From hw@gilawa.com Tue Jun 13 13:11:32 2023 From: Henning Westerholt To: sr-users@lists.kamailio.org Subject: [SR-Users] Re: Cannot achieve rtpengine HA Date: Tue, 13 Jun 2023 13:11:08 +0000 Message-ID: In-Reply-To: =?utf-8?q?=3CDB9PR05MB820385F7C6B1083A0D8BA9BFE155A=40DB9PR05MB?= =?utf-8?q?8203=2Eeurprd05=2Eprod=2Eoutlook=2Ecom=3E?= MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0147470631==" --===============0147470631== Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Hello, you can play with the rtpengine module timeout parameter to get a faster dete= ction of down rtpengine daemons. https://kamailio.org/docs/modules/5.5.x/modules/rtpengine.html#rtpengine.p.rt= pengine_tout_ms If you have some other detection/failover logic, it could also just deactivat= e the rtpengine on the affected kamailio services by RPC commands. Cheers, Henning -- Henning Westerholt - https://skalatan.de/blog/ Kamailio services - https://gilawa.com From: Arsenijs Kabihno Sent: Dienstag, 13. Juni 2023 13:57 To: Kamailio (SER) - Users Mailing List Subject: [SR-Users] Re: Cannot achieve rtpengine HA Hi, thank you! Probably the subject of this email is slightly misleading, the actual issue t= hat we have is that kamailio sticks to the rtpengine node selected in the beg= inning of the call. If rtpengine node goes down and there are active calls, kamailio continues to= try to communicate with it, when it is already inactive: Jun 12 07:01:26 ap07stvp.xxxxxxx.xxx docker[9373]: 27(33) ERROR: {2 3 INVITE = jdl9kq6fv51jcv1agk2e} rtpengine [rtpengine.c:3070]: send_rtpp_command(): can'= t send command "offer" to RTPEngine Jun 12 07:01:26 ap07stvp.xxxxxxx.xxx docker[9373]: 54(60) INFO: {1 3 ACK jdl9= kq6fv51jcv1agk2e}