2 <?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
5 * Licensed to the Apache Software Foundation (ASF) under one
6 * or more contributor license agreements. See the NOTICE file
7 * distributed with this work for additional information
8 * regarding copyright ownership. The ASF licenses this file
9 * to you under the Apache License, Version 2.0 (the
10 * "License"); you may not use this file except in compliance
11 * with the License. You may obtain a copy of the License at
13 * http://www.apache.org/licenses/LICENSE-2.0
15 * Unless required by applicable law or agreed to in writing, software
16 * distributed under the License is distributed on an "AS IS" BASIS,
17 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
18 * See the License for the specific language governing permissions and
19 * limitations under the License.
26 The important configs. are listed near the top. You should change
27 at least the setting for hbase.tmp.dir. Other settings will change
28 dependent on whether you are running hbase in standalone mode or
29 distributed. See the hbase reference guide for requirements and
30 guidance making configuration.
32 This file does not contain all possible configurations. The file would be
33 much larger if it carried everything. The absent configurations will only be
34 found through source code reading. The idea is that such configurations are
35 exotic and only those who would go to the trouble of reading a particular
36 section in the code would be knowledgeable or invested enough in ever wanting
37 to alter such configurations, so we do not list them here. Listing all
38 possible configurations would overwhelm and obscure the important.
42 <!--Configs you will likely change are listed here at the top of the file.
45 <name>hbase.tmp.dir</name>
46 <value>${java.io.tmpdir}/hbase-${user.name}</value>
47 <description>Temporary directory on the local filesystem.
48 Change this setting to point to a location more permanent
49 than '/tmp', the usual resolve for java.io.tmpdir, as the
50 '/tmp' directory is cleared on machine restart.</description>
53 <name>hbase.rootdir</name>
54 <value>${hbase.tmp.dir}/hbase</value>
55 <description>The directory shared by region servers and into
56 which HBase persists. The URL should be 'fully-qualified'
57 to include the filesystem scheme. For example, to specify the
58 HDFS directory '/hbase' where the HDFS instance's namenode is
59 running at namenode.example.org on port 9000, set this value to:
60 hdfs://namenode.example.org:9000/hbase. By default, we write
61 to whatever ${hbase.tmp.dir} is set too -- usually /tmp --
62 so change this configuration or else all data will be lost on
63 machine restart.</description>
66 <name>hbase.cluster.distributed</name>
68 <description>The mode the cluster will be in. Possible values are
69 false for standalone mode and true for distributed mode. If
70 false, startup will run all HBase and ZooKeeper daemons together
71 in the one JVM.</description>
74 <name>hbase.zookeeper.quorum</name>
75 <!--Ideally we should be using a hostname here instead of IP address. Please refer to
76 https://issues.apache.org/jira/browse/HBASE-23764 for why we switched to IP address. Should be
77 changed once we fix the underlying ZK issue.-->
78 <value>127.0.0.1</value>
79 <description>Comma separated list of servers in the ZooKeeper ensemble
80 (This config. should have been named hbase.zookeeper.ensemble).
81 For example, "host1.mydomain.com,host2.mydomain.com,host3.mydomain.com".
82 By default this is set to localhost for local and pseudo-distributed modes
83 of operation. For a fully-distributed setup, this should be set to a full
84 list of ZooKeeper ensemble servers. If HBASE_MANAGES_ZK is set in hbase-env.sh
85 this is the list of servers which hbase will start/stop ZooKeeper on as
86 part of cluster start/stop. Client-side, we will take this list of
87 ensemble members and put it together with the hbase.zookeeper.property.clientPort
88 config. and pass it into zookeeper constructor as the connectString
89 parameter.</description>
91 <!--The above are the important configurations for getting hbase up
95 <name>zookeeper.recovery.retry.maxsleeptime</name>
97 <description>Max sleep time before retry zookeeper operations in milliseconds,
98 a max time is needed here so that sleep time won't grow unboundedly
102 <name>hbase.local.dir</name>
103 <value>${hbase.tmp.dir}/local/</value>
104 <description>Directory on the local filesystem to be used
105 as a local storage.</description>
108 <!--Master configurations-->
110 <name>hbase.master.port</name>
112 <description>The port the HBase Master should bind to.</description>
115 <name>hbase.master.info.port</name>
117 <description>The port for the HBase Master web UI.
118 Set to -1 if you do not want a UI instance run.</description>
121 <name>hbase.master.info.bindAddress</name>
122 <value>0.0.0.0</value>
123 <description>The bind address for the HBase Master web UI
127 <name>hbase.master.logcleaner.plugins</name>
128 <value>org.apache.hadoop.hbase.master.cleaner.TimeToLiveLogCleaner,org.apache.hadoop.hbase.master.cleaner.TimeToLiveProcedureWALCleaner</value>
129 <description>A comma-separated list of BaseLogCleanerDelegate invoked by
130 the LogsCleaner service. These WAL cleaners are called in order,
131 so put the cleaner that prunes the most files in front. To
132 implement your own BaseLogCleanerDelegate, just put it in HBase's classpath
133 and add the fully qualified class name here. Always add the above
134 default log cleaners in the list.</description>
137 <name>hbase.master.logcleaner.ttl</name>
138 <value>600000</value>
139 <description>How long a WAL remain in the archive ({hbase.rootdir}/oldWALs) directory,
140 after which it will be cleaned by a Master thread. The value is in milliseconds.</description>
143 <name>hbase.master.procedurewalcleaner.ttl</name>
144 <value>604800000</value>
145 <description>How long a Procedure WAL will remain in the
146 archive directory, after which it will be cleaned
147 by a Master thread. The value is in milliseconds.</description>
150 <name>hbase.master.hfilecleaner.plugins</name>
151 <value>org.apache.hadoop.hbase.master.cleaner.TimeToLiveHFileCleaner</value>
152 <description>A comma-separated list of BaseHFileCleanerDelegate invoked by
153 the HFileCleaner service. These HFiles cleaners are called in order,
154 so put the cleaner that prunes the most files in front. To
155 implement your own BaseHFileCleanerDelegate, just put it in HBase's classpath
156 and add the fully qualified class name here. Always add the above
157 default hfile cleaners in the list as they will be overwritten in
158 hbase-site.xml.</description>
161 <name>hbase.procedure.store.region.hfilecleaner.plugins</name>
162 <value>org.apache.hadoop.hbase.master.cleaner.TimeToLiveHFileCleaner</value>
163 <description>A comma-separated list of BaseHFileCleanerDelegate invoked by
164 the RegionProcedureStore HFileCleaner service. These HFiles cleaners are
165 called in order, so put the cleaner that prunes the most files in front. To
166 implement your own BaseHFileCleanerDelegate, just put it in HBase's classpath
167 and add the fully qualified class name here. Always add the above
168 default hfile cleaners in the list as they will be overwritten in
169 hbase-site.xml.</description>
172 <name>hbase.master.infoserver.redirect</name>
174 <description>Whether or not the Master listens to the Master web
175 UI port (hbase.master.info.port) and redirects requests to the web
176 UI server shared by the Master and RegionServer. Config. makes
177 sense when Master is serving Regions (not the default).</description>
180 <name>hbase.master.fileSplitTimeout</name>
181 <value>600000</value>
182 <description>Splitting a region, how long to wait on the file-splitting
183 step before aborting the attempt. Default: 600000. This setting used
184 to be known as hbase.regionserver.fileSplitTimeout in hbase-1.x.
185 Split is now run master-side hence the rename (If a
186 'hbase.master.fileSplitTimeout' setting found, will use it to
187 prime the current 'hbase.master.fileSplitTimeout'
188 Configuration.</description>
191 <!--RegionServer configurations-->
193 <name>hbase.regionserver.port</name>
195 <description>The port the HBase RegionServer binds to.</description>
198 <name>hbase.regionserver.info.port</name>
200 <description>The port for the HBase RegionServer web UI
201 Set to -1 if you do not want the RegionServer UI to run.</description>
204 <name>hbase.regionserver.info.bindAddress</name>
205 <value>0.0.0.0</value>
206 <description>The address for the HBase RegionServer web UI</description>
209 <name>hbase.regionserver.info.port.auto</name>
211 <description>Whether or not the Master or RegionServer
212 UI should search for a port to bind to. Enables automatic port
213 search if hbase.regionserver.info.port is already in use.
214 Useful for testing, turned off by default.</description>
217 <name>hbase.regionserver.handler.count</name>
219 <description>Count of RPC Listener instances spun up on RegionServers.
220 Same property is used by the Master for count of master handlers.
221 Too many handlers can be counter-productive. Make it a multiple of
222 CPU count. If mostly read-only, handlers count close to cpu count
223 does well. Start with twice the CPU count and tune from there.</description>
226 <name>hbase.ipc.server.callqueue.handler.factor</name>
228 <description>Factor to determine the number of call queues.
229 A value of 0 means a single queue shared between all the handlers.
230 A value of 1 means that each handler has its own queue.</description>
233 <name>hbase.ipc.server.callqueue.read.ratio</name>
235 <description>Split the call queues into read and write queues.
236 The specified interval (which should be between 0.0 and 1.0)
237 will be multiplied by the number of call queues.
238 A value of 0 indicate to not split the call queues, meaning that both read and write
239 requests will be pushed to the same set of queues.
240 A value lower than 0.5 means that there will be less read queues than write queues.
241 A value of 0.5 means there will be the same number of read and write queues.
242 A value greater than 0.5 means that there will be more read queues than write queues.
243 A value of 1.0 means that all the queues except one are used to dispatch read requests.
245 Example: Given the total number of call queues being 10
246 a read.ratio of 0 means that: the 10 queues will contain both read/write requests.
247 a read.ratio of 0.3 means that: 3 queues will contain only read requests
248 and 7 queues will contain only write requests.
249 a read.ratio of 0.5 means that: 5 queues will contain only read requests
250 and 5 queues will contain only write requests.
251 a read.ratio of 0.8 means that: 8 queues will contain only read requests
252 and 2 queues will contain only write requests.
253 a read.ratio of 1 means that: 9 queues will contain only read requests
254 and 1 queues will contain only write requests.
258 <name>hbase.ipc.server.callqueue.scan.ratio</name>
260 <description>Given the number of read call queues, calculated from the total number
261 of call queues multiplied by the callqueue.read.ratio, the scan.ratio property
262 will split the read call queues into small-read and long-read queues.
263 A value lower than 0.5 means that there will be less long-read queues than short-read queues.
264 A value of 0.5 means that there will be the same number of short-read and long-read queues.
265 A value greater than 0.5 means that there will be more long-read queues than short-read queues
266 A value of 0 or 1 indicate to use the same set of queues for gets and scans.
268 Example: Given the total number of read call queues being 8
269 a scan.ratio of 0 or 1 means that: 8 queues will contain both long and short read requests.
270 a scan.ratio of 0.3 means that: 2 queues will contain only long-read requests
271 and 6 queues will contain only short-read requests.
272 a scan.ratio of 0.5 means that: 4 queues will contain only long-read requests
273 and 4 queues will contain only short-read requests.
274 a scan.ratio of 0.8 means that: 6 queues will contain only long-read requests
275 and 2 queues will contain only short-read requests.
279 <name>hbase.regionserver.msginterval</name>
281 <description>Interval between messages from the RegionServer to Master
282 in milliseconds.</description>
285 <name>hbase.regionserver.logroll.period</name>
286 <value>3600000</value>
287 <description>Period at which we will roll the commit log regardless
288 of how many edits it has.</description>
291 <name>hbase.regionserver.logroll.errors.tolerated</name>
293 <description>The number of consecutive WAL close errors we will allow
294 before triggering a server abort. A setting of 0 will cause the
295 region server to abort if closing the current WAL writer fails during
296 log rolling. Even a small value (2 or 3) will allow a region server
297 to ride over transient HDFS errors.</description>
300 <name>hbase.regionserver.hlog.reader.impl</name>
301 <value>org.apache.hadoop.hbase.regionserver.wal.ProtobufLogReader</value>
302 <description>The WAL file reader implementation.</description>
305 <name>hbase.regionserver.hlog.writer.impl</name>
306 <value>org.apache.hadoop.hbase.regionserver.wal.ProtobufLogWriter</value>
307 <description>The WAL file writer implementation.</description>
310 <name>hbase.regionserver.global.memstore.size</name>
312 <description>Maximum size of all memstores in a region server before new
313 updates are blocked and flushes are forced. Defaults to 40% of heap (0.4).
314 Updates are blocked and flushes are forced until size of all memstores
315 in a region server hits hbase.regionserver.global.memstore.size.lower.limit.
316 The default value in this configuration has been intentionally left empty in order to
317 honor the old hbase.regionserver.global.memstore.upperLimit property if present.
321 <name>hbase.regionserver.global.memstore.size.lower.limit</name>
323 <description>Maximum size of all memstores in a region server before flushes
324 are forced. Defaults to 95% of hbase.regionserver.global.memstore.size
325 (0.95). A 100% value for this value causes the minimum possible flushing
326 to occur when updates are blocked due to memstore limiting. The default
327 value in this configuration has been intentionally left empty in order to
328 honor the old hbase.regionserver.global.memstore.lowerLimit property if
333 <name>hbase.systemtables.compacting.memstore.type</name>
335 <description>Determines the type of memstore to be used for system tables like
336 META, namespace tables etc. By default NONE is the type and hence we use the
337 default memstore for all the system tables. If we need to use compacting
338 memstore for system tables then set this property to BASIC/EAGER
342 <name>hbase.regionserver.optionalcacheflushinterval</name>
343 <value>3600000</value>
345 Maximum amount of time an edit lives in memory before being automatically flushed.
346 Default 1 hour. Set it to 0 to disable automatic flushing.
350 <name>hbase.regionserver.dns.interface</name>
351 <value>default</value>
352 <description>The name of the Network Interface from which a region server
353 should report its IP address.</description>
356 <name>hbase.regionserver.dns.nameserver</name>
357 <value>default</value>
358 <description>The host name or IP address of the name server (DNS)
359 which a region server should use to determine the host name used by the
360 master for communication and display purposes.</description>
363 <name>hbase.regionserver.region.split.policy</name>
364 <value>org.apache.hadoop.hbase.regionserver.SteppingSplitPolicy</value>
366 A split policy determines when a region should be split. The various
367 other split policies that are available currently are BusyRegionSplitPolicy,
368 ConstantSizeRegionSplitPolicy, DisabledRegionSplitPolicy,
369 DelimitedKeyPrefixRegionSplitPolicy, KeyPrefixRegionSplitPolicy, and
370 SteppingSplitPolicy. DisabledRegionSplitPolicy blocks manual region splitting.
374 <name>hbase.regionserver.regionSplitLimit</name>
377 Limit for the number of regions after which no more region splitting
378 should take place. This is not hard limit for the number of regions
379 but acts as a guideline for the regionserver to stop splitting after
380 a certain limit. Default is set to 1000.
384 <!--ZooKeeper configuration-->
386 <name>zookeeper.session.timeout</name>
388 <description>ZooKeeper session timeout in milliseconds. It is used in two different ways.
389 First, this value is used in the ZK client that HBase uses to connect to the ensemble.
390 It is also used by HBase when it starts a ZK server and it is passed as the 'maxSessionTimeout'.
391 See https://zookeeper.apache.org/doc/current/zookeeperProgrammers.html#ch_zkSessions.
392 For example, if an HBase region server connects to a ZK ensemble that's also managed
393 by HBase, then the session timeout will be the one specified by this configuration.
394 But, a region server that connects to an ensemble managed with a different configuration
395 will be subjected that ensemble's maxSessionTimeout. So, even though HBase might propose
396 using 90 seconds, the ensemble can have a max timeout lower than this and it will take
397 precedence. The current default maxSessionTimeout that ZK ships with is 40 seconds, which is lower than
402 <name>zookeeper.znode.parent</name>
403 <value>/hbase</value>
404 <description>Root ZNode for HBase in ZooKeeper. All of HBase's ZooKeeper
405 files that are configured with a relative path will go under this node.
406 By default, all of HBase's ZooKeeper file paths are configured with a
407 relative path, so they will all go under this directory unless changed.
411 <name>zookeeper.znode.acl.parent</name>
413 <description>Root ZNode for access control lists.</description>
416 <name>hbase.zookeeper.dns.interface</name>
417 <value>default</value>
418 <description>The name of the Network Interface from which a ZooKeeper server
419 should report its IP address.</description>
422 <name>hbase.zookeeper.dns.nameserver</name>
423 <value>default</value>
424 <description>The host name or IP address of the name server (DNS)
425 which a ZooKeeper server should use to determine the host name used by the
426 master for communication and display purposes.</description>
429 The following three properties are used together to create the list of
430 host:peer_port:leader_port quorum servers for ZooKeeper.
433 <name>hbase.zookeeper.peerport</name>
435 <description>Port used by ZooKeeper peers to talk to each other.
436 See https://zookeeper.apache.org/doc/r3.4.10/zookeeperStarted.html#sc_RunningReplicatedZooKeeper
437 for more information.</description>
440 <name>hbase.zookeeper.leaderport</name>
442 <description>Port used by ZooKeeper for leader election.
443 See https://zookeeper.apache.org/doc/r3.4.10/zookeeperStarted.html#sc_RunningReplicatedZooKeeper
444 for more information.</description>
446 <!-- End of properties used to generate ZooKeeper host:port quorum list. -->
449 Beginning of properties that are directly mapped from ZooKeeper's zoo.cfg.
450 All properties with an "hbase.zookeeper.property." prefix are converted for
451 ZooKeeper's configuration. Hence, if you want to add an option from zoo.cfg,
452 e.g. "initLimit=10" you would append the following to your configuration:
454 <name>hbase.zookeeper.property.initLimit</name>
459 <name>hbase.zookeeper.property.initLimit</name>
461 <description>Property from ZooKeeper's config zoo.cfg.
462 The number of ticks that the initial synchronization phase can take.</description>
465 <name>hbase.zookeeper.property.syncLimit</name>
467 <description>Property from ZooKeeper's config zoo.cfg.
468 The number of ticks that can pass between sending a request and getting an
469 acknowledgment.</description>
472 <name>hbase.zookeeper.property.dataDir</name>
473 <value>${hbase.tmp.dir}/zookeeper</value>
474 <description>Property from ZooKeeper's config zoo.cfg.
475 The directory where the snapshot is stored.</description>
478 <name>hbase.zookeeper.property.clientPort</name>
480 <description>Property from ZooKeeper's config zoo.cfg.
481 The port at which the clients will connect.</description>
484 <name>hbase.zookeeper.property.maxClientCnxns</name>
486 <description>Property from ZooKeeper's config zoo.cfg.
487 Limit on number of concurrent connections (at the socket level) that a
488 single client, identified by IP address, may make to a single member of
489 the ZooKeeper ensemble. Set high to avoid zk connection issues running
490 standalone and pseudo-distributed.</description>
492 <!-- End of properties that are directly mapped from ZooKeeper's zoo.cfg -->
494 <!--Client configurations-->
496 <name>hbase.client.write.buffer</name>
497 <value>2097152</value>
498 <description>Default size of the BufferedMutator write buffer in bytes.
499 A bigger buffer takes more memory -- on both the client and server
500 side since server instantiates the passed write buffer to process
501 it -- but a larger buffer size reduces the number of RPCs made.
502 For an estimate of server-side memory-used, evaluate
503 hbase.client.write.buffer * hbase.regionserver.handler.count</description>
506 <name>hbase.client.pause</name>
508 <description>General client pause value. Used mostly as value to wait
509 before running a retry of a failed get, region lookup, etc.
510 See hbase.client.retries.number for description of how we backoff from
511 this initial pause amount and how this pause works w/ retries.</description>
514 <name>hbase.client.pause.cqtbe</name>
516 <description>Whether or not to use a special client pause for
517 CallQueueTooBigException (cqtbe). Set this property to a higher value
518 than hbase.client.pause if you observe frequent CQTBE from the same
519 RegionServer and the call queue there keeps full</description>
522 <name>hbase.client.retries.number</name>
524 <description>Maximum retries. Used as maximum for all retryable
525 operations such as the getting of a cell's value, starting a row update,
526 etc. Retry interval is a rough function based on hbase.client.pause. At
527 first we retry at this interval but then with backoff, we pretty quickly reach
528 retrying every ten seconds. See HConstants#RETRY_BACKOFF for how the backup
529 ramps up. Change this setting and hbase.client.pause to suit your workload.</description>
532 <name>hbase.client.max.total.tasks</name>
534 <description>The maximum number of concurrent mutation tasks a single HTable instance will
535 send to the cluster.</description>
538 <name>hbase.client.max.perserver.tasks</name>
540 <description>The maximum number of concurrent mutation tasks a single HTable instance will
541 send to a single region server.</description>
544 <name>hbase.client.max.perregion.tasks</name>
546 <description>The maximum number of concurrent mutation tasks the client will
547 maintain to a single Region. That is, if there is already
548 hbase.client.max.perregion.tasks writes in progress for this region, new puts
549 won't be sent to this region until some writes finishes.</description>
552 <name>hbase.client.perserver.requests.threshold</name>
553 <value>2147483647</value>
554 <description>The max number of concurrent pending requests for one server in all client threads
555 (process level). Exceeding requests will be thrown ServerTooBusyException immediately to prevent
556 user's threads being occupied and blocked by only one slow region server. If you use a fix
557 number of threads to access HBase in a synchronous way, set this to a suitable value which is
558 related to the number of threads will help you. See
559 https://issues.apache.org/jira/browse/HBASE-16388 for details.</description>
562 <name>hbase.client.scanner.caching</name>
563 <value>2147483647</value>
564 <description>Number of rows that we try to fetch when calling next
565 on a scanner if it is not served from (local, client) memory. This configuration
566 works together with hbase.client.scanner.max.result.size to try and use the
567 network efficiently. The default value is Integer.MAX_VALUE by default so that
568 the network will fill the chunk size defined by hbase.client.scanner.max.result.size
569 rather than be limited by a particular number of rows since the size of rows varies
570 table to table. If you know ahead of time that you will not require more than a certain
571 number of rows from a scan, this configuration should be set to that row limit via
572 Scan#setCaching. Higher caching values will enable faster scanners but will eat up more
573 memory and some calls of next may take longer and longer times when the cache is empty.
574 Do not set this value such that the time between invocations is greater than the scanner
575 timeout; i.e. hbase.client.scanner.timeout.period</description>
578 <name>hbase.client.keyvalue.maxsize</name>
579 <value>10485760</value>
580 <description>Specifies the combined maximum allowed size of a KeyValue
581 instance. This is to set an upper boundary for a single entry saved in a
582 storage file. Since they cannot be split it helps avoiding that a region
583 cannot be split any further because the data is too large. It seems wise
584 to set this to a fraction of the maximum region size. Setting it to zero
585 or less disables the check.</description>
588 <name>hbase.server.keyvalue.maxsize</name>
589 <value>10485760</value>
590 <description>Maximum allowed size of an individual cell, inclusive of value and all key
591 components. A value of 0 or less disables the check.
592 The default value is 10MB.
593 This is a safety setting to protect the server from OOM situations.
597 <name>hbase.client.scanner.timeout.period</name>
599 <description>Client scanner lease period in milliseconds.</description>
602 <name>hbase.client.localityCheck.threadPoolSize</name>
606 <!--Miscellaneous configuration-->
608 <name>hbase.bulkload.retries.number</name>
610 <description>Maximum retries. This is maximum number of iterations
611 to atomic bulk loads are attempted in the face of splitting operations
612 0 means never give up.</description>
615 <name>hbase.master.balancer.maxRitPercent</name>
617 <description>The max percent of regions in transition when balancing.
618 The default value is 1.0. So there are no balancer throttling. If set this config to 0.01,
619 It means that there are at most 1% regions in transition when balancing.
620 Then the cluster's availability is at least 99% when balancing.</description>
623 <name>hbase.balancer.period</name>
624 <value>300000</value>
625 <description>Period at which the region balancer runs in the Master.</description>
628 <name>hbase.normalizer.period</name>
629 <value>300000</value>
630 <description>Period at which the region normalizer runs in the Master.</description>
633 <name>hbase.normalizer.min.region.count</name>
635 <description>configure the minimum number of regions</description>
638 <name>hbase.normalizer.min.region.merge.age</name>
640 <description>configure the minimum age in days for region before it is considered for merge while
641 normalizing</description>
644 <name>hbase.regions.slop</name>
646 <description>Rebalance if any regionserver has average + (average * slop) regions.
647 The default value of this parameter is 0.001 in StochasticLoadBalancer (the default load balancer),
648 while the default is 0.2 in other load balancers (i.e., SimpleLoadBalancer).</description>
651 <name>hbase.server.thread.wakefrequency</name>
653 <description>In master side, this config is the period used for FS related behaviors:
654 checking if hdfs is out of safe mode, setting or checking hbase.version file,
655 setting or checking hbase.id file. Using default value should be fine.
656 In regionserver side, this config is used in several places: flushing check interval,
657 compaction check interval, wal rolling check interval. Specially, admin can tune
658 flushing and compaction check interval by hbase.regionserver.flush.check.period
659 and hbase.regionserver.compaction.check.period. (in milliseconds)</description>
662 <name>hbase.regionserver.flush.check.period</name>
663 <value>${hbase.server.thread.wakefrequency}</value>
664 <description>It determines the flushing check period of PeriodicFlusher in regionserver.
665 If unset, it uses hbase.server.thread.wakefrequency as default value.
666 (in milliseconds)</description>
669 <name>hbase.regionserver.compaction.check.period</name>
670 <value>${hbase.server.thread.wakefrequency}</value>
671 <description>It determines the compaction check period of CompactionChecker in regionserver.
672 If unset, it uses hbase.server.thread.wakefrequency as default value.
673 (in milliseconds)</description>
676 <name>hbase.server.versionfile.writeattempts</name>
679 How many times to retry attempting to write a version file
680 before just aborting. Each attempt is separated by the
681 hbase.server.thread.wakefrequency milliseconds.</description>
684 <name>hbase.hregion.memstore.flush.size</name>
685 <value>134217728</value>
687 Memstore will be flushed to disk if size of the memstore
688 exceeds this number of bytes. Value is checked by a thread that runs
689 every hbase.server.thread.wakefrequency.</description>
692 <name>hbase.hregion.percolumnfamilyflush.size.lower.bound.min</name>
693 <value>16777216</value>
695 If FlushLargeStoresPolicy is used and there are multiple column families,
696 then every time that we hit the total memstore limit, we find out all the
697 column families whose memstores exceed a "lower bound" and only flush them
698 while retaining the others in memory. The "lower bound" will be
699 "hbase.hregion.memstore.flush.size / column_family_number" by default
700 unless value of this property is larger than that. If none of the families
701 have their memstore size more than lower bound, all the memstores will be
702 flushed (just as usual).
706 <name>hbase.hregion.preclose.flush.size</name>
707 <value>5242880</value>
709 If the memstores in a region are this size or larger when we go
710 to close, run a "pre-flush" to clear out memstores before we put up
711 the region closed flag and take the region offline. On close,
712 a flush is run under the close flag to empty memory. During
713 this time the region is offline and we are not taking on any writes.
714 If the memstore content is large, this flush could take a long time to
715 complete. The preflush is meant to clean out the bulk of the memstore
716 before putting up the close flag and taking the region offline so the
717 flush that runs under the close flag has little to do.</description>
720 <name>hbase.hregion.memstore.block.multiplier</name>
723 Block updates if memstore has hbase.hregion.memstore.block.multiplier
724 times hbase.hregion.memstore.flush.size bytes. Useful preventing
725 runaway memstore during spikes in update traffic. Without an
726 upper-bound, memstore fills such that when it flushes the
727 resultant flush files take a long time to compact or split, or
728 worse, we OOME.</description>
731 <name>hbase.hregion.memstore.mslab.enabled</name>
734 Enables the MemStore-Local Allocation Buffer,
735 a feature which works to prevent heap fragmentation under
736 heavy write loads. This can reduce the frequency of stop-the-world
737 GC pauses on large heaps.
741 <name>hbase.hregion.memstore.mslab.chunksize</name>
742 <value>2097152</value>
743 <description>The maximum byte size of a chunk in the MemStoreLAB. Unit: bytes</description>
746 <name>hbase.regionserver.offheap.global.memstore.size</name>
748 <description>The amount of off-heap memory all MemStores in a RegionServer may use.
749 A value of 0 means that no off-heap memory will be used and all chunks in MSLAB
750 will be HeapByteBuffer, otherwise the non-zero value means how many megabyte of
751 off-heap memory will be used for chunks in MSLAB and all chunks in MSLAB will be
752 DirectByteBuffer. Unit: megabytes.
756 <name>hbase.hregion.memstore.mslab.max.allocation</name>
757 <value>262144</value>
758 <description>The maximal size of one allocation in the MemStoreLAB, if the desired byte
759 size exceed this threshold then it will be just allocated from JVM heap rather than MemStoreLAB.
763 <name>hbase.hregion.max.filesize</name>
764 <value>10737418240</value>
766 Maximum HFile size. If the sum of the sizes of a region's HFiles has grown to exceed this
767 value, the region is split in two.</description>
770 <name>hbase.hregion.majorcompaction</name>
771 <value>604800000</value>
772 <description>Time between major compactions, expressed in milliseconds. Set to 0 to disable
773 time-based automatic major compactions. User-requested and size-based major compactions will
774 still run. This value is multiplied by hbase.hregion.majorcompaction.jitter to cause
775 compaction to start at a somewhat-random time during a given window of time. The default value
776 is 7 days, expressed in milliseconds. If major compactions are causing disruption in your
777 environment, you can configure them to run at off-peak times for your deployment, or disable
778 time-based major compactions by setting this parameter to 0, and run major compactions in a
779 cron job or by another external mechanism.</description>
782 <name>hbase.hregion.majorcompaction.jitter</name>
784 <description>A multiplier applied to hbase.hregion.majorcompaction to cause compaction to occur
785 a given amount of time either side of hbase.hregion.majorcompaction. The smaller the number,
786 the closer the compactions will happen to the hbase.hregion.majorcompaction
787 interval.</description>
790 <name>hbase.hstore.compactionThreshold</name>
792 <description> If more than this number of StoreFiles exist in any one Store
793 (one StoreFile is written per flush of MemStore), a compaction is run to rewrite all
794 StoreFiles into a single StoreFile. Larger values delay compaction, but when compaction does
795 occur, it takes longer to complete.</description>
798 <name>hbase.regionserver.compaction.enabled</name>
800 <description>Enable/disable compactions on by setting true/false.
801 We can further switch compactions dynamically with the
802 compaction_switch shell command.</description>
805 <name>hbase.hstore.flusher.count</name>
807 <description> The number of flush threads. With fewer threads, the MemStore flushes will be
808 queued. With more threads, the flushes will be executed in parallel, increasing the load on
809 HDFS, and potentially causing more compactions. </description>
812 <name>hbase.hstore.blockingStoreFiles</name>
814 <description> If more than this number of StoreFiles exist in any one Store (one StoreFile
815 is written per flush of MemStore), updates are blocked for this region until a compaction is
816 completed, or until hbase.hstore.blockingWaitTime has been exceeded.</description>
819 <name>hbase.hstore.blockingWaitTime</name>
821 <description> The time for which a region will block updates after reaching the StoreFile limit
822 defined by hbase.hstore.blockingStoreFiles. After this time has elapsed, the region will stop
823 blocking updates even if a compaction has not been completed.</description>
826 <name>hbase.hstore.compaction.min</name>
828 <description>The minimum number of StoreFiles which must be eligible for compaction before
829 compaction can run. The goal of tuning hbase.hstore.compaction.min is to avoid ending up with
830 too many tiny StoreFiles to compact. Setting this value to 2 would cause a minor compaction
831 each time you have two StoreFiles in a Store, and this is probably not appropriate. If you
832 set this value too high, all the other values will need to be adjusted accordingly. For most
833 cases, the default value is appropriate. In previous versions of HBase, the parameter
834 hbase.hstore.compaction.min was named hbase.hstore.compactionThreshold.</description>
837 <name>hbase.hstore.compaction.max</name>
839 <description>The maximum number of StoreFiles which will be selected for a single minor
840 compaction, regardless of the number of eligible StoreFiles. Effectively, the value of
841 hbase.hstore.compaction.max controls the length of time it takes a single compaction to
842 complete. Setting it larger means that more StoreFiles are included in a compaction. For most
843 cases, the default value is appropriate.</description>
846 <name>hbase.hstore.compaction.min.size</name>
847 <value>134217728</value>
848 <description>A StoreFile (or a selection of StoreFiles, when using ExploringCompactionPolicy)
849 smaller than this size will always be eligible for minor compaction.
850 HFiles this size or larger are evaluated by hbase.hstore.compaction.ratio to determine if
851 they are eligible. Because this limit represents the "automatic include" limit for all
852 StoreFiles smaller than this value, this value may need to be reduced in write-heavy
853 environments where many StoreFiles in the 1-2 MB range are being flushed, because every
854 StoreFile will be targeted for compaction and the resulting StoreFiles may still be under the
855 minimum size and require further compaction. If this parameter is lowered, the ratio check is
856 triggered more quickly. This addressed some issues seen in earlier versions of HBase but
857 changing this parameter is no longer necessary in most situations. Default: 128 MB expressed
858 in bytes.</description>
861 <name>hbase.hstore.compaction.max.size</name>
862 <value>9223372036854775807</value>
863 <description>A StoreFile (or a selection of StoreFiles, when using ExploringCompactionPolicy)
864 larger than this size will be excluded from compaction. The effect of
865 raising hbase.hstore.compaction.max.size is fewer, larger StoreFiles that do not get
866 compacted often. If you feel that compaction is happening too often without much benefit, you
867 can try raising this value. Default: the value of LONG.MAX_VALUE, expressed in bytes.</description>
870 <name>hbase.hstore.compaction.ratio</name>
872 <description>For minor compaction, this ratio is used to determine whether a given StoreFile
873 which is larger than hbase.hstore.compaction.min.size is eligible for compaction. Its
874 effect is to limit compaction of large StoreFiles. The value of hbase.hstore.compaction.ratio
875 is expressed as a floating-point decimal. A large ratio, such as 10, will produce a single
876 giant StoreFile. Conversely, a low value, such as .25, will produce behavior similar to the
877 BigTable compaction algorithm, producing four StoreFiles. A moderate value of between 1.0 and
878 1.4 is recommended. When tuning this value, you are balancing write costs with read costs.
879 Raising the value (to something like 1.4) will have more write costs, because you will
880 compact larger StoreFiles. However, during reads, HBase will need to seek through fewer
881 StoreFiles to accomplish the read. Consider this approach if you cannot take advantage of
882 Bloom filters. Otherwise, you can lower this value to something like 1.0 to reduce the
883 background cost of writes, and use Bloom filters to control the number of StoreFiles touched
884 during reads. For most cases, the default value is appropriate.</description>
887 <name>hbase.hstore.compaction.ratio.offpeak</name>
889 <description>Allows you to set a different (by default, more aggressive) ratio for determining
890 whether larger StoreFiles are included in compactions during off-peak hours. Works in the
891 same way as hbase.hstore.compaction.ratio. Only applies if hbase.offpeak.start.hour and
892 hbase.offpeak.end.hour are also enabled.</description>
895 <name>hbase.hstore.time.to.purge.deletes</name>
897 <description>The amount of time to delay purging of delete markers with future timestamps. If
898 unset, or set to 0, all delete markers, including those with future timestamps, are purged
899 during the next major compaction. Otherwise, a delete marker is kept until the major compaction
900 which occurs after the marker's timestamp plus the value of this setting, in milliseconds.
904 <name>hbase.offpeak.start.hour</name>
906 <description>The start of off-peak hours, expressed as an integer between 0 and 23, inclusive.
907 Set to -1 to disable off-peak.</description>
910 <name>hbase.offpeak.end.hour</name>
912 <description>The end of off-peak hours, expressed as an integer between 0 and 23, inclusive. Set
913 to -1 to disable off-peak.</description>
916 <name>hbase.regionserver.thread.compaction.throttle</name>
917 <value>2684354560</value>
918 <description>There are two different thread pools for compactions, one for large compactions and
919 the other for small compactions. This helps to keep compaction of lean tables (such as
920 hbase:meta) fast. If a compaction is larger than this threshold, it
921 goes into the large compaction pool. In most cases, the default value is appropriate. Default:
922 2 x hbase.hstore.compaction.max x hbase.hregion.memstore.flush.size (which defaults to 128MB).
923 The value field assumes that the value of hbase.hregion.memstore.flush.size is unchanged from
924 the default.</description>
927 <name>hbase.regionserver.majorcompaction.pagecache.drop</name>
929 <description>Specifies whether to drop pages read/written into the system page cache by
930 major compactions. Setting it to true helps prevent major compactions from
931 polluting the page cache, which is almost always required, especially for clusters
932 with low/moderate memory to storage ratio.</description>
935 <name>hbase.regionserver.minorcompaction.pagecache.drop</name>
937 <description>Specifies whether to drop pages read/written into the system page cache by
938 minor compactions. Setting it to true helps prevent minor compactions from
939 polluting the page cache, which is most beneficial on clusters with low
940 memory to storage ratio or very write heavy clusters. You may want to set it to
941 false under moderate to low write workload when bulk of the reads are
942 on the most recently written data.</description>
945 <name>hbase.hstore.compaction.kv.max</name>
947 <description>The maximum number of KeyValues to read and then write in a batch when flushing or
948 compacting. Set this lower if you have big KeyValues and problems with Out Of Memory
949 Exceptions Set this higher if you have wide, small rows. </description>
952 <name>hbase.storescanner.parallel.seek.enable</name>
955 Enables StoreFileScanner parallel-seeking in StoreScanner,
956 a feature which can reduce response latency under special conditions.</description>
959 <name>hbase.storescanner.parallel.seek.threads</name>
962 The default thread pool size if parallel-seeking feature enabled.</description>
965 <name>hfile.block.cache.policy</name>
967 <description>The eviction policy for the L1 block cache (LRU or TinyLFU).</description>
970 <name>hfile.block.cache.size</name>
972 <description>Percentage of maximum heap (-Xmx setting) to allocate to block cache
973 used by a StoreFile. Default of 0.4 means allocate 40%.
974 Set to 0 to disable but it's not recommended; you need at least
975 enough cache to hold the storefile indices.</description>
978 <name>hfile.block.index.cacheonwrite</name>
980 <description>This allows to put non-root multi-level index blocks into the block
981 cache at the time the index is being written.</description>
984 <name>hfile.index.block.max.size</name>
985 <value>131072</value>
986 <description>When the size of a leaf-level, intermediate-level, or root-level
987 index block in a multi-level block index grows to this size, the
988 block is written out and a new block is started.</description>
991 <name>hbase.bucketcache.ioengine</name>
993 <description>Where to store the contents of the bucketcache. One of: offheap,
994 file, files, mmap or pmem. If a file or files, set it to file(s):PATH_TO_FILE.
995 mmap means the content will be in an mmaped file. Use mmap:PATH_TO_FILE. 'pmem'
996 is bucket cache over a file on the persistent memory device.
997 Use pmem:PATH_TO_FILE.
998 See http://hbase.apache.org/book.html#offheap.blockcache for more information.
1002 <name>hbase.hstore.compaction.throughput.lower.bound</name>
1003 <value>52428800</value>
1004 <description>The target lower bound on aggregate compaction throughput, in bytes/sec. Allows
1005 you to tune the minimum available compaction throughput when the
1006 PressureAwareCompactionThroughputController throughput controller is active. (It is active by
1007 default.)</description>
1010 <name>hbase.hstore.compaction.throughput.higher.bound</name>
1011 <value>104857600</value>
1012 <description>The target upper bound on aggregate compaction throughput, in bytes/sec. Allows
1013 you to control aggregate compaction throughput demand when the
1014 PressureAwareCompactionThroughputController throughput controller is active. (It is active by
1015 default.) The maximum throughput will be tuned between the lower and upper bounds when
1016 compaction pressure is within the range [0.0, 1.0]. If compaction pressure is 1.0 or greater
1017 the higher bound will be ignored until pressure returns to the normal range.</description>
1020 <name>hbase.bucketcache.size</name>
1022 <description>A float that EITHER represents a percentage of total heap memory
1023 size to give to the cache (if < 1.0) OR, it is the total capacity in
1024 megabytes of BucketCache. Default: 0.0</description>
1027 <name>hbase.bucketcache.bucket.sizes</name>
1029 <description>A comma-separated list of sizes for buckets for the bucketcache.
1030 Can be multiple sizes. List block sizes in order from smallest to largest.
1031 The sizes you use will depend on your data access patterns.
1032 Must be a multiple of 256 else you will run into
1033 'java.io.IOException: Invalid HFile block magic' when you go to read from cache.
1034 If you specify no values here, then you pick up the default bucketsizes set
1035 in code (See BucketAllocator#DEFAULT_BUCKET_SIZES).
1039 <name>hfile.format.version</name>
1041 <description>The HFile format version to use for new files.
1042 Version 3 adds support for tags in hfiles (See http://hbase.apache.org/book.html#hbase.tags).
1043 Also see the configuration 'hbase.replication.rpc.codec'.
1047 <name>hfile.block.bloom.cacheonwrite</name>
1048 <value>false</value>
1049 <description>Enables cache-on-write for inline blocks of a compound Bloom filter.</description>
1052 <name>io.storefile.bloom.block.size</name>
1053 <value>131072</value>
1054 <description>The size in bytes of a single block ("chunk") of a compound Bloom
1055 filter. This size is approximate, because Bloom blocks can only be
1056 inserted at data block boundaries, and the number of keys per data
1057 block varies.</description>
1060 <name>hbase.rs.cacheblocksonwrite</name>
1061 <value>false</value>
1062 <description>Whether an HFile block should be added to the block cache when the
1063 block is finished.</description>
1066 <name>hbase.rpc.timeout</name>
1067 <value>60000</value>
1068 <description>This is for the RPC layer to define how long (millisecond) HBase client applications
1069 take for a remote call to time out. It uses pings to check connections
1070 but will eventually throw a TimeoutException.</description>
1073 <name>hbase.client.operation.timeout</name>
1074 <value>1200000</value>
1075 <description>Operation timeout is a top-level restriction (millisecond) that makes sure a
1076 blocking operation in Table will not be blocked more than this. In each operation, if rpc
1077 request fails because of timeout or other reason, it will retry until success or throw
1078 RetriesExhaustedException. But if the total time being blocking reach the operation timeout
1079 before retries exhausted, it will break early and throw SocketTimeoutException.</description>
1082 <name>hbase.cells.scanned.per.heartbeat.check</name>
1083 <value>10000</value>
1084 <description>The number of cells scanned in between heartbeat checks. Heartbeat
1085 checks occur during the processing of scans to determine whether or not the
1086 server should stop scanning in order to send back a heartbeat message to the
1087 client. Heartbeat messages are used to keep the client-server connection alive
1088 during long running scans. Small values mean that the heartbeat checks will
1089 occur more often and thus will provide a tighter bound on the execution time of
1090 the scan. Larger values mean that the heartbeat checks occur less frequently
1094 <name>hbase.rpc.shortoperation.timeout</name>
1095 <value>10000</value>
1096 <description>This is another version of "hbase.rpc.timeout". For those RPC operation
1097 within cluster, we rely on this configuration to set a short timeout limitation
1098 for short operation. For example, short rpc timeout for region server's trying
1099 to report to active master can benefit quicker master failover process.</description>
1102 <name>hbase.ipc.client.tcpnodelay</name>
1104 <description>Set no delay on rpc socket connections. See
1105 http://docs.oracle.com/javase/1.5.0/docs/api/java/net/Socket.html#getTcpNoDelay()</description>
1108 <name>hbase.regionserver.hostname</name>
1110 <description>This config is for experts: don't set its value unless you really know what you are doing.
1111 When set to a non-empty value, this represents the (external facing) hostname for the underlying server.
1112 See https://issues.apache.org/jira/browse/HBASE-12954 for details.</description>
1115 <name>hbase.regionserver.hostname.disable.master.reversedns</name>
1116 <value>false</value>
1117 <description>This config is for experts: don't set its value unless you really know what you are doing.
1118 When set to true, regionserver will use the current node hostname for the servername and HMaster will
1119 skip reverse DNS lookup and use the hostname sent by regionserver instead. Note that this config and
1120 hbase.regionserver.hostname are mutually exclusive. See https://issues.apache.org/jira/browse/HBASE-18226
1121 for more details.</description>
1123 <!-- The following properties configure authentication information for
1124 HBase processes when using Kerberos security. There are no default
1125 values, included here for documentation purposes -->
1127 <name>hbase.master.keytab.file</name>
1129 <description>Full path to the kerberos keytab file to use for logging in
1130 the configured HMaster server principal.</description>
1133 <name>hbase.master.kerberos.principal</name>
1135 <description>Ex. "hbase/_HOST@EXAMPLE.COM". The kerberos principal name
1136 that should be used to run the HMaster process. The principal name should
1137 be in the form: user/hostname@DOMAIN. If "_HOST" is used as the hostname
1138 portion, it will be replaced with the actual hostname of the running
1139 instance.</description>
1142 <name>hbase.regionserver.keytab.file</name>
1144 <description>Full path to the kerberos keytab file to use for logging in
1145 the configured HRegionServer server principal.</description>
1148 <name>hbase.regionserver.kerberos.principal</name>
1150 <description>Ex. "hbase/_HOST@EXAMPLE.COM". The kerberos principal name
1151 that should be used to run the HRegionServer process. The principal name
1152 should be in the form: user/hostname@DOMAIN. If "_HOST" is used as the
1153 hostname portion, it will be replaced with the actual hostname of the
1154 running instance. An entry for this principal must exist in the file
1155 specified in hbase.regionserver.keytab.file</description>
1157 <!-- Additional configuration specific to HBase security -->
1159 <name>hadoop.policy.file</name>
1160 <value>hbase-policy.xml</value>
1161 <description>The policy configuration file used by RPC servers to make
1162 authorization decisions on client requests. Only used when HBase
1163 security is enabled.</description>
1166 <name>hbase.superuser</name>
1168 <description>List of users or groups (comma-separated), who are allowed
1169 full privileges, regardless of stored ACLs, across the cluster.
1170 Only used when HBase security is enabled.</description>
1173 <name>hbase.auth.key.update.interval</name>
1174 <value>86400000</value>
1175 <description>The update interval for master key for authentication tokens
1176 in servers in milliseconds. Only used when HBase security is enabled.</description>
1179 <name>hbase.auth.token.max.lifetime</name>
1180 <value>604800000</value>
1181 <description>The maximum lifetime in milliseconds after which an
1182 authentication token expires. Only used when HBase security is enabled.</description>
1185 <name>hbase.ipc.client.fallback-to-simple-auth-allowed</name>
1186 <value>false</value>
1187 <description>When a client is configured to attempt a secure connection, but attempts to
1188 connect to an insecure server, that server may instruct the client to
1189 switch to SASL SIMPLE (unsecure) authentication. This setting controls
1190 whether or not the client will accept this instruction from the server.
1191 When false (the default), the client will not allow the fallback to SIMPLE
1192 authentication, and will abort the connection.</description>
1195 <name>hbase.ipc.server.fallback-to-simple-auth-allowed</name>
1196 <value>false</value>
1197 <description>When a server is configured to require secure connections, it will
1198 reject connection attempts from clients using SASL SIMPLE (unsecure) authentication.
1199 This setting allows secure servers to accept SASL SIMPLE connections from clients
1200 when the client requests. When false (the default), the server will not allow the fallback
1201 to SIMPLE authentication, and will reject the connection. WARNING: This setting should ONLY
1202 be used as a temporary measure while converting clients over to secure authentication. It
1203 MUST BE DISABLED for secure operation.</description>
1206 <name>hbase.display.keys</name>
1208 <description>When this is set to true the webUI and such will display all start/end keys
1209 as part of the table details, region names, etc. When this is set to false,
1210 the keys are hidden.</description>
1213 <name>hbase.coprocessor.enabled</name>
1215 <description>Enables or disables coprocessor loading. If 'false'
1216 (disabled), any other coprocessor related configuration will be ignored.
1220 <name>hbase.coprocessor.user.enabled</name>
1222 <description>Enables or disables user (aka. table) coprocessor loading.
1223 If 'false' (disabled), any table coprocessor attributes in table
1224 descriptors will be ignored. If "hbase.coprocessor.enabled" is 'false'
1225 this setting has no effect.
1229 <name>hbase.coprocessor.region.classes</name>
1231 <description>A comma-separated list of region observer or endpoint coprocessors
1232 that are loaded by default on all tables. For any override coprocessor method,
1233 these classes will be called in order. After implementing your own Coprocessor,
1234 add it to HBase's classpath and add the fully qualified class name here.
1235 A coprocessor can also be loaded on demand by setting HTableDescriptor or the
1236 HBase shell.</description>
1239 <name>hbase.coprocessor.master.classes</name>
1241 <description>A comma-separated list of
1242 org.apache.hadoop.hbase.coprocessor.MasterObserver coprocessors that are
1243 loaded by default on the active HMaster process. For any implemented
1244 coprocessor methods, the listed classes will be called in order. After
1245 implementing your own MasterObserver, just put it in HBase's classpath
1246 and add the fully qualified class name here.</description>
1249 <name>hbase.coprocessor.abortonerror</name>
1251 <description>Set to true to cause the hosting server (master or regionserver)
1252 to abort if a coprocessor fails to load, fails to initialize, or throws an
1253 unexpected Throwable object. Setting this to false will allow the server to
1254 continue execution but the system wide state of the coprocessor in question
1255 will become inconsistent as it will be properly executing in only a subset
1256 of servers, so this is most useful for debugging only.</description>
1259 <name>hbase.rest.port</name>
1261 <description>The port for the HBase REST server.</description>
1264 <name>hbase.rest.readonly</name>
1265 <value>false</value>
1266 <description>Defines the mode the REST server will be started in. Possible values are:
1267 false: All HTTP methods are permitted - GET/PUT/POST/DELETE.
1268 true: Only the GET method is permitted.</description>
1271 <name>hbase.rest.threads.max</name>
1273 <description>The maximum number of threads of the REST server thread pool.
1274 Threads in the pool are reused to process REST requests. This
1275 controls the maximum number of requests processed concurrently.
1276 It may help to control the memory used by the REST server to
1277 avoid OOM issues. If the thread pool is full, incoming requests
1278 will be queued up and wait for some free threads.</description>
1281 <name>hbase.rest.threads.min</name>
1283 <description>The minimum number of threads of the REST server thread pool.
1284 The thread pool always has at least these number of threads so
1285 the REST server is ready to serve incoming requests.</description>
1288 <name>hbase.rest.support.proxyuser</name>
1289 <value>false</value>
1290 <description>Enables running the REST server to support proxy-user mode.</description>
1292 <property skipInDoc="true">
1293 <name>hbase.defaults.for.version</name>
1294 <value>@@@VERSION@@@</value>
1295 <description>This defaults file was compiled for version ${project.version}. This variable is used
1296 to make sure that a user doesn't have an old version of hbase-default.xml on the
1297 classpath.</description>
1300 <name>hbase.defaults.for.version.skip</name>
1301 <value>false</value>
1302 <description>Set to true to skip the 'hbase.defaults.for.version' check.
1303 Setting this to true can be useful in contexts other than
1304 the other side of a maven generation; i.e. running in an
1305 IDE. You'll want to set this boolean to true to avoid
1306 seeing the RuntimeException complaint: "hbase-default.xml file
1307 seems to be for and old version of HBase (\${hbase.version}), this
1308 version is X.X.X-SNAPSHOT"</description>
1311 <name>hbase.table.lock.enable</name>
1313 <description>Set to true to enable locking the table in zookeeper for schema change operations.
1314 Table locking from master prevents concurrent schema modifications to corrupt table
1315 state.</description>
1318 <name>hbase.table.max.rowsize</name>
1319 <value>1073741824</value>
1321 Maximum size of single row in bytes (default is 1 Gb) for Get'ting
1322 or Scan'ning without in-row scan flag set. If row size exceeds this limit
1323 RowTooBigException is thrown to client.
1327 <name>hbase.thrift.minWorkerThreads</name>
1329 <description>The "core size" of the thread pool. New threads are created on every
1330 connection until this many threads are created.</description>
1333 <name>hbase.thrift.maxWorkerThreads</name>
1335 <description>The maximum size of the thread pool. When the pending request queue
1336 overflows, new threads are created until their number reaches this number.
1337 After that, the server starts dropping connections.</description>
1340 <name>hbase.thrift.maxQueuedRequests</name>
1342 <description>The maximum number of pending Thrift connections waiting in the queue. If
1343 there are no idle threads in the pool, the server queues requests. Only
1344 when the queue overflows, new threads are added, up to
1345 hbase.thrift.maxQueuedRequests threads.</description>
1348 <name>hbase.regionserver.thrift.framed</name>
1349 <value>false</value>
1350 <description>Use Thrift TFramedTransport on the server side.
1351 This is the recommended transport for thrift servers and requires a similar setting
1352 on the client side. Changing this to false will select the default transport,
1353 vulnerable to DoS when malformed requests are issued due to THRIFT-601.
1357 <name>hbase.regionserver.thrift.framed.max_frame_size_in_mb</name>
1359 <description>Default frame size when using framed transport, in MB</description>
1362 <name>hbase.regionserver.thrift.compact</name>
1363 <value>false</value>
1364 <description>Use Thrift TCompactProtocol binary serialization protocol.</description>
1367 <name>hbase.rootdir.perms</name>
1369 <description>FS Permissions for the root data subdirectory in a secure (kerberos) setup.
1370 When master starts, it creates the rootdir with this permissions or sets the permissions
1371 if it does not match.</description>
1374 <name>hbase.wal.dir.perms</name>
1376 <description>FS Permissions for the root WAL directory in a secure(kerberos) setup.
1377 When master starts, it creates the WAL dir with this permissions or sets the permissions
1378 if it does not match.</description>
1381 <name>hbase.data.umask.enable</name>
1382 <value>false</value>
1383 <description>Enable, if true, that file permissions should be assigned
1384 to the files written by the regionserver</description>
1387 <name>hbase.data.umask</name>
1389 <description>File permissions that should be used to write data
1390 files when hbase.data.umask.enable is true</description>
1393 <name>hbase.snapshot.enabled</name>
1395 <description>Set to true to allow snapshots to be taken / restored / cloned.</description>
1398 <name>hbase.snapshot.restore.take.failsafe.snapshot</name>
1400 <description>Set to true to take a snapshot before the restore operation.
1401 The snapshot taken will be used in case of failure, to restore the previous state.
1402 At the end of the restore operation this snapshot will be deleted</description>
1405 <name>hbase.snapshot.restore.failsafe.name</name>
1406 <value>hbase-failsafe-{snapshot.name}-{restore.timestamp}</value>
1407 <description>Name of the failsafe snapshot taken by the restore operation.
1408 You can use the {snapshot.name}, {table.name} and {restore.timestamp} variables
1409 to create a name based on what you are restoring.</description>
1412 <name>hbase.snapshot.working.dir</name>
1414 <description>Location where the snapshotting process will occur. The location of the
1415 completed snapshots will not change, but the temporary directory where the snapshot
1416 process occurs will be set to this location. This can be a separate filesystem than
1417 the root directory, for performance increase purposes. See HBASE-21098 for more
1418 information</description>
1421 <name>hbase.server.compactchecker.interval.multiplier</name>
1423 <description>The number that determines how often we scan to see if compaction is necessary.
1424 Normally, compactions are done after some events (such as memstore flush), but if
1425 region didn't receive a lot of writes for some time, or due to different compaction
1426 policies, it may be necessary to check it periodically. The interval between checks is
1427 hbase.server.compactchecker.interval.multiplier multiplied by
1428 hbase.server.thread.wakefrequency.</description>
1431 <name>hbase.lease.recovery.timeout</name>
1432 <value>900000</value>
1433 <description>How long we wait on dfs lease recovery in total before giving up.</description>
1436 <name>hbase.lease.recovery.dfs.timeout</name>
1437 <value>64000</value>
1438 <description>How long between dfs recover lease invocations. Should be larger than the sum of
1439 the time it takes for the namenode to issue a block recovery command as part of
1440 datanode; dfs.heartbeat.interval and the time it takes for the primary
1441 datanode, performing block recovery to timeout on a dead datanode; usually
1442 dfs.client.socket-timeout. See the end of HBASE-8389 for more.</description>
1445 <name>hbase.column.max.version</name>
1447 <description>New column family descriptors will use this value as the default number of versions
1448 to keep.</description>
1451 <name>dfs.client.read.shortcircuit</name>
1452 <value>false</value>
1454 If set to true, this configuration parameter enables short-circuit local
1459 <name>dfs.domain.socket.path</name>
1462 This is a path to a UNIX domain socket that will be used for
1463 communication between the DataNode and local HDFS clients, if
1464 dfs.client.read.shortcircuit is set to true. If the string "_PORT" is
1465 present in this path, it will be replaced by the TCP port of the DataNode.
1466 Be careful about permissions for the directory that hosts the shared
1467 domain socket; dfsclient will complain if open to other users than the HBase user.
1471 <name>hbase.dfs.client.read.shortcircuit.buffer.size</name>
1472 <value>131072</value>
1473 <description>If the DFSClient configuration
1474 dfs.client.read.shortcircuit.buffer.size is unset, we will
1475 use what is configured here as the short circuit read default
1476 direct byte buffer size. DFSClient native default is 1MB; HBase
1477 keeps its HDFS files open so number of file blocks * 1MB soon
1478 starts to add up and threaten OOME because of a shortage of
1479 direct memory. So, we set it down from the default. Make
1480 it > the default hbase block size set in the HColumnDescriptor
1481 which is usually 64k.
1485 <name>hbase.regionserver.checksum.verify</name>
1488 If set to true (the default), HBase verifies the checksums for hfile
1489 blocks. HBase writes checksums inline with the data when it writes out
1490 hfiles. HDFS (as of this writing) writes checksums to a separate file
1491 than the data file necessitating extra seeks. Setting this flag saves
1492 some on i/o. Checksum verification by HDFS will be internally disabled
1493 on hfile streams when this flag is set. If the hbase-checksum verification
1494 fails, we will switch back to using HDFS checksums (so do not disable HDFS
1495 checksums! And besides this feature applies to hfiles only, not to WALs).
1496 If this parameter is set to false, then hbase will not verify any checksums,
1497 instead it will depend on checksum verification being done in the HDFS client.
1501 <name>hbase.hstore.bytes.per.checksum</name>
1502 <value>16384</value>
1504 Number of bytes in a newly created checksum chunk for HBase-level
1505 checksums in hfile blocks.
1509 <name>hbase.hstore.checksum.algorithm</name>
1510 <value>CRC32C</value>
1512 Name of an algorithm that is used to compute checksums. Possible values
1513 are NULL, CRC32, CRC32C.
1517 <name>hbase.client.scanner.max.result.size</name>
1518 <value>2097152</value>
1519 <description>Maximum number of bytes returned when calling a scanner's next method.
1520 Note that when a single row is larger than this limit the row is still returned completely.
1521 The default value is 2MB, which is good for 1ge networks.
1522 With faster and/or high latency networks this value should be increased.
1526 <name>hbase.server.scanner.max.result.size</name>
1527 <value>104857600</value>
1528 <description>Maximum number of bytes returned when calling a scanner's next method.
1529 Note that when a single row is larger than this limit the row is still returned completely.
1530 The default value is 100MB.
1531 This is a safety setting to protect the server from OOM situations.
1535 <name>hbase.status.published</name>
1536 <value>false</value>
1538 This setting activates the publication by the master of the status of the region server.
1539 When a region server dies and its recovery starts, the master will push this information
1540 to the client application, to let them cut the connection immediately instead of waiting
1545 <name>hbase.status.publisher.class</name>
1546 <value>org.apache.hadoop.hbase.master.ClusterStatusPublisher$MulticastPublisher</value>
1548 Implementation of the status publication with a multicast message.
1552 <name>hbase.status.listener.class</name>
1553 <value>org.apache.hadoop.hbase.client.ClusterStatusListener$MulticastListener</value>
1555 Implementation of the status listener with a multicast message.
1559 <name>hbase.status.multicast.address.ip</name>
1560 <value>226.1.1.3</value>
1562 Multicast address to use for the status publication by multicast.
1566 <name>hbase.status.multicast.address.port</name>
1567 <value>16100</value>
1569 Multicast port to use for the status publication by multicast.
1573 <name>hbase.dynamic.jars.dir</name>
1574 <value>${hbase.rootdir}/lib</value>
1576 The directory from which the custom filter JARs can be loaded
1577 dynamically by the region server without the need to restart. However,
1578 an already loaded filter/co-processor class would not be un-loaded. See
1579 HBASE-1936 for more details.
1581 Does not apply to coprocessors.
1585 <name>hbase.security.authentication</name>
1586 <value>simple</value>
1588 Controls whether or not secure authentication is enabled for HBase.
1589 Possible values are 'simple' (no authentication), and 'kerberos'.
1593 <name>hbase.rest.filter.classes</name>
1594 <value>org.apache.hadoop.hbase.rest.filter.GzipFilter</value>
1596 Servlet filters for REST service.
1600 <name>hbase.master.loadbalancer.class</name>
1601 <value>org.apache.hadoop.hbase.master.balancer.StochasticLoadBalancer</value>
1603 Class used to execute the regions balancing when the period occurs.
1604 See the class comment for more on how it works
1605 http://hbase.apache.org/devapidocs/org/apache/hadoop/hbase/master/balancer/StochasticLoadBalancer.html
1606 It replaces the DefaultLoadBalancer as the default (since renamed
1607 as the SimpleLoadBalancer).
1611 <name>hbase.master.loadbalance.bytable</name>
1612 <value>false</value>
1613 <description>Factor Table name when the balancer runs.
1618 <name>hbase.master.normalizer.class</name>
1619 <value>org.apache.hadoop.hbase.master.normalizer.SimpleRegionNormalizer</value>
1621 Class used to execute the region normalization when the period occurs.
1622 See the class comment for more on how it works
1623 http://hbase.apache.org/devapidocs/org/apache/hadoop/hbase/master/normalizer/SimpleRegionNormalizer.html
1627 <name>hbase.rest.csrf.enabled</name>
1628 <value>false</value>
1630 Set to true to enable protection against cross-site request forgery (CSRF)
1634 <name>hbase.rest-csrf.browser-useragents-regex</name>
1635 <value>^Mozilla.*,^Opera.*</value>
1637 A comma-separated list of regular expressions used to match against an HTTP
1638 request's User-Agent header when protection against cross-site request
1639 forgery (CSRF) is enabled for REST server by setting
1640 hbase.rest.csrf.enabled to true. If the incoming User-Agent matches
1641 any of these regular expressions, then the request is considered to be sent
1642 by a browser, and therefore CSRF prevention is enforced. If the request's
1643 User-Agent does not match any of these regular expressions, then the request
1644 is considered to be sent by something other than a browser, such as scripted
1645 automation. In this case, CSRF is not a potential attack vector, so
1646 the prevention is not enforced. This helps achieve backwards-compatibility
1647 with existing automation that has not been updated to send the CSRF
1652 <name>hbase.security.exec.permission.checks</name>
1653 <value>false</value>
1655 If this setting is enabled and ACL based access control is active (the
1656 AccessController coprocessor is installed either as a system coprocessor
1657 or on a table as a table coprocessor) then you must grant all relevant
1658 users EXEC privilege if they require the ability to execute coprocessor
1659 endpoint calls. EXEC privilege, like any other permission, can be
1660 granted globally to a user, or to a user on a per table or per namespace
1661 basis. For more information on coprocessor endpoints, see the coprocessor
1662 section of the HBase online manual. For more information on granting or
1663 revoking permissions using the AccessController, see the security
1664 section of the HBase online manual.
1668 <name>hbase.procedure.regionserver.classes</name>
1670 <description>A comma-separated list of
1671 org.apache.hadoop.hbase.procedure.RegionServerProcedureManager procedure managers that are
1672 loaded by default on the active HRegionServer process. The lifecycle methods (init/start/stop)
1673 will be called by the active HRegionServer process to perform the specific globally barriered
1674 procedure. After implementing your own RegionServerProcedureManager, just put it in
1675 HBase's classpath and add the fully qualified class name here.
1679 <name>hbase.procedure.master.classes</name>
1681 <description>A comma-separated list of
1682 org.apache.hadoop.hbase.procedure.MasterProcedureManager procedure managers that are
1683 loaded by default on the active HMaster process. A procedure is identified by its signature and
1684 users can use the signature and an instant name to trigger an execution of a globally barriered
1685 procedure. After implementing your own MasterProcedureManager, just put it in HBase's classpath
1686 and add the fully qualified class name here.</description>
1689 <name>hbase.coordinated.state.manager.class</name>
1690 <value>org.apache.hadoop.hbase.coordination.ZkCoordinatedStateManager</value>
1691 <description>Fully qualified name of class implementing coordinated state manager.</description>
1694 <name>hbase.regionserver.storefile.refresh.period</name>
1697 The period (in milliseconds) for refreshing the store files for the secondary regions. 0
1698 means this feature is disabled. Secondary regions sees new files (from flushes and
1699 compactions) from primary once the secondary region refreshes the list of files in the
1700 region (there is no notification mechanism). But too frequent refreshes might cause
1701 extra Namenode pressure. If the files cannot be refreshed for longer than HFile TTL
1702 (hbase.master.hfilecleaner.ttl) the requests are rejected. Configuring HFile TTL to a larger
1703 value is also recommended with this setting.
1707 <name>hbase.region.replica.replication.enabled</name>
1708 <value>false</value>
1710 Whether asynchronous WAL replication to the secondary region replicas is enabled or not.
1711 If this is enabled, a replication peer named "region_replica_replication" will be created
1712 which will tail the logs and replicate the mutations to region replicas for tables that
1713 have region replication > 1. If this is enabled once, disabling this replication also
1714 requires disabling the replication peer using shell or Admin java class.
1715 Replication to secondary region replicas works over standard inter-cluster replication.
1719 <name>hbase.http.filter.initializers</name>
1720 <value>org.apache.hadoop.hbase.http.lib.StaticUserWebFilter</value>
1722 A comma separated list of class names. Each class in the list must extend
1723 org.apache.hadoop.hbase.http.FilterInitializer. The corresponding Filter will
1724 be initialized. Then, the Filter will be applied to all user facing jsp
1725 and servlet web pages.
1726 The ordering of the list defines the ordering of the filters.
1727 The default StaticUserWebFilter add a user principal as defined by the
1728 hbase.http.staticuser.user property.
1732 <name>hbase.security.visibility.mutations.checkauths</name>
1733 <value>false</value>
1735 This property if enabled, will check whether the labels in the visibility
1736 expression are associated with the user issuing the mutation
1740 <name>hbase.http.max.threads</name>
1743 The maximum number of threads that the HTTP Server will create in its
1748 <name>hbase.replication.rpc.codec</name>
1749 <value>org.apache.hadoop.hbase.codec.KeyValueCodecWithTags</value>
1751 The codec that is to be used when replication is enabled so that
1752 the tags are also replicated. This is used along with HFileV3 which
1753 supports tags in them. If tags are not used or if the hfile version used
1754 is HFileV2 then KeyValueCodec can be used as the replication codec. Note that
1755 using KeyValueCodecWithTags for replication when there are no tags causes no harm.
1759 <name>hbase.replication.source.maxthreads</name>
1762 The maximum number of threads any replication source will use for
1763 shipping edits to the sinks in parallel. This also limits the number of
1764 chunks each replication batch is broken into. Larger values can improve
1765 the replication throughput between the master and slave clusters. The
1766 default of 10 will rarely need to be changed.
1769 <!-- Static Web User Filter properties. -->
1771 <name>hbase.http.staticuser.user</name>
1772 <value>dr.stack</value>
1774 The user name to filter as, on static web filters
1775 while rendering content. An example use is the HDFS
1776 web UI (user to be used for browsing files).
1780 <name>hbase.regionserver.handler.abort.on.error.percent</name>
1782 <description>The percent of region server RPC threads failed to abort RS.
1783 -1 Disable aborting; 0 Abort if even a single handler has died;
1784 0.x Abort only when this percent of handlers have died;
1785 1 Abort only all of the handers have died.</description>
1787 <!-- Mob properties. -->
1789 <name>hbase.mob.file.cache.size</name>
1792 Number of opened file handlers to cache.
1793 A larger value will benefit reads by providing more file handlers per mob
1794 file cache and would reduce frequent file opening and closing.
1795 However, if this is set too high, this could lead to a "too many opened file handlers"
1796 The default value is 1000.
1800 <name>hbase.mob.cache.evict.period</name>
1803 The amount of time in seconds before the mob cache evicts cached mob files.
1804 The default value is 3600 seconds.
1808 <name>hbase.mob.cache.evict.remain.ratio</name>
1811 The ratio (between 0.0 and 1.0) of files that remains cached after an eviction
1812 is triggered when the number of cached mob files exceeds the hbase.mob.file.cache.size.
1813 The default value is 0.5f.
1817 <name>hbase.master.mob.cleaner.period</name>
1818 <value>86400</value>
1820 The period that MobFileCleanerChore runs. The unit is second.
1821 The default value is one day. The MOB file name uses only the date part of
1822 the file creation time in it. We use this time for deciding TTL expiry of
1823 the files. So the removal of TTL expired files might be delayed. The max
1824 delay might be 24 hrs.
1828 <name>hbase.mob.major.compaction.region.batch.size</name>
1831 The max number of a MOB table regions that is allowed in a batch of the mob compaction. By
1832 setting this number to a custom value, users can control the overall effect of a major
1833 compaction of a large MOB-enabled table.
1834 Default is 0 - means no limit - all regions of a MOB table will be compacted at once
1838 <name>hbase.mob.compaction.chore.period</name>
1839 <value>604800</value>
1841 The period that MobCompactionChore runs. The unit is second.
1842 The default value is one week.
1846 <name>hbase.snapshot.master.timeout.millis</name>
1847 <value>300000</value>
1849 Timeout for master for the snapshot procedure execution.
1853 <name>hbase.snapshot.region.timeout</name>
1854 <value>300000</value>
1856 Timeout for regionservers to keep threads in snapshot request pool waiting.
1860 <name>hbase.rpc.rows.warning.threshold</name>
1863 Number of rows in a batch operation above which a warning will be logged.
1867 <name>hbase.master.wait.on.service.seconds</name>
1869 <description>Default is 5 minutes. Make it 30 seconds for tests. See
1870 HBASE-19794 for some context.</description>
1873 <name>hbase.master.cleaner.snapshot.interval</name>
1874 <value>1800000</value>
1876 Snapshot Cleanup chore interval in milliseconds.
1877 The cleanup thread keeps running at this interval
1878 to find all snapshots that are expired based on TTL
1883 <name>hbase.master.snapshot.ttl</name>
1886 Default Snapshot TTL to be considered when the user does not specify TTL while
1887 creating snapshot. Default value 0 indicates FOREVERE - snapshot should not be
1888 automatically deleted until it is manually deleted
1892 <name>hbase.master.regions.recovery.check.interval</name>
1893 <value>1200000</value>
1895 Regions Recovery Chore interval in milliseconds.
1896 This chore keeps running at this interval to
1897 find all regions with configurable max store file ref count
1902 <name>hbase.regions.recovery.store.file.ref.count</name>
1905 Very large number of ref count on a compacted
1906 store file indicates that it is a ref leak
1907 on that object(compacted store file).
1908 Such files can not be removed after
1909 it is invalidated via compaction.
1910 Only way to recover in such scenario is to
1911 reopen the region which can release
1912 all resources, like the refcount,
1913 leases, etc. This config represents Store files Ref
1914 Count threshold value considered for reopening
1915 regions. Any region with compacted store files
1916 ref count > this value would be eligible for
1917 reopening by master. Here, we get the max
1918 refCount among all refCounts on all
1919 compacted away store files that belong to a
1920 particular region. Default value -1 indicates
1921 this feature is turned off. Only positive
1922 integer value should be provided to
1923 enable this feature.
1927 <name>hbase.regionserver.slowlog.ringbuffer.size</name>
1930 Default size of ringbuffer to be maintained by each RegionServer in order
1931 to store online slowlog responses. This is an in-memory ring buffer of
1932 requests that were judged to be too slow in addition to the responseTooSlow
1933 logging. The in-memory representation would be complete.
1934 For more details, please look into Doc Section:
1935 Get Slow Response Log from shell
1939 <name>hbase.regionserver.slowlog.buffer.enabled</name>
1940 <value>false</value>
1942 Indicates whether RegionServers have ring buffer running for storing
1943 Online Slow logs in FIFO manner with limited entries. The size of
1944 the ring buffer is indicated by config: hbase.regionserver.slowlog.ringbuffer.size
1945 The default value is false, turn this on and get latest slowlog
1946 responses with complete data.