[asterisk-bugs] [JIRA] (ASTERISK-30142) Missing port in Contact header in response to incoming requests
Asterisk Team (JIRA)
noreply at issues.asterisk.org
Mon Jul 18 08:13:09 CDT 2022
[ https://issues.asterisk.org/jira/browse/ASTERISK-30142?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Asterisk Team updated ASTERISK-30142:
-------------------------------------
Status: Waiting for Feedback (was: Waiting for Feedback)
> Missing port in Contact header in response to incoming requests
> ---------------------------------------------------------------
>
> Key: ASTERISK-30142
> URL: https://issues.asterisk.org/jira/browse/ASTERISK-30142
> Project: Asterisk
> Issue Type: Bug
> Security Level: None
> Components: pjproject/pjsip
> Affects Versions: 18.12.1
> Environment: Debian Bullseye
> Reporter: Morten Sølvberg
> Assignee: Asterisk Team
> Severity: Critical
>
> I have recently updated all our tenants from v. 18.0.1 to 18.12.1, and I see behaviour that I haven't encountered before.
> When receiving incoming INVITEs, then the Contact header reported in the 183 and 200 OK responses does not contain the TCP listening port.
> For one tenant I use the TCP port 5118, but when the Contact header does not contain that specific port, then the remote party sends the ACK and subsequent BYE to port 5060 - which I am not iistening on.
> If I enable rewrite_contact on the Endpoint that receives the INVITE, then suddenly the port appears - however not sure why..
> Please note that all outbound INVITES and OPTIONS are sent with the correct port in the Contact header - regardless of whether contact_rewrite is enabled for the endpoint.
> I am quite certain that I haven't seen this behaviour prior to the upgrade to v. 18.12.1.
> In same upgrade I also updated the Debian version from Buster to Bullseye, but I can't see how this would affect Asterisk.
> It is not a major issue for me, if I can trust that enabling rewrite_contatct for all endpoints will cause Asterisk to include the listening port in the Contact header, but I would really like to know why.
> Is this a known bug/change or could it be a configuration issue?
> Just FYI: I do not REGISTER at the remote host.
> For the transport I use the below configuration:
> [transport-tcp]
> type=transport
> protocol=tcp
> bind=0.0.0.0:5118
> tos=cs3
> cos=3
> allow_reload=true
> external_media_address=52.166.13.xxx
> external_signaling_address=52.166.13.xxx
> local_net=10.0.0.0/8
> local_net=172.16.0.0/12
> local_net=192.168.0.0/16
--
This message was sent by Atlassian JIRA
(v6.2#6252)
More information about the asterisk-bugs
mailing list