torsocks release pending: towards a release candidate
Hi, Nick and I have been working on a torsocks release. At this point, I think we're at the point where we want to declare a release candidate which if it has no blockers, we'll likely call it a release. I think we'll call it version 1.3 as that seems to have been the intended version. Here is the git repo for all pending changes: https://gitweb.torproject.org/torsocks.git/shortlog/refs/heads/pending-chang... I'll likely merge this pending-changes branch into master in the next day or three. I've gone through all of the torsocks bugs on the Google code site and closed them out. The few that were sorta valid I've decided to close as unsupported as they were either 0) working for some or 1) totally obscure use cases that we never explicitly wanted to support. I think this also includes all of the pending Debian fixes and thus Debian can use our mainline release without patches after the next release. If anyone is horrified, wants to suggest other patches for inclusion, or has anything else to say (related to Torsocks) please speak up! All the best, Jacob
On Sun, Jan 27, 2013 at 04:16:30AM +0000, Jacob Appelbaum wrote:
Hi,
Nick and I have been working on a torsocks release. At this point, I think we're at the point where we want to declare a release candidate which if it has no blockers, we'll likely call it a release. I think we'll call it version 1.3 as that seems to have been the intended version.
Here is the git repo for all pending changes:
https://gitweb.torproject.org/torsocks.git/shortlog/refs/heads/pending-chang...
I'll likely merge this pending-changes branch into master in the next day or three.
I've gone through all of the torsocks bugs on the Google code site and closed them out. The few that were sorta valid I've decided to close as unsupported as they were either 0) working for some or 1) totally obscure use cases that we never explicitly wanted to support.
I think this also includes all of the pending Debian fixes and thus Debian can use our mainline release without patches after the next release.
If anyone is horrified, wants to suggest other patches for inclusion, or has anything else to say (related to Torsocks) please speak up!
Does the new version of torsocks speak the optimistic data version of SOCKS that Tor now supports (start sending the data *before* you get back the "connection established" message from the OP)? https://trac.torproject.org/projects/tor/ticket/3711 - Ian
Ian Goldberg:
On Sun, Jan 27, 2013 at 04:16:30AM +0000, Jacob Appelbaum wrote:
Hi,
Nick and I have been working on a torsocks release. At this point, I think we're at the point where we want to declare a release candidate which if it has no blockers, we'll likely call it a release. I think we'll call it version 1.3 as that seems to have been the intended version.
Here is the git repo for all pending changes:
https://gitweb.torproject.org/torsocks.git/shortlog/refs/heads/pending-chang...
I'll likely merge this pending-changes branch into master in the next day or three.
I've gone through all of the torsocks bugs on the Google code site and closed them out. The few that were sorta valid I've decided to close as unsupported as they were either 0) working for some or 1) totally obscure use cases that we never explicitly wanted to support.
I think this also includes all of the pending Debian fixes and thus Debian can use our mainline release without patches after the next release.
If anyone is horrified, wants to suggest other patches for inclusion, or has anything else to say (related to Torsocks) please speak up!
Does the new version of torsocks speak the optimistic data version of SOCKS that Tor now supports (start sending the data *before* you get back the "connection established" message from the OP)?
Hi Ian, This version of Torsocks is the initial bug clean up that will lead us into fixing larger issues such as that one. We think this is a good idea as a few of those bugs are quite a lot of work and there were many outstanding patches scattered around the internet which we have now merged... All the best, Jake
On Sun, Jan 27, 2013 at 04:37:55PM +0000, Jacob Appelbaum wrote:
Hi Ian,
This version of Torsocks is the initial bug clean up that will lead us into fixing larger issues such as that one. We think this is a good idea as a few of those bugs are quite a lot of work and there were many outstanding patches scattered around the internet which we have now merged...
All the best, Jake
Sounds great. Out of curiosity, do we know what applications get most often used with torsocks? - Ian
Ian Goldberg:
On Sun, Jan 27, 2013 at 04:37:55PM +0000, Jacob Appelbaum wrote:
Hi Ian,
This version of Torsocks is the initial bug clean up that will lead us into fixing larger issues such as that one. We think this is a good idea as a few of those bugs are quite a lot of work and there were many outstanding patches scattered around the internet which we have now merged...
All the best, Jake
Sounds great.
Out of curiosity, do we know what applications get most often used with torsocks?
I use it with {ssh,netcat,telnet,tlsdate} on a daily basis. I used to use it with pidgin before I stopped using pidgin for anything other than OTR development. I plan to use it as a library for anonymous socket calls during future development - this should allow me to remove all calls to socket() or connect() from say, tlsdate. Robert and I started working on this a while ago but we were both sidetracked. All the best, Jake
Hi, Jacob Appelbaum wrote (27 Jan 2013 04:16:30 GMT) :
I think this also includes all of the pending Debian fixes
I confirm. Cheers, -- intrigeri | GnuPG key @ https://gaffer.ptitcanardnoir.org/intrigeri/intrigeri.asc | OTR fingerprint @ https://gaffer.ptitcanardnoir.org/intrigeri/otr.asc
I tried this branch on a PPC Debian stable (squeeze) machine, and it appears to work fine. I saw no DNS requests, see no warnings about res_ methods and at least one .onion worked. (And no compiler warnings, woo!) -- meejah
participants (4)
-
Ian Goldberg -
intrigeri -
Jacob Appelbaum -
meejah