No, we can't edit the comments so I suppose marking it private is the only alternative :-/
So, if unsetting the no_proxy variable doesn't work, the last resort is that python's urllib.proxy_bypass() is telling us to avoid the proxy for bzr.savannah.gnu.org ? That sounds quite suspicious and hard to believe... >From there, the alternatives are: - you try single-stepping with pdb if you have a basic python knowledge, the focus is on ProxyHandler.set_proxy method (which should be called for all requests whether or not a proxy exists) or just had some print statements there to better locate the root cause (the method is defined at line ~930 in _urllib2_wrappers.py), - wait for me to provide an instrumented branch for you to try (this will certainly be delayed a bit as I'm swamped with other bugs right now) I can't yet figure out what is happening by just reading the code :-/ None of the domains mentioned in the no_proxy variable should match bzr.savannah.gnu.org at least. -- Bzr 2.1.1 broken through HTTP proxy https://bugs.launchpad.net/bugs/586341 You received this bug notification because you are a member of Ubuntu Bugs, which is a direct subscriber. -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
