Honestly, no I haven't, for no other reason than I don't have
anything concrete to tell them. =/ I know how I'd feel if someone
came to me and said "something is broken in your stuff and I can't
figure out what it is, data being sent seems to be delayed by one
packet worth. it seems to happen with this and seems to happen with
that but not this and ....". =/ Basically I don't feel I have
enough information to talk to anyone about it. I have brought it up
in #twisted before but that was completely useless.
Daniel
On Dec 5, 2006, at 5:04 PM, Chris Carlin wrote:
Daniel Henninger wrote:
This is unfortunately a known issue that appears to only occur in
Twisted 2.0 on up. It's fine in Twisted 1. Basically what
happens is your message is delayed until the next message goes out
to AIM's network. So if you think of all of the messages going to
AIM as a queue, it's always "one step behind". I've yet to figure
out what causes it as I can watch the handling enter and leave my
own code and go out into twisted core functionality. I believe
it's my code that's somehow causing the problem, but without the
delay showing up in my own code, it's hard to track down. My
theory is that I'm overriding some variable that is used by
twisted core and am confusing it. Still, haven't had more time to
investigate and kept getting nowhere with it. Interestingly
enough, using the epoll reactor appears to solve the problem.
Daniel
Have you ever asked the Twisted guys, BTW?
~Chris
_______________________________________________
py-transports mailing list
[email protected]
http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports
_______________________________________________
py-transports mailing list
[email protected]
http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports
