Jan The SRV routing policy can be used to send LRQ's after the ENUM policy. This requires a _H323ls._udp. SRV records to locate the RAS port of the gatekeeper. Interesting it seems the h323 url (according to H.323 Annex O) only return a host address (ie call signalling), you need an SRV record to resolve the RAS port to send an LRQ. Anyway for those that are interested in testing the ENUM policy you can get a free ENUM at e164.org and you point the record to your gatekeeper h323:user@address:1721 (don't forget to allow unregistered callers in the ini) In general restricting ENUM, SRV and the new RDS (SRV records hosted in an assigned domain) to only ARQ and Setup should be OK as these are really local gatekeeper functions however I agree with your assertions with respect to the SQL policy and LRQ. Simon -----Original Message----- From: openh323gk-users-bounces@xxxxxxxxxxxxxxxxxxxxx [mailto:openh323gk-users-bounces@xxxxxxxxxxxxxxxxxxxxx]On Behalf Of Jan Willamowius Sent: Wednesday, 12 September 2007 4:38 AM To: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx Subject: Re: Problem with call setup using ENUM on GNUGK226 Hi, I think the problem is a bit deeper than that: Since the policies are a chain, all policies can end the routing chain if they have found a final destination. So it can't loop inside one GnuGk. If misconfigured, calls between multiple gatekeepers can loop, but thats probably true for all types of policies. I think the problem is how the ENUM policy is implemented as an AliasPolicy and thus can only rewrites aliases. OnARQ and OnSetup (ENUM is also active OnSetup), just rewriting is ok, because one of the following policies (eg. explicit or internal) can do the final routing - but there is no equivalent policy that will send out the LCF after ENUM is done. I ran into something similar when implementing the SQL policy OnLRQ. OnARQ and OnSetup0p just rewriting the alias is enough, but I think OnLRQ the policy has to add a route to the destination itself and not rely on a following policy. Regards, Jan Simon Horne wrote: > > Jan > > The ENUM policy never had LRQ support, it has always been only call > signalling. The ENUM policy was disabled from incoming LRQ queries due to > the possibility of infinite looping when using it with SRV so it is mow > restricted to only doing ENUM lookups for locally registered users (via > ARQ). > > The H323 URI, H323:user@host[:port] which is returned by the ENUM lookup is > by default (according to H.323 Annex O) only a host address imply call > signalling (on the specified TCP port, if omitted the default is 1720). > > >From H323 Annex O > O.9 Resolving an H.323 URL to IP Address using DNS > The host part of the H.323 URL can specify one of the following: > - The IP numeric address of an H.323 entity. > - The DNS name of a host which is an H.323 entity. (ie reach by Call > Signalling) > - The remote DNS domain containing H.323 SRV RRs. (ie handle by SRV policy > ie _h323ls._udp record (LRQ)) > > Simon > > > -----Original Message----- > From: openh323gk-users-bounces@xxxxxxxxxxxxxxxxxxxxx > [mailto:openh323gk-users-bounces@xxxxxxxxxxxxxxxxxxxxx]On Behalf Of Jan > Willamowius > Sent: Tuesday, 11 September 2007 6:11 AM > To: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx > Subject: Re: Problem with call setup using ENUM on > GNUGK226 > > > Hi Simon, > > can you please explain a bit more why GnuGk shouldn't rewrite the called > alias in an LRQ via ENUM ? > > I see that you deactivated ENUM for LRQs earlier this year, but I'm not > sure why this is neccessary. > > Regards, > Jan > > > Simon Horne wrote: > > > > ok firstly ENUM cannot resolve to an LS address (LRQ) it can only resolve > to > > CS address (Call Signalling of a gatekeeper or of an actual softphone) so > > GnuGk ENUM policy resolves +3912345678 to the call signalling address of > > gnugk2.it so the ARQ/ACF returns 5678@xxxxxxxxx telling softphone1 to go > > call gnugk2.it to find 5678. Now your softphone is not correctly rewriting > > 3912345678 with 5678 when it calls to gnugk2.it so the call gets rejected. > > > > There are several ways to fix it. > > 1. Get a softphone that does correctly rewrite the address like PacPhone > > www.pacphone.com which by default supports NRENUM.net lookups and is also > > available in italian. > > > > 2. create a E.164 rewrite policy in GnuGk2 > > http://www.gnugk.org/gnugk-manual-6.html#ss6.2 > > like this > > [RasSrv::RewriteE164] > > 3912345=5 > > which means if you receive a full E.164 number rewrite it to a local > number. > > ie get 3912345678 rewrite to 5678. > > > > 3. Look at adding the SRV policy (needs pwlib/OpenH323 CVS) to follow the > > ENUM policy. You will need to add H323 LS SRV record to the gnugk2.it DNS. > > See http://nrenum.net/how-to_naptr_srv.html#h323-srv on how to do that. > This > > will after the ENUM lookup send an LRQ to gnugk2.it which should then > return > > the call signalling address of softphone2 directly and avoid the problem > > with the rewrite with softphone1 and GnuGk2. > > > > I hope that helps. > > > > Simon > > > > > > > > > > -----Original Message----- > > From: openh323gk-users-bounces@xxxxxxxxxxxxxxxxxxxxx > > [mailto:openh323gk-users-bounces@xxxxxxxxxxxxxxxxxxxxx]On Behalf Of > > Andrea De Vita > > Sent: Monday, 10 September 2007 11:36 PM > > To: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx > > Cc: tf-ecs@xxxxxxxxxx > > Subject: Problem with call setup using ENUM on > > GNUGK226 > > > > > > Dear All, > > > > I have a problem with Gnugk 2.2.6 call setup using ENUM. > > > > The scenario is this: > > > > SoftPhone1 -----> GNUGK1 ------> GNUGK2 ------> SoftPhone2 > > > > xxxxxx gnugk1.it gnugk2.it 5678 > > > > > > 1) SoftPhone1 call +3912345678 (ARQ), > > 2) GNUGK1 resolve by ENUM lookup (nrenum.net) 3912345678 in 5678@xxxxxxxxx > > and contact via LRQ gnugk2.it > > 3) GNUGK2 reply with LCF > > 4) GNUGK1 send ACF to SoftPhone1 (with h323-ID:5678@xxxxxxxxx) > > 5) SoftPhone1 send call setup to GNUGK1 (with dialedDigit 3912345678) > > > > > > Problem: Now GNUGK1 send the call setup to GNUGK2 but with the original > > q.931 dialed digit 3912345678 rather than 5678 as resolved by ENUM. > > Then GNUGK2 reply with release complete, canceling the call, because > > 3912345678 is not known by GNUGK2. > > > > Did you face a similar problem? In case you have a solution please let > > me know. > > > > Thanks > > Andrea > > > > > > ------------------------------------------------------------------------- > > This SF.net email is sponsored by: Microsoft > > Defy all challenges. Microsoft(R) Visual Studio 2005. > > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > _______________________________________________________ > > > > Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx > > Archive: > > http://sourceforge.net/mailarchive/forum.php?forum_name=openh323gk-users > > Unsubscribe: http://lists.sourceforge.net/lists/listinfo/openh323gk-users > > Homepage: http://www.gnugk.org/ > > > > > > -- > > No virus found in this incoming message. > > Checked by AVG Free Edition. > > Version: 7.5.485 / Virus Database: 269.13.8/993 - Release Date: 6/09/2007 > > 3:18 PM > > > > > > > > ------------------------------------------------------------------------- > > This SF.net email is sponsored by: Microsoft > > Defy all challenges. Microsoft(R) Visual Studio 2005. > > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > > _______________________________________________________ > > > > Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx > > Archive: > http://sourceforge.net/mailarchive/forum.php?forum_name=openh323gk-users > > Unsubscribe: http://lists.sourceforge.net/lists/listinfo/openh323gk-users > > Homepage: http://www.gnugk.org/ > > > > > -- > Jan Willamowius, jan@xxxxxxxxxxxxxx, http://www.gnugk.org/ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________________ > > Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx > Archive: > http://sourceforge.net/mailarchive/forum.php?forum_name=openh323gk-users > Unsubscribe: http://lists.sourceforge.net/lists/listinfo/openh323gk-users > Homepage: http://www.gnugk.org/ > > > -- > No virus found in this incoming message. > Checked by AVG Free Edition. > Version: 7.5.485 / Virus Database: 269.13.8/993 - Release Date: 6/09/2007 > 3:18 PM > > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________________ > > Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx > Archive: http://sourceforge.net/mailarchive/forum.php?forum_name=openh323gk-users > Unsubscribe: http://lists.sourceforge.net/lists/listinfo/openh323gk-users > Homepage: http://www.gnugk.org/ > -- Jan Willamowius, jan@xxxxxxxxxxxxxx, http://www.gnugk.org/ ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________________ Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx Archive: http://sourceforge.net/mailarchive/forum.php?forum_name=openh323gk-users Unsubscribe: http://lists.sourceforge.net/lists/listinfo/openh323gk-users Homepage: http://www.gnugk.org/ -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.485 / Virus Database: 269.13.8/993 - Release Date: 6/09/2007 3:18 PM ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________________ Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx Archive: http://sourceforge.net/mailarchive/forum.php?forum_name=openh323gk-users Unsubscribe: http://lists.sourceforge.net/lists/listinfo/openh323gk-users Homepage: http://www.gnugk.org/
