Fully agree with Michal - it's not a GK problem and we tested this with some different Endpoints and got it working. To be honest, I'm not that sure if tiptel/innovaphone really is interesseted in getting the innovaphone endpoints interoperating with any other Gatekeeper then theirs....... You know, Cisco or MS style - taking all from the standards others have worked out and then chaging it a little bit so it gets prorietary and using the own market force to push that proprietary "standard" into the market.... And again, maybe it's only a different interpretation of H.235 Specification Appendix D (even Appendex D is one of the more concret part of the H323 protocol reference). Btw, you know that there are other endpoints tested and working good with GnuGK? I.e. we are using Endpoints with ISDN and analog interfaces so our users may connect their own, existing telephony infrastructure. That's an important benefit for users! Gruss Frank -----Ursprüngliche Nachricht----- Von: openh323gk-users-admin@xxxxxxxxxxxxxxxxxxxxx [mailto:openh323gk-users-admin@xxxxxxxxxxxxxxxxxxxxx] Im Auftrag von Zygmuntowicz Michal Gesendet: Montag, 15. November 2004 17:37 An: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx Betreff: Re: innovaphone ip200 cant register It isn't correct - 2.0.9 should send H.235 tokens (authentication information) back to the endpoint. And it does so - you can check RCF message for nestedCryptoTokens. This was checked with a few H.235 compiliant devices and it works fine. ----- Original Message ----- From: "Thomas Balsfulland" <tbals@xxxxxxxxx> Sent: Monday, November 15, 2004 5:10 PM > On Thu, Nov 11, 2004 at 03:05:22PM +0100, Frank Fischer wrote: >> >> One more line on H.235 ... There is no such thing as ONE H.235 >> implementation method but some different methods to do H.235 >> authentication. >> In 2.0.9 there is a H.235 method called "Baseline Sercurity Profil" (see >> Annex D) implemented. >> But: Don't believe Recommendations are Standard - some of them, >> especially >> in H.323 are formulated that open (could, should instead of need, must) >> that >> there are possibly as much different implementations as developers >> working >> on it ... sounds funny, but might get very close to the reality :-( > > hi there, > > now i got an answer from innovaphone. i sent my gatekeeper logs and the > traces from the ip200 (from gnugk v2.0.9-cvs, pwlib v1.8.1, openh323 > v1.15.1). > the answer was german, sorry for bad translation: > > we have analyse your traces. we send a correct h.235 authentification and > the gatekeeper send no correct authentification back. the authentification > mechanism from h.235 is symetric, so the gatekeeper must register also at > the ip200 - but it dont do that. thats the reason why we ignore the > registrationconfirm. and we must ignore it to avoid > man-in-the-middle-attack. > the method of authentification is given with rrq, so the gatekeeper cant > do anything else. greetings from innovaphone. christoph kuenkel. > > is this correct? is this a bug of the gatekeeper? > > regards > > thomas ------------------------------------------------------- This SF.Net email is sponsored by: InterSystems CACHE FREE OODBMS DOWNLOAD - A multidimensional database that combines robust object and relational technologies, making it a perfect match for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8 _______________________________________________________ List: Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx Archive: http://sourceforge.net/mailarchive/forum.php?forum_id=8549 Homepage: http://www.gnugk.org/ ------------------------------------------------------- This SF.Net email is sponsored by: InterSystems CACHE FREE OODBMS DOWNLOAD - A multidimensional database that combines robust object and relational technologies, making it a perfect match for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8 _______________________________________________________ List: Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx Archive: http://sourceforge.net/mailarchive/forum.php?forum_id?49 Homepage: http://www.gnugk.org/
