The cause of the high messageSendTime's in the current unstable is: With PeerHandler, if we queue a message, we queue it to a node rather than a connection. So we queue a message when no connections are free. Then we wait for any connection to become free. Since they are all sending trailers, and since all our attempts to open a connection fail, either because of misconfiguration, severe overload, or deliberate sabotage (in order to reduce the effect of vast numbers of incoming connections), it remains on the queue. The reason it doesn't happen on builds without PeerHandler (such as stable) is that we wouldn't queue the message at all if all the ConnectionHandlers were sending trailers, we would open a new connection THEN queue the message on that - a much shorter time ON THE QUEUE. -- Matthew J Toseland - [EMAIL PROTECTED] Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so.
signature.asc
Description: Digital signature
_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
