asterisk users - Jan 2006 - Lockups since upgrade 1.2.3 - anyone else? Any ideas?

Boy oh boy. This blows. I upgraded to 1.2.2 from 1.0.9, and of course had
the timebomb bug. Immediately after upgrading to 1.2.3 we were ok, for 24
hours or so.

Since upgrading to 1.2.3, though, the whole system has locked up twice. Once
on Thursday, and then about a half hour ago. The server would reply to a
ping, but no ssh login, no local console login - just locked up. This ain't
good for business.

I have scoured the logs, and find no errors. Not even right before/around
the time of the crash.

I am worried that 1.2.3 is not as stable as 1.0.9 (or 1.0.10, though we
never ran that version). Is there a needed step aside from "make; make
install" that I missed when upgrading? Has anyone else had similar
problems?
Or, if I submit other info, would someone have a clue as to what to look at?
We run a TDM400P with 3 FXO modules, and about 15 SIP Cisco 79XX phones
here. Any help is appreciated, this cannot continue.

Sincerely,

Brent A. Torrenga
brent.torrenga@torrenga.com

Torrenga Engineering, Inc.
907 Ridge Road
Munster, Indiana 46321-1771

219.836.8918x325 Voice
219.836.1138 Facsimile
www.torrenga.com

Another piece to the puzzle, for what it's worth:

The last moments before the crash, an incoming Zap call was answered by a
SIP phone, parked, and then picked up by another SIP phone. During the
picked up conversation, the audio was reported to me to be patchy, described
as "cell phone" like. It is known that the calling party was not on a
cell
phone, but on a land line. ALSO, in the CDR CSV file, there is no mention of
the call having been taken off park, as though the patchy call never
happened. I confirmed this by looking at the specific Cisco phone, and
seeing the last call that was made, and at what time. Does this speak to any
suspect source of the issue?


Sincerely,

Brent A. Torrenga
brent.torrenga@torrenga.com

Torrenga Engineering, Inc.
907 Ridge Road
Munster, Indiana 46321-1771

219.836.8918x325 Voice
219.836.1138 Facsimile
www.torrenga.com

Hi Brent - 
> Boy oh boy. This blows. I upgraded to 1.2.2 from 1.0.9, and of course had
> the timebomb bug. Immediately after upgrading to 1.2.3 we were ok, for 24
> hours or so.
> 
> Since upgrading to 1.2.3, though, the whole system has locked up twice.
Once
> on Thursday, and then about a half hour ago. The server would reply to a
> ping, but no ssh login, no local console login - just locked up. This
ain't
> good for business.

We've been doing fine with 1.2.3 so far.  No problems reported, though I
only have it deployed in a small office.  Definitely no lock-ups.

On the asterisk side, just a basic question - did you make sure to remove
the old modules so the new 1.2.3 versions got installed?

As far as the lockups, maybe it is coincidental?  I've never had asterisk
(even the crazy CVS versions) lock a whole OS like that.  I have had
machines running asterisk lock up, but it always turned out to be caused by
something else like bad hardware, or unrelated network problems.

- Noah

Same here on a fresh install on a test machine, no TDM card, just fresh
install using fresh install of CentOS-4.2.ServerCD-i386.iso.

I thought it was more likely when we accessed Flash Operator Panel.

Want to upgrade from 1.0.9 but now a bit more worried.

James


-----Original Message-----
From: asterisk-users-bounces@lists.digium.com
[mailto:asterisk-users-bounces@lists.digium.com] On Behalf Of Brent Torrenga
Sent: Saturday, 28 January 2006 5:42 AM
To: asterisk-users@lists.digium.com
Subject: [Asterisk-Users] Lockups since upgrade 1.2.3 - anyone else?
Anyideas?

Boy oh boy. This blows. I upgraded to 1.2.2 from 1.0.9, and of course had
the timebomb bug. Immediately after upgrading to 1.2.3 we were ok, for 24
hours or so.

Since upgrading to 1.2.3, though, the whole system has locked up twice. Once
on Thursday, and then about a half hour ago. The server would reply to a
ping, but no ssh login, no local console login - just locked up. This ain't
good for business.

I have scoured the logs, and find no errors. Not even right before/around
the time of the crash.

I am worried that 1.2.3 is not as stable as 1.0.9 (or 1.0.10, though we
never ran that version). Is there a needed step aside from "make; make
install" that I missed when upgrading? Has anyone else had similar
problems?
Or, if I submit other info, would someone have a clue as to what to look at?
We run a TDM400P with 3 FXO modules, and about 15 SIP Cisco 79XX phones
here. Any help is appreciated, this cannot continue.

Sincerely,

Brent A. Torrenga
brent.torrenga@torrenga.com

Torrenga Engineering, Inc.
907 Ridge Road
Munster, Indiana 46321-1771

219.836.8918x325 Voice
219.836.1138 Facsimile
www.torrenga.com

_______________________________________________
--Bandwidth and Colocation provided by Easynews.com --

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

It looks like my modules are all up to date - they are all dated the 25th
Jan - aka "Black Wednesday".

What really bugs me about this is the lack of useful info from any logs. The
last call to take place, the call that gets distorted, has no entry. This
has to indicate something, no?
>Hmm - I'd do as others have suggested and move the
/usr/lib/asterisk/modules  directory to another, and do a make
clean;make;make install
>If you have app_rxfax.so installed then you must have customised your
original makefile, and not the 1.2.3 makefile, which would suggest that
these modules are from a previous asterisk version.
>Let us know how you get on.
>Julian.
>Dan Littlejohn wrote:
>> On 1/27/06, Julian Lyndon-Smith <asterisk@dotr.com> wrote:
>>> These modules are not part of the standard 1.2.3 release - did you 
>>> also install the 1.2.3 release of the asterisk-addons package ?
>>>
>>> If * is loading older modules (which it probably is because of your
>>> config files) then it may cause grief ;)
>>>
>>> My .2p worth. Probably not helpful, but maybe, just maybe ....
>>>
>>> Julian
>>>
>>> Dan Littlejohn wrote:
>>>> On 1/27/06, Noah Miller <noah@rosecompanies.com> wrote:
>>>>> Hi Brent -

Sincerely,

Brent A. Torrenga
brent.torrenga@torrenga.com

Torrenga Engineering, Inc.
907 Ridge Road
Munster, Indiana 46321-1771

219.836.8918x325 Voice
219.836.1138 Facsimile
www.torrenga.com

On Jan 29, 2006, at 1:24 PM, Michiel van Baak wrote:
> On 13:09, Sun 29 Jan 06, Martin Joseph wrote:
>> I removed the following to get it starting up again:
>>
>> app_enumlookup.so
>> app_groupcount.so
>> app_md5.so
>> app_txtcidname.so
>> func_cut.so
>
> Both the README and the UPGRADE listed that those functions
> became obsolete and were replaced by dialplan functions.
> It got me too the first time, but after reading some more
> docs and the release messages all got clear to me.
> Now everytime I upgrade I first move away the
> module/app/func dir just to be sure.

Thanks,  duh!  I guess RTFM is the answer...

Marty

This thread seems to have gone off the original topic...

Brent Torrenga wrote:
> Boy oh boy. This blows. I upgraded to 1.2.2 from 1.0.9, and of course
had 
> the timebomb bug. Immediately after upgrading to 1.2.3 we were ok, for
24 
> hours or so. 
> Since upgrading to 1.2.3, though, the whole system has locked up
twice. Once 
> on Thursday, and then about a half hour ago. The server would reply to
a 
> ping, but no ssh login, no local console login - just locked up. This
ain't 
> good for business. 
And James Sturges wrote:
> Same here on a fresh install on a test machine, no TDM card, just
fresh 
> install using fresh install of CentOS-4.2.ServerCD-i386.iso. 
> I thought it was more likely when we accessed Flash Operator Panel. 
We have had the same symptoms - all TCP/UDP services become
inaccessible, but ping still works - twice in the last three days.
We're running 1.2.3 on Debian, kernel 2.4.27.  SIP only - no PRI cards.
Nothing of any significane in the logs, as far as I can see.

Will try upgrading to 1.2.4 - has that helped either of you two?

Regards

David