Find a better value for MAX_REND_FAILURES

For a while, MAX_REND_FAILURES was set to 30. That's clearly wrong; in #4241 (moved) we set it to 8. But is 8 the right value? We should investigate actual failure rates and pick a value that's more in tune with what we're actually seeing in the wild.

changed milestone to %Tor: 0.2.6.x-final

added component::core tor/tor milestone::Tor: 0.2.6.x-final parent::15463 priority::medium resolution::implemented sponsor::R status::closed tor-hs type::defect labels

Trac:
Keywords: N/A deleted, SponsorR added

Trac:
Parent: N/A to #15463 (moved)

We decided that this ticket will host easy short-term changes, and we will use #15557 (moved) for more intelligent retry logic fixes.

For this ticket, please see branch bug11447 in my repo. It merely changes MAX_REND_FAILURES to 1, so that the HS has one try and one retry to make its rendezvous circuits work.

Trac:
Status: new to needs_review

It's very difficult to actually measure the failure rate in the wild but with #15463 (moved) being seen lately bringing this one to 1 is saner for the HS and Tor network than the current value of 8 which is a pretty high amount of retry. Without DoS defense or counter measures, keeping this one low is ideal in this context.

ACK on the patch.

Taking this in 0.2.6 and later.

Trac:
Resolution: N/A to implemented
Milestone: Tor: 0.2.??? to Tor: 0.2.6.x-final
Status: needs_review to closed

Trac:
Keywords: SponsorR deleted, N/A added
Sponsor: N/A to SponsorR

closed

mentioned in issue #15463 (moved)

mentioned in issue #15557 (moved)

moved to tpo/core/tor#11447 (closed)

mentioned in issue tpo/core/tor#15557 (closed)