[mnet-devel] debugging notes

[Zooko O'Whielacronx]

> > I just observed the bug that we hypothesized today on IRC: that the 
> > "don't ask again" patch causes the second asking to fail.  So when the 
> > peerman switch is off, we need the "cache responses" feature.  When the 
> > peerman switch is on, the "don't ask again" feature is off.
> are you sure that this wasn't better before you added that demultiplex 
> line?

Yes.  The demultiplexing in question is for when one higher-level invocation of 
peerman.lookup_contact_info() results in several messages sent out, one to each 
metatracker.  We need to "demultiplex" the resulting "lookup contact info 
responses", so if 3 responses come back all containing the new contact info, 
peerman calls up the higher-level invoker once instead of 3 times.

This is different than when there are two invocations of 
peerman.lookup_contact_info() for the same peer.

Your new idea is to make a "demultiplexer" for that, so that multiple nearly-
simultaneous invocations of peerman.lookup_contact_info() for the same peer 
result in only one "lookup contact info" message (per metatracker).

--Z



-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
mnet-devel mailing list
mnet-devel at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mnet-devel