Can the T.106 timer be changed to prevent broken calls?

[573dawn]

Hi All, I am seeing the MSDSE Media exception error on many dropped calls. I understand that this is an H450/T.106 timer problem because I always get the "local timer expired" with these. Is there a way to increase the time?

I have verified that both ends of the IP line are set to H450. I also have them set for G711 ULAW 64k, should I consider using auto-select instead?

Thanks, Dawn

 

[ProviderReborn]

I have had nothing but negative results with the autoselect of a VoIP line. I would advice not using this option Dawn.

Regards
Lee

 

[573dawn]

TheProvider, thanks I won't change it then

 

[mattKnight]

Were the PtP links installed especially for this project or were they in place beforehand?

Have the cisco routers been configured with VOIP in mind? Do they have any sort of firewall, Access lists or similar on them? At the moment, I'd look closer at the data links.

Have the cisco's had stuff like CDP turned off?

Take Care

Matt
If at first you don't succeed, skydiving is not for you.

 

[intrigrant]

This timer cannot be changed.
i reviewed this problem and i think it is related to a network problem. The messages are send from the master gatekeeper through IP port 1719 to the slave gatekeeper which should reply on port 1720.
The master gatekeeper controls the state of the media stream so either the messages from port 1719 do not reach the slave gatekeeper or the responds on port 1720 do not reach the master gatekeeper.
If the problem is not in all the calls then the VCM card is defective in the problem site, let it be replaced.
The VCM channels are used either sequential or rotary i don't know. In hardware there are four or five VCM channels on one chip depending of the type of VCM card.
If the first chip is ok but the second is not and the channels are used sequentially then you will have this problem on the fifth or sixth call from the problem site.
If the channels are used rotary then the problem happens once every x calls, it should show some kind of pattern.

 

[573dawn]

mattKnight, the PTPs were put in place before the phone system, we have a company-wide software program that all locations use. They are very underutilized.

I have configured a policy-map on the routers between the main and trouble site, but I'm not sure it is working properly, not all of the allegedly configured traffic shows up if you look at the stats. I need to correct this and just haven't gotten to it. The other sites don't have the policy-map in place because their IOS won't support it and I'm unsure how else to do it.

There are no firewalls to worry about, and could you please tell me what CDP is? I know I should know and I am just drawing a blank

 

[573dawn]

intrigrant, that is very enlightening. I will see if I can determine the rotation. It certainly isn't every call, or I would have to have bodyguards <g>

What might be the best way to differentiate between a network issue and a VCM issue? We have 8 VCMs at the trouble site, 4 each at th other 2 remote sites, and 16 at our main site.

 

[mattKnight]

There are no firewalls to worry about, and could you please tell me what CDP is? I know I should know and I am just drawing a blank

I didn't expect there to be a firewall on a PtP link (but you never know)

CDP - Cisco Discovery protocol. This allows Cisco units to communicate with each other and share information about their configuration. It is generally recommended that CDP be turned off (especially for VOIP links). Also I'd look at whether RIP etc are in use (I'd expect not, but you never know)

I doubt that RIP or CDP are causing your issues though..

Take Care

Matt
If at first you don't succeed, skydiving is not for you.

 

[573dawn]

mattKnight, thank you and CDP is not running, nor have I implemented any routing protocols, static routing only.