[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #20459 [Core Tor/Tor]: ewma_cmp_cmux never considers policies different
#20459: ewma_cmp_cmux never considers policies different
--------------------------+------------------------------------
Reporter: pastly | Owner:
Type: defect | Status: needs_review
Priority: High | Milestone: Tor: 0.3.0.x-final
Component: Core Tor/Tor | Version: Tor: 0.2.6.2-alpha
Severity: Normal | Resolution:
Keywords: | Actual Points:
Parent ID: | Points:
Reviewer: | Sponsor:
--------------------------+------------------------------------
Changes (by pastly):
* status: merge_ready => needs_review
Comment:
FWIW, I've watched the order the scheduler picks channels. A few months
ago, based on memory, the scheduler would pick channels in an order
something like this:
3, 2, 5, 3, 2, 5, 3, 2, 3, 2, 3, 2, 3, 3, 3, 3
With this fix, it picks channels in an order something like this:
11, 7, 7, 1, 1, 1, 1, 1
I encourage us to verify the before behavior as memory <<< logs.
Of course, this tells us nothing about how circuits are picked within a
channel, which is very important.
I'm moving this out of merge_ready because I think we have too many
questions right now. Hopefully needs_review is a more accurate
description.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/20459#comment:7>
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