Re: phantom Hang-ups

mymite060708 · ‎01-26-2007

Very randomly a call will ring on an agents for phone 1 ring then stop and hang up

The call will show up as a missed call on the agent?s phone

The agent automatically gets put into ?ready state? (in normal operation if they miss a call they are put into work state

The caller says the phone just hang up

I have been through all the logs and can?t see any issues

Any know bugs?

tapk5 · ‎01-29-2007

I got the same problem in one of after hour queue. 1 ring then hang up, customer provided some examples, which in historical report, are labeled as 'abandoned', but it is hardly believed so many calls are abandoned like that, roughly 20% calls are affeted.

As I can see, IPCC doesn't its' job, route calls to IP phone. I would suggest spending some time in CCM traces for these calls.

Keep me post if you got something.

Wei

william.george · ‎01-30-2007

Similar Issues here.

symptoms included:

1) agent phone rings only once, then no body there

2) when issue occurs, it happend 4 - 6 times at that phone

3) intermittent in nature

We've had this problem and were able to trace it down to a delay with a "call forward busy" action on a DN which forwards to our ICD trigger if the human desk can't take the call. We were further able to isolate this to a 2821 ISR IOS gateway, which when taken out of service caused the problem to go away. We subsequently provisioned a PRI onto this gateway and allowed only calls into a test DID to pass through the gateway, hit a test DN, CFWD Busy to our ICD trigger, and were able to duplicatte the issue and confirm that it only occurs on calls thru the IOS gateway. Approx. 1 - 8 calls fails with this issue in our test ing. Calls to the same ICD trigger through another gateway (6608 card) do not have issues.

Cisco verified in CCM/JTAPI/IPCC traces that they saw a 7 second delay in the transfer operation on these calls. Call through the same gateway had exhibited occasional delay in when users pressed the "transfer" softkey on IP phones as well, which further confirms that the issues was not script or IPCC related, but a gateway/mgcp/ccm issue.

Cause of gateway issues is not known at this time, however we are curious to get it figured out correct the configurtions issues which are to blame.

william.george · ‎01-30-2007

I'm happy to chat about this at bgeorge@lifetimefitness.com

johnnylingo · ‎01-30-2007

There seems to be a lot of weird problems with 2800 gateways using MGCP going on out there.

Was this gateway MGCP, and what IOS version?

william.george · ‎01-30-2007

Yes, it's a 2821 Chassis and four (4) VWIC version 2 T1 interfaces with MGCP controlled PRI's on them. Issue occurs regardless of how many PRI's are enabled.

DSP's = one (1) 48, one (1) 32, and one (1) 16.

From an overall performance / capacity point of view the thing works great even with a ton of calls on it, it's just this one issue which I'm sure we'll figure out here relatively soon.

One note that makes me think it's a software bug or similar, is that this is the only one of over 60 other MGCP controlled PRI gateways on the 2821 platform that is running a 12.4 release with similar configs. All others are running 12.3T variations, and they are working OK. I've not downgraded and tested yet cause I want to give TAC a chance to analyze the traces and .cap files to see if they can figure it out.

*** Version ***

Cisco IOS Software, 2800 Software (C2800NM-IPVOICEK9-M), Version 12.4(9)T, RELEASE SOFTWARE (fc1)

Technical Support: http://www.cisco.com/techsupport

Compiled Fri 16-Jun-06 22:22 by prod_rel_team

ROM: System Bootstrap, Version 12.3(8r)T7, RELEASE SOFTWARE (fc1)

DOUG DAVIDSON · ‎01-31-2007

I am experiencing the same type of issue, but it has happened on non-ACD calls and I am not running a 2811 as a gateway. I have a Cat6503 with the CMM blade, which has a 6 port T1 card that my PRIs are attached to.

Cisco IOS Software, Cat6K-lc Software (wscmm-I6S-M), Version 12.3(14)T6, RELEASE

SOFTWARE (fc2)

Technical Support: http://www.cisco.com/techsupport

Compiled Thu 05-Jan-06 11:49 by dchih

johnnylingo · ‎02-01-2007

Interesting...we've got 4 2821s running 12.3(11)T10, mix of VWIC-2MFT-T1, VWIC-1MFT-T1, and VWIC-4FXO. I have not observed any problems. CCM version is 4.1(3)SR1

If anyone can match this to a bug, please post a followup. I had planned to upgrade to 12.4 mainline but will certainly hold off for now.

tapk5 · ‎02-01-2007

It is hard to say whether this is caused by specifi model of voice gateway or IOS verison. The troubled number is in 3825 with IOS 12.4(11), while I never hear problem for the numbers on 2811/2821 so far.

Wei

tapk5 · ‎02-12-2007

We captured syslog from Voice gateway, and these calls seem disconnected from Telco side, with "normal call clearing" event, just like, callers dropped calls.

We opened Cisco TAC case, and Cisco asked to turn "disable alerting progress indicator" service paramter to true in CCM, it does help, the number of dropped calls decreases significantly. Cisco suspects one or more switches in PSTN network reject this signaling.

We are still talking to Telco to see if this is the case.

Wei

johnnylingo · ‎02-15-2007

Thanks for the update!

loevans · ‎02-16-2007

Wei,

Can you post the case number? I'd like to have one of my Cisco SE buddies look into this. I'm having the same problem at two different call centers in two different cities with three different physical router models as MGCP gateways.

Thanks!

tapk5 · ‎02-16-2007

it is 605294039.

william.george · ‎03-07-2007

I've downgraded my code on this 2821

From: (C2800NM-IPVOICEK9-M), Version 12.4(9)T, RELEASE SOFTWARE (fc1)

To:(C2800NM-IPVOICE-M), Version 12.3(14)T7, RELEASE SOFTWARE (fc2)

Problem seems to have gone away. I have not put it into production yet, but on the 12.4 code, I could reproduce the issue pretty easily while on the 12.3 code I've not been able to make it happen in my using the same tests. I'll put this back in production next week and post a followup after a week of live trial .

johnnylingo · ‎04-05-2007

I'd love to hear the details on this. I had the exact same problem today, and downgrading from 12.3(11)T11 to 12.3(11)T10 cleared everything up. Again, this is a 2821 w/ VWIC-2MFT-T1 on CCM 4.1(3)sr4b.

This seems to be a highly significant bug, but I still have not seen anybody post an ID.