On Tue, Sep 13, 2005 at 11:57:36PM -0700, Vanessa wrote: > As for this one: > > Got a trailer chunk ahead of our time! > > It seems this is caused by a reestablished session (with that I refer to > a connection between peers) where a packet from the previous session > with the same peer is considered part of the current session (mux mess). > This sometimes causes almost all sessions to be closed and sometimes it > is more than one packets from the previous session. This applies to both > an inbound session that is closed and opened again by the same peer soon > after and an outbound session that is closed and opened again to the > same peer soon after. Maybe putting in a longer delay than the constant > I saw would be a workable workaround. Is it delayed finalization or some > state that remains after a session got closed?
You mean when a connection is closed, some packets are lost and these result in the trailers currently being sent over it failing? It sounds to me like several hundred nodes have you in their routing tables, due to it being really popular, due to its large datastore (what's your connection speed?). This would mean that they constantly fight over your node. But still, we won't drop newbie connections quickly, short of a bug. So presumably these are nodes which are in your routing table. -- Matthew J Toseland - toad at amphibian.dyndns.org Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 189 bytes Desc: Digital signature URL: <https://emu.freenetproject.org/pipermail/support/attachments/20050914/dce3a000/attachment.pgp>
