Am 22.06.19 um 17:18 schrieb Sergey Ilinykh:
In response to Council Minutes 2019-06-19
quote from
https://wiki.xmpp.org/web/XEP-Remarks/XEP-0260:_Jingle_SOCKS5_Bytestreams_Transport_Method
Another problem with early (before accept) transport replace is the fact we
have to send the same offer twice. For example we have S5B and IBB. The
lousy s5b implementation can only gather s5b proxy candidates so it may
fail before we sent initial offer (session/content accept). So after proxy
discovery failure we may want to send transport-replace request to IBB
which will contain everything needed for IBB negotiation (at least block
size). Then we have to repeat transport offer with session/content-accept
which will force the remote party to reinitialize IBB transport what looks
like a bad practice, which may be even worse with other transports. To make
things right it has to be allowed to send session/content-accept without
transport element if it was accepted earlier.
To solve this we have https://github.com/xsf/xeps/pull/793 but may be it
require more actions.
Let's discuss.
Moving over from github. Peter, Lance and me looked at a very similar
case back in 2015, fallback from ice-udp to raw-udp. It also does a
transport-replace before session-accept.
https://xmpp.org/extensions/xep-0371.html#fallback
has the full protocol flow. Its a bit unclear to me now why Romeo would
send a ice-udp transport to the gateway which is unlikely to announce
this in disco... (which I assume is the way to pick the transport even
though 0166 is silent here)
We kept this within the current jingle specification by including an
empty namespaced
<transport xmlns='urn:xmpp:jingle:transports:raw-udp:1'/>
in the session-accept (example 15) which basically says "reminder: we
agreed on this transport, hopefully you remember the details".
For IBB I would recommend a similar thing but keeping the sid attribute,
the initiator can then verify that this refers to an already active
session that does not need to be reinitialized.
(I just hope we're using this pattern consistently...)
cheers
Philipp
(and thanks for the poke Dave!)
_______________________________________________
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: [email protected]
_______________________________________________
