I am not sure if the errors in the opensips logs are at the same time when the TCP connection breaks or not . But I can see some errors in the logs below .
* ERROR:core:tcpconn_async_connect: poll error: flags 1c ERROR:core:tcpconn_async_connect: failed to retrieve SO_ERROR [server=10.m.n.o:5060] (111) Connection refused ERROR:core:proto_tcp_send: async TCP connect failed ERROR:tm:msg_send: send() to 10.m.n.o:5060 for proto tcp/2 failed ERROR:tm:t_uac: attempt to send to 'sip:10.m.n.o:5060' failed INFO:core:probe_max_sock_buff: using snd buffer of 416 kb INFO:core:init_sock_keepalive: TCP keepalive enabled on socket 44* *10.m.n.o:5060 - this is one of the gateway IPs in my dr_gateways table . * These error logs are resolved automatically as well . I can see these logs randomly once in 2/3 days in my opensips logs . I have not mentioned any parameters for the proto_tcp module . Just loaded the module in the config . Do I need to specify some config level parameter as well ? *Thanks & Regards* *Sasmita Panda* *Senior Network Testing and Software Engineer* *3CLogic , ph:07827611765* On Fri, Feb 14, 2025 at 8:52 PM Bogdan-Andrei Iancu <bog...@opensips.org> wrote: > Hi, > > OpenSIPS listens on 5060 for incoming TCP conns, but when connecting (to > other server) it uses a random / ephemeral port as source of the outbound > TCP conn - this is normal, how TCP works in most of the cases. > > When OpenSIPS is not able to connect anymore, do you see any errors in the > logs? > > Regards, > > Bogdan-Andrei Iancu > > OpenSIPS Founder and Developer > https://www.opensips-solutions.com > https://www.siphub.com > > On 13.02.2025 08:59, Sasmita Panda wrote: > > On the same server I can see something different as well . > > I have only configured 5060 ports for TCP and UDP in the listen parameter > . > But I can see openisps established connections from different ports rather > than 5060 . How is that possible ? > > tcp 0 0 172.x.y.z:5060 10.m.n.o:42560 ESTABLISHED > 24449/opensips > tcp 0 0 172.x.y.z:5060 10.a.b.c:35507 ESTABLISHED > 24448/opensips > tcp 0 0 172.x.y.z:58609 10.m.n.o:5060 ESTABLISHED > 24450/opensips > tcp 0 0 172.x.y.z:57323 10.a.b.c:5060 ESTABLISHED > 24449/opensips > > Here 172.x.y.z is my opensips server and others are gateway IPs . I can > understand the TCP connection from the 5060 port . > How do opensips select other ports here ? > Is this the default behaviour of opensips ? Or in the latest version the > behaviour changes ? > > > > > > > *Thanks & Regards* > *Sasmita Panda* > *Senior Network Testing and Software Engineer* > *3CLogic , ph:07827611765* > > > On Thu, Feb 13, 2025 at 10:44 AM Sasmita Panda <spa...@3clogic.com> wrote: > >> Hi , >> >> >> I have a server running on opensips 2.2 version (Yes that is very old I >> know , we are planning to upgrade that in near future ) . >> >> On that server I have set up the gateways with TCP socket connection . >> Openisps while sending outbound calls to the gateway give the call in TCP . >> >> From the last 2/3 months we observed , randomly sometime in 7days or >> 10/12 days the connection won't get established and calls get affected . >> >> Once we restart the opensips service things start working fine . So here >> is opensips blocking something ? or is it on my linux machine blocking the >> connection ? >> >> For the proto_tcp module I have not configured any parameters . just >> loaded the module . I am hoping the default values must be in use . >> >> Can anybody suggest to me what I need to debug here ? >> >> *Thanks & Regards* >> *Sasmita Panda* >> *Senior Network Testing and Software Engineer* >> *3CLogic , ph:07827611765* >> > > _______________________________________________ > Users mailing > listUsers@lists.opensips.orghttp://lists.opensips.org/cgi-bin/mailman/listinfo/users > > >
_______________________________________________ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
