Hi Russell, > I think that preventing random failures from Debian build infrastructure > is the best I can do for the time being
Well, not if the upstream library code itself is broken wrt to the timezones and its not just some peculiarity of the test. In fact, if the code is broken then breaking in Debian is actually "correct", from a funny point of view. Anyway, worth checking with upstream before patching over the issue within Debian just in case they never respond and this (potentially) falls through the gaps IMHO. :) Regards, -- ,''`. : :' : Chris Lamb `. `'` la...@debian.org / chris-lamb.co.uk `-