Tor 0.2.9.5-alpha is released, with numerous bugfixes

by nickm | November 8, 2016

Tor 0.2.9.5-alpha fixes numerous bugs discovered in the previous alpha version. We believe one or two probably remain, and we encourage everyone to test this release.
You can download the source from the usual place on the website. Packages should be available over the next several days. Remember to check the signatures!
Please note: This is an alpha release. You should only try this one if you are interested in tracking Tor development, testing new features, making sure that Tor still builds on unusual platforms, or generally trying to hunt down bugs. If you want a stable experience, please stick to the stable releases.
Below are the changes since 0.2.9.4-alpha.

Changes in version 0.2.9.5-alpha - 2016-11-08

  • Major bugfixes (client performance):
    • Clients now respond to new application stream requests immediately when they arrive, rather than waiting up to one second before starting to handle them. Fixes part of bug 19969; bugfix on 0.2.8.1-alpha.
  • Major bugfixes (client reliability):
    • When Tor leaves standby because of a new application request, open circuits as needed to serve that request. Previously, we would potentially wait a very long time. Fixes part of bug 19969; bugfix on 0.2.8.1-alpha.

 

  • Major bugfixes (download scheduling):
    • When using an exponential backoff schedule, do not give up on downloading just because we have failed a bunch of times. Since each delay is longer than the last, retrying indefinitely won't hurt. Fixes bug 20536; bugfix on 0.2.9.1-alpha.
    • If a consensus expires while we are waiting for certificates to download, stop waiting for certificates.
    • If we stop waiting for certificates less than a minute after we started downloading them, do not consider the certificate download failure a separate failure. Fixes bug 20533; bugfix on 0.2.0.9-alpha.
    • Remove the maximum delay on exponential-backoff scheduling. Since we now allow an infinite number of failures (see ticket 20536), we must now allow the time to grow longer on each failure. Fixes part of bug 20534; bugfix on 0.2.9.1-alpha.
    • Make our initial download delays closer to those from 0.2.8. Fixes another part of bug 20534; bugfix on 0.2.9.1-alpha.
    • When determining when to download a directory object, handle times after 2038 if the operating system supports them. (Someday this will be important!) Fixes bug 20587; bugfix on 0.2.8.1-alpha.
    • When using exponential backoff in test networks, use a lower exponent, so the delays do not vary as much. This helps test networks bootstrap consistently. Fixes bug 20597; bugfix on 20499.
  • Minor features (geoip):
    • Update geoip and geoip6 to the November 3 2016 Maxmind GeoLite2 Country database.
  • Minor bugfixes (client directory scheduling):
    • Treat "relay too busy to answer request" as a failed request and a reason to back off on our retry frequency. This is safe now that exponential backoffs retry indefinitely, and avoids a bug where we would reset our download schedule erroneously. Fixes bug 20593; bugfix on 0.2.9.1-alpha.
  • Minor bugfixes (client, logging):
    • Remove a BUG warning in circuit_pick_extend_handshake(). Instead, assume all nodes support EXTEND2. Use ntor whenever a key is available. Fixes bug 20472; bugfix on 0.2.9.3-alpha.
    • On DNSPort, stop logging a BUG warning on a failed hostname lookup. Fixes bug 19869; bugfix on 0.2.9.1-alpha.
  • Minor bugfixes (hidden services):
    • When configuring hidden services, check every hidden service directory's permissions. Previously, we only checked the last hidden service. Fixes bug 20529; bugfix the work to fix 13942 in 0.2.6.2-alpha.
  • Minor bugfixes (portability):
    • Fix compilation with OpenSSL 1.1 and less commonly-used CPU architectures. Closes ticket 20588.
    • Use ECDHE ciphers instead of ECDH in tortls tests. LibreSSL has removed the ECDH ciphers which caused the tests to fail on platforms which use it. Fixes bug 20460; bugfix on 0.2.8.1-alpha.
    • Fix implicit conversion warnings under OpenSSL 1.1. Fixes bug 20551; bugfix on 0.2.1.1-alpha.
  • Minor bugfixes (relay bootstrap):
    • Ensure relays don't make multiple connections during bootstrap. Fixes bug 20591; bugfix on 0.2.8.1-alpha.
  • Minor bugfixes (relay):
    • Work around a memory leak in OpenSSL 1.1 when encoding public keys. Fixes bug 20553; bugfix on 0.0.2pre8.
    • Avoid a small memory leak when informing worker threads about rotated onion keys. Fixes bug 20401; bugfix on 0.2.6.3-alpha.
    • Do not try to parallelize workers more than 16x without the user explicitly configuring us to do so, even if we do detect more than 16 CPU cores. Fixes bug 19968; bugfix on 0.2.3.1-alpha.
  • Minor bugfixes (single onion services):
    • Start correctly when creating a single onion service in a directory that did not previously exist. Fixes bug 20484; bugfix on 0.2.9.3-alpha.
  • Minor bugfixes (testing):
    • Avoid a unit test failure on systems with over 16 detectable CPU cores. Fixes bug 19968; bugfix on 0.2.3.1-alpha.
  • Documentation:
    • Clarify that setting HiddenServiceNonAnonymousMode requires you to also set "SOCKSPort 0". Fixes bug 20487; bugfix on 0.2.9.3-alpha.
    • Module-level documentation for several more modules. Closes tickets 19287 and 19290.

Comments

Please note that the comment area below has been archived.

Anonymous (not verified) said:

November 08, 2016

Permalink

Though it's a highly technical matter, Tor community is doing good for society. Please keep it up.

Anonymous (not verified) said:

November 08, 2016

Permalink

Thank you so much for all your time and effort, everyone in this great project.
Is https://panopticlick.eff.org/ accurate for Tor Browser Bundle, or just mainstream browsers?
Specifically, how much privacy is lost by disabling all cookies, all javascript, images, two of the above, or all three of the above?
Please keep up the good work and don't let the public opinion against all things private get you down.

Anonymous (not verified) said:

November 09, 2016

Permalink

Any official statement concerning recent Europol and the Five Eyes capture of hidden services?

Anonymous (not verified) said:

November 09, 2016

Permalink

Im wondering how debugging an alpha version works with Tor.
Its all client based? (do I have to generate my own debugging framework?)
Or, do the network have capabilities to debug errors from alpha clients? If yes, are there "prefered debug (entry) nodes"?

Anonymous (not verified) said:

November 12, 2016

Permalink

Hello,

I am using Tor 6.0.5

Starting this week ie about November 8 2016 onwards Tor the slowest I have known even if I reload or use a new circuit with a new isp number or close down and restart. In other words just very slow no matter what! Is Tor being interferred with to slow it down?

cheers

Anonymous (not verified) said:

November 16, 2016

Permalink

Tor 0.2.9.5-alpha fixes numerous bugs discovered in the previous alpha version. We believe one or two probably remain, and we encourage everyone to test this release.

You can download the source from the usual place on the website. Packages should be available over the next several days. Remember to check the signatures!

Please note: This is an alpha release. You should only try this one if you are interested in tracking Tor development, testing new features, making sure that Tor still builds on unusual platforms, or generally trying to hunt down bugs. If you want a stable experience, please stick to the stable releases.

Below are the changes since 0.2.9.4-alpha.
thanksssssss

Anonymous (not verified) said:

November 16, 2016

Permalink

Something wrong with Tor network (~3 times slower):
Your network connection speed appears to have changed. Resetting timeout to 60s after 18 timeouts and 1000 buildtimes.