<div>There other ways to validate H323 fallback. Make sure if running H323 on the MGCP box as a fallback to ensure the MGCP to add the session target and destination pattern to the dial peers. </div> <div> </div> <div>! a typical MGCP DP:</div> <div>dial-peer voice 1 pots<BR> application mgcpapp<BR> port 1/0:0</div> <div> </div> <div>! a DP with h323 support</div> <div>ccm-manager switchback immediate<BR>ccm-manager fallback-mgcp <BR>ccm-manager mgcp<BR></div> <div>dial-peer voice 1 pots<BR> application mgcpapp<BR> destination-pattern 91..........<BR> port 1/0:0</div> <div>!</div> <div>dial-peer voice 555 voip <BR> application mgcpapp<BR> destination pattern 555...<BR> session-target ipv4:172.20.21.8<BR> codec g711ulaw</div> <div><BR> </div> <div>To test fallback you can delete the MGCP GW in the CCM and attempt to process the calls. <BR></div> <div>Peter Casanave</div> <div><BR><B><I>Ted Nugent
<tednugent69@yahoo.com></I></B> wrote:</div> <BLOCKQUOTE class=replbq style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid">I'm not 100% certain on this but I beleive that the<BR>mechanism for initiating a fallback to h323 is a loss<BR>of MGCP keepalives to callmanager (TCP 2428 I think?).<BR>Which isn't likely to occur if CM is up and running.<BR>Unless of course as Ryan mentioned MGCP went<BR>completely belly up. However in most MGCP bugs that<BR>I've come across the gateway is still (at least<BR>partially) registered with CM and its doubtful that<BR>this would cause a fallback to H323.<BR>As you stated I guess it would really depends on the<BR>severity of the MGCP failure for it to be worth it.<BR><BR><BR>--- Jonathan Charles <JONVOIP@GMAIL.COM>wrote:<BR><BR>> I am not seriously looking to DO anything... I am<BR>> proposing an idea<BR>> and seeing if it is viable.<BR>> <BR>> The goal is maximum redundancy and to that end,
I<BR>> was thinking running<BR>> MGCP and H.323 on the same gateway. If MGCP fails,<BR>> go with H.323<BR>> (while MGCP is active, all the H.323 commands are<BR>> disabled anyway).<BR>> <BR>> To implement this on the CCM side, I was going to<BR>> put the gateways in<BR>> the system twice, once as MGCP, once as H.323. Then<BR>> create two<BR>> route-groups, one for each and then list the MGCP<BR>> first int he RL.<BR>> <BR>> The theory seems sound, but will MGCP fail enough<BR>> for it to go back<BR>> into H.323 mode? Obviously, if MGCP is down from the<BR>> CCM side, CCM<BR>> will work its way down the route-list to the H.323<BR>> entry and try that.<BR>> But inbound calls may not be aware that MGCP has<BR>> failed enough and<BR>> won't try H.323 (as a guess... I am not sure if this<BR>> is going to<BR>> work).<BR>> <BR>> <BR>> Jonathan<BR>> <BR>> On 7/17/06, Ted Nugent
<TEDNUGENT69@YAHOO.COM><BR>> wrote:<BR>> > I guess I did misunderstand part of the question<BR>> > judging from the other responses... Can you just<BR>> > clarify... you're looking to run both MGCP and<BR>> H323 on<BR>> > the same gateways? Both Protocols on the same<BR>> gateway<BR>> > on a seperate trunks? Or are you just looking at<BR>> using<BR>> > an MGCP gateway as primary and a second H323 as a<BR>> > backup?<BR>> ><BR>> ><BR>> > --- Jonathan Charles <JONVOIP@GMAIL.COM>wrote:<BR>> ><BR>> > > So, I want to use MGCP as my primary gateway<BR>> > > protocol and H.323 as a<BR>> > > backup (no SRST, no remote sites).<BR>> > ><BR>> > > The plan is to add the gateways to CCM as H.323<BR>> AND<BR>> > > MGCP and have two<BR>> > > route-groups, one for MGCP and one for H.323 and<BR>> > > then combine them<BR>> > >
into a route-list with MGCP first.<BR>> > ><BR>> > > The problem is that I want to strip the 9 for<BR>> MGCP<BR>> > > (on the route-group<BR>> > > or device) and not do so for H.323.<BR>> > ><BR>> > > First off, as a design is this retarded? Second,<BR>> > > should I strip the 9<BR>> > > on the gateway or change the H.323 dial-peers to<BR>> not<BR>> > > include the 9 and<BR>> > > strip it on the route list?<BR>> > ><BR>> > ><BR>> > ><BR>> > ><BR>> > > Jonathan<BR>> > > _______________________________________________<BR>> > > cisco-voip mailing list<BR>> > > cisco-voip@puck.nether.net<BR>> > ><BR>> https://puck.nether.net/mailman/listinfo/cisco-voip<BR>> > ><BR>> ><BR>> ><BR>> > __________________________________________________<BR>> > Do You Yahoo!?<BR>> >
Tired of spam? Yahoo! Mail has the best spam<BR>> protection around<BR>> > http://mail.yahoo.com<BR>> ><BR>> _______________________________________________<BR>> cisco-voip mailing list<BR>> cisco-voip@puck.nether.net<BR>> https://puck.nether.net/mailman/listinfo/cisco-voip<BR>> <BR><BR><BR>__________________________________________________<BR>Do You Yahoo!?<BR>Tired of spam? Yahoo! Mail has the best spam protection around <BR>http://mail.yahoo.com <BR>_______________________________________________<BR>cisco-voip mailing list<BR>cisco-voip@puck.nether.net<BR>https://puck.nether.net/mailman/listinfo/cisco-voip<BR></BLOCKQUOTE><BR><p> 
<hr size=1>Love cheap thrills? Enjoy PC-to-Phone <a href="http://us.rd.yahoo.com/mail_us/taglines/postman9/*http://us.rd.yahoo.com/evt=39666/*http://messenger.yahoo.com/"> calls to 30+ countries</a> for just 2¢/min with Yahoo! Messenger with Voice.