[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #6341 [Tor Relay]: connection_or_flush_from_first_active_circuit() does wrong thing when ewma_enabled
#6341: connection_or_flush_from_first_active_circuit() does wrong thing when
ewma_enabled
-----------------------+----------------------------------------------------
Reporter: arma | Owner:
Type: defect | Status: needs_review
Priority: normal | Milestone: Tor: 0.2.3.x-final
Component: Tor Relay | Version:
Keywords: | Parent:
Points: | Actualpoints:
-----------------------+----------------------------------------------------
Comment(by robgjansen):
Replying to [comment:22 arma]:
> Let me see if I'm interpreting the results correctly. The graphs show
that ewma-unpatched outperforms the other two -- their requests get to the
first byte quicker and also get to the last byte quicker. But the data in
your comment show that far fewer requests are made. Is it fair to say that
the clients in the ewma-unpatched case put less load on the network
because they're not making as many requests total?
This is absolutely correct. Only ~15 GiB worth of completed downloads
happened in the unpatched case as opposed to ~40 GiB in the patched case.
> Is this 'not making as many requests' correlated to 'not finishing some
of their requests in a timely fashion'?
Could be. Though, the performance graphs shows that the downloads that did
complete were fairly fast. We can't really draw conclusions directly from
the unpatched performance because the load is so much different.
I wonder if clients were simply 'crashing' in the unpatched network,
causing them to no longer make requests. In one of your previous comments,
you mention that you expected some simulated Tors to crash. Did you mean
in the unpatched or patched case?
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/6341#comment:23>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs