|Summary:||ASTERISK-07910: No bridge call|
|Reporter:||Alvaro Ivan Parres Peredo (arabe)||Labels:|
|Date Opened:||2006-10-10 12:13:49||Date Closed:||2006-12-18 09:34:48.000-0600|
|Environment:||Attachments:||( 0) full|
1) 6 Queues
2) 10 Agentes all in the 6 Queues.
Some times when all the calls start been queue and all the agents are No in Use but any call is assign to any agent.
This problems presents in more stress moments.
If we tray to do a show queues the CLI never answerd... but continue answereng the show channels command.
|Comments:||By: Alvaro Ivan Parres Peredo (arabe) 2006-10-10 13:28:03|
We have see that this is like a
multiple exclusion problem.
If we have a call in 6 queues, all queues with same weight, and one free agent in all queues we have this situation..
As if all queues want to take the agent.
By: Serge Vecher (serge-v) 2006-10-13 13:21:29
can you please enable debug output in logger.conf for console, do set debug 4 and set verbose 4 in CLI and produce some logs to look at?
By: Alvaro Ivan Parres Peredo (arabe) 2006-10-17 11:30:02
O.k let me mount the same scenario into a demo, because the system is a production system.
By: Alvaro Ivan Parres Peredo (arabe) 2006-10-17 12:01:22
Today we have the same situation with version 1.2.5 we are checking for any other cause (static...) because after 2 days that the systems was working well we
install a new UPS and the system fail again.
By: Anthony LaMantia (alamantia) 2006-10-17 16:33:39
you should really see if this is reproducable agins tthe latest svn branch of 1.2 , and when you try to reproduce the error please include some of the logging output.
By: Alvaro Ivan Parres Peredo (arabe) 2006-10-17 19:33:44
I have upload a FULL log
from the last Call bridge to a Agent
and to the kill we had to do to the asterisk for restart
becouse we can?t even send the command restart o something
By: Serge Vecher (serge-v) 2006-10-23 13:24:03
what's the status with 1.2.13?
By: Alvaro Ivan Parres Peredo (arabe) 2006-10-24 09:53:51
we going to try to mount a demo scenario with 1.2.13
It little hard to reproduce the situation.
By: Alvaro Ivan Parres Peredo (arabe) 2006-10-25 13:26:47
We have installed the version 1.2.13 with another server:
18 Agents into 2 queues. All Agents in both queues.
And today we have 2 times the same situation.
Is import to note that this server have only 512 MB of RAM
We are upgrading today to 2 GB of RAM
The original client, is also with version 1.2.13 and have not
present any trouble in the day.
By: Alvaro Ivan Parres Peredo (arabe) 2006-10-25 13:43:33
Is import to say that when we have different Agents
for each Queue...
This never happend.
By: Serge Vecher (serge-v) 2006-10-26 10:07:52
I see that you are using custom patches, like chan_unicall on your system -- we can't help much in scenarios where non-stock Asterisk distribution is used.
By: Alvaro Ivan Parres Peredo (arabe) 2006-10-26 13:50:23
Yes we are using chan_unicall.
Let me try to mount the escenario with all iax or sip channels
with no chan_unicall patch.
By: Alvaro Ivan Parres Peredo (arabe) 2006-11-08 17:41:37.000-0600
We have end 15 days of test with our 2 clients
with problem with the next results:
Client A (7 queues, 10 agents, max 30 incomming calls):
Review the electrical Instalation
found problems on the ground.
After fixing this all get working fine.
Client B (2 queues, 30 agentes, max 60 incomming calls):
We increment the Memory to 1.5 GB
And the problems continue but each 7 or 10 days.
By: Serge Vecher (serge-v) 2006-11-14 15:01:59.000-0600
have you removed chan_unicall?
By: Alvaro Ivan Parres Peredo (arabe) 2006-12-16 12:57:55.000-0600
Today we are having one month with out downs, we have made 2 acctions:
In place A: We make a verification of the electrical installation
and detect that it was wrong (No ground). We repair this and
all the problems get solve.
In place B: We put more RAM memory from 512 to 2 GB and the service
So i think we can close this bug.
By: Serge Vecher (serge-v) 2006-12-18 09:34:47.000-0600
not a bug in Asterisk