Re: [asterisk-users] kore dump

2007-06-29 Thread Ed Nuñez
For anyone interested on the crashes I was experiencing when using ChanSpy
from SIP extension to SIP extensions with the group option.  For the last
couple of days, I’ve been monitoring from Zap extensions to SIP extensions,
and the system has not crashed once.  The problem only happens when I spy
from SIP.

 

 

From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Vadim
Berezniker
Sent: Tuesday, June 26, 2007 2:36 PM
To: Asterisk Users Mailing List - Non-Commercial Discussion
Subject: Re: [asterisk-users] kore dump

 

use the safe_asterisk script

 

it will restart asterisk if it crashes and it enables core dumps (your core
size limit is probably set to 0 when you start asterisk).

 

From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Ed Nuñez
Sent: Tuesday, June 26, 2007 2:22 PM
To: Asterisk Users Mailing List - Non-Commercial Discussion;
[EMAIL PROTECTED]
Subject: [asterisk-users] kore dump

 

I am running Asterisk 1.4.5 and addons 1.4.1 in a CentOS 5 Server.

 

My PBX has experienced several core dumps the last couple of days and I am
not sure if this is what’s causing it, but it always seems to happen when a
particular extension on a grandstream phone uses ChanSpy SIP group.

 

I have not been able to locate where the core dump file is being saved.   I
can’t find it in my TMP directory.

 

I would also like to know if Asterisk can be setup to automatically re start
if there is a core dump.  I was thinking of setting up a cron job to launch
Asterisk every minute.  If it’s running, no harm done, and if it crashes,
the cron job will make sure that it’s started every 60 seconds.

 

Any suggestions?

 

 

Thank you

 

Ed Nuñez

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] kore dump

2007-06-29 Thread Russell Bryant
Ed Nuñez wrote:
 For anyone interested on the crashes I was experiencing when using 
 ChanSpy from SIP extension to SIP extensions with the group option.  For 
 the last couple of days, I’ve been monitoring from Zap extensions to SIP 
 extensions, and the system has not crashed once.  The problem only 
 happens when I spy from SIP.

First, just to be safe, give 1.4.6 a try.  If you still have a problem, please 
report it to bugs.digium.com  We would be happy to help figure it out and get 
it 
fixed.

To get the backtrace:

1) Run make menuselect, go to Compiler Flags, select DONT_OPTIMIZE.

2) # make clean ; make ; make install

3) Run asterisk with the -g argument

4) gdb /usr/sbin/asterisk core.12345

4.i)  (gdb) bt
4.ii) (gdb) bt full

-- 
Russell Bryant
Software Engineer
Digium, Inc.

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] kore dump

2007-06-27 Thread Ed Nuñez
What is a god Windows application to read core dump files?



-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of J. Oquendo
Sent: Tuesday, June 26, 2007 4:26 PM
To: Asterisk Users Mailing List - Non-Commercial Discussion
Subject: Re: [asterisk-users] kore dump

Vadim Berezniker wrote:

 use the safe_asterisk script

 it will restart asterisk if it crashes and it enables core dumps (your 
 core size limit is probably set to 0 when you start asterisk).

 *From:* [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] *On Behalf Of *Ed 
 Nuñez
 *Sent:* Tuesday, June 26, 2007 2:22 PM
 *To:* Asterisk Users Mailing List - Non-Commercial Discussion; 
 [EMAIL PROTECTED]
 *Subject:* [asterisk-users] kore dump

 I am running Asterisk 1.4.5 and addons 1.4.1 in a CentOS 5 Server.

 My PBX has experienced several core dumps the last couple of days and 
 I am not sure if this is what’s causing it, but it always seems to 
 happen when a particular extension on a grandstream phone uses ChanSpy 
 SIP group.

 I have not been able to locate where the core dump file is being 
 saved. I can’t find it in my TMP directory.

 I would also like to know if Asterisk can be setup to automatically re 
 start if there is a core dump. I was thinking of setting up a cron job 
 to launch Asterisk every minute. If it’s running, no harm done, and if 
 it crashes, the cron job will make sure that it’s started every 60 
 seconds.

 Any suggestions?

 Thank you

 Ed Nuñez

 --
 --

 ___
 --Bandwidth and Colocation Provided by http://www.api-digital.com--

 asterisk-users mailing list
 To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users

If that fails you could always try something like:
*/2 * * * * /bin/ps -C /usr/bin/asterisk || { /usr/bin/asterisk  }

or so...

--

J. Oquendo
http://pgp.mit.edu:11371/pks/lookup?op=getsearch=0x1383A743
echo infiltrated.net|sed 's/^/sil@/g' 

Wise men talk because they have something to say; fools, because they have
to say something. -- Plato





___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] kore dump

2007-06-27 Thread William Moore
On 6/27/07, Ed Nuñez [EMAIL PROTECTED] wrote:
 What is a god Windows application to read core dump files?

No.  Core files must be examined on the same system that created them.

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] kore dump

2007-06-27 Thread Jared Smith
On 6/27/07, Ed Nuñez [EMAIL PROTECTED] wrote:
 What is a god Windows application to read core dump files?

The core files are meant to be read by the gdb debugger on the machine
in which the crash happened, so that gdb can look at the debugging
symbols in the code and the system libraries.  A core file by itself
is pretty useless, so I doubt anyone has written a Windows application
to read core dump files.

-Jared

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] kore dump

2007-06-27 Thread John Faubion
What is a god Windows application to read core dump files?

Microsoft jokes aside, I would seriously doubt there could be a good Windows
application for analyzing core dumps. Due to the OS specific nature of core
dumps, the need to have the source files, debugger and more, would make it
difficult. I'm not saying there isn't one, I've just never heard of one.
When developing software modules, I've had some success using crash on
Fedora systems. Though as a whole system, a review of the logs to see what
was changed just prior to getting the core dumps has been more effective at
isolating the problem than the analysis of the core dump.

John


___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


[asterisk-users] kore dump

2007-06-26 Thread Ed Nuñez
I am running Asterisk 1.4.5 and addons 1.4.1 in a CentOS 5 Server.

 

My PBX has experienced several core dumps the last couple of days and I am not 
sure if this is what's causing it, but it always seems to happen when a 
particular extension on a grandstream phone uses ChanSpy SIP group.

 

I have not been able to locate where the core dump file is being saved.   I 
can't find it in my TMP directory.

 

I would also like to know if Asterisk can be setup to automatically re start if 
there is a core dump.  I was thinking of setting up a cron job to launch 
Asterisk every minute.  If it's running, no harm done, and if it crashes, the 
cron job will make sure that it's started every 60 seconds.

 

Any suggestions?

 

 

Thank you

 

Ed Nuñez

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] kore dump

2007-06-26 Thread John Faubion
  I would also like to know if Asterisk can be setup to automatically re
start if there is a core dump.

Sure! You should already have the required script. Just run it from
safe_asterisk. Here is a link with more info:
http://www.asteriskdocs.org/modules/tinycontent/content/docbook/current/docs
-html/x389.html


John
___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] kore dump

2007-06-26 Thread Vadim Berezniker
use the safe_asterisk script

 

it will restart asterisk if it crashes and it enables core dumps (your core 
size limit is probably set to 0 when you start asterisk).

 

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ed Nuñez
Sent: Tuesday, June 26, 2007 2:22 PM
To: Asterisk Users Mailing List - Non-Commercial Discussion; [EMAIL PROTECTED]
Subject: [asterisk-users] kore dump

 

I am running Asterisk 1.4.5 and addons 1.4.1 in a CentOS 5 Server.

 

My PBX has experienced several core dumps the last couple of days and I am not 
sure if this is what's causing it, but it always seems to happen when a 
particular extension on a grandstream phone uses ChanSpy SIP group.

 

I have not been able to locate where the core dump file is being saved.   I 
can't find it in my TMP directory.

 

I would also like to know if Asterisk can be setup to automatically re start if 
there is a core dump.  I was thinking of setting up a cron job to launch 
Asterisk every minute.  If it's running, no harm done, and if it crashes, the 
cron job will make sure that it's started every 60 seconds.

 

Any suggestions?

 

 

Thank you

 

Ed Nuñez

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] kore dump

2007-06-26 Thread Jared Smith
On 6/26/07, Ed Nuñez [EMAIL PROTECTED] wrote:
 I have not been able to locate where the core dump file is being saved.   I
 can't find it in my TMP directory.

Check the directory in which you're starting Asterisk.  It doesn't
sound like you're using the Red Hat initscript to start Asterisk, so
you're probably starting it manually.  I'd check there first.

 I would also like to know if Asterisk can be setup to automatically re start
 if there is a core dump.  I was thinking of setting up a cron job to launch
 Asterisk every minute.  If it's running, no harm done, and if it crashes,
 the cron job will make sure that it's started every 60 seconds.

Check out the safe_asterisk script that's installed with Asterisk.
Again, I recommend you check out the initscript, which calls
safe_asterisk and restarts Asterisk if it crashes.  It also ensures
that your core dumps end up in /tmp where you can find them more
easily.  On Red Hat and CentOS, you can install the initscript by
typing make config in the Asterisk source directory.

-Jared

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] kore dump

2007-06-26 Thread Eric Lubow
Ed,

   I am having a problem with Asterisk frequently crashing on me as
well.  I just run it under supervise:
   http://cr.yp.to/daemontools/supervise.html

   This way it will be restarted if svc determines it isn't running.

Eric

On Tue, 2007-06-26 at 13:22 -0500, Ed Nuñez wrote:
 I am running Asterisk 1.4.5 and addons 1.4.1 in a CentOS 5 Server.
 
  
 
 My PBX has experienced several core dumps the last couple of days and
 I am not sure if this is what’s causing it, but it always seems to
 happen when a particular extension on a grandstream phone uses ChanSpy
 SIP group.
 
  
 
 I have not been able to locate where the core dump file is being
 saved.   I can’t find it in my TMP directory.
 
  
 
 I would also like to know if Asterisk can be setup to automatically re
 start if there is a core dump.  I was thinking of setting up a cron
 job to launch Asterisk every minute.  If it’s running, no harm done,
 and if it crashes, the cron job will make sure that it’s started every
 60 seconds.
 
  
 
 Any suggestions?
 
  
 
  
 
 Thank you
 
  
 
 Ed Nuñez
 
 
 ___
 --Bandwidth and Colocation Provided by http://www.api-digital.com--
 
 asterisk-users mailing list
 To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
-- 
Eric Lubow
LinkExperts, Inc.
Systems Administrator
e: [EMAIL PROTECTED]
w: www.linkexperts.com
p: 212.542.5201


___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] kore dump

2007-06-26 Thread J. Oquendo

Vadim Berezniker wrote:


use the safe_asterisk script

it will restart asterisk if it crashes and it enables core dumps (your 
core size limit is probably set to 0 when you start asterisk).


*From:* [EMAIL PROTECTED] 
[mailto:[EMAIL PROTECTED] *On Behalf Of *Ed Nuñez

*Sent:* Tuesday, June 26, 2007 2:22 PM
*To:* Asterisk Users Mailing List - Non-Commercial Discussion; 
[EMAIL PROTECTED]

*Subject:* [asterisk-users] kore dump

I am running Asterisk 1.4.5 and addons 1.4.1 in a CentOS 5 Server.

My PBX has experienced several core dumps the last couple of days and 
I am not sure if this is what’s causing it, but it always seems to 
happen when a particular extension on a grandstream phone uses ChanSpy 
SIP group.


I have not been able to locate where the core dump file is being 
saved. I can’t find it in my TMP directory.


I would also like to know if Asterisk can be setup to automatically re 
start if there is a core dump. I was thinking of setting up a cron job 
to launch Asterisk every minute. If it’s running, no harm done, and if 
it crashes, the cron job will make sure that it’s started every 60 
seconds.


Any suggestions?

Thank you

Ed Nuñez



___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


If that fails you could always try something like:
*/2 * * * * /bin/ps -C /usr/bin/asterisk || { /usr/bin/asterisk  }

or so...

--

J. Oquendo
http://pgp.mit.edu:11371/pks/lookup?op=getsearch=0x1383A743
echo infiltrated.net|sed 's/^/sil@/g' 


Wise men talk because they have something to say;
fools, because they have to say something. -- Plato




smime.p7s
Description: S/MIME Cryptographic Signature
___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Re: [asterisk-users] kore dump

2007-06-26 Thread Luki
 I am having a problem with Asterisk frequently crashing on me as
 well.  I just run it under supervise:

But that's just a band-aid. If it crashes, it takes all calls with it.
Hardly a good thing, unless you only have 1 call at a time -- then
it's probably no the end of the world.

I still don't know what's up with the crashes but here are two
observations I made:

1) I moved the same installation from one hardware to another. On the
former hardware it would crash every 2 weeks, on average. On the new
hardware, it has not yet crashed and it has 9 weeks of uptime. Same
call volume, same devices, same network. I'm running asterisk
chroot'ed so all libraries, binaries, config files, etc. are
identical. Only the hardware and kernel are different.

2) The same old hardware has been in service for 3 years and no other
programs crash on it. Ever. It's no unusual seeing uptime for say
qmail, samba or bind of 200+ days. I have therefore reasons to believe
that the hardware is OK.

So go figure. And BTW, the crashes (based on the core dumps) are
always at a different place. There is no consistency. Right now I'm
just glad it no longer core dump on me :).

--Luki

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] kore dump

2007-06-26 Thread Mojo with Horan Company, LLC
just an idea, but maybe qmail, samba, and bind have a smaller memory 
footprint than an in-use asterisk? can you take the hardware offline 
long enough for a memtest?

Moj

Luki wrote:
 It's no unusual seeing uptime for say
 qmail, samba or bind of 200+ days. 

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users


Re: [asterisk-users] kore dump

2007-06-26 Thread Luki
 just an idea, but maybe qmail, samba, and bind have a smaller memory
 footprint than an in-use asterisk?

No, probably not. Asterisk's is about 20-40 MB depending on the number
of extensions, etc. Smbd's is similar, bind's is actually 90 MB (with
about 600 zones).


 can you take the hardware offline long enough for a memtest?

The machine has been retired (routine upgrade cycle). But I hardly
doubt that was the problem. My guess is it was somehow related to
limited CPU power (thread switching, interrupts, or whatnot). The old
hardware was single CPU and a lot slower.

--Luki

___
--Bandwidth and Colocation Provided by http://www.api-digital.com--

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users