I will have a try at updating the draft to reflect both options 1 and 2
below. I'm hoping we can have both mechanisms without introducing
significant complexity. If this turns out not to be true, we may need
to pick only one option.
Thanks,
Alan
[email protected] wrote:
From: Jason Fischl <[email protected]>
To summarize, there are 3 options:
1) client publishes all state changes only
2) optimization for servers that maintain call state and clients only
need to publish when seizing a line. This requires a mechanism for the
server to signal to the client that it does not need to do publish all
state changes.
3) clients only publish for line seize.
Please comment on these 3 options. The current draft describes option 2.
Clearly 2 is the best, if a suitable mechanism can be defined.
3 is totally unworkable in a system that has a classic (that is,
non-call-stateful, or "RFC 3261") proxy. Since sipX has a
non-call-stateful proxy, I find it undesirable.
Dale
_______________________________________________
BLISS mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/bliss
_______________________________________________
BLISS mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/bliss
