cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1469
Views
15
Helpful
10
Replies

Multiple PRI Configuration

jasondrisc
Level 1
Level 1

Hello,

I apologize if this is a dup post but I haven't been unable to find the info I'm looking for exactly.

We currently have one PRI connected to a VWIC-2MFT-T1 configured as h.323. We have a new provider bringing in two PRI's. What I want to do is enable the card on an NM-HDV2-2T1/E1 and configure the controllers as MGCP for the two new PRI's and do some testing before we make the switch. Once the switch is complete I'd remove the h.323.

I have a TAC case open which the engineer suggested NFAS. I don't think this is neccessary with just two PRI's. My thought was to configure the two controllers for the new PRI's and route calls to both, if the first becomes saturated, the next call(s) would go to the second.

Any suggestions/recomendations or suggested documentation would be appreciated.

Thanks,

Jason

1 Accepted Solution

Accepted Solutions

thisisshanky
Level 11
Level 11

Jason,

Dont sweat your time and money behind NFAS unless you have 4 or more PRIs where you dont want to pay X amount of dollars per month per D-channel. NFAS is good, but has its downsides too. Without ordering a primary and backup d-channel and for some reason if both those t1s are down, your whole trunk group could go down. In your case you could definitely do first t1 as primary d channel and second as backup d-channel in the nfas group. NFAS is not supported with MGCP, so you will have to stay at MGCP. Also if you ever want to get out of NFAS and go back to individual D-channels, Telco's charge you hefty $$ to do that ($2k usually).

Go ahead and simply use individual D-channels for these new T1s and use MGCP with Callmanager. Life is a lot easier with MGCP (but also make sure you use a stable IOS, as I have had some nasty experiences with MGCP with recent IOSes).

Here is a good guide.

http://www.cisco.com/en/US/products/sw/voicesw/ps556/products_tech_note09186a00807f8b70.shtml

Once you configure the two controller ports, you can create a route group and add the gateways in any order you want and add route group to a route list. Point your route patterns for PSTN access to this route list. When calls are placed outbound, CM will use this route list and use the first gateway port listed in the route group as primary. When the first gateway port is fully loaded, CM will automatically send subsequent calls through the second gateway port in the route group.

HTH

Sankar

PS: please remember to rate posts!

Sankar Nair
UC Solutions Architect
Pacific Northwest | CDW
CCIE Collaboration #17135 Emeritus

View solution in original post

10 Replies 10

thisisshanky
Level 11
Level 11

Jason,

Dont sweat your time and money behind NFAS unless you have 4 or more PRIs where you dont want to pay X amount of dollars per month per D-channel. NFAS is good, but has its downsides too. Without ordering a primary and backup d-channel and for some reason if both those t1s are down, your whole trunk group could go down. In your case you could definitely do first t1 as primary d channel and second as backup d-channel in the nfas group. NFAS is not supported with MGCP, so you will have to stay at MGCP. Also if you ever want to get out of NFAS and go back to individual D-channels, Telco's charge you hefty $$ to do that ($2k usually).

Go ahead and simply use individual D-channels for these new T1s and use MGCP with Callmanager. Life is a lot easier with MGCP (but also make sure you use a stable IOS, as I have had some nasty experiences with MGCP with recent IOSes).

Here is a good guide.

http://www.cisco.com/en/US/products/sw/voicesw/ps556/products_tech_note09186a00807f8b70.shtml

Once you configure the two controller ports, you can create a route group and add the gateways in any order you want and add route group to a route list. Point your route patterns for PSTN access to this route list. When calls are placed outbound, CM will use this route list and use the first gateway port listed in the route group as primary. When the first gateway port is fully loaded, CM will automatically send subsequent calls through the second gateway port in the route group.

HTH

Sankar

PS: please remember to rate posts!

Sankar Nair
UC Solutions Architect
Pacific Northwest | CDW
CCIE Collaboration #17135 Emeritus

Sankar,

Thank you very much for the information and quick response!! Not only was it informative, it's what I was hoping to hear.

One last question I forgot to include in my post. When I enable the card on the NM-HDV2-2T1/E1, how should I set clocking during my testing phase? Right now I have the following:

network-clock-participate slot 1 **NM-HDV2-2T1/E1**

network-clock-participate wic 0

no network-clock-participate wic 1

no network-clock-participate wic 2

network-clock-select 1 T1 0/0/0 **Current PRI**

My thought was to add the following for the new PRI's/Controllers:

network-clock-select 2 T1 1/0

network-clock-select 3 T1 1/1

Would that be correct?

Jason

Yes add backup clock select commands from your remaining T1s. Thats good.

Sankar Nair
UC Solutions Architect
Pacific Northwest | CDW
CCIE Collaboration #17135 Emeritus

Thanks again!! I'll be sure to rate the post.

Rob Huffman
Hall of Fame
Hall of Fame

Hi Sankar,

Another very informative post! You could write an excellent book with the answers you detail here. 5 points from Calgary :)

Best regards,

Rob

Thanks for the rating and kind words Rob. I wish I had good writing skills, i would have become the next Agatha Christie.

After re-reading through my post, i see i should add an errata to the post..

I meant to say "you will have to stay at H323" instead of MGCP

Sankar Nair
UC Solutions Architect
Pacific Northwest | CDW
CCIE Collaboration #17135 Emeritus

5 from my side !! very informative

Thank you for the rating Neeraj, I feel rich already. :)

Sankar Nair
UC Solutions Architect
Pacific Northwest | CDW
CCIE Collaboration #17135 Emeritus

Sankar,

I think I found where you corrected yourself but just to confirm I'm not misunderstanding. I can have both the h.323 gateway active as well as the two new MGCP gateways running concurrently, correct?

Jason

Yes, that should work as long as you use H323 to handle calls from one t1 port and MGCP to handle calls from another t1 port. MGCP will over ride H323 if you use both protocols to control the same t1 port.

Also if you look at remote sites hanging off a Centralized CM cluster, and if you use MGCP as the protocol, when wan connection is broken, the router has to be configured with fallback commands to revert to H323 protocol to route calls in and out through the local PRI. So you will have to use a mix of mgcp and h323 configs.

Regardless you may have a H323 gateway config in place for your nfas configs and you may have some analog devices on an fxs port that you may want to use MGCP. Either ways you can always mix and match H323 / MGCP/SCCP on the same router as long as you dont use it for the same port.

HTH

Sankar

Sankar Nair
UC Solutions Architect
Pacific Northwest | CDW
CCIE Collaboration #17135 Emeritus
Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: