Re: call setup through parent GK in proxy mode

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Thank you Jan for the ForwardOnFacility hint!

It works as expected (the Local endpoint does not get the AlternateAddress and does not try to
connect to the parent GK anymore).
However now the local GK itself is trying to re-setup the call in a weird way:

 12:52:42.865 3       ProxyChannel.cxx(1023)  Q931s   Received: Setup CRV=29770 from LOCALENDPOINT:5559
 12:52:42.866 2             gkacct.cxx(950)   GKACCT  Successfully logged event 1 for call no. 10
 12:52:42.866 3       ProxyChannel.cxx(2862)  Q931s   Call 10 is NAT type 0
 12:52:42.866 1       ProxyChannel.cxx(870)   Call 10: h245Routed=1 proxy=1
 12:52:42.866 3       ProxyChannel.cxx(887)   GK      Call 10 proxy enabled
 12:52:42.870 3       ProxyChannel.cxx(4677)  Q931    Connect to REMOTEENDPOINT:1720 from HOMEGATEKEEPER:47014 successful
 12:52:42.878 3       ProxyChannel.cxx(1023)  Q931d   Received: CallProceeding CRV=29770 from REMOTEENDPOINT:1720
 12:52:42.880 3       ProxyChannel.cxx(1023)  Q931d   Received: Facility CRV=29770 from REMOTEENDPOINT:1720
 12:52:42.881 3       ProxyChannel.cxx(1576)  Q931d   Call 10 is forwarded to PARENTGATEKEEPER:1720
 12:52:42.881 3       ProxyChannel.cxx(2862)  Q931s   Call 10 is NAT type 0
 12:52:42.881 1       ProxyChannel.cxx(870)   Call 10: h245Routed=1 proxy=1
 12:52:42.881 3       ProxyChannel.cxx(887)   GK      Call 10 proxy enabled
 12:52:42.883 3       ProxyChannel.cxx(4711)  Q931    Connect to PARENTGATEKEEPER:1720 from HOMEGATEKEEPER:47015 successful
 12:52:42.889 3       ProxyChannel.cxx(1023)  Q931s   Received: Setup CRV=29770 from PARENTGATEKEEPER:56973
 12:52:42.890 2       ProxyChannel.cxx(2079)  Q931s   Warning: socket (PARENTGATEKEEPER:56973) already attached for callid [.....]
 12:52:42.890 2             RasTbl.cxx(3854)  CDR     ignore not connected call
 12:52:42.890 2             gkacct.cxx(950)   GKACCT  Successfully logged event 2 for call no. 10
 12:52:42.890 3           yasocket.cxx(576)   Q931s   Delete socket PARENTGATEKEEPER:56973
 12:52:42.981 3       ProxyChannel.cxx(1023)  Q931d   Received: ReleaseComplete CRV=29770 from PARENTGATEKEEPER:1720

note the Q931s socket already attached Warning!
The problem seems still to be the Facility packet.
It comes from the remote endpoint. So it's maybe a configuration issue there? However I think the GK is doing something wrong in this situation.
At the local Endpoint I get an immediate Disconnect message.
Normally (when everything is working) I see an "Alerting" Q931d Message instead of the Facility message.



Some excerpt of our gatekeeper.ini that I consider relevant for the issue:

[RoutedMode]
CallSignalHandlerNumber=5
RemoveH245AddressOnTunneling=0
AcceptNeighborsCalls=1
AcceptUnregisteredCalls=1
SupportNATedEndpoints=1

DropCallsByReleaseComplete=1

ForwardOnFacility=1

[Proxy]
Enable=1

[RasSrv::LRQFeatures]
AlwaysForwardLRQ=1

IncludeDestinationInfoInLCF=0
CiscoGKCompatible=1


Maybe there is another config option I need to set (or unset?)

Sorry if this message does not appear in the original thread, it's surely meant as a follow-up.

Thank you for any help!
Tobias



Jan Willamowius wrote:

> Thats what ForwardOnFacility is for. By default its off, but when you
> turn it on its supposed to handle this situation.
>
> [RoutedMode]
> ForwardOnFacility=1
>
> Regards,
> Jan
>
>
>
> > Simon Horne wrote:
> > Tobias
>
> I have seen this too.
> The problem is that once an endpoint is registered you cannot call directly
> and the call must go through the gatekeeper.  When you call the endpoint you
> receive back a facility message with routeCalltoGatekeeper which instructs
> the caller to drop this call and call the gatekeeper.  In the facility
> message contains the gatekeeper address is the alternateAddress field. I
> think there also should be an alternateAlias field so the call to the
> gatekeeper knows where to forward on the call.  I think this might be a bug.
> We just need to find the time to debug the issue.
>
> Simon
>
>
>
> -----Original Message-----
> From: Tobias Marquart [mailto:tobimarq@xxxxxx]
> Sent: Wednesday, January 26, 2011 8:30 AM
> To: Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx
> Subject:  call setup through parent GK in proxy mode
>
> Hi all,
>
> we have a strange call setup behavior in a special situation. I hope that it
> has to do with the gnugk config so that we can somehow fix it.
>
> We use gnugk in proxy mode (without any problems). Our gk is connected to a
> gatekeeper hierarchy (one neighbor/parent gk).
>
> Consider 2 different types of external endpoints I try to call from a local
> endpoint:
>
> 1) Endpoints that are not in any way connected to our Gatekeeper hierarchy,
> I call them by IP-Address
> 2) Endpoints that are connected to (=known in) our Gatekeeper hierarchy, but
> not registered directly with our GK.
>
> The first type of endpoint works perfectly.
> The second type works perfectly when dialing via h323alias/E.164 However
> when trying to call these endpoints by using the IP-Address, the problem
> starts:
> First we realized an "unreachable destination" error at the local endpoint.
> This is in a somehow "closed" and NATted network environment.
> (when not behind NAT/firewall restrictions, the situation I describe below
> is the same, but the call setup nevertheless works).
>
> We realized that the local endpoint tries to connect to the parent
> gatekeeper directly (at a very late point in time of the call setup).
>
> Further analysis:
> The local endpoint gets a "h.225 CS facility" packet from our gnugk at an
> early point in the call setup.
> In this packet there is mentioned the IP-Address of the parent GK as
> "alternativeAddress".
> This facility packet does not appear in any of the other call setup
> scenarios (in the working scenarios there occurs an "Alerting" packet
> instead).
> The admissionRequest/admissionConfirm log entries on gnugk have absolutely
> identical flags and options regardless of the call setup scenario.
>
> Gnugk just seems not not be able to do a "normal" call setup in that special
> case. The problem occurs with several remote endpoints, so it seems to be a
> general issue.
> I have detailed gnugk-logs and wireshark packet traces (from the local
> endpoint) available (via PM) in case this is necessary. However I hope that
> this behavior is somehow known (and can be fixed easily).
>
> Btw: The situation happens also with the very newest gnugk release.
>
> I very much hope you can help me somehow, tia, Tobias


------------------------------------------------------------------------------
Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
Finally, a world-class log management solution at an even better price-free!
Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
February 28th, so secure your free ArcSight Logger TODAY! 
http://p.sf.net/sfu/arcsight-sfd2d
_______________________________________________________

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/


[Index of Archives]     [SIP]     [Open H.323]     [Gnu Gatekeeper]     [Asterisk PBX]     [ISDN Cause Codes]     [Yosemite News]

  Powered by Linux