+1 * Signature: ok * Checksum : ok * Rat check (17.0.19): ok - mvn clean apache-rat:check -D surefire.Xmx=10g -D surefire.rerunFailingTestsCount=3 -D surefire.firstPartForkCount=0.25C -D surefire.secondPartForkCount=0.25C * Built from source (17.0.19): ok - mvn clean install -D surefire.Xmx=10g -D surefire.rerunFailingTestsCount=3 -D surefire.firstPartForkCount=0.25C -D surefire.secondPartForkCount=0.25C -DskipTests * Unit tests pass (17.0.19): failed - mvn clean package -P runAllTests -D surefire.Xmx=12g -Dsurefire.timeout=1800 -D surefire.rerunFailingTestsCount=3 -D surefire.firstPartForkCount=0.25C -D surefire.secondPartForkCount=0.25C -Dsurefire.rerunFailingTestsCount=3 * Basic Web UI: ok * Run LTT with 1M rows: ok * run standalone ITBLL: ok
-Stephen On Sat, Jul 25, 2026 at 7:47 AM 张铎(Duo Zhang) <[email protected]> wrote: > I've finished a rolling upgrade test from 2.6.6 to 3.0.0RC0. > > I set up two 2.6.6 clusters, set up a peer on each cluster to > replicate all tables to the other cluster. > Both clusters have 2 masters and 5 regionservers, and share the same > zookeeper and hdfs cluster. > I created two tables, ltt and ltt2, which have 150 and 100 regions. > I used LoadTestTool on cluster A to write 100M rows to the ltt table, > and also used LoadTestTool on cluster B to write 100M rows to the ltt2 > table. > > During the loading, I rolling upgraded the two clusters, first cluster > A, and then cluster B. > I used the standard rolling upgrade process, first upgrading backup > master, then active master, and then upgrade regionservers one by one, > unloading and then loading all the regions on the upgraded region > servers. > > Both upgrade finished successfully, and after the two LTT write > finished, I used LTT read to read both ltt and ltt2 table on both > clusters, all finished without errors. > I also used shell to count ltt and ltt2 tables on both clusters, all 4 > tables have 100M rows. > > Dávid Paksy <[email protected]> 于2026年7月22日周三 20:36写道： > > > > > +1 (non-binding) > > I checked: > > > > * Signature: ok > > * Checksum : ok > > * LICENSE and NOTICE: In place > > * Rat check (17.0.19): ok > > - mvn clean apache-rat:check -D skipTests > > * Built from source (17.0.19): ok > > - mvn clean install -D skipTests -DskipTests > > > > Started a standalone cluster - OK > > Checked logs and web UI pages, looks good. > > > > Run these shell commands - OK > > > > create_namespace 'nj' > > create 'nj:test', 'cf', {NUMREGIONS => 3, SPLITALGO => 'HexStringSplit'} > > describe 'nj:test' > > put 'nj:test', 'r1', 'cf:1', 'v1' > > put 'nj:test', '1', 'cf:1', 'v1' > > put 'nj:test', 'z00', 'cf:1', 'v3' > > scan 'nj:test' > > get 'nj:test', 'r1' > > flush 'nj:test' > > major_compact 'nj:test' > > get_splits 'nj:test' > > split 'nj:test', '0' > > get_splits 'nj:test' > > > > Executed LoadTestTool - OK > > bin/hbase ltt -num_keys 10000 -write 100:100 -num_tables 10 -multiput > > > > > > Xiao Liu <[email protected]> ezt írta (időpont: 2026. júl. 21., K, > > 10:43): > > > > > +1 (non-binding) > > > I checked: > > > > > > * Signature: OK > > > * Checksum: OK > > > * LICENSE and NOTICE: In place > > > * Rat check (17.0.12): OK > > > - mvn clean apache-rat:check > > > * Build from source (17.0.12): OK > > > - mvn clean install -DskipTests > > > > > > > > > Start a mini cluster in standalone mode, succeeded. > > > Check logs and web pages, it looks good. > > > > > > Run these shell commands, looks fine. > > > create 'test', 'cf' > > > list 'test' > > > describe 'test' > > > put 'test', 'row1', 'cf:a', 'value1' > > > put 'test', 'row2', 'cf:b', 'value2' > > > put 'test', 'row3', 'cf:c', 'value3' > > > scan 'test' > > > get 'test', 'row1' > > > flush 'test' > > > major_compact 'test' > > > scan 'test' > > > disable 'test' > > > enable 'test' > > > disable 'test' > > > drop 'test' > > > > > > > > > Run LTT to load 10000 rows, Passed > > > ./bin/hbase ltt -write 20:1024 -num_keys 10000 > > > > > > Run these shell commands, looks fine > > > flush 'cluster_test' > > > major_compact 'cluster_test' > > > scan 'cluster_test', {LIMIT=>10} > > > count 'cluster_test' > > > > > > > > > > > > On 2026/07/20 15:20:45 Duo Zhang wrote: > > > > I've passed a 10B ITBLL run on this RC. > > > > > > > > I will do some upgrading tests too. > > > > > > > > Duo Zhang <[email protected]> 于2026年7月18日周六 17:00写道： > > > > > > > > > > Please vote on this Apache hbase release candidate, > > > > > hbase-3.0.0RC0 > > > > > > > > > > The VOTE will remain open for at least 72 hours. > > > > > > > > > > [ ] +1 Release this package as Apache hbase 3.0.0 > > > > > [ ] -1 Do not release this package because ... > > > > > > > > > > The tag to be voted on is 3.0.0RC0: > > > > > > > > > > https://github.com/apache/hbase/tree/3.0.0RC0 > > > > > > > > > > This tag currently points to git reference > > > > > > > > > > da418afa6973bf4222231c1d233410d8c18ac7ec > > > > > > > > > > The release files, including signatures, digests, as well as > CHANGES.md > > > > > and RELEASENOTES.md included in this RC can be found at: > > > > > > > > > > https://dist.apache.org/repos/dist/dev/hbase/3.0.0RC0/ > > > > > > > > > > Maven artifacts are available in a staging repository at: > > > > > > > > > > > > > > https://repository.apache.org/content/repositories/orgapachehbase-1620/ > > > > > > > > > > Artifacts were signed with the 0x9AD2AE49 key which can be found > in: > > > > > > > > > > https://downloads.apache.org/hbase/KEYS > > > > > > > > > > 3.0.0 is the first release for our 3.x major release line, which > > > includes the > > > > > following big feature/changes: > > > > > Drop JDK8/JDK11 support, now the minimum JDK version is 17 > > > > > Drop hadoop 2.x support > > > > > Synchronous Replication > > > > > OpenTelemetry Tracing > > > > > Distributed MOB Compaction > > > > > Backup and Restore > > > > > Move RSGroup balancer to core > > > > > Reimplement sync client on async client > > > > > CPEPs on shaded protobuf > > > > > Move the logging framework from log4j to log4j2 > > > > > Decouple region replication and general replication framework, > and > > > > > also make region replication can work when SKIP_WAL is used > > > > > A new file system based replication peer storage > > > > > Used hbase table instead of zookeeper for tracking hbase > replication > > > queue > > > > > Yield SCP and TRSP to improve MTTR when restarting large cluster > > > > > Upgrade to junit5 > > > > > > > > > > To reduce the artifacts size, now we will not include tests related > > > jars in > > > > > our tarball. If you want to run hbase-it by yourself, you need to > > > build with > > > > > assemblyForTesting profile by your own. See HBASE-29192 for more > > > details. > > > > > > > > > > And this time we will not include CHANGES.md and RELEASENOTE.md > > > > > in our source code, you can find it on the download site. For > getting > > > these > > > > > two files for old releases, please go to > > > > > > > > > > https://archive.apache.org/dist/hbase/ > > > > > > > > > > To learn more about Apache hbase, please see > > > > > > > > > > http://hbase.apache.org/ > > > > > > > > > > Thanks, > > > > > Your HBase Release Manager > > > > > > > >
