[Freeswitch-users] Handling temporary WS breakages more gracefully

David P davidswalkabout at gmail.com
Mon Jul 26 01:21:24 UTC 2021


When there is a temporary break in verto's WSS login, it does verto.purge()
and that seems to force a new createOffer/createAnswer negotiation and
re-execution of the dialplan if the connection is re-established.

Is there a reason why renegotiation of the media connections is necessary?

In our case we've tried to minimize disruptions to the callers by using
verto's support for reattaching/recovering sessions, but that seems to
complicate things when the dialplan is re-executed...it seems to result in
simultaneous verto calls with the same users. And some of these sessions
seem to trigger media_timeout despite there being media (resulting in
sudden loss of media in recordings).

Does it seem that I'm interpreting things consistently with the design of
verto and FS? If so, then it seems that we should disable the verto.purge()
so the current ICE connections are kept and the dialplan is not
re-executed, and rely on media_timeout to detect abandoned sessions.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freeswitch.org/pipermail/freeswitch-users/attachments/20210726/d09b2db5/attachment.html>


More information about the FreeSWITCH-users mailing list