On Tue, Jul 30, 2013 at 07:25:14PM +0000, Animesh Chaturvedi wrote: > > Folks code freeze was on 7/29 and starting today we are moving ACS 42 into > limited updates in preparation for RC on 8/19. Code changes are now limited > to fixes to blocker and critical bugs only. I need to follow up on doc and > translations. > > > There is a tremendous effort going on since last couple of weeks to fix > issues found during the QA cycle. In last week alone 175 new issues were > created and 250 were resolved. The current unresolved issue count stands at > 300+ with 60 blocker and critical. > > Given the high number of blocker and critical and that QA is still filing > many defects I will follow what Chip did for 4.1 [1]. Up until when we cut > the first RC on (2013-08-19) committers should continue to check in fixes to > 4.2 branch for blocker and critical issues. Please do any fixes as cleanly as > possible (i.e., squash related commits and make sure to avoid messy merge > commits as much as possible). >
The best type of commit for keeping things clean is a cherry-pick. Merging into the release branch is messy to track as the RM. > > For Contributors if you have a fix that needs to go into 4.2, please email > the list with the subject tag [ACS42] and note the review that contains the > patch. I'll pick whichever I can, but will also rely on others with commit > rights to make it efficient. > > I will review all commits going into that branch, and will revert anything > that's outside the scope of that I outlined above (unless there is a > discussion and consensus on this list to include something else). > > There are large number of issues that are resolved but not verified yet. > Given that the number is huge with over 300 blocker and critical we need to > prioritize and verify the ones that were returned by developers as ( Invalid, > Incomplete, CanNotReproduce, Later, NotAProblem) first. I have created a > filter [2] to facilitate identifying these issues. This is no way means not > to verify other issues but it is to close on the ones that have the most > likelihood of being reopened during verification. > > [1] http://markmail.org/message/qkkxycablpsmogsx > [2] https://issues.apache.org/jira/issues/#?filter=12324570 > > > Thanks > Animesh >
