[off-list] On Tue, Oct 09, 2007 at 06:59:47PM -0400, Roger Dingledine wrote: > On Tue, Oct 09, 2007 at 06:54:33PM -0400, Roger Dingledine wrote: > > > I think that's the "missing" element. There is no way that I know of > > > for a circuit to refuse any new TCP connection while still staying > > > open for existing ones. > > > > Actually, that feature exists right now. Look for calls to > > we_are_hibernating() in various places in the code (command.c and > > connection_edge.c in particular). > > I think it might be smart to instead check "are we hibernating, or are > we not a server anymore" in these various places. I'd be happy to apply > a patch for that if one appeared. :) > > It's not quite what Robert asked for originally though, since it won't > affect a very active stream that stays very active. > > Perhaps a combination of refusing new requests plus throttling via > RelayBandwidthRate would be sufficient? Added to the bugtracker as bug 525 so we don't forget it. yrs, -- Nick Mathewson
Attachment:
pgpgVymMhKKyt.pgp
Description: PGP signature