Trying to troubleshoot an issue with a 7960 phone, Call Manager 4.13 sr1 and the phone load 7.2.2.
The problem seems to happen on conference calls. The IP Phone user will be in a conference call to a PSTN, 2801 on a pri through an MFT-T1, and all of a sudden the other parties on the conference cannot hear the IP Phone users. However the IP Phone user can still hear the conference.
They have to hang up and reinitiate the call.
I just upgraded to the 7.2.3 firmware and will see if that makes a difference tomorrow.
I tried swapping phones.
I tried swapping the ethernet cable in the wall, (two separate cables in the wall, just went from A to B.)
It's connected to a 3560 POE switch in a voice vlan. I have auto qos voip cisco-phone on the switch.
I also pulled the Publisher Conference Bridge from the Media Resource Group. Try to narrow down which conference resource might be causing the problem.
Check wait for far end H.245 terminal capability set in the H.323 gateway configuration page.
This field applies only to H.323 devices.
This check box specifies that Cisco CallManager waits to receive the far-end H.245 Terminal Capability Set before it sends its H.245 Terminal Capability Set. By default, the system checks this check box. To specify that Cisco CallManager should initiate capabilities exchange, uncheck this check box.
I am having the exact same problem, though not just on conference calls. I would love to find a solution as its driving us crazy! Our phones were connected through a simple router to a DSL connection, and then through a cable connection. All to no avail. Same problem.
We think our problem relates to a bug with Third party Call Manager servers and MGCP/Skinny. I'm digging up the bug ID right now. The Adaptec driver seems to be causing MGCP and some skinny devices to re-register every 5 hours. I've been assured from our Cisco friends that this issue is top of the list and they hope to have a new driver from Adaptec shortly.
Even though there is no workaround listed below, they had us extend the skinny registration for the phones to 60 seconds. Nothing exists like that for MGCP, so we converted to h.323.
Can't say it solved all the issue though as I have some complaints about voice quality going across the PRI right now. Hoping that is a different problem though.
If you're not using Call Manager third party servers, then it sounds like you may have another issue. Are you running different CODECS across the link, G.729 and G.711. Your remote site is g.729 and the main site is g.711. Easy way to narrow it down is to put the remote site on G.711 so it can use the Software Conferencing resources of Call Manager. Just remember this bumps up your bandwidth requirements of your call. Usually when I hear about this problem it's transcoding and/or speed/duplex issues.
I know I've seen several models of cable modems only negotiate half duplex even though Full Duplex may be available. This could lead to one-way audio issues. I would verify on all ends of the connection, 100/FULL is hard coded where possible. If you can't do 100/Full, then 10/Full. If you can't do this, find some more equipment that will give you that type of control.
MGCP GW unregisters from primary CCM, fallback to backup CCM and switch back to Primary in 60 seconds, this event happens very 5 hours.
At the same time of MGCP GW bouncing, Page faults/sec of ccm process and %Disktime counters in performance monitor spikes as well
SDL/CCM trace files have 33-37 seconds gap/blank when the MGCP GW bouncing, and nothing is written down in the trace files.
This problem only observed on Callmanager installed on MCS-7825-H1 or HP DL320 G3 server. With this server model, %Disktime spikes every 5 hours even though without Callmanager installed. And IP phones are not reported having this issue but Unity vo
ice mail ports, MoH, MTP devices are reported by customer to bouncing very 5 hours as well
Further Problem Description:
The MGCP GW unregister from CCM because Callmanager fail to respond to the MGCP NTFY keepalive message.
The spike of Page faults/sec and %Disktime can be observed in performance counters log every 5 hours even though the performance counter indicate there are enough available memory.
The %Disktime spike even though there is no CCM application installed.
Please rate this release note:
If you gave this release note a low rating, please tell us why.
Select One The information is not clearThe information is incompleteThe information is not related to my issue
Are you getting this error “Installer User Interface Mode Not Supported. The installer cannot run in this UI mode. To specify the interface mode, use the -i command-line option, followed by the UI mode identifier. The value UI mode identifiers...
The below trick might come handy when you have to add a new node to a cluster but you don't have or is unsure of the security password for the publisher. This procedure has been around for ages.
1) Login into the CLI of the Publisher.