Remove broken fallback directory mirrors

Some time around 0.3.0.?-rc, we should remove all non-functional fallback directories from the December 2016 list (or rebuild the entire list).

I'll create a draft branch here, but the script should be re-run just before the release (and ideally, operators should be contacted and given an opportunity to fix any errors).

changed milestone to %Tor: 0.3.2.x-final

added 028-backport 029-backport TorCoreTeam201703 component::core tor/tor fallback milestone::Tor: 0.3.2.x-final parent::21564 points::1 priority::medium resolution::fixed severity::normal status::closed type::defect labels

See my branch bad-fallbacks-201701-028, which needs:

• operators to be contacted,
• the script to be re-run just before 0.3.0.?-rc

Trac:
Status: new to needs_revision

The operator of fallback directory mirror FA3415659444AE006E7E9E5375E82F29700CFDFD has emailed me to ask for it to be removed.

The fallback check yesterday already detected an issue and removed it from the list in my branch bad-fallbacks-201701-028.

Trac:
Status: needs_revision to needs_review

No, this needs revision: operators need to be contacted, and then the script should be run again.

Trac:
Status: needs_review to needs_revision

The whitelist and script changes are in #21121 (moved). The branch is now bad-fallbacks-201702-028, but I'll wait a week or two after contacting operators and re-run the script.

Trac:
fallback_dirs_2017-02-14-2310_60c4a1ed35.log

Log for fallback directory check existing

(I won't email tor-team, 31 emails is a few too many.)

I emailed 12 operators about 15 relays. (The remaining 16 were long-term downtime or low averages, which operators can't fix.)

11 relays were on very old or non-recommended versions. 2 relays had added IPv6 addresses. 1 relay required a manual IPv6 address update due to #21460 (moved).

If everyone upgrades, we should be back to around ~160 fallbacks in a week or two when the list is regenerated.

Replying to teor:

2 relays had added IPv6 addresses. 1 relay required a manual IPv6 address update due to #21460 (moved).

All these 3 relays will require their IPv6 addresses to be added manually. I'll do that when I get confirmation from their operators that the IPv6 addresses are permanent.

8C00FA7369A7A308F6A137600F0FA07990D9D451 confirmed [2001:bc8:225f:142:6c69:7461:7669:73]:9001

9BF04559224F0F1C3C953D641F1744AF0192543A confirmed [2a01:4f8:190:34f0::2]:4223

I need to add both these addresses to the whitelist, and manually add them to the generated fallback list.

Trac:
Keywords: N/A deleted, TorCoreTeam201703 added

I also need to remove DF3AED4322B1824BF5539AE54B2D1B38E080FF05 from the list, as the operator has opted-out.

Trac:
Milestone: Tor: 0.3.0.x-final to Tor: 0.3.1.x-final

I also need to remove F80FDE27EFCB3F6A7B4E2CC517133DBFFA78BA2D and 4FEE77AFFD157BBCF2D896AE417FBF647860466C as they have been shut down.

This won't make the 0.3.1 freeze, and isn't needed yet, because 90% of the fallback directories are still running, which gives us 99.9% client success when trying 3 fallbacks.

https://consensus-health.torproject.org/graphs.html#fallbackdirgraphs

Trac:
Milestone: Tor: 0.3.1.x-final to Tor: 0.3.2.x-final

The operator of 5.35.251.247 will replace it with 91.250.84.156, possibly with new keys. We can handle this using the normal rebuild process.

We also want to remove kitten2=2EBD117806EE43C3CC885A8F1E4DC60F207E7D3E due to #22268 (moved).

Trac:
Parent: N/A to #21564 (moved)

The whitelist and blacklist changes here are in my branch fallback-inputs, and will be dealt with in #21121 (moved).

I rebuilt the list rather than dropping old fallbacks. I'll handle that in #21564 (moved).

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

closed

changed time estimate to 8h

mentioned in issue #21564 (moved)

moved to tpo/core/tor#21283 (closed)