Hello, according tcpdump-traces on the linux machine and log entries in syslog (see below) the 200 OK passed the SIP Router (this means - the call was accounted and afterwards the bogus event was noted):
Jun 10 18:29:09 debian /usr/sbin/kamailio[20026]: INFO: <script>: ---xlog: a call with sip:116...@10.16.10.99 is coming in Jun 10 18:29:12 debian /usr/sbin/kamailio[20027]: NOTICE: acc [acc.c:275]: ACC: transaction answered: timestamp=1276208952;method=INVITE;from_tag=3120233282-36939680-1276187634579;to_tag=3338003860-56654088-1276187447718;call_id=741466613-36939680-1276187634...@10.16.10.152;code=200;reason=OK;src_user=116201;src_domain=10.16.10.99;dst_user=116202;dst_domain=10.16.10.90;src_ip=10.16.10.152 Jun 10 18:29:12 debian /usr/sbin/kamailio[20025]: CRITICAL: dialog [dlg_hash.c:591]: bogus event 6 in state 2 for dlg 0xa7eb3ea8 [2285:1225504260] with clid '741466613-36939680-1276187634...@10.16.10.152' and tags '3120233282-36939680-1276187634579' '' However, this is not the one and only event / state combination I found in syslog. There is also an entry with event 6 in state 1 present: Jun 10 19:35:49 debian /usr/sbin/kamailio[20026]: INFO: <script>: ---xlog: a call with sip:116...@10.16.10.99 is coming in Jun 10 19:35:50 debian /usr/sbin/kamailio[20027]: NOTICE: acc [acc.c:275]: ACC: transaction answered: timestamp=1276212950;method=INVITE;from_tag=3805046041-37017968-1276191635112;to_tag=befc3f7326747dei0;call_id=789390150-37017968-1276191635...@10.16.10.152;code=200;reason=OK;src_user=116201;src_domain=10.16.10.99;dst_user=116002;dst_domain=10.16.10.205;src_ip=10.16.10.152 Jun 10 19:35:50 debian /usr/sbin/kamailio[20026]: CRITICAL: dialog [dlg_hash.c:591]: bogus event 6 in state 1 for dlg 0xa7e979b8 [1648:1273002173] with clid '789390150-37017968-1276191635...@10.16.10.152' and tags '3805046041-37017968-1276191635112' '' But tomorrow I will test the setting - as recommended by you - to reduce kamailio to a single worker and check if the problem will keep repreduceable. Regards, Klaus Feichtinger > Bogus event 6 in state 2 means ACK received in early state. > This is a little bit bizarre, because it seems that the dialog did not > see the 200ok passing by. > Try to use a single worker (children=1) and see if you can reproduce > the problem. > Also, check that the 200ok is really going through the proxy. > > Regards, > Ovidiu Sas > > On Thu, Jun 10, 2010 at 6:44 AM, Klaus Feichtinger > <klaus.feichtin...@gmx.net> wrote: > >> Hello list, >> >> under special circumstances I have a problem with the DIALOG module of >> SIP-Router/Kamailio version 3.0.2. The dialog module is used in combination >> with presence, presence_xml, pua, pua_usrloc and pua_dialoginfo modules + >> db_mysql modules. The db_mode is set to "write_through", because of >> redundancy requirements. >> >> From my point of view this problem has something to do with timing in >> general. E.g. a so called "data call" has a duration of about 50 ms only; >> INVITE....BYE). That problem occurs on this short calls only. Because of the >> problems in the dialog state machine, NOTIFY messages are incorrect. Even >> when the call is already finished, the NOTIFY message includes the state >> "confirmed" and causes a wrong status indication.... >> >> The detailed error message can be seen here: >> "Jun 9 16:00:39 debian /usr/sbin/kamailio[14992]: CRITICAL: dialog >> [dlg_hash.c:591]: bogus event 6 in state 2 for dlg [dialog-ID] with clid >> [Call-ID] and tags '1299370188-28358304-1276092068837' ''" >> >> >> I found in old mails of the developer list another error that looked nearly >> the same, but it differed in the event and state of the dialog state machine >> (the thread can be found under the link >> http://www.mail-archive.com/de...@lists.kamailio.org/msg03234.html). That >> bug should - according the information that I've found in the list - be >> solved. Therefore I will ask you: is the error as displayed above another >> well known error / bug, which (maybe) should already be solved? Is it a new >> bug? >> >> Thanks for any information. >> >> regards, >> >> Klaus Feichtinger >> >> _______________________________________________ >> sr-dev mailing list >> sr-dev@lists.sip-router.org >> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev >> >> > _______________________________________________ sr-dev mailing list sr-dev@lists.sip-router.org http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
