[asterisk-bugs] [JIRA] (ASTERISK-28669) chan_sip: Device states lost when sip reload

[Joshua Elson (JIRA)]

    [ https://issues.asterisk.org/jira/browse/ASTERISK-28669?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=249243#comment-249243 ] 

Joshua Elson commented on ASTERISK-28669:
-----------------------------------------

Having had a great deal of experience with chan_sip and realtime, you more or less never should operationally use a sip reload with realtime. No matter the rt* settings, you will end up causing all peers to be removed from memory until another action causes them to be looked up again from realtime and/or the device re-registers, but as this bug points out, existing device state is lost. In my view, this behavior is by design.

Practically speaking though, you should never really need a sip reload in the course of “normal” operations. If you need to update cached peer information, a sip prune XXX will get you what you want in a far more surgical way. Other lower level operations rolled up in the reload command can be done in other ways or scheduled when peers are not in use.

> chan_sip: Device states lost when sip reload
> --------------------------------------------
>
>                 Key: ASTERISK-28669
>                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-28669
>             Project: Asterisk
>          Issue Type: Bug
>      Security Level: None
>          Components: Channels/chan_sip/General
>    Affects Versions: 13.26.0, GIT, 16.6.0
>            Reporter: cagdas kopuz
>            Assignee: Unassigned
>         Attachments: asterisk-debuglog.txt, asterisk logs.txt, realtime.txt
>
>
> Device states of agent lost when sip reload and after second call came same agent. After that, agent status forgot first call state and turn to second call's state. But first call still same states. By the way, queues' member states also changed. Log is attached.



--
This message was sent by Atlassian JIRA
(v6.2#6252)