On 10 October 2011 18:37, Eli Zaretskii <[email protected]> wrote: > It seems something's wrong with savannah's bzr server. First, a user > reported that he is unable to "bzr pull", see > > http://lists.gnu.org/archive/html/emacs-devel/2011-10/msg00283.html > > I tried a fresh "bzr branch" on fencepost, and it worked for me, both > with bzr:// and nosmart+bzr://. > > However, on a different system, not so close to the server as > fencepost, I do see a similar problem, albeit after 59:58 minutes of > continuous work (could it be there's a 1-hour timeout there?):
That number does seem like more than a coincidence. jameinel tells me he recalls hearing of a one-hour timeout on savannah? The traceback just indicates the client saw the server drop the connection while it was in the middle of pulling; unfortunately there's nothing more. Getting a log with -Dhpss will tell you more. m
