@lists.digium.com> Sent: Monday, December 08, 2003 2:30 PM Subject: Re: [Asterisk-Users] Asterisk freezing HELP> Hi TC, > > I followed your instructions. Today I managed to lock up one of our > Asterisk boxes with an incoming SIP call. > > It locks up right after seeing I this line: > Dec 8 13:48:34 WARNING[4101]: File chan_sip.c, Line 456 (retrans_pkt): > Maximum retries exceeded on call fc19b3c44627f4d6@192.168.0.160 for seqno > 1058 (Response) > > I was able to follow down to STEP #7 of your Deadlock debug Instructions, > but the file in step #8 is not present. > > In any case here is the relevant thread that is locked: > Thread 6 (Thread 4101 (LWP 8372)): > #0 0x420292e5 in sigsuspend () from /lib/i686/libc.so.6 > #1 0x40039679 in __pthread_wait_for_restart_signal () from > /lib/i686/libpthread.so.0 > #2 0x4003b4b9 in __pthread_alt_lock () from /lib/i686/libpthread.so.0 > #3 0x4003816e in pthread_mutex_lock () from /lib/i686/libpthread.so.0 > #4 0x41be8235 in retrans_pkt (data=0x8138e50) at > ../include/asterisk/lock.h:89 > #5 0x08051e05 in ast_sched_runq (con=0x80eeac8) at sched.c:355 > #6 0x41bfba86 in do_monitor (data=0x0) at chan_sip.c:5409 > #7 0x40036fef in pthread_start_thread () from /lib/i686/libpthread.so.0 > > (gdb) thread 6 > [Switching to thread 6 (Thread 4101 (LWP 8372))]#0 0x420292e5 insigsuspend> () from /lib/i686/libc.so.6 > (gdb) frame 2 > #2 0x4003b4b9 in __pthread_alt_lock () from /lib/i686/libpthread.so.0 > (gdb) > > Is this enough to open a ticket in the BUG tracker? > > Is there anything else I should look for? I have not restared Asteriskyet.> > Thanks, > Andres. > > > ----- Original Message ----- > From: "TC" <trclark@shaw.ca> > To: <asterisk-users@lists.digium.com> > Sent: Friday, December 05, 2003 4:47 PM > Subject: Re: [Asterisk-Users] Asterisk freezing HELP > > > > > > > > The only suspicious thing is that they always seem to freeze when the > SIP > > > channel is right around this line: > > > if (pkt->owner->owner) { > > > /* XXX Potential deadlocK?? XXX */ > > > ast_queue_hangup(pkt->owner->owner, 0); > > > ast_mutex_unlock(&pkt->owner->owner->lock); > > Would you be able to follow the How to Deug a dead lock next time > > & post the info to bugs.digium.com ?? > > > > > > _______________________________________________ > > Asterisk-Users mailing list > > Asterisk-Users@lists.digium.com > > http://lists.digium.com/mailman/listinfo/asterisk-users > > > > > _______________________________________________ > Asterisk-Users mailing list > Asterisk-Users@lists.digium.com > http://lists.digium.com/mailman/listinfo/asterisk-users
TeleSIP wrote:>UPDATE: >We were able to consistently reproduce this problem using a Grandstream >phone with buggy firmware. > >Mark Spencer logged into our Asterisk and identified the issue. He said it >was a typo in an ast_mutex_lock. After fixing it, the problem seems to have >been solved. We have now repeated about 100 calls and no lockup (with the >buggy phone we were able to lock it up in under 7 calls). > >CVS should now reflect his fix. > >And by the way, do not use firmware > 1.0.4.18 on GS phones. It contains a >nasty SIP Port bug. > >Regards, >Andres. > >One bug I found on my GS ATA adapter. I had it pointing to my * server for ntp. I did not have ntpd running. You could start a call, but during the call it would do a ntp request. The server was sending an ICMP message back and then the adapter would terminate the call. I fired up ntpd and all is well. -- Bob Knight [-w] the work option bk@minusw.com 925-449-9163