[OpenSIPS-Users] Some calls fail with tcp_connect_blocking error on version 2.1.1

Owais Ahmad millennium.bug at gmail.com
Tue Jun 21 12:51:07 CEST 2016


Hi
​
Răzvan,

I have tested with your recommendations. Here are my findings:

i) Disabling tcp_accept_aliases increases frequency of this error.
ii) Upgrading to 2.1.3 does not resolve the issue. On this new version, I
also tested tcp_accept_aliases=0/1. I get the exact same error I posted
previously.

Is there a way we can control the number of retries opensips should make if
there is a UAC connect failure? Saying that because I know for sure that B
party has a working network connection.

I have configured the following, in case it helps you point me to the
actual issue:

maxbuffer=65536
tcp_children=100
tcp_accept_aliases=0
tcp_connect_timeout=3
tcp_keepalive=0
modparam("proto_tls", "tls_max_msg_chunks", 8)
modparam("proto_tls", "tls_handshake_timeout", 3)
modparam("proto_tls", "tls_send_timeout", 7)

Regards,
Owais

On Tue, Jun 21, 2016 at 1:16 PM, Răzvan Crainea <razvan at opensips.org> wrote:

> Hi, Owais!
>
> You should consider upgrade OpenSIPS to a newer version (at least 2.1.3),
> because there were some fixes done related to this issue. If you cannot
> upgrade right now, try to disable the auto TCP aliasing[1] (set it to 0).
>
> [1] http://www.opensips.org/Documentation/Script-CoreParameters-2-1#toc90
>
> Best regards,
>
> ​​
> Răzvan Crainea
> OpenSIPS Solutionswww.opensips-solutions.com
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20160621/9751b219/attachment.htm>


More information about the Users mailing list