[asterisk-users] Asterisk using 200% CPU and then crashing...

2007-10-17 Thread Carlos Chavez
We have a customer that has Asterisk 1.4.12.1, Zaptel 1.4.5.1,
Asterisk-Addons 1.4.3. running on a Dell Poweredge 1900 server (Dual
Core Xeon, 4gb RAM, 500gb Raid 5).  Until a month ago they had two
TE120P cards and everything was working fine.  Since they needed to add
a third E1 line we decided to change one of the TE120P cards with a
TE210P.  After the change we had a couple of crashes (server completely
locks up).  At first we thought we had a defective card so we changed it
but kept having the same problem.  We have tried a TE411P, a Sangoma
A104x (PCI express), an Openvox D210P and D410P and still the server
crashes about three times a week.  I can see the Asterisk process go
from around 10% usage to 100%, then 200% and the server hangs after
that.

Why would inserting a multiport card affect Asterisk and the server?
How can I debug this situation?  I do not have enough slots to insert
three single cards of the same type so I need the multiport card to
work.  When Asterisk goes above 100% it will start ignoring commands you
issue on the CLI but calls keep coming and going.  At this moment I have
to watch the server all day and if it goes above 100% I restart Asterisk
when we have a low call volume.  Obviously I cannot keep doing this
forever.

Could this be a problem with the Server?  Maybe a Kernel problem?  So
far all I have been willing to do is put acpi=off as a parameter on
the kernel because I read that sometimes ACPI causes problems.  Same
problem still.  I have not been able to determine a pattern to when
Asterisk will start using more CPU, sometimes it will go four days of
heavy use without a problem and then start acting up on a Saturday with
very few users.  Any ideas will be appreciated.

-- 
Telecomunicaciones Abiertas de México S.A. de C.V.
Carlos Chávez Prats
Director de Tecnología
+52-55-91169161 ext 2001


signature.asc
Description: This is a digitally signed message part
___
--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] Asterisk using 200% CPU and then crashing...

2007-10-17 Thread Doug Lytle
Carlos Chavez wrote:
 crashes about three times a week.  I can see the Asterisk process go
 from around 10% usage to 100%, then 200% and the server hangs after
 that.

   Why would inserting a multiport card affect Asterisk and the server?
 How can I debug this situation?  I do not have enough slots to insert
   

Carlos,

Just as a temporary suggestion, move back to the 2 cards and see if you 
are still experiencing the issues.  If you still have the issues, you'll 
know to look elsewhere.

Doug

-- 
 
Ben Franklin quote:

Those who would give up Essential Liberty to purchase a little Temporary 
Safety, deserve neither Liberty nor Safety.



___
--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] Asterisk using 200% CPU and then crashing...

2007-10-17 Thread Atis Lezdins
On Wednesday 17 October 2007 19:09:23 Carlos Chavez wrote:
 Why would inserting a multiport card affect Asterisk and the
 server? How can I debug this situation?  I do not have enough slots to
 insert three single cards of the same type so I need the multiport card to
 work.  When Asterisk goes above 100% it will start ignoring commands you
 issue on the CLI but calls keep coming and going.  At this moment I have to
 watch the server all day and if it goes above 100% I restart Asterisk when
 we have a low call volume.  Obviously I cannot keep doing this forever.

I recently had the same problems. Just that wasnt't related to installing any 
new hardware. You can check out the issue 
http://bugs.digium.com/view.php?id=10775

Could you provide your OS and glibc version? Also - can you try to disable 
IAX?

Regards,
Atis

-- 
Atis Lezdins
VoIP Developer,
IQ Labs Inc.
[EMAIL PROTECTED]
Skype: atis.lezdins
Cell Phone: +371 28806004
Work phone: +1 800 7502835

___
--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] Asterisk using 200% CPU and then crashing...

2007-10-17 Thread Carlos Chavez
On Wed, 2007-10-17 at 21:03 +0300, Atis Lezdins wrote:
 On Wednesday 17 October 2007 19:09:23 Carlos Chavez wrote:
  Why would inserting a multiport card affect Asterisk and the
  server? How can I debug this situation?  I do not have enough slots to
  insert three single cards of the same type so I need the multiport card to
  work.  When Asterisk goes above 100% it will start ignoring commands you
  issue on the CLI but calls keep coming and going.  At this moment I have to
  watch the server all day and if it goes above 100% I restart Asterisk when
  we have a low call volume.  Obviously I cannot keep doing this forever.
 
 I recently had the same problems. Just that wasnt't related to installing any 
 new hardware. You can check out the issue 
 http://bugs.digium.com/view.php?id=10775
 
 Could you provide your OS and glibc version? Also - can you try to disable 
 IAX?
 

The server is running CentOS 5 with all upgrades applied.  I have
glibc-2.5-12 installed.  If I could only test the individual cards for
two days before having to install the Sangoma A104x again and in those
two days I did not have any problems.  

We are not using IAX at the moment so I can disable it from
modules.conf the next time I have to restart Asterisk.

Right now we have two E1 links to the phone company and one E1 link to
a Panasonic PBX.  There are around 40 SIP phones connected to the
server.  Six of those phones are used for a small call center with one
queue.  All E1 links use Unicall.

-- 
Telecomunicaciones Abiertas de México S.A. de C.V.
Carlos Chávez Prats
Director de Tecnología
+52-55-91169161 ext 2001


signature.asc
Description: This is a digitally signed message part
___
--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] Asterisk using 200% CPU and then crashing...

2007-10-17 Thread Philipp Kempgen
Carlos Chavez wrote:
   We have a customer that has Asterisk 1.4.12.1, Zaptel 1.4.5.1,
 Asterisk-Addons 1.4.3. running on a Dell Poweredge 1900 server (Dual
 Core Xeon, 4gb RAM, 500gb Raid 5).  Until a month ago they had two
 TE120P cards and everything was working fine.  Since they needed to add
 a third E1 line we decided to change one of the TE120P cards with a
 TE210P.  After the change we had a couple of crashes (server completely
 locks up).  At first we thought we had a defective card so we changed it
 but kept having the same problem.  We have tried a TE411P, a Sangoma
 A104x (PCI express), an Openvox D210P and D410P and still the server
 crashes about three times a week.  I can see the Asterisk process go
 from around 10% usage to 100%, then 200% and the server hangs after
 that.

The Dell PowerEdge 1600 is listed on
http://www.digium.com/en/docs/misc/compatibility_notes.php
Not sure about the 1900.

However your problem might be completely unrelated to
any Dell specific issues.

Regards,
  Philipp Kempgen

-- 
amooma GmbH - Bachstr. 126 - 56566 Neuwied - http://www.amooma.de
Let's use IT to solve problems and not to create new ones.
  Asterisk? - http://www.das-asterisk-buch.de

Geschäftsführer: Stefan Wintermeyer
Handelsregister: Neuwied B 14998

___
--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] Asterisk using 200% CPU and then crashing...

2007-10-17 Thread Tilghman Lesher
On Wednesday 17 October 2007 13:22:00 Carlos Chavez wrote:
   The server is running CentOS 5 with all upgrades applied.  I have
 glibc-2.5-12 installed.  If I could only test the individual cards for
 two days before having to install the Sangoma A104x again and in those
 two days I did not have any problems.

Your version of glibc screams out as the possible problem.  Try upgrading
to glibc-2.6.

-- 
Tilghman

___
--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