| Dario De Agostini 2005-05-05, 7:45 am |
| Dear List,
i've been reported a strange bug...
i have an a/v application which uses netstream.send to "propagate" =
actions between clients...
this choice have been made because i wanted to keep actions and voice =
synchronized.
everything works nicely...
we have a main server with 3 "proxy" which replicates everything to 36 =
users total.
(12 users for each proxy)
without going into architecture details we have the chance that one of =
the 36 users connect to the main server to publish a streaming which has =
to be seen from the other 35 clients.
In addition to this we have some accounts (presenter and others) which =
connect to main server and publish content...
the bug we have been reported is about a STRANGE lag which affects only =
commands sent over netstreams...
so , sometimes, we have the main presenter (which gets the published =
streams directly from main server) which receive audio and video =
PERFECTLY but receive .send commands after several MINUTES!
After a while (we have not been able to understand exactly "when") he =
gets ALL the .send messages toghether (i guess something like the server =
side "reset" parameter on Stream.play function).
After receiving all the commands it starts working like expected (no =
lag) ... until the next lag situation.
I've read many person here complaining about weird FCS "lag" which has =
to be solved with next release... i'd like to know if this case apply to =
the situation.
Is this a "code" problem or an FCS problem ?
we MUST find a solution.
Thanks in advance!
Dario De Agostini
=-----------------------------------------------------------
Supported by Fig Leaf Software - http://www.figleaf.com
=-----------------------------------------------------------
To change your subscription options or search the archive:
http://chattyfig.figleaf.com/mailman/listinfo/flashcomm
|