Hi,
According to Roger's blog post [1], and to various sources on IRC
(quoted below), we should ship Tor 0.2.4..17-rc or newer in Tails
0.20.1. Roger writes "Step two is to get more users using the NTor
circuit-level handshake, which is new in Tor 0.2.4".
I trust these opinions, and I have been happily using Tor 0.2.4 for
a few months, so I'm not *that* scared to move to a major new version
of Tor in a point-release (without any RC). I'm scared, still, but it
does not look like we have much choice here. The good news is that we
have 10 days between the freeze and the final release, so if we have
some unofficial (and untested) RC image tested in the wild, then
there's room to gain some confidence.
Using the tor-0.2.4.x-squeeze distribution at deb.torproject.org
should do the trick.
Thoughts? Any adverse effects I didn't think of?
The 0.20.1 freeze is in four days from now, so I'll make the changes
tomorrow unless someone objects -- potential reviewers, stay tuned!
[1]
https://blog.torproject.org/blog/how-to-handle-millions-new-tor-clients
nickm: "I would _generally_ recommend the latest 0.2.4 for nearly
anything, but I don't have a good sense of how to keep stuff like
tails stable. maybe armadev has thoughts"
Lunar^: "go for 0.2.4. with the patches that are being deployed by the
relay operators right now, 0.2.4 will really work better"
Cheers,
--
intrigeri
| GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc
| OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc