[cisco-voip] SIP to PRI transcoding

Blake Pfankuch - Mailing List blake.mailinglist at pfankuch.me
Tue May 20 20:02:49 EDT 2014


Great question.  In this case the 3825's are not actually attached to Call Manager as these 2 devices exist in a network that cannot connect to our UCM environment.  The Fax cluster (used accurately) is layer 2 connected to these 3825's which then talk out to the SIP carrier via switched private line.  As this falls into the Production environment, we are trying to keep it from even having to talk to UCM.

Thanks,
Blake

-----Original Message-----
From: Peter Slow [mailto:peter.slow at gmail.com] 
Sent: Tuesday, May 20, 2014 5:44 PM
To: Blake Pfankuch - Mailing List
Cc: Cisco VOIP (cisco-voip at puck.nether.net)
Subject: Re: [cisco-voip] SIP to PRI transcoding

Hi Blake,
     Why are you using the CUBEs for terminating the T.38 in this case? they shoudl sort of just be acting as TDM gateways for the purpose of terminating the t.38. How many clusters do you have and where are the fax server(s) and planned TDM gateways located?

Fax server --> CUCM ---> PRI theoretically requires no transcoding.
did i miss something?
Actually, even fax server -> cucm-> cube -> gateway shouldnt require it.

...trying to pass your t.38 through an MTP of any sort is going to make your life difficult. what's mandating that?

i also don't entirely disagree with James regarding getting a new carrier, but you'll still have better luck with a local gateway.

-Peter

On Mon, May 19, 2014 at 6:19 PM, Blake Pfankuch - Mailing List <blake.mailinglist at pfankuch.me> wrote:
> I am working with a faxing system that is giving me more nightmares 
> than I can even count…  I am working with a software system that sends 
> FoIP using
> T.38 and V.17 faxing.  It does not support G711 fallback, and that has 
> been giving us fits like you would not believe our upstream carrier.  
> The specific issue lies in paths which do not support T.38 and are 
> trying to do
> G711 fallback.  I have a pair of 3825 router configured as UBE’s in an 
> HSRP configuration.
>
>
>
> I am looking to mitigate some of these issues by implementing a couple 
> of TDM PRI’s (one to each device).  These PRI’s will be configured in 
> an active/passive failover group from the carrier side uplinked into 
> the existing 3825 UBE’s.
>
>
>
> My question is on the conversion as I have never done this before.  I 
> know this will pull a DSP (or a couple) for each transcoding session, 
> so I have 8 PVDM 64’s I am going to throw in these routers.  I still 
> need to be able to support SIP in case of a PRI failure as both PRI’s 
> come in on the same channelized DS3.
>
>
>
> Has anyone had to do this before?  Specifically with T.38.  Had good 
> luck with it or should I look towards the solution I planned for next 
> year which was a pair of ASR1001 routers to act as SBC’s and handle these functions.
>
>
>
> Thanks in Advance,
>
> Blake
>
>
> _______________________________________________
> cisco-voip mailing list
> cisco-voip at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>



More information about the cisco-voip mailing list