[cisco-voip] FW: Call Admission Control and ISDN Calls

[O'Brien, Neil]

Actually - I've just realised that the location on the phone I was
testing with was wrong.  Even though the CSS for the PSTN was correct,
I'm assuming that because Call Manager saw the phone in one location and
it breaking out of another this is why it was deducting the 64K.

 

Also, when I realised this, I decided to check all the other phones for
that site and one phone (which must have been moved from another
location) had the wrong location/CSS/PSTN.  So every call this phone
made, would go out through the WAN at 64K ( I think).

 

There another question - If  calls between regions are g729, and a phone
from location A is routed out via the WAN through location B, does it go
as g711 or g729.  Testing would suggest that it goes at g711??

 

Thanks,

 

Neil

 

 

 

From: O'Brien, Neil 
Sent: 29 January 2010 20:37
To: cisco-voip at puck.nether.net
Subject: Call Admission Control and ISDN Calls

 

Hi,

 

I'm going to try and explain this as best I can.

 

We have Call Manager 7 cluster in a centralised call processing model
with 9 satellite sites.  All outbound calls from each site breakout
locally to that site, be it via ISDN BRI or PRI.

 

Each route list for each site has a send entry for another site.  In the
event local outbound calls fail, they will go via the WAN and breakout
from the other site.

 

On one site (I'm worried now that it might be on the other sites as
well) I've been troubleshooting for a while now an issue whereby when
only one channel of a BRI is being used, all the rest of the out bound
calls start going out through the other site, even though there is a
second channel on the active BRI and 2 other BRI (4 channels) available.

 

Removing the backup paths from the route list and using a single BRI I
started debugging and came to the conclusion that the second channel
wasn't being utilised but after some more digging I realised that I was
getting "Not Enough Bandwidth" when a second call was attempted.  This
pointed me at Call Admission Control.

 

The site in question only has a 128k link into it, and I have the
location bandwidth set accordingly.  But when I up the bandwidth
allocation or set it to unlimited for that location, calls start going
out on all BRI's  no problem.

 

Now I'm confused - I thought the CAC only tracked calls between
locations, deducting bandwidth accordingly.  But for some reason, it's
deducting 64K (an ISDN channel) from this locations bandwidth
allocation, even though the call is breaking out locally.

 

I'd appreciate anyone's help on this.  I don't know yet if it's just
this site, I'm going to start testing when I send this mail.

 

Thanks in advance,

 

Neil

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/cisco-voip/attachments/20100129/7ed0b508/attachment.html>