[tor-bugs] #15744 [Tor]: Is 16384 introductions a sane limit for IP rotation?

[Tor Bug Tracker & Wiki]

#15744: Is 16384 introductions a sane limit for IP rotation?
-------------------------+--------------------------------
     Reporter:  dgoulet  |      Owner:
         Type:  defect   |     Status:  new
     Priority:  normal   |  Milestone:  Tor: 0.2.7.x-final
    Component:  Tor      |    Version:
   Resolution:           |   Keywords:  SponsorR, tor-hs
Actual Points:           |  Parent ID:  #15714
       Points:           |
-------------------------+--------------------------------

Comment (by asn):

 Some thoughts on graphs:

 - If we want more data for this ticket, we are mainly interested in how
 the the histogram of introductions per circuit looks like.

 - We should use the lifetime information to figure out info on the churn
 of hidden services. For this, the histogram of the lifetime of circuits
 should be helpful. The idea, is that if we see many ultra-volatile hidden
 services, maybe we should make our protocols more friendly to this use
 case.

 - Another, more experimental graph, would be how the value of each bin of
 the histogram changes over time. I imagine this would not be very
 revealing, but maybe it reveals some information on the edge-cases we
 observed. Like, maybe ultra busy hidden services only appear on weekends
 or something. Unfortunately, we probably don't have a big enough data set
 to make useful observations from this.

 - Of course, if we had the non-aggregated stats, we could also check how
 long did 16384+ intro circuits last. This could help us understand if such
 circuits are getting DoSed (they lasted 5 minutes), or they just see
 regular client activity. But it's ok that we don't have this info.

 tl;dr The histograms of both data sets are enough for now.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/15744#comment:11>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online