[asterisk-bugs] [JIRA] (ASTERISK-22428) [patch] SIP unregister does not fully unregister when using Realtime sip peers

[Ben Smithurst (JIRA)]

    [ https://issues.asterisk.org/jira/browse/ASTERISK-22428?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=209781#comment-209781 ] 

Ben Smithurst edited comment on ASTERISK-22428 at 8/30/13 8:47 AM:
-------------------------------------------------------------------

I've attached the patch I used - [^realtime_unregister.diff] - different approach, but seems to work.

One difference is that Michael's patch still puts 'Expires: 120' in the response to unregister, but mine says 'Expires: 0'.
                
      was (Author: bensmithurst):
    I've attached the patch I used - different approach, but seems to work.

One difference is that Michael's patch still puts 'Expires: 120' in the response to unregister, but mine says 'Expires: 0'.
                  
> [patch] SIP unregister does not fully unregister when using Realtime sip peers
> ------------------------------------------------------------------------------
>
>                 Key: ASTERISK-22428
>                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-22428
>             Project: Asterisk
>          Issue Type: Bug
>      Security Level: None
>    Affects Versions: 1.8.23.0, 11.5.0
>            Reporter: Ben Smithurst
>            Assignee: Ben Smithurst
>            Severity: Minor
>         Attachments: after.pcap, asterisk-22428-dont-update-unregistered-peer.diff, before.pcap, console_after.txt, console_before.txt, realtime_unregister.diff
>
>
> When a SIP phone unregisters, this does not seem to be fully saved back to the DB when using realtime.  I see that the ipaddr and port fields are cleared, but regseconds and fullcontact are not, so the registration still works.  Asterisk seems to know it is unregistering, as it logs 'Unregistering SIP xxx' on the console.
> In console_before.txt you can see that after unregister, a call is placed, and the call is routed to the extension, but in console_after.txt it immediately says everyone is busy/congested, as expected.
> I have also attached before/after PCAPs and the simple patch I have used to solve this.  The problem is present, and the patch works, in both 11.5.0, and the head of branches/11 (SVN-branch-11-r397758).

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.asterisk.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira