On Oct 11, 2012, at 10:04 PM, Edison Su <edison...@citrix.com> wrote:
> Both of these issues we found today are not blocker and the fixes are trivial. > We don't need another round test from QA team: the fix for bug CloudStack-316 > is agreed by Wido, Marcus and I, and tested by Wido, another bug fixes are > related to legal. Agreed. If we abort, we can immediately cut another release candidate and start another vote. No need for the test engineers in the community to due a full round of testing prior to seeing a new vote. I would, however, love any future vote to include test engineers personally voting based on testing the official RC artifacts! > Seems recasting another round voting is inevitable? > I'm going to leave this thread open until the morning to gather more opinions and comments. However, I think you are right. In all likelihood, I'll abort tomorrow morning and create a new RC and vote thread. > Sent from my iPhone > > On Oct 11, 2012, at 5:47 PM, "Simon Weller" <swel...@ena.com> wrote: > >> >> >> <On Thu, Oct 11, 2012 at 7:56 PM, Wido den Hollander <w...@widodh.nl> wrote: >>> >>> >>> On 10/11/2012 03:21 AM, Chip Childers (ASF) wrote: >>>> >>>> Hi All, >>>> >>>> I would like to call a vote for Apache CloudStack (Incubating) Release >>>> 4.0.0-incubating. >>>> >>>> Instructions for Validating and Testing the artifacts can be found here: >>>> >>>> https://cwiki.apache.org/confluence/display/CLOUDSTACK/CloudStack+4.0+test+procedure >>>> >>>> We encourage the whole community to download and test these release >>>> artifacts so that any critical issues can be resolved before the >>>> release is made. Everyone is free to vote on this release, so please >>>> give it a shot! >>> >>> >>> Sorry for stopping the +1 spree, but I'm going to vote -1. >>> >>> While testing the artifact I ran into CLOUDSTACK-316 (which I created). >>> >>> QA (nofi) only tested with cloudbr0 in their setup and never tested usage >>> with other traffic labels. >>> >>> After talking with Edison and Marcus we came up with commit >>> 513b680d96d07fd44479995ac5eb6358725c9421 which resolves it. >>> >>> The problem in my setup was that adding the host would fail and you'd have >>> to figure out what was going wrong. >>> >>> I understood what was going wrong, but this would/could scare a new user >>> away since a very simple use-case didn't work during the wizard. >>> >>> So my vote is -1 on this artifact. >>> >>> Wido >>> >>> Wido, >>> >>> Thanks for the detailed testing. >>> >>> IMO, I believe we would be best served aborting round 1 and >>> regenerating a new release artifact with the fix for your issue. I >>> would also include the fixes for at least CLOUDSTACK-314 (Citrix >>> license header remains in >>> test/integration/component/test_allocation_states.py) and >>> CLOUDSTACK-302 (New Features Are Added to ReleaseNotes). Both of >>> these are low risk changes, but important documentation and legal >>> issues. >>> >>> I'm not going to abort the vote immediately, but I would like to get >>> opinions from other community members on this topic. >>> >>> -chip >> >> This is a good catch. We have always used cloudbr0, so we didn't test this >> in our lab either. This definitely needs to be addressed as it's going to >> cause lots of pain for new KVM users. >> >> I'm changing my vote to -1. >> >> - Si >> >> <snip> >
