Genesys CTI User Forum
Genesys CTI User Forum => Genesys CTI Technical Discussion => Topic started by: lekie on October 14, 2009, 04:20:48 AM
-
Dear All,
Have someone use to face the problem about that numeric that is show on ccpulse+
it is stuck in callwaiting for a long time?and it will be gone by itsalf ,it was not get abandon call also.
And another question is here are multi-site,I got the problem about load balancing call.
The call of customer go to agent who is logged in on site A more than site B
while site B still has agent was waiting for next call longer than site A.
It seem like a site A can recieved call more than site B.
How can I check the log or have any stat release to upgrade this bug?
Now i am using stat server version 7.6.100.20, usr version 7.5.002.02
Thank you in advance.
Best regard,
Lekie
-
hi lekie i faced this issue too, try 2 open a SR to trobleshot but TS ask alway for a connid so as u understand i can not provide it.
However i read on manual that its posible to create a script to clear this stucked calls.
But i did not test it.
Hope this help u.
-
Could you provide me how to create a script to clear or what is the document that I can make it by myself
Thanks you in advance.
Cheers,
Lekie
-
Under TServer document check the Stuck Calls option, be carefull with it as it will consider all calls without events within the range that you define as ghost, including the valid ones like long conversations and will "kill" them from TServer and will no longer report to StatServer or any other but call will be still active at PBX side
-
Lekie,
There is description of T-Server call clear script in "Framework 7.6 Management Layer User's Guide". It describes how it works and what is needed to run it. However, this script is able to clear stuck calls on T-Server level only. It's possible you're facing stuck call on StatServer side only. In that case the only solution is restart of StatServer.
R.
-
Thanks for all information
Cheers,
Lekie
-
To all:
as far as I recall, stuck call option does not apply to Virtual Queue (please correct me if I am wrong), therefore, it is of little use in most cases where EventAbandoned or EventDistributed has not been delivered to StatServer.
I remember discussing this here a very long time ago, but what we do is force verbose log in STATSERVER for about a minute, and then go through the log and find the report of calls in queues and time entered. Based on that report in the log, you can then get ConnID and then use TestPhone to issue EventAbandoned with that ConnID and ThisDN=Queue in question. Works like a charm every time.
If there is a better way to do it for VQs, I definitely woudl love to know it. The problem we have experienced was really profound in 7.2.xxxx versions but we also had it in 7.5xxx and above as well.
-
I had a similar problem a while back, I don't really know if this is your case ... For me the problem was solved after setting off TAT(Trunk Anti-Tromboning) on the RCAP param in PABX.
My exact version of T-Server is 7.6.008.01 and the PBX is “T-Server for Nortel Communication Server 1000 with SCCS/MLS”.
Best regards
-
Hi All,
In order to avoid stuck calls on VQ's you should configure "call_kpl_time" option on URS (introduced in 7.5 version).
Also there is option "check-stuck-calls" on StatServer, so you don't have to restart SS to clear stuck calls.
Regards,
-
thanks you so much Koldun :)