You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hbase.apache.org by zh...@apache.org on 2022/06/27 14:56:28 UTC
svn commit: r55310 [3/5] - /release/hbase/3.0.0-alpha-3/
Added: release/hbase/3.0.0-alpha-3/RELEASENOTES.md
==============================================================================
--- release/hbase/3.0.0-alpha-3/RELEASENOTES.md (added)
+++ release/hbase/3.0.0-alpha-3/RELEASENOTES.md Mon Jun 27 14:56:28 2022
@@ -0,0 +1,6660 @@
+
+<!---
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements. See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership. The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License. You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+-->
+# HBASE 3.0.0-alpha-3 Release Notes
+
+These release notes cover new developer and user-facing incompatibilities, important issues, features, and major improvements.
+
+
+---
+
+* [HBASE-26956](https://issues.apache.org/jira/browse/HBASE-26956) | *Major* | **ExportSnapshot tool supports removing TTL**
+
+ExportSnapshot tool support removing TTL of snapshot. If we use the ExportSnapshot tool to recover snapshot with TTL from cold storage to hbase cluster, we can set \`-reset-ttl\` to prevent snapshot from being deleted immediately.
+
+
+---
+
+* [HBASE-26167](https://issues.apache.org/jira/browse/HBASE-26167) | *Major* | **Allow users to not start zookeeper and dfs cluster when using TestingHBaseCluster**
+
+Introduce two new methods when creating a TestingHBaseClusterOption.
+
+public Builder useExternalDfs(String uri)
+public Builder useExternalZooKeeper(String connectString)
+
+Users can use these two methods to specify external zookeeper or HDFS cluster to be used by the TestingHBaseCluster.
+
+
+---
+
+* [HBASE-27108](https://issues.apache.org/jira/browse/HBASE-27108) | *Blocker* | **Revert HBASE-25709**
+
+HBASE-25709 caused a regression for scans that result in a large number of rows and has been reverted in this release.
+
+
+---
+
+* [HBASE-15519](https://issues.apache.org/jira/browse/HBASE-15519) | *Major* | **Add per-user metrics**
+
+Adds per-user metrics for reads/writes to each RegionServer. These metrics are exported by default. hbase.regionserver.user.metrics.enabled can be used to disable the feature if desired for any reason.
+
+
+---
+
+* [HBASE-26923](https://issues.apache.org/jira/browse/HBASE-26923) | *Minor* | **PerformanceEvaluation support encryption option**
+
+Add a new command line argument: --encryption to enable encryptopn in PerformanceEvaluation tool.
+
+Usage:
+ encryption Encryption type to use (AES, ...). Default: 'NONE'"
+
+Examples:
+ To run a AES encryption sequentialWrite:
+ $ bin/hbase org.apache.hadoop.hbase.PerformanceEvaluation --table=xxx --encryption='AES' sequentialWrite 10
+
+
+---
+
+* [HBASE-26983](https://issues.apache.org/jira/browse/HBASE-26983) | *Major* | **Upgrade JRuby to 9.3.4.0**
+
+Updates the version of JRuby that ships with HBase for the HBase shell to 9.3.4.0.
+
+Note that this changes the supported version of Ruby for HBase shell integration to Ruby 2.6.
+
+
+---
+
+* [HBASE-26993](https://issues.apache.org/jira/browse/HBASE-26993) | *Major* | **Make the new framework for region replication could work for SKIP\_WAL**
+
+In this issue we make the new region replication framework introduced by HBASE-26233 could also replicate for table which DURABILITY is Durability.SKIP\_WAL or for individual Mutation which is Durability.SKIP\_WAL.
+It is implemented entirely on the basis of HBASE-26233 and requires no additional configuration.
+
+
+---
+
+* [HBASE-26649](https://issues.apache.org/jira/browse/HBASE-26649) | *Major* | **Support meta replica LoadBalance mode for RegionLocator#getAllRegionLocations()**
+
+When setting 'hbase.locator.meta.replicas.mode' to "LoadBalance" at HBase client, RegionLocator#getAllRegionLocations() now load balances across all Meta Replica Regions. Please note, results from non-primary meta replica regions may contain stale data.
+
+
+---
+
+* [HBASE-27055](https://issues.apache.org/jira/browse/HBASE-27055) | *Minor* | **Add additional comments when using HBASE\_TRACE\_OPTS with standalone mode**
+
+hbase-env.sh has been updated with an optional configuration HBASE\_OPTS for standalone mode
+
+# export HBASE\_OPTS="${HBASE\_OPTS} ${HBASE\_TRACE\_OPTS} -Dotel.resource.attributes=service.name=hbase-standalone"
+
+
+---
+
+* [HBASE-26342](https://issues.apache.org/jira/browse/HBASE-26342) | *Major* | **Support custom paths of independent configuration and pool for hfile cleaner**
+
+Configure the custom hifle paths (under archive directory), e.g. data/default/testTable1,data/default/testTable2 by "hbase.master.hfile.cleaner.custom.paths".
+Configure hfile cleaner classes for the custom paths by "hbase.master.hfilecleaner.custom.paths.plugins".
+Configure the shared pool size of custom hfile cleaner paths by "hbase.cleaner.custom.hfiles.pool.size".
+
+
+---
+
+* [HBASE-27047](https://issues.apache.org/jira/browse/HBASE-27047) | *Minor* | **Fix typo for metric drainingRegionServers**
+
+Fix typo for metric drainingRegionServers. Change metric name from draininigRegionServers to drainingRegionServers.
+
+
+---
+
+* [HBASE-25465](https://issues.apache.org/jira/browse/HBASE-25465) | *Minor* | **Use javac --release option for supporting cross version compilation**
+
+When compiling with java 11 and above, we will use --release 8 to maintain java 8 compatibility.
+Also upgrade jackson to 2.13.1 because in hbase-thirdparty 4.1.0 we shade jackson 2.13.1.
+
+
+---
+
+* [HBASE-27024](https://issues.apache.org/jira/browse/HBASE-27024) | *Major* | **The User API and Developer API links are broken on hbase.apache.org**
+
+Upgrade maven-site-plugin to 3.12.0, maven-javadoc-plugin to 3.4.0.
+
+
+---
+
+* [HBASE-26986](https://issues.apache.org/jira/browse/HBASE-26986) | *Major* | **Trace a one-shot execution of a Master procedure**
+
+Individual executions of procedures are now wrapped in a tracing span. No effort is made to coordinate multiple executions back to a common PID.
+
+
+---
+
+* [HBASE-27013](https://issues.apache.org/jira/browse/HBASE-27013) | *Major* | **Introduce read all bytes when using pread for prefetch**
+
+Introduce optional flag hfile.pread.all.bytes.enabled for pread that must read full bytes with the next block header. This feature is specially helpful when users are running HBase with Blob storage like S3 and Azure Blob storage. Especially when using HBase with S3A and set fs.s3a.experimental.input.fadvise=sequential, it can save input stream from seeking backward that spent more time on storage connection reset time.
+
+
+---
+
+* [HBASE-26899](https://issues.apache.org/jira/browse/HBASE-26899) | *Major* | **Run spotless:apply**
+
+Run spotless:apply to format our code base.
+When viewing 'git blame', you may find a file has a large amount lines are modified by the commit of HBASE-26899, so you need to go back to the commit before this commit, and viewing 'git blame' again.
+
+
+---
+
+* [HBASE-26617](https://issues.apache.org/jira/browse/HBASE-26617) | *Major* | **Use spotless to reduce the pain on fixing checkstyle issues**
+
+Use spotless to format our java file and pom file, using the hbase\_eclipse\_formatter.xml and eclipse.importerorder file under our dev-support directory.
+On all branches, the ratchetFrom is set the commit just before the commit which introduces the spotless plugin, so we will only format the files which are touched in later commits.
+From now on, you should type mvn spotless:apply before generating a PR, the spotless plugin will fix most of the format issues for you.
+
+
+---
+
+* [HBASE-22349](https://issues.apache.org/jira/browse/HBASE-22349) | *Major* | **Stochastic Load Balancer skips balancing when node is replaced in cluster**
+
+StochasticLoadBalancer now respects the hbase.regions.slop configuration value as another factor in determining whether to attempt a balancer run. If any regionserver has a region count outside of the target range, the balancer will attempt to balance. Using the default 0.2 value, the target range is 80%-120% of the average (mean) region count per server. Whether the balancer will ultimately move regions will still depend on the weights of StochasticLoadBalancer's cost functions.
+
+
+---
+
+* [HBASE-26581](https://issues.apache.org/jira/browse/HBASE-26581) | *Minor* | **Add metrics around failed replication edits**
+
+Adds new metrics sink.failedBatches and source.failedBatches to replication metrics, allowing one to track failures on both sides of a replication stream. As with other source metrics, the new source.failedBatches is reported globally and by peer id.
+
+
+---
+
+* [HBASE-26807](https://issues.apache.org/jira/browse/HBASE-26807) | *Major* | **Unify CallQueueTooBigException special pause with CallDroppedException**
+
+Introduces a new config "hbase.client.pause.server.overloaded", deprecating old "hbase.client.pause.cqtbe". The new config specifies a special pause time to use when the client receives an exception from the server indicating that it is overloaded. Currently this applies to CallQueueTooBigException and CallDroppedException.
+
+
+---
+
+* [HBASE-26942](https://issues.apache.org/jira/browse/HBASE-26942) | *Minor* | **cache region locations when getAllRegionLocations()**
+
+Calling AsyncRegionLocator.getAllRegionLocations() will now add all of those returned region locations into the meta cache for the client.
+
+
+---
+
+* [HBASE-26891](https://issues.apache.org/jira/browse/HBASE-26891) | *Minor* | **Make MetricsConnection scope configurable**
+
+Adds a new "hbase.client.metrics.scope" config which allows users to define a custom scope for each Connection's metric instance. The default scope has also been changed to include the clusterId of the Connection, which should help differentiate metrics for processes connecting to multiple clusters. The scope is added to the ObjectName for JMX beans, so can be used to query for metrics for a particular connection. Using a custom scope might be useful in cases where you maintain separate Connections for writes vs reads. In that case you can set the scope appropriately and differentiate metrics for each.
+
+
+---
+
+* [HBASE-26067](https://issues.apache.org/jira/browse/HBASE-26067) | *Major* | **Change the way on how we track store file list**
+
+Introduces the StoreFileTracker interface to HBase. This is a server-side interface which abstracts how a Store (column family) knows what files should be included in that Store. Previously, HBase relied on a listing the directory a Store used for storage to determine the files which should make up that Store.
+
+After this feature, there are two implementations of StoreFileTrackers. The first (and default) implementation is listing the Store directory. The second is a new implementation which records files which belong to a Store within each Store. Whenever the list of files that make up a Store change, this metadata file will be updated.
+
+This feature is notable in that it better enables HBase to function on storage systems which do not provide the typical posix filesystem semantics, most importantly, those which do not implement a file rename operation which is atomic. Storage systems which do not implement atomic renames often implement a rename as a copy and delete operation which amplifies the I/O costs by 2x.
+
+At scale, this feature should have a 2x reduction in I/O costs when using storage systems that do not provide atomic renames, most importantly in HBase compactions and memstore flushes. See the corresponding section, "Store File Tracking", in the HBase book for more information on how to use this feature.
+
+
+---
+
+* [HBASE-26618](https://issues.apache.org/jira/browse/HBASE-26618) | *Minor* | **Involving primary meta region in meta scan with CatalogReplicaLoadBalanceSimpleSelector**
+
+When META replica LoadBalance mode is enabled at client-side, clients will try to read from one META region first. If META location is from any non-primary META regions, in case of errors, it will fall back to the primary META region.
+
+
+---
+
+* [HBASE-26245](https://issues.apache.org/jira/browse/HBASE-26245) | *Major* | **Store region server list in master local region**
+
+A typical HBase deployment on cloud is to store the data other than WAL on OSS, and store the WAL data on a special HDFS cluster. A common operation is to rebuild the cluster with fresh new zk cluster and HDFS cluster, with only the old rootdir on OSS. But it requires extra manual steps since we rely on the WAL directory to find out previous live region servers, so we can schedule SCP to bring regions online.
+After this issue, now it is possible to rebuild the cluster without extra manual steps as we will also store the previous live region servers in master local region.
+But notice that you'd better stop masters first and then region servers when rebuilding, as some tests show that if there are some pending procedures, the new clusters may still hang.
+
+
+---
+
+* [HBASE-26810](https://issues.apache.org/jira/browse/HBASE-26810) | *Major* | **Add dynamic configuration support for system coprocessors**
+
+Now dynamic configuration, configurign hbase-site and then call update\_all\_config, can also applied for the following properties.
+
+hbase.coprocessor.master.classes
+hbase.coprocessor.region.classes
+hbase.coprocessor.regionserver.classes
+hbase.coprocessor.user.region.classes
+
+
+---
+
+* [HBASE-25895](https://issues.apache.org/jira/browse/HBASE-25895) | *Major* | **Implement a Cluster Metrics JSON endpoint**
+
+Introduces a REST+JSON endpoint on the master info server port, which is used to render the "Region Visualizer" section of the Master Status page. When enabled, access to this API is gated by authentication only, just like the Master Status page. This API is considered InterfaceAudience.Private, can change or disappear without notice.
+
+
+---
+
+* [HBASE-26323](https://issues.apache.org/jira/browse/HBASE-26323) | *Major* | **Introduce a SnapshotProcedure**
+
+Introduce a snapshot procedure to snapshot tables. Set hbase.snapshot.procedure.enabled to false to disable snapshot procedure.
+
+
+---
+
+* [HBASE-26822](https://issues.apache.org/jira/browse/HBASE-26822) | *Major* | **Revert the changes on hbase-daemon.sh after switching to log4j2 properties file**
+
+Remove the HBASE\_ROOT\_LOGGER\_LEVEL, HBASE\_ROOT\_LOGGER\_APPENDER, HBASE\_SECURITY\_LOGGER\_LEVEL and HBASE\_SECURITY\_LOGGER\_APPENDER environment variables as log4j 2.17.2 supports set level and appenders at once.
+You can still use HBASE\_ROOT\_LOGGER and HBASE\_SECURITY\_LOGGER environment variables.
+
+
+---
+
+* [HBASE-26552](https://issues.apache.org/jira/browse/HBASE-26552) | *Major* | **Introduce retry to logroller to avoid abort**
+
+For retrying to roll log, the wait timeout is limited by "hbase.regionserver.logroll.wait.timeout.ms",
+and the max retry time is limited by "hbase.regionserver.logroll.retries".
+Do not retry to roll log is the default behavior.
+
+
+---
+
+* [HBASE-26723](https://issues.apache.org/jira/browse/HBASE-26723) | *Major* | **Switch to use log4j2.properties file to configure log4j2**
+
+Use log4j2.properties file instead of log4j2.xml.
+
+With LOG4J2-3341 in place, we could still make use of the old syntax where we could specify level and appenders at once, so we do not need to do splitting in our bash script and also make it less hurt for our users to upgrade.
+
+
+---
+
+* [HBASE-26640](https://issues.apache.org/jira/browse/HBASE-26640) | *Major* | **Reimplement master local region initialization to better work with SFT**
+
+Introduced a 'hbase.master.store.region.file-tracker.impl' config to specify the store file tracker implementation for master local region.
+
+If not present, master local region will use the cluster level store file tracker implementation.
+
+
+---
+
+* [HBASE-26673](https://issues.apache.org/jira/browse/HBASE-26673) | *Major* | **Implement a shell command for change SFT implementation**
+
+Introduced two shell commands for change table's or family's sft:
+
+change\_sft:
+ Change table's or table column family's sft. Examples:
+ hbase\> change\_sft 't1','FILE'
+ hbase\> change\_sft 't2','cf1','FILE'
+
+change\_sft\_all:
+ Change all of the tables's sft matching the given regex:
+ hbase\> change\_sft\_all 't.\*','FILE'
+ hbase\> change\_sft\_all 'ns:.\*','FILE'
+ hbase\> change\_sft\_all 'ns:t.\*','FILE'
+
+
+---
+
+* [HBASE-26742](https://issues.apache.org/jira/browse/HBASE-26742) | *Major* | **Comparator of NOT\_EQUAL NULL is invalid for checkAndMutate**
+
+The semantics of checkAndPut for null(or empty) value comparator is changed, the old match is always true.
+But we should consider that EQUAL or NOT\_EQUAL for null check is a common usage, so the semantics of checkAndPut for matching null is correct now.
+There is rare use of LESS or GREATER null, so keep the semantics for them.
+
+
+---
+
+* [HBASE-26688](https://issues.apache.org/jira/browse/HBASE-26688) | *Major* | **Threads shared EMPTY\_RESULT may lead to unexpected client job down.**
+
+Result#advance with empty cell list will always return false but not raise NoSuchElementException when called multiple times.
+This is a behavior change so it is an 'incompatible change', but since it will not introduce any compile error and the old behavior is 'broken', so we also fix it for current release branches.
+
+
+---
+
+* [HBASE-26714](https://issues.apache.org/jira/browse/HBASE-26714) | *Major* | **Introduce path configuration for system coprocessors**
+
+After HBASE-26714, users of hbase can now set system coprocessor with a path of the JAR file other than the files in default classpath. E.g. when setting hbase.coprocessor.region.classes/hbase.coprocessor.regionserver.classes/hbase.coprocessor.user.region.classes/hbase.coprocessor.master.classes, user can set a comma-separated list with the format of className\|priority\|path, where the path could be a file path of the supported filesystem.
+
+
+---
+
+* [HBASE-26473](https://issues.apache.org/jira/browse/HBASE-26473) | *Major* | **Introduce \`db.hbase.container\_operations\` span attribute**
+
+<!-- markdown --> Introduces an HBase-specific tracing attribute called `db.hbase.container_operations`. This attribute contains a list of table operations contained in the batch/list/envelope operation, allowing an operator to seek, for example, all `PUT` operations, even they occur inside of a `BATCH` or `COMPARE_AND_SET`.
+
+
+---
+
+* [HBASE-26587](https://issues.apache.org/jira/browse/HBASE-26587) | *Major* | **Introduce a new Admin API to change SFT implementation**
+
+Introduced two admin methods:
+
+void modifyTableStoreFileTracker(TableName, String) throws IOException;
+void modifyColumnFamilyStoreFileTracker(TableName, byte[], String) throws IOException;
+
+You can use these two methods to change the store file tracker implementation for a table or a family, where you just need to specify the final store file tracker implementation, and do not need to take care of the 'migration' intermediate state, at master side we will take care of it for you.
+
+
+---
+
+* [HBASE-26661](https://issues.apache.org/jira/browse/HBASE-26661) | *Major* | **Remove deprecated methods in MasterObserver**
+
+Removed the below methods in MasterObserver
+
+void preModifyTable(ObserverContext\<MasterCoprocessorEnvironment\>, TableName, TableDescriptor) throws IOException;
+
+void postModifyTable(ObserverContext\<MasterCoprocessorEnvironment\>, TableName, TableDescriptor) throws IOException;
+
+void preModifyTableAction(ObserverContext\<MasterCoprocessorEnvironment\>, TableName, TableDescriptor) throws IOException;
+
+void postCompletedModifyTableAction(ObserverContext\<MasterCoprocessorEnvironment\>, TableName, TableDescriptor) throws IOException;
+
+void preModifyNamespace(ObserverContext\<MasterCoprocessorEnvironment\>, NamespaceDescriptor) throws IOException;
+
+void postModifyNamespace(ObserverContext\<MasterCoprocessorEnvironment\>, NamespaceDescriptor) throws IOException;
+
+
+---
+
+* [HBASE-26469](https://issues.apache.org/jira/browse/HBASE-26469) | *Critical* | **correct HBase shell exit behavior to match code passed to exit**
+
+<!-- markdown -->
+User input handling has been refactored to make use of IRB sessions directly and the HBase shell attempts to ensure user provided calls to exit are able to convey failure and success.
+
+Those scripting use of the HBase shell should be aware that the exit code may have changed:
+ * a 0 code, or no code, passed to a call to exit from stdin in non-interactive mode will now exit cleanly. in prior versions this would have exited with an error and non-zero exit code. (note that in HBase 2.4.x this call will still result in a non-zero exit code)
+ * for other combinations of passing in an initialization script or reading from stdin with using the non-interactive flag, the exit code being 0 or non-0 should now line up with releases prior to 2.4, which is a change in behavior compared to versions 2.4.0 - 2.4.9.
+
+Please see the issue details for a table of expected exit codes.
+
+
+---
+
+* [HBASE-26631](https://issues.apache.org/jira/browse/HBASE-26631) | *Major* | **Upgrade junit to 4.13.2**
+
+Upgrade junit to 4.13.2 for addressing CVE-2020-15250.
+
+
+---
+
+* [HBASE-26233](https://issues.apache.org/jira/browse/HBASE-26233) | *Major* | **The region replication framework should not be built upon the general replication framework**
+
+In this issue we re-implement the 'Async WAL Replication' mechanism fo region replication, to decouple with the general replication framwork.
+
+Now, we extend the MVCC write entry to be able to carry an action once it is completed, so we can attach the WAL edits to this action and send them out directly after the write entry is completed, without touch the actual WAL system. So the special 'region\_replica\_replication' peer is also useless.
+
+We also introduce a new replicateToReplica method at region server side for receiving the WAL edits, which has a much simpler logic than the old replay method.
+
+After this issue, we do not need to treat META table specially in region replication, as said above, we do not depend on the general replication framework any more. But to avoid effecting the usage too much, the 'config hbase.region.replica.replication.catalog.enabled' is kept, you still need to enable this flag if you want to enable 'Async WAL Replication' support for META table.
+
+Rolling upgrading with 'Async WAL Replication enabled is supported. During rolling upgrading, the 'region\_replica\_replication' will be removed automatically after master is upgraded. And at region server side, if the new replicateToReplica method is not available when it tries to replicate to secondary replicas, it will fallback to use replay method automatically.
+
+Please related sections in our ref guide are:
+
+http://hbase.apache.org/book.html#async.wal.replication
+http://hbase.apache.org/book.html#async.wal.replication.meta
+http://hbase.apache.org/book.html#\_async\_wal\_replication\_for\_meta\_table\_as\_of\_hbase\_3\_0\_0
+
+
+---
+
+* [HBASE-26347](https://issues.apache.org/jira/browse/HBASE-26347) | *Major* | **Support detect and exclude slow DNs in fan-out of WAL**
+
+This issue provides the method to detect slow datanodes by checking the packets processing time of each datanode connected by the WAL. When a datanode is considered slow, the datanode will be added to an exclude cache on the regionserver, and every stream created will exclude all the cached slow datanodes in a configured period. The exclude logic cooperate with the log rolling logic, will react more sensitively to the lower slow datanodes, whatever there is hardware failure or hotspots.
+
+hbase.regionserver.async.wal.max.exclude.datanode.countï¼default 3ï¼and hbase.regionserver.async.wal.exclude.datanode.info.ttl.hour (default 6) means no more than 3 slow datanodes will be excluded on one regionserver, and the exclude cache for the slow datanodes is valid in 6 hours.
+
+There are two conditions used to determine whether a datanode is slow,
+1. For small packet, we just have a simple time limit(configured by hbase.regionserver.async.wal.datanode.slow.packet.process.time.millis, default 6s), without considering the size of the packet.
+
+2. For large packet, we will calculate the speed, and check if the speed (configured by hbase.regionserver.async.wal.datanode.slow.packet.speed.min.kbs, default 20KB/s) is too slow.
+
+The large and small split point is configured by hbase.regionserver.async.wal.datanode.slow.check.speed.packet.data.length.min (default 64KB).
+
+
+---
+
+* [HBASE-26079](https://issues.apache.org/jira/browse/HBASE-26079) | *Major* | **Use StoreFileTracker when splitting and merging**
+
+Changes MergeTableRegionProcedure and SplitTableRegionProcedure to use the underlying StoreFileTracker implementation to select the valid files on regions to be merged/split, rather than direct listing those from file system. It also changes merge/split commit logic in HRegionFileSystem, to add the links/refs on resulting split/merged regions to the StoreFileTracker.
+
+
+
+# HBASE 3.0.0-alpha-2 Release Notes
+
+These release notes cover new developer and user-facing incompatibilities, important issues, features, and major improvements.
+
+
+---
+
+* [HBASE-26606](https://issues.apache.org/jira/browse/HBASE-26606) | *Major* | **Upgrade log4j2 to 2.17.0**
+
+Upgrade log4j2 to 2.17.0 to address CVE-2021-45105.
+
+
+---
+
+* [HBASE-26542](https://issues.apache.org/jira/browse/HBASE-26542) | *Minor* | **Apply a \`package\` to test protobuf files**
+
+The protobuf structures used in test are all now scoped by the package name \`hbase.test.pb\`.
+
+
+---
+
+* [HBASE-26572](https://issues.apache.org/jira/browse/HBASE-26572) | *Major* | **Upgrade to log4j 2.16.0**
+
+Upgrade log4j2 to 2.16.0 to address CVE-2021-45046.
+
+
+---
+
+* [HBASE-26554](https://issues.apache.org/jira/browse/HBASE-26554) | *Minor* | **Introduce a new parameter in jmx servlet to exclude the specific mbean**
+
+introduce a new parameter 'excl' in jmx servlet to exclude the specific mbean
+
+
+---
+
+* [HBASE-26557](https://issues.apache.org/jira/browse/HBASE-26557) | *Major* | **log4j2 has a critical RCE vulnerability**
+
+Upgrade log4j2 to 2.15.0 for addressing CVE-2021-44228.
+
+
+---
+
+* [HBASE-26524](https://issues.apache.org/jira/browse/HBASE-26524) | *Major* | **Support remove coprocessor by class name via alter table command**
+
+This is an incompatible change for TableDescriptorBuilder#removeCoprocessor, if coprocessor does not exist and removeCoprocessor is being called, the new behavior is emitting an IllegalArgumentException instead of previously do nothing
+
+From now on, alter command introduces a new method table\_remove\_coprocessor that operator can remove table coprocessor by class name. Please see the usage as below
+
+\* remove a single table coprocessor
+alter 't1', METHOD =\> 'table\_remove\_coprocessor', CLASSNAME =\> 'org.apache.hadoop.hbase.coprocessor.SimpleRegionObserver'
+
+\* remove multiple coprocessors
+alter 't1', METHOD =\> 'table\_remove\_coprocessor', CLASSNAME =\> ['org.apache.hadoop.hbase.coprocessor.SimpleRegionObserver', 'org.apache.hadoop.hbase.coprocessor.Export']
+
+
+---
+
+* [HBASE-26529](https://issues.apache.org/jira/browse/HBASE-26529) | *Minor* | **Document HBASE-26524 to section of Dynamic Unloading**
+
+added the basic command usage of \`table\_remove\_coprocessor\` to the definition guide at src/main/asciidoc/\_chapters/cp.adoc.
+
+
+hbase\> alter 'users', METHOD =\> 'table\_remove\_coprocessor', CLASSNAME =\>
+ 'org.myname.hbase.Coprocessor.RegionObserverExample'
+
+
+---
+
+* [HBASE-26512](https://issues.apache.org/jira/browse/HBASE-26512) | *Major* | **Make timestamp format configurable in HBase shell scan output**
+
+HBASE-23930 changed the formatting of the timestamp attribute on each Cell as displayed by the HBase shell to be formatted as an ISO-8601 string rather that milliseconds since the epoch. Some users may have logic which expects the timestamp to be displayed as milliseconds since the epoch. This change introduces the configuration property hbase.shell.timestamp.format.epoch which controls whether the shell will print an ISO-8601 formatted timestamp (the default "false") or milliseconds since the epoch ("true").
+
+
+---
+
+* [HBASE-26484](https://issues.apache.org/jira/browse/HBASE-26484) | *Major* | **Update vote.tmpl in our create-release scripts to link KEYS from downloads.a.o instead of dist.a.o**
+
+Use downloads.a.o instead of dist.a.o in our vote.tmpl for releasing.
+
+
+---
+
+* [HBASE-26426](https://issues.apache.org/jira/browse/HBASE-26426) | *Major* | **Remove 'Review Board' section from the site**
+
+HBase now uses github PR for submitting and reviewing patches, so remove Review board link on our website.
+
+
+---
+
+* [HBASE-26363](https://issues.apache.org/jira/browse/HBASE-26363) | *Major* | **OpenTelemetry configuration support for per-process service names**
+
+<!-- markdown -->Each HBase process can have its own `service.name`, a value that can be completely customized by the operator. See the comment and examples in conf/hbase-env.sh.
+
+
+---
+
+* [HBASE-26362](https://issues.apache.org/jira/browse/HBASE-26362) | *Major* | **Upload mvn site artifacts for nightly build to nightlies**
+
+Now we will upload the site artifacts to nightlies for nightly build as well as pre commit build.
+
+
+---
+
+* [HBASE-26316](https://issues.apache.org/jira/browse/HBASE-26316) | *Minor* | **Per-table or per-CF compression codec setting overrides**
+
+It is now possible to specify codec configuration options as part of table or column family schema definitions. The configuration options will only apply to the defined scope. For example:
+
+ hbase\> create 'sometable', \\
+ { NAME =\> 'somefamily', COMPRESSION =\> 'ZSTD' }, \\
+ CONFIGURATION =\> { 'hbase.io.compress.zstd.level' =\> '9' }
+
+
+---
+
+* [HBASE-26329](https://issues.apache.org/jira/browse/HBASE-26329) | *Major* | **Upgrade commons-io to 2.11.0**
+
+Upgraded commons-io to 2.11.0.
+
+
+---
+
+* [HBASE-26186](https://issues.apache.org/jira/browse/HBASE-26186) | *Major* | **jenkins script for caching artifacts should verify cached file before relying on it**
+
+Add a '--verify-tar-gz' option to cache-apache-project-artifact.sh for verifying whether the cached file can be parsed as a gzipped tarball.
+Use this option in our nightly job to avoid failures on broken cached hadoop tarballs.
+
+
+---
+
+* [HBASE-26339](https://issues.apache.org/jira/browse/HBASE-26339) | *Major* | **SshPublisher will skip uploading artifacts if the build is failure**
+
+Now we will mark build as unstable instead of failure when the yetus script returns error. This is used to solve the problem that the SshPublisher jenkins plugin will skip uploading artifacts if the build is marked as failure. In fact, the test output will be more important when there are UT failures.
+
+
+---
+
+* [HBASE-26317](https://issues.apache.org/jira/browse/HBASE-26317) | *Major* | **Publish the test logs for pre commit jenkins job to nightlies**
+
+Now we will upload test\_logs.zip for our pre commit jobs to nightlies to save space on jenkins node. You can see the test\_logs.txt to get the actual url of the test\_logs.zip, or visit https://nightlies.apache.org/hbase directly to find the artifacts.
+
+
+---
+
+* [HBASE-26313](https://issues.apache.org/jira/browse/HBASE-26313) | *Major* | **Publish the test logs for our nightly jobs to nightlies.apache.org**
+
+Now we will upload test\_logs.zip for our nightly jobs to nightlies to save space on jenkins node. You can see the test\_logs.txt to get the actual url of the test\_logs.zip, or visit https://nightlies.apache.org/hbase directly to find the artifacts.
+
+
+---
+
+* [HBASE-26318](https://issues.apache.org/jira/browse/HBASE-26318) | *Major* | **Publish test logs for flaky jobs to nightlies**
+
+Now we will upload the surefire output for our flaky test jobs to nightlies to save space on jenkins node. You can see the test\_logs.txt to get the actual url of the surefire output, or visit https://nightlies.apache.org/hbase directly to find the artifacts.
+
+
+---
+
+* [HBASE-26259](https://issues.apache.org/jira/browse/HBASE-26259) | *Major* | **Fallback support to pure Java compression**
+
+This change introduces provided compression codecs to HBase as
+ new Maven modules. Each module provides compression codec support that formerly required Hadoop native codecs, which in turn relies on native code integration, which may or may not be available on a given hardware platform or in an operational environment. We now provide codecs in the HBase distribution for users whom for whatever reason cannot or do not wish to deploy the Hadoop native codecs.
+
+
+---
+
+* [HBASE-26321](https://issues.apache.org/jira/browse/HBASE-26321) | *Major* | **Post blog to hbase.apache.org on SCR cache sizing**
+
+Pushed blog at https://blogs.apache.org/hbase/entry/an-hbase-hdfs-short-circuit
+
+
+---
+
+* [HBASE-26270](https://issues.apache.org/jira/browse/HBASE-26270) | *Minor* | **Provide getConfiguration method for Region and Store interface**
+
+Provide 'getReadOnlyConfiguration' method for Store and Region interface
+
+
+---
+
+* [HBASE-26273](https://issues.apache.org/jira/browse/HBASE-26273) | *Major* | **TableSnapshotInputFormat/TableSnapshotInputFormatImpl should use ReadType.STREAM for scanning HFiles**
+
+HBase's MapReduce API which can operate over HBase snapshots will now default to using ReadType.STREAM instead of ReadType.DEFAULT (which is PREAD) as a result of this change. HBase developers expect that STREAM will perform significantly better for average Snapshot-based batch jobs. Users can restore the previous functionality (using PREAD) by updating their code to explicitly set a value of \`ReadType.PREAD\` on the \`Scan\` object they provide to TableSnapshotInputFormat, or by setting the configuration property "hbase.TableSnapshotInputFormat.scanner.readtype" to "PREAD" in hbase-site.xml.
+
+
+---
+
+* [HBASE-25288](https://issues.apache.org/jira/browse/HBASE-25288) | *Major* | **Make MasterRpcServices not extends RSRpcServices and also HMaster not extends HRegionServer**
+
+HMaster is not a sub class for HRegionServer now, and also MasterRpcServices is not a sub class for RSRpcServices.
+
+MasterRpcServices still implements the AdminService interface, but only updateConfiguration, getRegionInfo, clearSlowLogsResponses and getLogEntries can be called, for other methods in AdminService, you will get a DoNotRetryIOException when calling against master.
+
+
+---
+
+* [HBASE-25891](https://issues.apache.org/jira/browse/HBASE-25891) | *Major* | **Remove dependence on storing WAL filenames for backup**
+
+\* Remove dependence on storing WAL filenames for backup in backup:system meta table
+
+
+---
+
+* [HBASE-26276](https://issues.apache.org/jira/browse/HBASE-26276) | *Major* | **Allow HashTable/SyncTable to perform rawScan when comparing cells**
+
+Added --rawScan option to HashTable job, which allows HashTable/SyncTable to perform raw scans. If this property is omitted, it defaults to false. When used together with --versions set to a high value, SyncTable will fabricate delete markers to all old versions still hanging (not cleaned yet by major compaction), avoiding the inconsistencies reported in HBASE-21596.
+
+
+---
+
+* [HBASE-26230](https://issues.apache.org/jira/browse/HBASE-26230) | *Major* | **Start an in process HRegionServer in maintenance mode**
+
+When enabling maintenance mode, now we will start an in process region server instead of letting HMaster carry the system regions.
+
+
+---
+
+* [HBASE-26147](https://issues.apache.org/jira/browse/HBASE-26147) | *Major* | **Add dry run mode to hbase balancer**
+
+This change adds new API to the Admin interface for triggering Region balancing on a cluster. A new BalanceRequest object was introduced which allows for configuring a dry run of the balancer (compute a plan without enacting it) and running the balancer in the presence of RITs. Corresponding API was added to the HBase shell as well.
+
+
+---
+
+* [HBASE-26103](https://issues.apache.org/jira/browse/HBASE-26103) | *Major* | **conn.getBufferedMutator(tableName) leaks thread executors and other problems (for master branch)**
+
+Deprecate (unused) BufferedMutatorParams#pool and BufferedMutatorParams#getPool
+
+
+---
+
+* [HBASE-26204](https://issues.apache.org/jira/browse/HBASE-26204) | *Major* | **VerifyReplication should obtain token for peerQuorumAddress too**
+
+VerifyReplication obtains tokens even if the peer quorum parameter is used. VerifyReplication with peer quorum can be used for secure clusters also.
+
+
+---
+
+* [HBASE-26180](https://issues.apache.org/jira/browse/HBASE-26180) | *Major* | **Introduce a initial refresh interval for RpcConnectionRegistry**
+
+Introduced a 'hbase.client.bootstrap.initial\_refresh\_delay\_secs' config to control the first refresh delay for bootstrap nodes. The default value is 1/10 of periodic refresh interval.
+
+
+---
+
+* [HBASE-26173](https://issues.apache.org/jira/browse/HBASE-26173) | *Major* | **Return only a sub set of region servers as bootstrap nodes**
+
+Introduced a 'hbase.client.bootstrap.node.limit' config to limit the max number of bootstrap nodes we return to client. The default value is 10.
+
+
+---
+
+* [HBASE-26182](https://issues.apache.org/jira/browse/HBASE-26182) | *Major* | **Allow disabling refresh of connection registry endpoint**
+
+Set 'hbase.client.bootstrap.refresh\_interval\_secs' to -1 can disable refresh of connection registry endpoint.
+
+
+---
+
+* [HBASE-26212](https://issues.apache.org/jira/browse/HBASE-26212) | *Minor* | **Allow AuthUtil automatic renewal to be disabled**
+
+This change introduces a configuration property "hbase.client.keytab.automatic.renewal" to control AuthUtil, the class which automatically tries to perform Kerberos ticket renewal in client applications. This configuration property defaults to "true", meaning that AuthUtil will automatically attempt to renew Kerberos tickets per its capabilities. Those who want AuthUtil to not renew client Kerberos tickets can set this property to be "false".
+
+
+---
+
+* [HBASE-26174](https://issues.apache.org/jira/browse/HBASE-26174) | *Major* | **Make rpc connection registry the default registry on 3.0.0**
+
+Now RpcConnectionRegistry is the default ConnectionRegistry.
+
+
+---
+
+* [HBASE-26172](https://issues.apache.org/jira/browse/HBASE-26172) | *Major* | **Deprecate MasterRegistry**
+
+MasterRegistry is deprecated. Please use RpcConnectionRegistry instead.
+
+
+---
+
+* [HBASE-26193](https://issues.apache.org/jira/browse/HBASE-26193) | *Major* | **Do not store meta region location as permanent state on zookeeper**
+
+Introduce a new 'info' family in master local region for storing the location of meta regions.
+We will still mirror the location of meta regions to ZooKeeper, for backwards compatibility. But now you can also clean the meta location znodes(usually prefixed with 'meta-region-server') on ZooKeeper without mess up the cluster state. You can get a clean restart of the cluster, and after restarting, we will mirror the location of meta regions to ZooKeeper again.
+
+
+---
+
+* [HBASE-24842](https://issues.apache.org/jira/browse/HBASE-24842) | *Minor* | **make export snapshot report size can be config**
+
+Set new config snapshot.export.report.size to size at which you want to see reporting.
+
+
+---
+
+* [HBASE-24652](https://issues.apache.org/jira/browse/HBASE-24652) | *Minor* | **master-status UI make date type fields sortable**
+
+Makes RegionServer 'Start time' sortable in the Master UI
+
+
+---
+
+* [HBASE-26200](https://issues.apache.org/jira/browse/HBASE-26200) | *Major* | **Undo 'HBASE-25165 Change 'State time' in UI so sorts (#2508)' in favor of HBASE-24652**
+
+Undid showing RegionServer 'Start time' in ISO-8601 format. Revert.
+
+
+---
+
+* [HBASE-6908](https://issues.apache.org/jira/browse/HBASE-6908) | *Major* | **Pluggable Call BlockingQueue for HBaseServer**
+
+Can pass in a FQCN to load as the call queue implementation.
+
+Standardized arguments to the constructor are the max queue length, the PriorityFunction, and the Configuration.
+
+PluggableBlockingQueue abstract class provided to help guide the correct constructor signature.
+
+Hard fails with PluggableRpcQueueNotFound if the class fails to load as a BlockingQueue\<CallRunner\>
+
+Upstreaming on behalf of Hubspot, we are interested in defining our own custom RPC queue and don't want to get involved in necessarily upstreaming internal requirements/iterations.
+
+
+---
+
+* [HBASE-26196](https://issues.apache.org/jira/browse/HBASE-26196) | *Major* | **Support configuration override for remote cluster of HFileOutputFormat locality sensitive**
+
+Allow any configuration for the remote cluster in HFileOutputFormat2 that could be useful the different configuration from the job's configuration is necessary to connect the remote cluster, for instance, non-secure vs secure.
+
+
+---
+
+* [HBASE-26150](https://issues.apache.org/jira/browse/HBASE-26150) | *Major* | **Let region server also carry ClientMetaService**
+
+Introduced a RpcConnectionRegistry.
+
+Config 'hbase.client.bootstrap.servers' to set up the initial bootstrap nodes. The registry will refresh the bootstrap server periodically or on errors. Notice that, masters and region servers both implement the necessary rpc interface so you are free to config either masters or region servers as the initial bootstrap nodes.
+
+
+---
+
+* [HBASE-26160](https://issues.apache.org/jira/browse/HBASE-26160) | *Minor* | **Configurable disallowlist for live editing of loglevels**
+
+Adds a new hbase.ui.logLevels.readonly.loggers config which takes a comma-separated list of logger names. Similar to log4j configurations, the logger names can be prefixes or a full logger name. The log level of read only loggers cannot be changed via the logLevel UI or setlevel CLI. This is useful for securing sensitive loggers, such as the SecurityLogger used for audit logs.
+
+
+---
+
+* [HBASE-26154](https://issues.apache.org/jira/browse/HBASE-26154) | *Minor* | **Provide exception metric for quota exceeded and throttling**
+
+Adds "exceptions.quotaExceeded" and "exceptions.rpcThrottling" to HBase server and Thrift server metrics.
+
+
+---
+
+* [HBASE-26098](https://issues.apache.org/jira/browse/HBASE-26098) | *Major* | **Support passing a customized Configuration object when creating TestingHBaseCluster**
+
+Now TestingHBaseClusterOption support passing a Configuration object when building.
+
+
+---
+
+* [HBASE-26146](https://issues.apache.org/jira/browse/HBASE-26146) | *Minor* | **Allow custom opts for hbck in hbase bin**
+
+Adds HBASE\_HBCK\_OPTS environment variable to bin/hbase for passing extra options to hbck/hbck2. Defaults to HBASE\_SERVER\_JAAS\_OPTS if specified, or HBASE\_REGIONSERVER\_OPTS.
+
+
+---
+
+* [HBASE-26081](https://issues.apache.org/jira/browse/HBASE-26081) | *Major* | **Copy HBTU to hbase-testing-util, rename the HBTU related classes in hbase-server and mark them as IA.LimitedPrivate**
+
+Copy HBaseCommonTestingUtility, HBaseZKTestingUtility, HBaseTestingUtility, HBaseCluster, MiniHBaseCluster, StartMiniClusterOption to hbase-testing-util, and mark them as Deprecated. They will be removed in 4.0.0. End users should use TestingHBaseCluster for writing UTs in the future.
+
+And in hbase-server and related modules, these classes are renamed.
+HBaseCommonTestingUtility -\> HBaseCommonTestingUtil
+HBaseZKTestingUtility -\> HBaseZKTestingUtil
+HBaseTestingUtility -\> HBaseTestingUtil
+HBaseCluster -\> HBaseClusterInterface
+MiniHBaseCluster -\> SingleProcessHBaseCluster
+StartMiniClusterOption -\> StartTestingClusterOption
+And all these classes are marked as IA.LimitedPrivate("Phoenix"), and IS.Evolving. Except Phoenix, other end users should not use them any more.
+
+
+---
+
+* [HBASE-25986](https://issues.apache.org/jira/browse/HBASE-25986) | *Minor* | **Expose the NORMALIZARION\_ENABLED table descriptor through a property in hbase-site**
+
+New config: hbase.table.normalization.enabled
+
+Default value: false
+
+Description: This config is used to set default behaviour of normalizer at table level. To override this at table level one can set NORMALIZATION\_ENABLED at table descriptor level and that property will be honored. Of course, this property at table level can only work if normalizer is enabled at cluster level using "normalizer\_switch true" command.
+
+
+---
+
+* [HBASE-26090](https://issues.apache.org/jira/browse/HBASE-26090) | *Major* | **Remove the deprecated methods in Scan which should be removed in 3.0.0**
+
+The following methods have been removed:
+
+public Scan setTimeStamp(long timestamp);
+public Scan setSmall(boolean small);
+public boolean isSmall();
+
+The first method is just a typo, use setTimestamp instead. And for small scan, just use setLimit to control the number of records to return, and use setReadType to use pread when scanning. And for the one trip rpc optimizatoin, now the openScanner call will always return results if any, and it will close the scanner when limit is reached. So in this case the default implementation is already 'one rpc', we do not need introduce a special one any more.
+
+
+---
+
+* [HBASE-26082](https://issues.apache.org/jira/browse/HBASE-26082) | *Major* | **Mark LocalHBaseCluster as IA.Private**
+
+LocalHBaseCluster is now IA.Private.
+
+In general, It should not be IA.Public. It is only used by HMaster to start a local cluster, as well as in HBTU to start a local cluster for test. End users should use HBTU(or its alternate classes) instead of using LocalHBaseCluster.
+
+
+---
+
+* [HBASE-26080](https://issues.apache.org/jira/browse/HBASE-26080) | *Major* | **Implement a new mini cluster class for end users**
+
+Introduce a new TestingHBaseCluster for end users to start a mini hbase cluster in tests.
+
+After starting the cluster, you can create a Connection with the returned Configuration instance for accessing the cluster.
+
+Besides the mini hbase cluster, TestingHBaseCluster will also start a zookeeper cluster and dfs cluster. But you can not control the zookeeper and dfs cluster separately, as for end users, you do not need to test the behavior of HBase cluster when these systems are broken.
+
+We provide methods for start/stop master and region server, and also the whole hbase cluster. Notice that, we do not provide methods to get the address for masters and regionservers, or locate a region, you could use the Admin interface to do this.
+
+
+
+# HBASE 3.0.0-alpha-1 Release Notes
+
+These release notes cover new developer and user-facing incompatibilities, important issues, features, and major improvements.
+
+
+---
+
+* [HBASE-22923](https://issues.apache.org/jira/browse/HBASE-22923) | *Major* | **hbase:meta is assigned to localhost when we downgrade the hbase version**
+
+Introduced new config: hbase.min.version.move.system.tables
+
+When the operator uses this configuration option, any version between
+the current cluster version and the value of "hbase.min.version.move.system.tables"
+does not trigger any auto-region movement. Auto-region movement here
+refers to auto-migration of system table regions to newer server versions.
+It is assumed that the configured range of versions does not require special
+handling of moving system table regions to higher versioned RegionServer.
+This auto-migration is done by AssignmentManager#checkIfShouldMoveSystemRegionAsync().
+Example: Let's assume the cluster is on version 1.4.0 and we have
+set "hbase.min.version.move.system.tables" as "2.0.0". Now if we upgrade
+one RegionServer on 1.4.0 cluster to 1.6.0 (\< 2.0.0), then AssignmentManager will
+not move hbase:meta, hbase:namespace and other system table regions
+to newly brought up RegionServer 1.6.0 as part of auto-migration.
+However, if we upgrade one RegionServer on 1.4.0 cluster to 2.2.0 (\> 2.0.0),
+then AssignmentManager will move all system table regions to newly brought
+up RegionServer 2.2.0 as part of auto-migration done by
+AssignmentManager#checkIfShouldMoveSystemRegionAsync().
+
+
+---
+
+* [HBASE-25902](https://issues.apache.org/jira/browse/HBASE-25902) | *Critical* | **Add missing CFs in meta during HBase 1 to 2.3+ Upgrade**
+
+While upgrading cluster from 1.x to 2.3+ versions, after the active master is done setting it's status as 'Initialized', it attempts to add 'table' and 'repl\_barrier' CFs in meta. Once CFs are added successfully, master is aborted with PleaseRestartMasterException because master has missed certain initialization events (e.g ClusterSchemaService is not initialized and tableStateManager fails to migrate table states from ZK to meta due to missing CFs). Subsequent active master initialization is expected to be smooth.
+In the presence of multi masters, when one of them becomes active for the first time after upgrading to HBase 2.3+, it is aborted after fixing CFs in meta and one of the other backup masters will take over and become active soon. Hence, overall this is expected to be smooth upgrade if we have backup masters configured. If not, operator is expected to restart same master again manually.
+
+
+---
+
+* [HBASE-26029](https://issues.apache.org/jira/browse/HBASE-26029) | *Critical* | **It is not reliable to use nodeDeleted event to track region server's death**
+
+Introduce a new step in ServerCrashProcedure to move the replication queues of the dead region server to other live region servers, as this is the only reliable way to get the death event of a region server.
+The old ReplicationTracker related code have all been purged as they are not used any more.
+
+
+---
+
+* [HBASE-25877](https://issues.apache.org/jira/browse/HBASE-25877) | *Major* | **Add access check for compactionSwitch**
+
+Now calling RSRpcService.compactionSwitch, i.e, Admin.compactionSwitch at client side, requires ADMIN permission.
+This is an incompatible change but it is also a bug, as we should not allow any users to disable compaction on a regionserver, so we apply this to all active branches.
+
+
+---
+
+* [HBASE-25993](https://issues.apache.org/jira/browse/HBASE-25993) | *Major* | **Make excluded SSL cipher suites configurable for all Web UIs**
+
+Add "ssl.server.exclude.cipher.list" configuration to excluded cipher suites for the http server started by the InfoServer.
+
+
+---
+
+* [HBASE-25920](https://issues.apache.org/jira/browse/HBASE-25920) | *Major* | **Support Hadoop 3.3.1**
+
+Fixes to make unit tests pass and to make it so an hbase built from branch-2 against a 3.3.1RC can run on a 3.3.1RC small cluster.
+
+
+---
+
+* [HBASE-25969](https://issues.apache.org/jira/browse/HBASE-25969) | *Major* | **Cleanup netty-all transitive includes**
+
+We have an (old) netty-all in our produced artifacts. It is transitively included from hadoop. It is needed by MiniMRCluster referenced from a few MR tests in hbase. This commit adds netty-all excludes everywhere else but where tests will fail unless the transitive is allowed through. TODO: move MR and/or MR tests out of hbase core.
+
+
+---
+
+* [HBASE-22708](https://issues.apache.org/jira/browse/HBASE-22708) | *Major* | **Remove the deprecated methods in Hbck interface**
+
+Removed method 'scheduleServerCrashProcedure' of hbck.java in hbase-client module. This method was deprecated since 2.2.1 and declare to be removed in 3.0.0.
+
+
+---
+
+* [HBASE-25963](https://issues.apache.org/jira/browse/HBASE-25963) | *Major* | **HBaseCluster should be marked as IA.Public**
+
+Change HBaseCluster to IA.Public as its sub class MiniHBaseCluster is IA.Public.
+
+
+---
+
+* [HBASE-25649](https://issues.apache.org/jira/browse/HBASE-25649) | *Major* | **Complete the work on moving all the balancer related classes to hbase-balancer module**
+
+Introduced a ClusterInfoProvider as a bridge for LoadBalancer implementation to get cluster information and operate on the cluster, so the LoadBalancer implementation does not need to depend on HMaster/MasterServices directly. So then we could move most LoadBalancer related code to hbase-balancer module. MasterClusterInfoProvider is the actual implementation class for ClusterInfoProvider.
+
+Moved most unit tests for load balancer to hbase-balancer module, unless it requires starting a mini cluster.
+
+RSGroupBasedLoadBalancer and related code are still in hbase-server module. As we still have lots of migration and compatibility code, which makes it deeply tied togather with HMaster/MasterServices. We could try to move it again in 4.0.0, when we have purged the migration and compatibility code.
+
+
+---
+
+* [HBASE-25841](https://issues.apache.org/jira/browse/HBASE-25841) | *Minor* | **Add basic jshell support**
+
+This change adds a new command \`hbase jshell\` command-line interface. It launches an interactive JShell session with HBase on the classpath, as well as a the client package already imported.
+
+
+---
+
+* [HBASE-25894](https://issues.apache.org/jira/browse/HBASE-25894) | *Major* | **Improve the performance for region load and region count related cost functions**
+
+In CostFromRegionLoadFunction, now we will only recompute the cost for a given region server in regionMoved function, instead of computing all the costs every time.
+Introduced a DoubleArrayCost for better abstraction, and also try to only compute the final cost on demand as the computation is also a bit expensive.
+
+
+---
+
+* [HBASE-25869](https://issues.apache.org/jira/browse/HBASE-25869) | *Major* | **WAL value compression**
+
+WAL storage can be expensive, especially if the cell values represented in the edits are large, consisting of blobs or significant lengths of text. Such WALs might need to be kept around for a fairly long time to satisfy replication constraints on a space limited (or space-contended) filesystem.
+
+Enable WAL compression and, with this feature, WAL value compression, to save space in exchange for slightly higher WAL append latencies. The degree of performance impact will depend on the compression algorithm selection. SNAPPY or ZSTD are recommended algorithms, if native codec support is available. SNAPPY may even provide an overall improvement in WAL commit latency, so is the best choice. GZ can be a reasonable fallback where native codec support is not available.
+
+To enable WAL compression, value compression, and select the desired algorithm, edit your site configuration like so:
+
+\<!-- to enable compression --\>
+\<property\>
+ \<name\>hbase.regionserver.wal.enablecompression\</name\>
+ \<value\>true\</value\>
+\</property\>
+
+\<!-- to enable value compression --\>
+\<property\>
+ \<name\>hbase.regionserver.wal.value.enablecompression\</name\>
+ \<value\>true\</value\>
+\</property\>
+
+\<!-- choose the value compression algorithm â\>
+\<property\>
+ \<name\>hbase.regionserver.wal.value.compression.type\</name\>
+ \<value\>snappy\</value\>
+\</property\>
+
+
+---
+
+* [HBASE-25682](https://issues.apache.org/jira/browse/HBASE-25682) | *Major* | **Add a new command to update the configuration of all RSs in a RSGroup**
+
+Added updateConfiguration(String groupName) admin interface & update\_rsgroup\_config to the HBase shell to reload a subset of configuration on all servers in the rsgroup.
+
+
+---
+
+* [HBASE-25032](https://issues.apache.org/jira/browse/HBASE-25032) | *Major* | **Do not assign regions to region server which has not called regionServerReport yet**
+
+<!-- markdown -->
+
+After this change a region server can only accept regions (as seen by master) after it's first report to master is sent successfully. Prior to this change there could be cases where the region server finishes calling regionServerStartup but is actually stuck during initialization due to issues like misconfiguration and master tries to assign regions and they are stuck because the region server is in a weird state and not ready to serve them.
+
+
+---
+
+* [HBASE-25826](https://issues.apache.org/jira/browse/HBASE-25826) | *Major* | **Revisit the synchronization of balancer implementation**
+
+Narrow down the public facing API for LoadBalancer by removing balanceTable and setConf methods.
+Redesigned the initilization sequence to simplify the initialization code. Now all the setters are just 'setter', all the initialization work are moved to initialize method.
+Rename setClusterMetrics to updateClusterMetrics, as it will be called periodically while other setters will only be called once before initialization.
+Add javadoc for LoadBalancer class to mention how to do synchronization on implementation classes.
+
+
+---
+
+* [HBASE-25834](https://issues.apache.org/jira/browse/HBASE-25834) | *Major* | **Remove balanceTable method from LoadBalancer interface**
+
+Remove balanceTable method from LoadBalancer interface as we never call it outside balancer implementation.
+Mark balanceTable method as protected in BaseLoadBalancer.
+Mark balanceCluster method as final in BaseLoadBalancer, the implementation classes should not override it anymore, just implement the balanceTable method is enough.
+
+
+---
+
+* [HBASE-22120](https://issues.apache.org/jira/browse/HBASE-22120) | *Major* | **Replace HTrace with OpenTelemetry**
+
+In this issue we change our tracing system from HTrace to OpenTelemetry.
+The HTrace dependencies are banned(transitive dependencies are still allowed as hadoop still depends on them), the imports of htrace related classes are also banned.
+We add OpenTelemtry support for our RPC system, which means all the rpc methods will be traced on both client side and server side.
+Most methods in Table interface are also traced, except scan and coprocessor related methods. As now the scan implementation is always 'async prefetch', we haven't find a suitable way to represent this relationship between the foreground and background spans yet.
+At server side, due to the same reason, we only use a span to record the time of the WAL sync operation, without tracing into the background sync thread.
+And we do not trace the next method of RegionScanner, as a scan rpc call may lead to thousands of RegionScanner.next calls, which could slow down the rpc call even when tracing is disabled.
+On how to enable tracing, please read the Tracing section in our refguide.
+https://hbase.apache.org/book.html#tracing
+
+
+---
+
+* [HBASE-25756](https://issues.apache.org/jira/browse/HBASE-25756) | *Minor* | **Support alternate compression for major and minor compactions**
+
+It is now possible to specify alternate compression algorithms for major or minor compactions, via new ColumnFamilyBuilder or HColumnDescriptor methods {get,set}{Major,Minor}CompressionType(), or shell schema attributes COMPRESSION\_COMPACT\_MAJOR or COMPRESSION\_COMPACT\_MINOR. This can be used to select a fast algorithm for frequent minor compactions and a slower algorithm offering better compression ratios for infrequent major compactions.
+
+
+---
+
+* [HBASE-25766](https://issues.apache.org/jira/browse/HBASE-25766) | *Major* | **Introduce RegionSplitRestriction that restricts the pattern of the split point**
+
+After HBASE-25766, we can specify a split restriction, "KeyPrefix" or "DelimitedKeyPrefix", to a table with the "hbase.regionserver.region.split\_restriction.type" property. The "KeyPrefix" split restriction groups rows by a prefix of the row-key. And the "DelimitedKeyPrefix" split restriction groups rows by a prefix of the row-key with a delimiter.
+
+For example:
+\`\`\`
+# Create a table with a "KeyPrefix" split restriction, where the prefix length is 2 bytes
+hbase\> create 'tbl1', 'fam', {CONFIGURATION =\> {'hbase.regionserver.region.split\_restriction.type' =\> 'KeyPrefix', 'hbase.regionserver.region.split\_restriction.prefix\_length' =\> '2'}}
+
+# Create a table with a "DelimitedKeyPrefix" split restriction, where the delimiter is a comma (,)
+hbase\> create 'tbl2', 'fam', {CONFIGURATION =\> {'hbase.regionserver.region.split\_restriction.type' =\> 'DelimitedKeyPrefix', 'hbase.regionserver.region.split\_restriction.delimiter' =\> ','}}
+\`\`\`
+
+Instead of specifying a split restriction to a table directly, we can also set the properties in hbase-site.xml. In this case, the specified split restriction is applied for all the tables.
+
+Note that the split restriction is also applied to a user-specified split point so that we don't allow users to break the restriction, which is different behavior from the existing KeyPrefixRegionSplitPolicy and DelimitedKeyPrefixRegionSplitPolicy.
+
+
+---
+
+* [HBASE-25290](https://issues.apache.org/jira/browse/HBASE-25290) | *Major* | **Remove table on master related code in balancer implementation**
+
+Removed the deprecated configs 'hbase.balancer.tablesOnMaster' and 'hbase.balancer.tablesOnMaster.systemTablesOnly', which means in 3.0.0 release, master can not carry regions any more.
+According to the compatibility guide, we should keep this till 4.0.0, but since it never works well, we decided to remove in 3.0.0 release.
+Notice that, when maintenance mode is on, HMaster could still carry system regions. This is the only exception.
+
+
+---
+
+* [HBASE-25775](https://issues.apache.org/jira/browse/HBASE-25775) | *Major* | **Use a special balancer to deal with maintenance mode**
+
+Introduced a MaintenanceLoadBalancer to be used only under maintenance mode. Typically you should not use it as your balancer implementation.
+
+
+---
+
+* [HBASE-25767](https://issues.apache.org/jira/browse/HBASE-25767) | *Major* | **CandidateGenerator.getRandomIterationOrder is too slow on large cluster**
+
+In the actual implementation classes of CandidateGenerator, now we just random select a start point and then iterate sequentially, instead of using the old way, where we will create a big array to hold all the integers in [0, num\_regions\_in\_cluster), shuffle the array, and then iterate on the array.
+The new implementation is 'random' enough as every time we just select one candidate. The problem for the old implementation is that, it will create an array every time when we want to get a candidate, if we have tens of thousands regions, we will create an array with tens of thousands length everytime, which causes big GC pressure and slow down the balancer execution.
+
+
+---
+
+* [HBASE-25744](https://issues.apache.org/jira/browse/HBASE-25744) | *Major* | **Change default of \`hbase.normalizer.merge.min\_region\_size.mb\` to \`0\`**
+
+Before this change, by default, the normalizer would exclude any region with a total \`storefileSizeMB\` \<= 1 from merge consideration. This changes the default so that these small regions will be merged away.
+
+
+---
+
+* [HBASE-25716](https://issues.apache.org/jira/browse/HBASE-25716) | *Major* | **The configured loggers in log4j2.xml will always be created**
+
+Added 'createOnDemand' for all the appenders to let them only create log files when necessary. And since log4j2 appender will always create the parent directory and the default location for http request log is /var/log/hbase, we commented out in the log4j2.xml so it will not try to create the directory by default. Users need to uncomment it when enabling http request log.
+
+
+---
+
+* [HBASE-25199](https://issues.apache.org/jira/browse/HBASE-25199) | *Minor* | **Remove HStore#getStoreHomedir**
+
+Moved the following methods from HStore to HRegionFileSystem
+
+- #getStoreHomedir(Path, RegionInfo, byte[])
+- #getStoreHomedir(Path, String, byte[])
+
+
+---
+
+* [HBASE-25174](https://issues.apache.org/jira/browse/HBASE-25174) | *Major* | **Remove deprecated fields in HConstants which should be removed in 3.0.0**
+
+Removed the following constants without replacement
+
+- HConstants#HBASE\_REGIONSERVER\_LEASE\_PERIOD\_KEY
+- HConstants#CP\_HTD\_ATTR\_KEY\_PATTERN
+- HConstants#CP\_HTD\_ATTR\_VALUE\_PATTERN
+- HConstants#CP\_HTD\_ATTR\_VALUE\_PARAM\_KEY\_PATTERN
+- HConstants#CP\_HTD\_ATTR\_VALUE\_PARAM\_VALUE\_PATTERN
+- HConstants#CP\_HTD\_ATTR\_VALUE\_PARAM\_PATTERN
+- HConstants#META\_QOS
+
+Moved the following constant into private scope
+
+- HConstants#OLDEST\_TIMESTAMP
+
+
+---
+
+* [HBASE-25685](https://issues.apache.org/jira/browse/HBASE-25685) | *Major* | **asyncprofiler2.0 no longer supports svg; wants html**
+
+If asyncprofiler 1.x, all is good. If asyncprofiler 2.x and it is hbase-2.3.x or hbase-2.4.x, add '?output=html' to get flamegraphs from the profiler.
+
+Otherwise, if hbase-2.5+ and asyncprofiler2, all works. If asyncprofiler1 and hbase-2.5+, you may have to add '?output=svg' to the query.
+
+
+---
+
+* [HBASE-19577](https://issues.apache.org/jira/browse/HBASE-19577) | *Major* | **Use log4j2 instead of log4j for logging**
+
+Use log4j2 instead of log4j for logging.
+Exclude log4j dependency from hbase and transitive dependencies, use log4j-1.2-api as test dependency for bridging as hadoop still need log4j for some reasons. Copy FileAppender implementation in hbase-logging as the ContainerLogAppender for YARN NodeManager extends it. All log4j.properties files have been replaced by log4j2.xml.
+For log4j2, there is no 'log4j.rootLogger' config, so we need to config level and appender separately, the system properties are now 'hbase.root.logger.level' and 'hbase.root.logger.appender', for security loggers they are 'hbase.security.logger.level' and 'hbase.security.logger.appender'. But for setting them from command line, you could still use something like 'HBASE\_ROOT\_LOGGER=INFO,console' as we will split it and set level and appender separately.
+
+
+---
+
+* [HBASE-25681](https://issues.apache.org/jira/browse/HBASE-25681) | *Major* | **Add a switch for server/table queryMeter**
+
+Adds "hbase.regionserver.enable.server.query.meter" and "hbase.regionserver.enable.table.query.meter" switches which are off by default.
+
+Note, these counters used to be ON by default; now they are off.
+
+
+---
+
+* [HBASE-25518](https://issues.apache.org/jira/browse/HBASE-25518) | *Major* | **Support separate child regions to different region servers**
+
+Config key for enable/disable automatically separate child regions to different region servers in the procedure of split regions. One child will be kept to the server where parent region is on, and the other child will be assigned to a random server.
+
+hbase.master.auto.separate.child.regions.after.split.enabled
+
+Default setting is false/off.
+
+
+---
+
+* [HBASE-25608](https://issues.apache.org/jira/browse/HBASE-25608) | *Major* | **Support HFileOutputFormat locality sensitive even destination cluster is different from source cluster**
+
+
+Added configurations to specify the ZK cluster key for remote cluster in HFileOutputFormat2.
+Default, input and output are to the cluster specified in Job configuration.
+Use HFileOutputformat2#configureRemoteCluster to have output go to a remote cluster.
+HFileOutputFormat2#configureIncrementalLoad(Job, Table, RegionLocator) configure them using Table's configuration.
+You can also configure them by calling HFileOutputFormat2#configureRemoteCluster explicitly.
+
+
+---
+
+* [HBASE-25665](https://issues.apache.org/jira/browse/HBASE-25665) | *Major* | **Disable reverse DNS lookup for SASL Kerberos client connection**
+
+New client side configuration \`hbase.unsafe.client.kerberos.hostname.disable.reversedns\` is added.
+
+This configuration is advanced for experts and you shouldn't specify unless you really what is this configuration and doing.
+HBase secure client using SASL Kerberos performs DNS reverse lookup to get hostname for server principal using InetAddress.getCanonicalHostName by default (false for this configuration).
+If you set true for this configuration, HBase client doen't perform DNS reverse lookup for server principal and use InetAddress.getHostName which is sent by HBase cluster instead.
+This helps your client application deploy under unusual network environment which DNS doesn't provide reverse lookup.
+Check the description of the JIRA ticket, HBASE-25665 carefully and check that this configuration fits your environment and deployment actually before enable this configuration.
+
+
+---
+
+* [HBASE-25374](https://issues.apache.org/jira/browse/HBASE-25374) | *Minor* | **Make REST Client connection and socket time out configurable**
+
+Configuration parameter to set rest client connection timeout
+
+"hbase.rest.client.conn.timeout" Default is 2 \* 1000
+
+"hbase.rest.client.socket.timeout" Default of 30 \* 1000
+
+
+---
+
+* [HBASE-25566](https://issues.apache.org/jira/browse/HBASE-25566) | *Major* | **RoundRobinTableInputFormat**
+
+Adds RoundRobinTableInputFormat, a subclass of TableInputFormat, that takes the TIF#getSplits list and resorts it so as to spread the InputFormats as broadly about the cluster as possible. RRTIF works to frustrate bunching of InputSplits on RegionServers to avoid the scenario where a few RegionServers are working hard fielding many InputSplits while others idle hosting a few or none.
+
+
+---
+
+* [HBASE-25587](https://issues.apache.org/jira/browse/HBASE-25587) | *Major* | **[hbck2] Schedule SCP for all unknown servers**
+
+Adds scheduleSCPsForUnknownServers to Hbck Service.
+
+
+---
+
+* [HBASE-25636](https://issues.apache.org/jira/browse/HBASE-25636) | *Minor* | **Expose HBCK report as metrics**
+
+Expose HBCK repost results in metrics, includes: "orphanRegionsOnRS", "orphanRegionsOnFS", "inconsistentRegions", "holes", "overlaps", "unknownServerRegions" and "emptyRegionInfoRegions".
+
+
+---
+
+* [HBASE-25582](https://issues.apache.org/jira/browse/HBASE-25582) | *Major* | **Support setting scan ReadType to be STREAM at cluster level**
+
+Adding a new meaning for the config 'hbase.storescanner.pread.max.bytes' when configured with a value \<0.
+In HBase 2.x we allow the Scan op to specify a ReadType (PREAD / STREAM/ DEFAULT). When Scan comes with DEFAULT read type, we will start scan with preads and later switch to stream read once we see we are scanning a total data size \> value of hbase.storescanner.pread.max.bytes. (This is calculated for data per region:cf). This config defaults to 4 x of HFile block size = 256 KB by default.
+This jira added a new meaning for this config when configured with a -ve value. In such case, for all scans with DEFAULT read type, we will start with STREAM read itself. (Switch at begin of the scan itself)
+
+
+---
+
+* [HBASE-25460](https://issues.apache.org/jira/browse/HBASE-25460) | *Major* | **Expose drainingServers as cluster metric**
+
+Exposed new jmx metrics: "draininigRegionServers" and "numDrainingRegionServers" to provide "comma separated names for regionservers that are put in draining mode" and "num of such regionservers" respectively.
+
+
+---
+
+* [HBASE-25615](https://issues.apache.org/jira/browse/HBASE-25615) | *Major* | **Upgrade java version in pre commit docker file**
+
+jdk8u232-b09 -\> jdk8u282-b08
+jdk-11.0.6\_10 -\> jdk-11.0.10\_9
+
+
+---
+
+* [HBASE-23887](https://issues.apache.org/jira/browse/HBASE-23887) | *Major* | **New L1 cache : AdaptiveLRU**
+
+Introduced new L1 cache: AdaptiveLRU. This is supposed to provide better performance than default LRU cache.
+Set config key "hfile.block.cache.policy" to "AdaptiveLRU" in hbase-site in order to start using this new cache.
+
+
+---
+
+* [HBASE-25375](https://issues.apache.org/jira/browse/HBASE-25375) | *Major* | **Provide a VM-based release environment**
+
+<!-- markdown -->
+Adds a Vagrant project for running releases under `dev-support/release-vm`. This is intended to be an environment wherein the `create-release` scripts can be run (in docker mode). See the directory's readme for details.
+
+
+---
+
+* [HBASE-25449](https://issues.apache.org/jira/browse/HBASE-25449) | *Major* | **'dfs.client.read.shortcircuit' should not be set in hbase-default.xml**
+
+The presence of HDFS short-circuit read configuration properties in hbase-default.xml inadvertently causes short-circuit reads to not happen inside of RegionServers, despite short-circuit reads being enabled in hdfs-site.xml.
+
+
+---
+
+* [HBASE-25333](https://issues.apache.org/jira/browse/HBASE-25333) | *Major* | **Add maven enforcer rule to ban VisibleForTesting imports**
+
+Ban the imports of guava VisiableForTesting, which means you should not use this annotation in HBase any more.
+For IA.Public and IA.LimitedPrivate classes, typically you should not expose any test related fields/methods there, and if you want to hide something, use IA.Private on the specific fields/methods.
+For IA.Private classes, if you want to expose something only for tests, use the RestrictedApi annotation from error prone, where it could cause a compilation error if someone break the rule in the future.
+
+
+---
+
+* [HBASE-25473](https://issues.apache.org/jira/browse/HBASE-25473) | *Major* | **[create-release] checkcompatibility.py failing with "KeyError: 'binary'"**
+
+Adds temporary exclude of hbase-shaded-testing-util from checkcompatibility so create-release can work again. Undo exclude when sub-issue is fixed.
+
+
+---
+
+* [HBASE-25441](https://issues.apache.org/jira/browse/HBASE-25441) | *Critical* | **add security check for some APIs in RSRpcServices**
+
+RsRpcServices APIs that can be accessed only through Admin rights:
+- stopServer
+- updateFavoredNodes
+- updateConfiguration
+- clearRegionBlockCache
+- clearSlowLogsResponses
+
+
+---
+
+* [HBASE-25432](https://issues.apache.org/jira/browse/HBASE-25432) | *Blocker* | **we should add security checks for setTableStateInMeta and fixMeta**
+
+setTableStateInMeta and fixMeta can be accessed only through Admin rights
+
+
+---
+
+* [HBASE-25318](https://issues.apache.org/jira/browse/HBASE-25318) | *Minor* | **Configure where IntegrationTestImportTsv generates HFiles**
+
+Added IntegrationTestImportTsv.generatedHFileFolder configuration property to override the default location in IntegrationTestImportTsv. Useful for running the integration test when HDFS Transparent Encryption is enabled.
+
+
+---
+
+* [HBASE-24751](https://issues.apache.org/jira/browse/HBASE-24751) | *Minor* | **Display Task completion time and/or processing duration on Web UI**
+
+Adds completion time to tasks display.
+
+
+---
+
+* [HBASE-25456](https://issues.apache.org/jira/browse/HBASE-25456) | *Critical* | **setRegionStateInMeta need security check**
+
+setRegionStateInMeta can be accessed only through Admin rights
+
+
+---
+
+* [HBASE-25451](https://issues.apache.org/jira/browse/HBASE-25451) | *Major* | **Upgrade commons-io to 2.8.0**
+
+Upgrade commons-io to 2.8.0. Remove deprecated IOUtils.closeQuietly call in code base.
+
+
+---
+
+* [HBASE-24764](https://issues.apache.org/jira/browse/HBASE-24764) | *Minor* | **Add support of adding base peer configs via hbase-site.xml for all replication peers.**
+
+<!-- markdown -->
+
+Adds a new configuration parameter "hbase.replication.peer.base.config" which accepts a semi-colon separated key=CSV pairs (example: k1=v1;k2=v2_1,v3...). When this configuration is set on the server side, these kv pairs are added to every peer configuration if not already set. Peer specific configuration overrides have precedence over the above default configuration. This is useful in cases when some configuration has to be set for all the peers by default and one does not want to add to every peer definition.
+
+
+---
+
+* [HBASE-25419](https://issues.apache.org/jira/browse/HBASE-25419) | *Major* | **Remove deprecated methods in RpcServer implementation**
+
+Removed all the call methods except call(RpcCall, MonitoredRPCHandler) in RpcServerInterface.
+
+
+---
+
+* [HBASE-24966](https://issues.apache.org/jira/browse/HBASE-24966) | *Major* | **The methods in AsyncTableRegionLocator should not throw IOException directly**
+
+Remove the throws part for the following 3 methods in AsyncTableRegionLocator:
+
+getStartKeys
+getEndKeys
+getStartEndKeys
+
+It is a mistake and we will never throw exception directly from these methods. The should get the exception from the returned CompletableFuture.
+
+In order to not introduce new methods and make more confusing, we just remove the throws part. It is an incompatible change, you may need to change your code to remove the catch section when you upgrade to HBase 3.x.
+
+
+---
+
+* [HBASE-25127](https://issues.apache.org/jira/browse/HBASE-25127) | *Major* | **Enhance PerformanceEvaluation to profile meta replica performance.**
+
+Three new commands are added to PE:
+
+metaWrite, metaRandomRead and cleanMeta.
+
+Usage example:
+hbase pe --rows=100000 metaWrite 1
+hbase pe --nomapreduce --rows=100000 metaRandomRead 32
+hbase pe --rows=100000 cleanMeta 1
+
+metaWrite and cleanMeta should be run with only 1 thread and the same number of rows so all the rows inserted will be cleaned up properly.
+
+metaRandomRead can be run with multiple threads. The rows option should set to within the range of rows inserted by metaWrite
+
+
+---
+
+* [HBASE-25237](https://issues.apache.org/jira/browse/HBASE-25237) | *Major* | **'hbase master stop' shuts down the cluster, not the master only**
+
+\`hbase master stop\` should shutdown only master by default.
+1. Help added to \`hbase master stop\`:
+To stop cluster, use \`stop-hbase.sh\` or \`hbase master stop --shutDownCluster\`
+
+2. Help added to \`stop-hbase.sh\`:
+stop-hbase.sh can only be used for shutting down entire cluster. To shut down (HMaster\|HRegionServer) use hbase-daemon.sh stop (master\|regionserver)
+
+
+---
+
+* [HBASE-25242](https://issues.apache.org/jira/browse/HBASE-25242) | *Critical* | **Add Increment/Append support to RowMutations**
+
+<!-- markdown -->
+
+After HBASE-25242, we can add Increment/Append operations to RowMutations and perform those operations atomically in a single row.
+
+This change alters a public API so that the `mutateRow` method in both the `Table` and `AsyncTable` classes now returns a `Result` object with the result of the passed operations. Previously these methods had no return values.
+
+Code that calls these methods which was compiled against an earlier version of the HBase client libraries will fail at runtime with a `NoSuchMethodError` when used with this release. If you are going to ignore the result of the passed operations you can simply recompile your application with an updated dependency and no additional changes.
+
+
+---
+
+* [HBASE-25263](https://issues.apache.org/jira/browse/HBASE-25263) | *Major* | **Change encryption key generation algorithm used in the HBase shell**
+
+Since the backward-compatible change we introduced in HBASE-25263, we use the more secure PBKDF2WithHmacSHA384 key generation algorithm (instead of PBKDF2WithHmacSHA1) to generate a secret key for HFile / WalFile encryption, when the user is defining a string encryption key in the hbase shell.
+
+
+---
+
+* [HBASE-24268](https://issues.apache.org/jira/browse/HBASE-24268) | *Minor* | **REST and Thrift server do not handle the "doAs" parameter case insensitively**
+
+This change allows the REST and Thrift servers to handle the "doAs" parameter case-insensitively, which is deemed as correct per the "specification" provided by the Hadoop community.
+
+
+---
+
+* [HBASE-25278](https://issues.apache.org/jira/browse/HBASE-25278) | *Minor* | **Add option to toggle CACHE\_BLOCKS in count.rb**
+
+A new option, CACHE\_BLOCKS, was added to the \`count\` shell command which will force the data for a table to be loaded into the block cache. By default, the \`count\` command will not cache any blocks. This option can serve as a means to for a table's data to be loaded into block cache on demand. See the help message on the count shell command for usage details.
+
+
+---
+
+* [HBASE-18070](https://issues.apache.org/jira/browse/HBASE-18070) | *Critical* | **Enable memstore replication for meta replica**
+
+"Async WAL Replication" [1] was added by HBASE-11183 "Timeline Consistent region replicas - Phase 2 design" but only for user-space tables. This feature adds "Async WAL Replication" for the hbase:meta table. It also adds a client 'LoadBalance' mode that has reads go to replicas first and to the primary only on fail so as to shed read load from the primary to alleviate \*hotspotting\* on the hbase:meta Region.
+
+Configuration is as it was for the user-space 'Async WAL Replication'. See [2] and [3] for details on how to enable.
+
+1. http://hbase.apache.org/book.html#async.wal.replication
+2. http://hbase.apache.org/book.html#async.wal.replication.meta
+3. http://hbase.apache.org/book.html#\_async\_wal\_replication\_for\_meta\_table\_as\_of\_hbase\_2\_4\_0
+
+
+---
+
+* [HBASE-25126](https://issues.apache.org/jira/browse/HBASE-25126) | *Major* | **Add load balance logic in hbase-client to distribute read load over meta replica regions.**
+
+See parent issue, HBASE-18070, release notes for how to enable.
+
+
+---
+
+* [HBASE-25026](https://issues.apache.org/jira/browse/HBASE-25026) | *Minor* | **Create a metric to track full region scans RPCs**
+
+Adds a new metric where we collect the number of full region scan requests at the RPC layer. This will be collected under "name" : "Hadoop:service=HBase,name=RegionServer,sub=Server"
+
+
+---
+
+* [HBASE-25253](https://issues.apache.org/jira/browse/HBASE-25253) | *Major* | **Deprecated master carrys regions related methods and configs**
+
+Since 2.4.0, deprecated all master carrys regions related methods(LoadBalancer,BaseLoadBalancer,ZNodeClearer) and configs(hbase.balancer.tablesOnMaster, hbase.balancer.tablesOnMaster.systemTablesOnly), they will be removed in 3.0.0.
+
+
+---
+
+* [HBASE-20598](https://issues.apache.org/jira/browse/HBASE-20598) | *Major* | **Upgrade to JRuby 9.2**
+
+<!-- markdown -->
+The HBase shell now relies on JRuby 9.2. This is a new major version change for JRuby. The most significant change is Ruby compatibility changed from Ruby 2.3 to Ruby 2.5. For more detailed changes please see [the JRuby release announcement for the start of the 9.2 series](https://www.jruby.org/2018/05/24/jruby-9-2-0-0.html) as well as the [general release announcement page for updates since that version](https://www.jruby.org/news).
+
+The runtime dependency versions present on the server side classpath for the Joni (now 2.1.31) and JCodings (now 1.0.55) libraries have also been updated to match those found in the JRuby version shipped with HBase. These version changes are maintenance releases and should be backwards compatible when updated in tandem.
+
+
+---
+
+* [HBASE-25181](https://issues.apache.org/jira/browse/HBASE-25181) | *Major* | **Add options for disabling column family encryption and choosing hash algorithm for wrapped encryption keys.**
+
+<!-- markdown -->
+This change adds options for disabling column family encryption and choosing hash algorithm for wrapped encryption keys. Changes are done such that defaults will keep the same behavior prior to this issue.
+
+Prior to this change HBase always used the MD5 hash algorithm to store a hash for encryption keys. This hash is needed to verify the secret key of the subject. (e.g. making sure that the same secrey key is used during encrypted HFile read and write). The MD5 algorithm is considered weak, and can not be used in some (e.g. FIPS compliant) clusters. Having a configurable hash enables us to use newer and more secure hash algorithms like SHA-384 or SHA-512 (which are FIPS compliant).
+
+The hash is set via the configuration option `hbase.crypto.key.hash.algorithm`. It should be set to a JDK `MessageDigest` algorithm like "MD5", "SHA-256" or "SHA-384". The default is "MD5" for backward compatibility.
+
+Alternatively, clusters which rely on an encryption at rest mechanism outside of HBase (e.g. those offered by HDFS) and wish to ensure HBase's encryption at rest system is inactive can set `hbase.crypto.enabled` to `false`.
+
+
+---
+
+* [HBASE-25238](https://issues.apache.org/jira/browse/HBASE-25238) | *Critical* | **Upgrading HBase from 2.2.0 to 2.3.x fails because of âMessage missing required fields: stateâ**
+
+Fixes master procedure store migration issues going from 2.0.x to 2.2.x and/or 2.3.x. Also fixes failed heartbeat parse during rolling upgrade from 2.0.x. to 2.3.x.
+
+
+---
+
+* [HBASE-25235](https://issues.apache.org/jira/browse/HBASE-25235) | *Major* | **Cleanup the deprecated methods in TimeRange**
+
+Removed all the public constructors of TimeRange, thus now we make TimeRange final. Also removed the withinTimeRange(byte[], int) method, just use withinTimeRange(long) instead.
+
+
+---
+
+* [HBASE-25212](https://issues.apache.org/jira/browse/HBASE-25212) | *Major* | **Optionally abort requests in progress after deciding a region should close**
+
+If hbase.regionserver.close.wait.abort is set to true, interrupt RPC handler threads holding the region close lock.
+
+Until requests in progress can be aborted, wait on the region close lock for a configurable interval (specified by hbase.regionserver.close.wait.time.ms, default 60000 (1 minute)). If we have failed to acquire the close lock after this interval elapses, if allowed (also specified by hbase.regionserver.close.wait.abort), abort the regionserver.
+
+We will attempt to interrupt any running handlers every hbase.regionserver.close.wait.interval.ms (default 10000 (10 seconds)) until either the close lock is acquired or we reach the maximum wait time.
+
+
+---
+
+* [HBASE-25173](https://issues.apache.org/jira/browse/HBASE-25173) | *Major* | **Remove owner related methods in TableDescriptor/TableDescriptorBuilder**
+
+Remove the OWNER field in the table creation statement, the relevant permissions will be automatically granted to the current active users of the client.
+
+
+---
+
+* [HBASE-25167](https://issues.apache.org/jira/browse/HBASE-25167) | *Major* | **Normalizer support for hot config reloading**
+
+<!-- markdown -->
+This patch adds [dynamic configuration](https://hbase.apache.org/book.html#dyn_config) support for the following configuration keys related to the normalizer:
+* hbase.normalizer.throughput.max_bytes_per_sec
+* hbase.normalizer.split.enabled
+* hbase.normalizer.merge.enabled
+* hbase.normalizer.min.region.count
+* hbase.normalizer.merge.min_region_age.days
+* hbase.normalizer.merge.min_region_size.mb
+
+
+---
+
+* [HBASE-25224](https://issues.apache.org/jira/browse/HBASE-25224) | *Major* | **Maximize sleep for checking meta and namespace regions availability**
+
+Changed the max sleep time during meta and namespace regions availability check to be 60 sec. Previously there was no such cap
+
+
+---
+
+* [HBASE-25197](https://issues.apache.org/jira/browse/HBASE-25197) | *Trivial* | **Remove SingletonCoprocessorService**
+
+Removed org.apache.hadoop.hbase.coprocessor.SingletonCoprocessorService without a replacement.
+
+
+---
+
+* [HBASE-25198](https://issues.apache.org/jira/browse/HBASE-25198) | *Minor* | **Remove RpcSchedulerFactory#create(Configuration, PriorityFunction)**
+
+Removed RpcSchedulerFactory#create(Configuration conf, PriorityFunction priority) without a replacement.
+
+
+---
+
+* [HBASE-24628](https://issues.apache.org/jira/browse/HBASE-24628) | *Major* | **Region normalizer now respects a rate limit**
+
+<!-- markdown -->
+Introduces a new configuration, `hbase.normalizer.throughput.max_bytes_per_sec`, for specifying a limit on the throughput of actions executed by the normalizer. Note that while this configuration value is in bytes, the minimum honored valued is `1,000,000`, or `1m`. Supports values configured using the human-readable suffixes honored by [`Configuration.getLongBytes`](https://hadoop.apache.org/docs/current/api/org/apache/hadoop/conf/Configuration.html#getLongBytes-java.lang.String-long-)
+
+
+---
+
+* [HBASE-24528](https://issues.apache.org/jira/browse/HBASE-24528) | *Major* | **Improve balancer decision observability**
+
+Retrieve latest balancer decisions made by LoadBalancers.
+
+Examples:
+ hbase\> get\_balancer\_decisions
+Retrieve recent balancer decisions with region plans
+
+ hbase\> get\_balancer\_decisions LIMIT =\> 10
+Retrieve 10 most recent balancer decisions with region plans
+
+
+Config change:
+
+hbase.master.balancer.decision.buffer.enabled:
+
+ Indicates whether active HMaster has ring buffer running for storing
+ balancer decisions in FIFO manner with limited entries. The size of
+ the ring buffer is indicated by config:
+ hbase.master.balancer.decision.queue.size
+
+
+---
+
+* [HBASE-14067](https://issues.apache.org/jira/browse/HBASE-14067) | *Major* | **bundle ruby files for hbase shell into a jar.**
+
+<!-- markdown -->
+The `hbase-shell` artifact now contains the ruby files that implement the hbase shell. There should be no downstream impact for users of the shell that rely on the `hbase shell` command.
+
+Folks that wish to include the HBase ruby classes defined for the shell in their own JRuby scripts should add the `hbase-shell.jar` file to their classpath rather than add `${HBASE_HOME}/lib/ruby` to their load paths.
+
+
+---
+
+* [HBASE-24875](https://issues.apache.org/jira/browse/HBASE-24875) | *Major* | **Remove the force param for unassign since it dose not take effect any more**
+
+<!-- markdown -->
[... 5039 lines stripped ...]