Re: Call Failover

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

 



Hi,
I've implemented a new switch
[RoutedMode]DisableRetryChecks=1
This will retry all calls regardless of FastStart or H.245 messages.But use this with care.
Regards,Jan
xx xx wrote:> Hi Jan,> what I have found is that when gnugk tries the 2nd route it sends "complete" setup message like this one for the first route, and if the far-end GW responds with "call proceeding" gnugk forward this message to the caller. > So for me the caller have no chance to confuse both routes because he did't know for it before. > This is also the case when I have disabled the check for h245.> > mortex> > >  >-------- Оригинално писмо -------->  >От:  Jan Willamowius <jan@xxxxxxxxxxxxxx>>  >Относно: Re:  Call Failover>  >До: openh323gk-users@xxxxxxxxxxxxxxxxxxxxx>  >Изпратено на: Сряда, 2006, Август 2 11:27:55 EEST>  >---------------------------------->  >>  >xx xx wrote:>  >> >  >> Hi Alex,>  >> I had the same situation. Actually the problem for me was that I have always received m_h245socket != NULL using some CPEs. With others I have no problems. To solve this I commented this checks and now gk works well - just checks the disconnect reason and thus I can easyily control when to fail over the call.>  >>>  >> Actually many carriers gives fast start response and still can't complete the call, so I think thes checks are quite restrictive and not should be done.>  >> >  >>  >I think your solution is too broad for general GnuGk code. Are you>  >sure your callers can establish the 2nd call after they have received>  >H.245 responses from the first failing call ? Or could this result in>  >multiple more retries that all just fail ?>  >>  >> My problem is that my gk crashes very often 1-2 times a day (with this checks and without them, no difference). Can you please share your experience including pwlib, openh323 version. >  >>  >I have some more reports of high load installations crashing with 2.2.4>  >while others run fine. The current guess is that there is a race>  >condition someplace where we handle multiple messages for the same call.>  >But so far no specific cause has been found.>  >>  >Regards,>  >Jan
-- Jan Willamowius, jan@xxxxxxxxxxxxxx, http://www.gnugk.org/
-------------------------------------------------------------------------Using Tomcat but need to do more? Need to support web services, security?Get stuff done quickly with pre-integrated technology to make your job easierDownload IBM WebSphere Application Server v.1.0.1 based on Apache Geronimohttp://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642_______________________________________________________
Posting: mailto:Openh323gk-users@xxxxxxxxxxxxxxxxxxxxxxxxxxxx: http://sourceforge.net/mailarchive/forum.php?forum_id=8549Unsubscribe: http://lists.sourceforge.net/lists/listinfo/openh323gk-usersHomepage: http://www.gnugk.org/


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

  Powered by Linux