Let test-network.sh tolerate some number of failures

This had been part of #29280 (moved), but teor notes correctly that it should be part of Chutney's test-network.sh

changed milestone to %Tor: 0.4.1.x-final

Trac:
Parent Ticket: #29280 (moved)

added 029-backport 034-backport 035-backport 040-backport CI PTs actualpoints::.3 component::core tor/chutney milestone::Tor: 0.4.1.x-final network-team-roadmap-2019-Q1Q2 owner::nickm parent::29280 points::.3 priority::medium resolution::fixed reviewer::teor severity::normal sponsor::19 status::closed type::enhancement labels

See my chutney branch test-network-forgiving-v2, with PR at https://github.com/teor2345/chutney/pull/2 . This branch is based on Teor's chutney-travis-v4 branch so that it can get CI itself. It should get rebased before merging. It also increments the allowed failures in the travis ci to 1.

Trac:
Reviewer: N/A to teor
Status: assigned to needs_review
Actualpoints: N/A to .3

Looks good to me. I made some comments, but I worked out the answers myself.

Trac:
Points: N/A to .3
Status: needs_review to merge_ready

Rebased as my test-network-forgiving-v2-rebased.

CI failed one job on 0.2.9, but the debug run succeeded: https://travis-ci.org/torproject/chutney/jobs/505547963

We might need to increase the allowed failures for 0.2.9, or put a sleep in between the allow failures tests. I opened #29761 (moved) to track the failures, so we can work out what to do here.

Merged to master as 9ef9ccc.

Trac:
Resolution: N/A to fixed
Status: merge_ready to closed

closed

changed time estimate to 2h 24m

added 2h 24m of time spent

mentioned in issue #29280 (moved)

moved to tpo/core/chutney#29748 (closed)