#20132: Let large client deployments use a local directory cache -----------------------------------+------------------------------ Reporter: teor | Owner: Type: enhancement | Status: new Priority: Medium | Milestone: Tor: 0.2.??? Component: Core Tor/Tor | Version: Severity: Normal | Resolution: Keywords: tor-hs, tor2web, rsos | Actual Points: Parent ID: | Points: 1 Reviewer: | Sponsor: -----------------------------------+------------------------------
Comment (by teor): Replying to [comment:1 arma]: > Replying to [ticket:20132 teor]: > > One of the things that concerns me about large tor client farms is that they download a ~1.5MB consensus per client per hour. > > It should in theory be compressed, so smaller than that. But yes, it is a lot over time, per client. > > > It would really help to have a client option for this in Tor, but there is a tradeoff - compromise that relay, and you own all the clients. > > In theory, this is what DirPort was for long ago. If you set up one Tor client with a DirPort, and set httpproxy on your other clients, then they'll ask their dir questions to the one central client. I guess this is a bit trickier nowadays with the begindir style questions, but that's a simple matter of turning off ... wait, it appears that we ripped out TunnelDirConns. Well, we used to have this functionality, back in the day. :) FallbackDir still has this effect - set up one relay (doesn't have to be public), add it as the FallbackDir for all your clients, your clients go to it first, then go to the authorities. Of course, if your FallbackDir dies, all your clients hammer the authorities. -- Ticket URL: <https://trac.torproject.org/projects/tor/ticket/20132#comment:2> Tor Bug Tracker & Wiki <https://trac.torproject.org/> The Tor Project: anonymity online _______________________________________________ tor-bugs mailing list tor-bugs@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs