#18809: Handle linked connections better during bootstrap ------------------------------------+------------------------------------ Reporter: teor | Owner: Type: defect | Status: new Priority: Medium | Milestone: Tor: 0.2.8.x-final Component: Tor | Version: Tor: 0.2.8.1-alpha Severity: Normal | Resolution: Keywords: must-fix-before-028-rc | Actual Points: Parent ID: | Points: Reviewer: | Sponsor: ------------------------------------+------------------------------------
Comment (by teor): Replying to [comment:2 arma]: > I believe the current code "works" in that it never generates any extra consensus fetches. So the functionality intended from #4483 is not present in 0.2.8. This isn't quite correct: the current code schedules additional consensus fetches more often when connections fail fast. But for slow or blackholed connections, it won't generate simultaneous connections. So it's an improvement over the 0.2.7 behaviour, but it doesn't fully implement the proposal. > I am not sure whether we will want to fix this in 0.2.8 -- I think implementing #4483 will be quite some lines of diff. I think we can fix most of this by checking for an existing consensus downloads in connection_dir_finished_flushing (#18808) and before we call connection_ap_attach_pending (this bug). Then we modify the code that counts consensus downloads to account for begindir requests, as per the description in the ticket. (This probably means some refactoring to do this counting in one place.) This should leave us with much cleaner code, because the complex checking & counting logic will be in one place. I think we can fix anything else in 0.2.9. > But I'm going to figure that one out later. In the mean time I'm accumulating patches on my bug18809 branch. Looking forward to it. -- Ticket URL: <https://trac.torproject.org/projects/tor/ticket/18809#comment:3> 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