[tor-bugs] #23318 [Core Tor/Tor]: compute_weighted_bandwidths: do not add 0.5 to final_weight

[Tor Bug Tracker & Wiki]

#23318: compute_weighted_bandwidths: do not add 0.5 to final_weight
-------------------------------------------------+-------------------------
 Reporter:  cypherpunks                          |          Owner:  nickm
     Type:  defect                               |         Status:
                                                 |  needs_review
 Priority:  Medium                               |      Milestone:  Tor:
                                                 |  0.3.2.x-final
Component:  Core Tor/Tor                         |        Version:  Tor:
                                                 |  0.2.4.3-alpha
 Severity:  Normal                               |     Resolution:
 Keywords:  path-selection, 029-backport,        |  Actual Points:
  030-backport, 031-backport, review-group-24    |
Parent ID:                                       |         Points:  1
 Reviewer:  asn                                  |        Sponsor:
                                                 |  SponsorQ
-------------------------------------------------+-------------------------
Changes (by teor):

 * status:  needs_revision => needs_review


Comment:

 The underlying issue here is how tor deals with a consensus that has all-
 zero bandwidths.

 Before the original patch:
 1. Tor calculates 1 as the weighted bandwidth for all the nodes
 2. frac_nodes_with_descriptors() returns a sensible value, so Tor
 bootstraps
 3. node_sl_choose_by_bandwidth() returns a badly-weighted value (I split
 off #24169 to deal with this), but Tor works anyway

 After the original patch:
 1. Tor calculates 0 as the weighted bandwidth for all the nodes
 2. frac_nodes_with_descriptors() returns 0, so Tor fails to bootstrap

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/23318#comment:20>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online