gateway FS informs it's client FS about users hanged up with a long delay
Hi, When I use two FreeSWITCH instances ('internal' and 'external'), all users register to the 'external' instance which acts as a gateway by 'internal' instance (which in turn is controlled by my applicaiton with commands sent by socket). When user hangs up, the 'hanged up' event is propagated to the 'internal' instance after a long time (~3 minutes) instead of being propagated immediately. What can cause this issue? Br/ Maciej Aniserowicz
Incorrect NAT configuration so one of the boxes is not actually getting a BYE. On Oct 5, 2009, at 3:13 AM, Maciej Aniserowicz wrote:
Hi, When I use two FreeSWITCH instances ('internal' and 'external'), all users register to the 'external' instance which acts as a gateway by 'internal' instance (which in turn is controlled by my applicaiton with commands sent by socket). When user hangs up, the 'hanged up' event is propagated to the 'internal' instance after a long time (~3 minutes) instead of being propagated immediately. What can cause this issue?
Both of the instances are run on the same machine, i just changed the default ports they use. Can anything else cause this strange behavior? MA Michael Jerris wrote:
Incorrect NAT configuration so one of the boxes is not actually getting a BYE.
On Oct 5, 2009, at 3:13 AM, Maciej Aniserowicz wrote:
Hi, When I use two FreeSWITCH instances ('internal' and 'external'), all users register to the 'external' instance which acts as a gateway by 'internal' instance (which in turn is controlled by my applicaiton with commands sent by socket). When user hangs up, the 'hanged up' event is propagated to the 'internal' instance after a long time (~3 minutes) instead of being propagated immediately. What can cause this issue?
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
-- View this message in context: http://n2.nabble.com/gateway-FS-informs-it-s-client-FS-about-users-hanged-up... Sent from the freeswitch-users mailing list archive at Nabble.com.
On Thu, Oct 8, 2009 at 6:18 AM, Maciej Aniserowicz < maciej.aniserowicz@gmail.com> wrote:
Both of the instances are run on the same machine, i just changed the default ports they use. Can anything else cause this strange behavior? MA
Did a packet capture yield any clues? That is, were you able to confirm that each instance sent and received all the packets that you believe they should have sent and received? The reason I ask is so that you don't end up chasing a ghost because you made an assumption somewhere in your troubleshooting. -MC
Michael Jerris wrote:
Incorrect NAT configuration so one of the boxes is not actually getting a BYE.
On Oct 5, 2009, at 3:13 AM, Maciej Aniserowicz wrote:
Hi, When I use two FreeSWITCH instances ('internal' and 'external'), all users register to the 'external' instance which acts as a gateway by 'internal' instance (which in turn is controlled by my applicaiton with commands sent by socket). When user hangs up, the 'hanged up' event is propagated to the 'internal' instance after a long time (~3 minutes) instead of being propagated immediately. What can cause this issue?
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
-- View this message in context: http://n2.nabble.com/gateway-FS-informs-it-s-client-FS-about-users-hanged-up... Sent from the freeswitch-users mailing list archive at Nabble.com.
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
Hello, The issue is resolved. I feel stupid, because Michael Jerris was right the first time. Setting external_rtp_ip and external_sip_ip to $${local_ip_v4} made it work. But the strange thing is: it SOMETIMES worked before without any delay, which 'should not be possible', because the original IP was my external ip and the BYE message was sent straight to it. And there is no way it could reach the target 'internal' FS, because it runs on virtual machine, and no ports are forwarded on my router. Any thoughts? Why this could (rarely) work even with the previous config? Thanks to both of you for your answers. MA ----- Original Message ----- From: mercutioviz (via Nabble) To: Maciej Aniserowicz Sent: Thursday, October 08, 2009 7:06 PM Subject: Re: [Freeswitch-users] gateway FS informs it's client FS about users hanged up with a long delay On Thu, Oct 8, 2009 at 6:18 AM, Maciej Aniserowicz <[hidden email]> wrote: Both of the instances are run on the same machine, i just changed the default ports they use. Can anything else cause this strange behavior? MA Did a packet capture yield any clues? That is, were you able to confirm that each instance sent and received all the packets that you believe they should have sent and received? The reason I ask is so that you don't end up chasing a ghost because you made an assumption somewhere in your troubleshooting. -MC Â Michael Jerris wrote: > > Incorrect NAT configuration so one of the boxes is not actually > getting a BYE. > > > On Oct 5, 2009, at 3:13 AM, Maciej Aniserowicz wrote: > >> Hi, >> When I use two FreeSWITCH instances ('internal' and 'external'), all >> users register to the 'external' instance which acts as a gateway by >> 'internal' instance (which in turn is controlled by my applicaiton >> with commands sent by socket). >> When user hangs up, the 'hanged up' event is propagated to the >> 'internal' instance after a long time (~3 minutes) instead of being >> propagated immediately. >> What can cause this issue? > > > _______________________________________________ > FreeSWITCH-users mailing list > [hidden email] > http://lists.freeswitch.org/mailman/listinfo/freeswitch-users > UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users > http://www.freeswitch.org > > -- View this message in context: http://n2.nabble.com/gateway-FS-informs-it-s-client-FS-about-users-hanged-up... Sent from the freeswitch-users mailing list archive at Nabble.com. _______________________________________________ FreeSWITCH-users mailing list [hidden email] http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org _______________________________________________ FreeSWITCH-users mailing list [hidden email] http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org ------------------------------------------------------------------------------ View message @ http://n2.nabble.com/gateway-FS-informs-it-s-client-FS-about-users-hanged-up... To unsubscribe from Re: gateway FS informs it's client FS about users hanged up with a long delay, click here. -- View this message in context: http://n2.nabble.com/gateway-FS-informs-it-s-client-FS-about-users-hanged-up... Sent from the freeswitch-users mailing list archive at Nabble.com.
On Oct 9, 2009, at 10:40 AM, Maciej Aniserowicz wrote:
Hello, The issue is resolved. I feel stupid, because Michael Jerris was right the first time. Setting external_rtp_ip and external_sip_ip to $${local_ip_v4} made it work. But the strange thing is: it SOMETIMES worked before without any delay, which 'should not be possible', because the original IP was my external ip and the BYE message was sent straight to it. And there is no way it could reach the target 'internal' FS, because it runs on virtual machine, and no ports are forwarded on my router. Any thoughts? Why this could (rarely) work even with the previous config?
Thanks to both of you for your answers.
MA
It could work at least in regards to rtp if you had something else on the other side that adjusted to incorrect rtp ip like freeswitch does, the bye probably never really worked. Mike
participants (3)
-
Maciej Aniserowicz
-
Michael Collins
-
Michael Jerris