On Mon, Apr 28, 2014 at 11:06 AM, George Kadianakis desnacked@riseup.netwrote:

Kevin P Dyer kpdyer@gmail.com writes:

...

Hi Mike,

It looks like [1] broke FTE [2].

Can you hold off on pushing any releases public, until we resolve #11629?

George - Can we revert obfsproxy back to 0.2.4, to give me time to sort this out?

Sure, that seems reasonable for now. Could you ensure that the TBB team is aware of this downgrade?

Actually, looks like Mike is already on the case! [2]

BTW, on the long term, I'm not sure if it's a good idea to assume that the internal networking API of obfsproxy is stable. It shouldn't change frequently, but it wouldn't surprise me if it will need to change again at some point.

Would it make sense to include in FTE the parts of obfsproxy you need? Of course, this has other issues, like the fact that we will need to fix obfsproxy bugs in FTE too.

I think this will be the best short-term solution, until we find a better way to refactor the common components into pyptlib or some other shared library.

Thanks and sorry for the troubles!

No worries.

-Kevin

[1] https://github.com/kpdyer/tor-browser-bundle/commit/778e6d65b14f7434c7dc9d0e... [2] https://gitweb.torproject.org/builders/tor-browser-bundle.git/commit/793a4cd...