It would be great if you can share the details, Jim. I ran the entire live test suite and there was only one failure in the metrics test... I'm curious about the failures you've seen. Many thanks for taking the time and helping with the release validation!
@demobox, +1 to file an issue to properly track this. Do you know if with our current plan we could have some Windows build in CloudBees? On Nov 29, 2017 4:10 AM, "Jim Spring" <jmspr...@gmail.com> wrote: I've seen a mix of azurecompute-arm failures (sorry, I'll enumerate tomorrow), but nothing to hold up the release. Small things like password requirements changing. -jim On Tue, Nov 28, 2017 at 5:52 PM, Andrew Phillips <aphill...@qrmedia.com> wrote: >> However, I am unclear on why the test fails since the filesystem blobstore >> should convert user input / >> to backend fs separator \ on Windows. That is to say, code which runs on >> Linux and Windows *should* >> behave identically, although in previous releases they did not. > > > PR to prevent the test failure here: > https://github.com/jclouds/jclouds/pull/1162 > > I'm not sure this change is correct, given that (as Andrew G pointed out) > the code is actually *not* expected to fail on Windows (and thus the test is > expected to pass). > > From a quick local test, this test was already failing in 2.0.2 (when it was > first added [1]), so the original implementation may simply have been > incomplete. Since we have't seen any related comments or issues, as far as I > am aware, I'd agree that this is probably not a reason to stop the release. > > Would it make sense to try to fix the underlying problem, or at least open a > bug to track this, though? > > Regards > > ap > > [1] > https://github.com/jclouds/jclouds/commit/6452960c72a5f8247acae12481b94c df9d8e4218#diff-9592ea1a2837c5f2cd419db6bde4fd5d
