On Mon, Jun 10, 2019 at 2:12 AM Guanghao Zhang <[email protected]> wrote:
> > > > We usually > > aggregate them so the release notes has all release notes for all > releases > > (see 2.1.x releases). > > > I checked the release notes for branch-2.1. It only contains the issues for > 2.1.*. So for 2.2.0, it should only contain the issues for 2.2.0? > > I thought it had all versions, just like hbase-2.0 branch and releases before 2.0.0. We are a little inconsistent here then. I started a separate thread to discuss what range for CHANGES.md. Thanks Guanghao, S > 张铎(Duo Zhang) <[email protected]> 于2019年6月9日周日 下午8:43写道： > > > +1(binding) > > Built from src: OK > > Checked sums & sigs: All matched > > Rat check: Passed > > Ran all UTs(openjdk8u202): Always, several UTs in hbase-server failed > with > > -PrunAllTests but passed when running sequentially inside the > hbase-server > > module. > > > > The compatibility report: Let me carry the notes from the vote thread for > > RC4 > > HBaseSnapshotException and RestoreSnapshotException: Just changed the > > parameter type from Exception to Throwable, should be fine. > > QuotaSettings: The constructor is protected and this is class is not > > supposed to be implemented by users > > QuotaSettingsFactory: The fromThrottle method is protected > > MultiRowRangeFilter.RowRange: Just moved the Comparable related stuff > to > > its super class, should be fine > > Permission: Used to be IA.Private so no problem > > LossyCounting: Marked as IA.Private in HBASE-21991 > > > > Artem Ervits <[email protected]> 于2019年6月7日周五 下午7:17写道： > > > > > +1 (non-binding) > > > > > > * Signature: ok > > > * Checksum : ok > > > * Rat check (1.8.0_212): ok > > > - mvn clean apache-rat:check > > > * Built from source (1.8.0_212): ok > > > - mvn clean install -DskipTests > > > > > > * Unit tests pass (1.8.0_212): failed > > > - mvn verify -P runAllTests > > > tests ran for 48 hours before my machine just shut down. > > > > > > Java 1M writes: OK > > > HBase shell: OK > > > Logs: OK > > > UI: OK > > > bulk loaded some data: OK > > > upgraded existing 2.1.5RC1 environment on a pseudodistributed 2.8.5 > > Hadoop > > > cluster: OK > > > binary artifact revision is Unknown > > > > > > > > > On Thu, Jun 6, 2019 at 6:19 PM Tak-Lon (Stephen) Wu < > [email protected]> > > > wrote: > > > > > > > +1 (non-binding) > > > > > > > > * Signature: ok > > > > * Checksum : ok > > > > * Rat check (1.8.0_212): ok > > > > - mvn clean apache-rat:check > > > > * Built from source (1.8.0_212): ok > > > > - mvn clean install -DskipTests > > > > * Unit tests pass (1.8.0_212): ok > > > > - mvn test -P runAllTests > > > > * On cluster tests with hadoop-2.8.5 (1.8.0_201): ok > > > > - basic operations > > > (create/put/get/scan/flush/list/disable/drop) > > > > - Run read/write LTT tests with 1m keys passed > > > > - hbase ltt -tn TestTable2 -write 10:16 -num_keys > 1000000 > > > > - hbase hbase ltt -tn TestTable2 -read 100 -num_keys > > 1000000 > > > > - ITBLL 1m passed > > > > - hbase > > > > org.apache.hadoop.hbase.test.IntegrationTestBigLinkedList -m > > > serverKilling > > > > loop 1 24 1000000 ${RANDOM} 1 > > > > > > > > *### Normal build* > > > > bash hbase-vote.sh -s > > > > https://dist.apache.org/repos/dist/dev/hbase/2.2.0RC5/ > > > > > > > > Two tests below reran and passed the second time, here > > > > TestCompactionWithCoprocessor is not part of flaky tests in > > > > > > > > > > > > > > https://builds.apache.org/view/H-L/view/HBase/job/HBase-Find-Flaky-Tests/job/branch-2.2/lastSuccessfulBuild/artifact/dashboard.html > > > > but IMO it's not a big issue because it seems an race condition issue > > at > > > > the point of assertion. > > > > > > > > TestProcedureAdmin.testAbortProcedureInterruptedNotAllowed > > > > TestCompactionWithCoprocessor>TestCompaction.testStopStartCompaction > -> > > > > this one seems not showing up on the nightly > > > > > > > > ERROR of TestCompactionWithCoprocessor > > > > TestCompaction.testStopStartCompaction:390 expected:<1> but was:<2> > > > > > > > > *### Check API compatibility report > > > > < > > > > > > > > > > https://dist.apache.org/repos/dist/dev/hbase/2.2.0RC5/api_compare_2.2.0RC5_to_2.1.4.html > > > > >* > > > > > > > > > > > > it seems like `source % of Compatibility` is 82.1% (still learning > how > > > this > > > > means to vote), but it's mostly related to those removed and added in > > the > > > > shaded > > > > library hbase-shaded-client-byo-hadoop-2.1.4 and > > > > hbase-shaded-mapreduce-2.1.4 > > > > > > > > On Thu, Jun 6, 2019 at 7:47 AM Stack <[email protected]> wrote: > > > > > > > > > On Thu, Jun 6, 2019 at 7:04 AM Guanghao Zhang <[email protected]> > > > > wrote: > > > > > > > > > > > ... > > > > > > > > > > > > > "All others ar dropped" means? It was generated by yetus. I > > > previously > > > > > > committed a wrong RELEASENOTES.md which is too big. So I replaced > > it > > > > with > > > > > > the new one which was generated by yetus. > > > > > > > > > > > > > > > > > > > > > No hurry Guanghao. Can wait till your holiday is done. > > > > > > > > > > Yetus generates the release notes/changes for one release. We > usually > > > > > aggregate them so the release notes has all release notes for all > > > > releases > > > > > (see 2.1.x releases). I see the concurrent discussion going on in > > > another > > > > > thread on release notes so maybe this is the way forward (though it > > > good > > > > > being able to see changes over time I'd say). > > > > > > > > > > S > > > > > > > > > > > > > > >
