[tor-bugs] #29427 [Core Tor/Tor]: kist: Poor performance with a small amount of sockets

Tor Bug Tracker & Wiki blackhole at torproject.org
Thu Oct 3 14:45:53 UTC 2019


#29427: kist: Poor performance with a small amount of sockets
-------------------------------------------------+-------------------------
 Reporter:  dgoulet                              |          Owner:  dgoulet
     Type:  defect                               |         Status:
                                                 |  needs_revision
 Priority:  Medium                               |      Milestone:  Tor:
                                                 |  0.4.3.x-final
Component:  Core Tor/Tor                         |        Version:  Tor:
                                                 |  0.3.2.1-alpha
 Severity:  Major                                |     Resolution:
 Keywords:  tor-sched, kist, regression?,        |  Actual Points:  0.1
  041-deferred-20190530, 042-should,             |
  BugSmashFund, 035-backport, 040-backport,      |
  041-backport, 042-backport                     |
Parent ID:                                       |         Points:  0.1
 Reviewer:  nickm                                |        Sponsor:
-------------------------------------------------+-------------------------
Changes (by dgoulet):

 * keywords:
     tor-sched, kist, regression?, 041-deferred-20190530, 042-should,
     BugSmashFund
     =>
     tor-sched, kist, regression?, 041-deferred-20190530, 042-should,
     BugSmashFund, 035-backport, 040-backport, 041-backport, 042-backport
 * status:  needs_review => needs_revision
 * milestone:  Tor: 0.4.2.x-final => Tor: 0.4.3.x-final


Comment:

 Replying to [comment:15 nickm]:
 > Setting myself as reviewer per discussion at meeting.
 >
 > One more question: is this something we want to think about potentially
 backporting?  If not, should it wait for 043 when we can treat it as a
 feature and add a new consensus parameter?

 Discussion with nickm on IRC. There is a good argument on preventing
 partitioning client/HS into two buckets of "performance".

 So because of this, we'll defer this to 043, add two consensus parameters
 (client and relay sched interval) and backport it back to 035. We'll have
 a good chunk of the 043 cycle to make sure it works properly.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/29427#comment:16>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list