+1 on all. I've also been doing lots of testing on TRUNK this past week and things look good to me as well.
This coming week I will be finalizing HBASE-3053 (multiple masters in unit tests) and building out some new unit tests based on it so we can officially close out some of the master/failover JIRAs still open against 0.90 and also gain more confidence in the new code as we move towards release. I will also work on getting the latest HBCK onto trunk by Wednesday. JG > -----Original Message----- > From: saint....@gmail.com [mailto:saint....@gmail.com] On Behalf Of > Stack > Sent: Sunday, October 03, 2010 11:01 AM > To: HBase Dev List > Subject: Feature freeze for 0.90 on Wednesday, October 5th, with branch > soon after? > > I've been running cluster tests of TRUNK over the last week or so and > IMO things are looking pretty good. I think its time to cut the first > 0.90 release candidate. > > Toward that end, I'd like to suggest that we feature freeze TRUNK on > Wednesday, October 5th. After Wednesday, I'd suggest that there > should be no more commits on TRUNK unless a stabilizing bug fix or a > documentation addition. Soon after Wednesday -- Friday the 7th? -- > I'd like to cut a 0.90 branch from which we'll make the first HBase > 0.90.0 release candidate. Any objections or suggestions on how best > to bring the 0.90 release home? > > Features missing that we are hoping to get into a 0.90 include fixup > of replication to match the new master, fixup of the hbsck tool again > to suit new master, and the initial coprocessor commit. Are there any > others? > > Thanks, > St.Ack
