· 7 years ago · Nov 25, 2018, 06:56 AM
1<?xml version="1.0" encoding="UTF-8" standalone="no"?>
2<?xml-stylesheet type="text/xsl" href="configuration.xsl"?><!--
3 Licensed to the Apache Software Foundation (ASF) under one or more
4 contributor license agreements. See the NOTICE file distributed with
5 this work for additional information regarding copyright ownership.
6 The ASF licenses this file to You under the Apache License, Version 2.0
7 (the "License"); you may not use this file except in compliance with
8 the License. You may obtain a copy of the License at
9
10 http://www.apache.org/licenses/LICENSE-2.0
11
12 Unless required by applicable law or agreed to in writing, software
13 distributed under the License is distributed on an "AS IS" BASIS,
14 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
15 See the License for the specific language governing permissions and
16 limitations under the License.
17--><configuration>
18 <!-- WARNING!!! This file is auto generated for documentation purposes ONLY! -->
19 <!-- WARNING!!! Any changes you make to this file will be ignored by Hive. -->
20 <!-- WARNING!!! You must make your changes in hive-site.xml instead. -->
21 <!-- Hive Execution Parameters -->
22 <property>
23 <name>hive.exec.script.wrapper</name>
24 <value/>
25 <description/>
26 </property>
27 <property>
28 <name>hive.exec.plan</name>
29 <value/>
30 <description/>
31 </property>
32 <property>
33 <name>hive.exec.stagingdir</name>
34 <value>.hive-staging</value>
35 <description>Directory name that will be created inside table locations in order to support HDFS encryption. This is replaces ${hive.exec.scratchdir} for query results with the exception of read-only tables. In all cases ${hive.exec.scratchdir} is still used for other temporary files, such as job plans.</description>
36 </property>
37 <property>
38 <name>hive.exec.scratchdir</name>
39 <value>/tmp/hive</value>
40 <description>HDFS root scratch dir for Hive jobs which gets created with write all (733) permission. For each connecting user, an HDFS scratch dir: ${hive.exec.scratchdir}/<username> is created, with ${hive.scratch.dir.permission}.</description>
41 </property>
42 <property>
43 <name>hive.exec.local.scratchdir</name>
44 <value>${system:java.io.tmpdir}/${system:user.name}</value>
45 <description>Local scratch space for Hive jobs</description>
46 </property>
47 <property>
48 <name>hive.downloaded.resources.dir</name>
49 <value>${system:java.io.tmpdir}/${hive.session.id}_resources</value>
50 <description>Temporary local directory for added resources in the remote file system.</description>
51 </property>
52 <property>
53 <name>hive.scratch.dir.permission</name>
54 <value>700</value>
55 <description>The permission for the user specific scratch directories that get created.</description>
56 </property>
57 <property>
58 <name>hive.exec.submitviachild</name>
59 <value>false</value>
60 <description/>
61 </property>
62 <property>
63 <name>hive.exec.submit.local.task.via.child</name>
64 <value>true</value>
65 <description>
66 Determines whether local tasks (typically mapjoin hashtable generation phase) runs in
67 separate JVM (true recommended) or not.
68 Avoids the overhead of spawning new JVM, but can lead to out-of-memory issues.
69 </description>
70 </property>
71 <property>
72 <name>hive.exec.script.maxerrsize</name>
73 <value>100000</value>
74 <description>
75 Maximum number of bytes a script is allowed to emit to standard error (per map-reduce task).
76 This prevents runaway scripts from filling logs partitions to capacity
77 </description>
78 </property>
79 <property>
80 <name>hive.exec.script.allow.partial.consumption</name>
81 <value>false</value>
82 <description>
83 When enabled, this option allows a user script to exit successfully without consuming
84 all the data from the standard input.
85 </description>
86 </property>
87 <property>
88 <name>stream.stderr.reporter.prefix</name>
89 <value>reporter:</value>
90 <description>Streaming jobs that log to standard error with this prefix can log counter or status information.</description>
91 </property>
92 <property>
93 <name>stream.stderr.reporter.enabled</name>
94 <value>true</value>
95 <description>Enable consumption of status and counter messages for streaming jobs.</description>
96 </property>
97 <property>
98 <name>hive.exec.compress.output</name>
99 <value>false</value>
100 <description>
101 This controls whether the final outputs of a query (to a local/HDFS file or a Hive table) is compressed.
102 The compression codec and other options are determined from Hadoop config variables mapred.output.compress*
103 </description>
104 </property>
105 <property>
106 <name>hive.exec.compress.intermediate</name>
107 <value>false</value>
108 <description>
109 This controls whether intermediate files produced by Hive between multiple map-reduce jobs are compressed.
110 The compression codec and other options are determined from Hadoop config variables mapred.output.compress*
111 </description>
112 </property>
113 <property>
114 <name>hive.intermediate.compression.codec</name>
115 <value/>
116 <description/>
117 </property>
118 <property>
119 <name>hive.intermediate.compression.type</name>
120 <value/>
121 <description/>
122 </property>
123 <property>
124 <name>hive.exec.reducers.bytes.per.reducer</name>
125 <value>256000000</value>
126 <description>size per reducer.The default is 256Mb, i.e if the input size is 1G, it will use 4 reducers.</description>
127 </property>
128 <property>
129 <name>hive.exec.reducers.max</name>
130 <value>1009</value>
131 <description>
132 max number of reducers will be used. If the one specified in the configuration parameter mapred.reduce.tasks is
133 negative, Hive will use this one as the max number of reducers when automatically determine number of reducers.
134 </description>
135 </property>
136 <property>
137 <name>hive.exec.pre.hooks</name>
138 <value/>
139 <description>
140 Comma-separated list of pre-execution hooks to be invoked for each statement.
141 A pre-execution hook is specified as the name of a Java class which implements the
142 org.apache.hadoop.hive.ql.hooks.ExecuteWithHookContext interface.
143 </description>
144 </property>
145 <property>
146 <name>hive.exec.post.hooks</name>
147 <value/>
148 <description>
149 Comma-separated list of post-execution hooks to be invoked for each statement.
150 A post-execution hook is specified as the name of a Java class which implements the
151 org.apache.hadoop.hive.ql.hooks.ExecuteWithHookContext interface.
152 </description>
153 </property>
154 <property>
155 <name>hive.exec.failure.hooks</name>
156 <value/>
157 <description>
158 Comma-separated list of on-failure hooks to be invoked for each statement.
159 An on-failure hook is specified as the name of Java class which implements the
160 org.apache.hadoop.hive.ql.hooks.ExecuteWithHookContext interface.
161 </description>
162 </property>
163 <property>
164 <name>hive.exec.query.redactor.hooks</name>
165 <value/>
166 <description>
167 Comma-separated list of hooks to be invoked for each query which can
168 tranform the query before it's placed in the job.xml file. Must be a Java class which
169 extends from the org.apache.hadoop.hive.ql.hooks.Redactor abstract class.
170 </description>
171 </property>
172 <property>
173 <name>hive.client.stats.publishers</name>
174 <value/>
175 <description>
176 Comma-separated list of statistics publishers to be invoked on counters on each job.
177 A client stats publisher is specified as the name of a Java class which implements the
178 org.apache.hadoop.hive.ql.stats.ClientStatsPublisher interface.
179 </description>
180 </property>
181 <property>
182 <name>hive.exec.parallel</name>
183 <value>false</value>
184 <description>Whether to execute jobs in parallel</description>
185 </property>
186 <property>
187 <name>hive.exec.parallel.thread.number</name>
188 <value>8</value>
189 <description>How many jobs at most can be executed in parallel</description>
190 </property>
191 <property>
192 <name>hive.mapred.reduce.tasks.speculative.execution</name>
193 <value>true</value>
194 <description>Whether speculative execution for reducers should be turned on. </description>
195 </property>
196 <property>
197 <name>hive.exec.counters.pull.interval</name>
198 <value>1000</value>
199 <description>
200 The interval with which to poll the JobTracker for the counters the running job.
201 The smaller it is the more load there will be on the jobtracker, the higher it is the less granular the caught will be.
202 </description>
203 </property>
204 <property>
205 <name>hive.exec.dynamic.partition</name>
206 <value>true</value>
207 <description>Whether or not to allow dynamic partitions in DML/DDL.</description>
208 </property>
209 <property>
210 <name>hive.exec.dynamic.partition.mode</name>
211 <value>strict</value>
212 <description>
213 In strict mode, the user must specify at least one static partition
214 in case the user accidentally overwrites all partitions.
215 In nonstrict mode all partitions are allowed to be dynamic.
216 </description>
217 </property>
218 <property>
219 <name>hive.exec.max.dynamic.partitions</name>
220 <value>1000</value>
221 <description>Maximum number of dynamic partitions allowed to be created in total.</description>
222 </property>
223 <property>
224 <name>hive.exec.max.dynamic.partitions.pernode</name>
225 <value>100</value>
226 <description>Maximum number of dynamic partitions allowed to be created in each mapper/reducer node.</description>
227 </property>
228 <property>
229 <name>hive.exec.max.created.files</name>
230 <value>100000</value>
231 <description>Maximum number of HDFS files created by all mappers/reducers in a MapReduce job.</description>
232 </property>
233 <property>
234 <name>hive.exec.default.partition.name</name>
235 <value>__HIVE_DEFAULT_PARTITION__</value>
236 <description>
237 The default partition name in case the dynamic partition column value is null/empty string or any other values that cannot be escaped.
238 This value must not contain any special character used in HDFS URI (e.g., ':', '%', '/' etc).
239 The user has to be aware that the dynamic partition value should not contain this value to avoid confusions.
240 </description>
241 </property>
242 <property>
243 <name>hive.lockmgr.zookeeper.default.partition.name</name>
244 <value>__HIVE_DEFAULT_ZOOKEEPER_PARTITION__</value>
245 <description/>
246 </property>
247 <property>
248 <name>hive.exec.show.job.failure.debug.info</name>
249 <value>true</value>
250 <description>
251 If a job fails, whether to provide a link in the CLI to the task with the
252 most failures, along with debugging hints if applicable.
253 </description>
254 </property>
255 <property>
256 <name>hive.exec.job.debug.capture.stacktraces</name>
257 <value>true</value>
258 <description>
259 Whether or not stack traces parsed from the task logs of a sampled failed task
260 for each failed job should be stored in the SessionState
261 </description>
262 </property>
263 <property>
264 <name>hive.exec.job.debug.timeout</name>
265 <value>30000</value>
266 <description/>
267 </property>
268 <property>
269 <name>hive.exec.tasklog.debug.timeout</name>
270 <value>20000</value>
271 <description/>
272 </property>
273 <property>
274 <name>hive.output.file.extension</name>
275 <value/>
276 <description>
277 String used as a file extension for output files.
278 If not set, defaults to the codec extension for text files (e.g. ".gz"), or no extension otherwise.
279 </description>
280 </property>
281 <property>
282 <name>hive.exec.mode.local.auto</name>
283 <value>false</value>
284 <description>Let Hive determine whether to run in local mode automatically</description>
285 </property>
286 <property>
287 <name>hive.exec.mode.local.auto.inputbytes.max</name>
288 <value>134217728</value>
289 <description>When hive.exec.mode.local.auto is true, input bytes should less than this for local mode.</description>
290 </property>
291 <property>
292 <name>hive.exec.mode.local.auto.input.files.max</name>
293 <value>4</value>
294 <description>When hive.exec.mode.local.auto is true, the number of tasks should less than this for local mode.</description>
295 </property>
296 <property>
297 <name>hive.exec.drop.ignorenonexistent</name>
298 <value>true</value>
299 <description>Do not report an error if DROP TABLE/VIEW/Index/Function specifies a non-existent table/view/index/function</description>
300 </property>
301 <property>
302 <name>hive.ignore.mapjoin.hint</name>
303 <value>true</value>
304 <description>Ignore the mapjoin hint</description>
305 </property>
306 <property>
307 <name>hive.file.max.footer</name>
308 <value>100</value>
309 <description>maximum number of lines for footer user can define for a table file</description>
310 </property>
311 <property>
312 <name>hive.resultset.use.unique.column.names</name>
313 <value>true</value>
314 <description>
315 Make column names unique in the result set by qualifying column names with table alias if needed.
316 Table alias will be added to column names for queries of type "select *" or
317 if query explicitly uses table alias "select r1.x..".
318 </description>
319 </property>
320 <property>
321 <name>fs.har.impl</name>
322 <value>org.apache.hadoop.hive.shims.HiveHarFileSystem</value>
323 <description>The implementation for accessing Hadoop Archives. Note that this won't be applicable to Hadoop versions less than 0.20</description>
324 </property>
325 <property>
326 <name>hive.metastore.warehouse.dir</name>
327 <value>/user/hive/warehouse</value>
328 <description>location of default database for the warehouse</description>
329 </property>
330 <property>
331 <name>hive.metastore.uris</name>
332 <value/>
333 <description>Thrift URI for the remote metastore. Used by metastore client to connect to remote metastore.</description>
334 </property>
335 <property>
336 <name>hive.metastore.fastpath</name>
337 <value>false</value>
338 <description>Used to avoid all of the proxies and object copies in the metastore. Note, if this is set, you MUST use a local metastore (hive.metastore.uris must be empty) otherwise undefined and most likely undesired behavior will result</description>
339 </property>
340 <property>
341 <name>hive.metastore.hbase.catalog.cache.size</name>
342 <value>50000</value>
343 <description>Maximum number of objects we will place in the hbase metastore catalog cache. The objects will be divided up by types that we need to cache.</description>
344 </property>
345 <property>
346 <name>hive.metastore.hbase.aggregate.stats.cache.size</name>
347 <value>10000</value>
348 <description>Maximum number of aggregate stats nodes that we will place in the hbase metastore aggregate stats cache.</description>
349 </property>
350 <property>
351 <name>hive.metastore.hbase.aggregate.stats.max.partitions</name>
352 <value>10000</value>
353 <description>Maximum number of partitions that are aggregated per cache node.</description>
354 </property>
355 <property>
356 <name>hive.metastore.hbase.aggregate.stats.false.positive.probability</name>
357 <value>0.01</value>
358 <description>Maximum false positive probability for the Bloom Filter used in each aggregate stats cache node (default 1%).</description>
359 </property>
360 <property>
361 <name>hive.metastore.hbase.aggregate.stats.max.variance</name>
362 <value>0.1</value>
363 <description>Maximum tolerable variance in number of partitions between a cached node and our request (default 10%).</description>
364 </property>
365 <property>
366 <name>hive.metastore.hbase.cache.ttl</name>
367 <value>600s</value>
368 <description>
369 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
370 Number of seconds for a cached node to be active in the cache before they become stale.
371 </description>
372 </property>
373 <property>
374 <name>hive.metastore.hbase.cache.max.writer.wait</name>
375 <value>5000ms</value>
376 <description>
377 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
378 Number of milliseconds a writer will wait to acquire the writelock before giving up.
379 </description>
380 </property>
381 <property>
382 <name>hive.metastore.hbase.cache.max.reader.wait</name>
383 <value>1000ms</value>
384 <description>
385 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
386 Number of milliseconds a reader will wait to acquire the readlock before giving up.
387 </description>
388 </property>
389 <property>
390 <name>hive.metastore.hbase.cache.max.full</name>
391 <value>0.9</value>
392 <description>Maximum cache full % after which the cache cleaner thread kicks in.</description>
393 </property>
394 <property>
395 <name>hive.metastore.hbase.cache.clean.until</name>
396 <value>0.8</value>
397 <description>The cleaner thread cleans until cache reaches this % full size.</description>
398 </property>
399 <property>
400 <name>hive.metastore.hbase.connection.class</name>
401 <value>org.apache.hadoop.hive.metastore.hbase.VanillaHBaseConnection</value>
402 <description>Class used to connection to HBase</description>
403 </property>
404 <property>
405 <name>hive.metastore.hbase.aggr.stats.cache.entries</name>
406 <value>10000</value>
407 <description>How many in stats objects to cache in memory</description>
408 </property>
409 <property>
410 <name>hive.metastore.hbase.aggr.stats.memory.ttl</name>
411 <value>60s</value>
412 <description>
413 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
414 Number of seconds stats objects live in memory after they are read from HBase.
415 </description>
416 </property>
417 <property>
418 <name>hive.metastore.hbase.aggr.stats.invalidator.frequency</name>
419 <value>5s</value>
420 <description>
421 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
422 How often the stats cache scans its HBase entries and looks for expired entries
423 </description>
424 </property>
425 <property>
426 <name>hive.metastore.hbase.aggr.stats.hbase.ttl</name>
427 <value>604800s</value>
428 <description>
429 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
430 Number of seconds stats entries live in HBase cache after they are created. They may be invalided by updates or partition drops before this. Default is one week.
431 </description>
432 </property>
433 <property>
434 <name>hive.metastore.hbase.file.metadata.threads</name>
435 <value>1</value>
436 <description>Number of threads to use to read file metadata in background to cache it.</description>
437 </property>
438 <property>
439 <name>hive.metastore.connect.retries</name>
440 <value>3</value>
441 <description>Number of retries while opening a connection to metastore</description>
442 </property>
443 <property>
444 <name>hive.metastore.failure.retries</name>
445 <value>1</value>
446 <description>Number of retries upon failure of Thrift metastore calls</description>
447 </property>
448 <property>
449 <name>hive.metastore.port</name>
450 <value>9083</value>
451 <description>Hive metastore listener port</description>
452 </property>
453 <property>
454 <name>hive.metastore.client.connect.retry.delay</name>
455 <value>1s</value>
456 <description>
457 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
458 Number of seconds for the client to wait between consecutive connection attempts
459 </description>
460 </property>
461 <property>
462 <name>hive.metastore.client.socket.timeout</name>
463 <value>600s</value>
464 <description>
465 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
466 MetaStore Client socket timeout in seconds
467 </description>
468 </property>
469 <property>
470 <name>hive.metastore.client.socket.lifetime</name>
471 <value>0s</value>
472 <description>
473 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
474 MetaStore Client socket lifetime in seconds. After this time is exceeded, client
475 reconnects on the next MetaStore operation. A value of 0s means the connection
476 has an infinite lifetime.
477 </description>
478 </property>
479 <property>
480 <name>javax.jdo.option.ConnectionPassword</name>
481 <value>mine</value>
482 <description>password to use against metastore database</description>
483 </property>
484 <property>
485 <name>hive.metastore.ds.connection.url.hook</name>
486 <value/>
487 <description>Name of the hook to use for retrieving the JDO connection URL. If empty, the value in javax.jdo.option.ConnectionURL is used</description>
488 </property>
489 <property>
490 <name>javax.jdo.option.Multithreaded</name>
491 <value>true</value>
492 <description>Set this to true if multiple threads access metastore through JDO concurrently.</description>
493 </property>
494 <property>
495 <name>javax.jdo.option.ConnectionURL</name>
496 <value>jdbc:derby:;databaseName=metastore_db;create=true</value>
497 <description>
498 JDBC connect string for a JDBC metastore.
499 To use SSL to encrypt/authenticate the connection, provide database-specific SSL flag in the connection URL.
500 For example, jdbc:postgresql://myhost/db?ssl=true for postgres database.
501 </description>
502 </property>
503 <property>
504 <name>hive.metastore.dbaccess.ssl.properties</name>
505 <value/>
506 <description>
507 Comma-separated SSL properties for metastore to access database when JDO connection URL
508 enables SSL access. e.g. javax.net.ssl.trustStore=/tmp/truststore,javax.net.ssl.trustStorePassword=pwd.
509 </description>
510 </property>
511 <property>
512 <name>hive.hmshandler.retry.attempts</name>
513 <value>10</value>
514 <description>The number of times to retry a HMSHandler call if there were a connection error.</description>
515 </property>
516 <property>
517 <name>hive.hmshandler.retry.interval</name>
518 <value>2000ms</value>
519 <description>
520 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
521 The time between HMSHandler retry attempts on failure.
522 </description>
523 </property>
524 <property>
525 <name>hive.hmshandler.force.reload.conf</name>
526 <value>false</value>
527 <description>
528 Whether to force reloading of the HMSHandler configuration (including
529 the connection URL, before the next metastore query that accesses the
530 datastore. Once reloaded, this value is reset to false. Used for
531 testing only.
532 </description>
533 </property>
534 <property>
535 <name>hive.metastore.server.max.message.size</name>
536 <value>104857600</value>
537 <description>Maximum message size in bytes a HMS will accept.</description>
538 </property>
539 <property>
540 <name>hive.metastore.server.min.threads</name>
541 <value>200</value>
542 <description>Minimum number of worker threads in the Thrift server's pool.</description>
543 </property>
544 <property>
545 <name>hive.metastore.server.max.threads</name>
546 <value>1000</value>
547 <description>Maximum number of worker threads in the Thrift server's pool.</description>
548 </property>
549 <property>
550 <name>hive.metastore.server.tcp.keepalive</name>
551 <value>true</value>
552 <description>Whether to enable TCP keepalive for the metastore server. Keepalive will prevent accumulation of half-open connections.</description>
553 </property>
554 <property>
555 <name>hive.metastore.archive.intermediate.original</name>
556 <value>_INTERMEDIATE_ORIGINAL</value>
557 <description>
558 Intermediate dir suffixes used for archiving. Not important what they
559 are, as long as collisions are avoided
560 </description>
561 </property>
562 <property>
563 <name>hive.metastore.archive.intermediate.archived</name>
564 <value>_INTERMEDIATE_ARCHIVED</value>
565 <description/>
566 </property>
567 <property>
568 <name>hive.metastore.archive.intermediate.extracted</name>
569 <value>_INTERMEDIATE_EXTRACTED</value>
570 <description/>
571 </property>
572 <property>
573 <name>hive.metastore.kerberos.keytab.file</name>
574 <value/>
575 <description>The path to the Kerberos Keytab file containing the metastore Thrift server's service principal.</description>
576 </property>
577 <property>
578 <name>hive.metastore.kerberos.principal</name>
579 <value>hive-metastore/_HOST@EXAMPLE.COM</value>
580 <description>
581 The service principal for the metastore Thrift server.
582 The special string _HOST will be replaced automatically with the correct host name.
583 </description>
584 </property>
585 <property>
586 <name>hive.metastore.sasl.enabled</name>
587 <value>false</value>
588 <description>If true, the metastore Thrift interface will be secured with SASL. Clients must authenticate with Kerberos.</description>
589 </property>
590 <property>
591 <name>hive.metastore.thrift.framed.transport.enabled</name>
592 <value>false</value>
593 <description>If true, the metastore Thrift interface will use TFramedTransport. When false (default) a standard TTransport is used.</description>
594 </property>
595 <property>
596 <name>hive.metastore.thrift.compact.protocol.enabled</name>
597 <value>false</value>
598 <description>
599 If true, the metastore Thrift interface will use TCompactProtocol. When false (default) TBinaryProtocol will be used.
600 Setting it to true will break compatibility with older clients running TBinaryProtocol.
601 </description>
602 </property>
603 <property>
604 <name>hive.metastore.token.signature</name>
605 <value/>
606 <description>The delegation token service name to match when selecting a token from the current user's tokens.</description>
607 </property>
608 <property>
609 <name>hive.cluster.delegation.token.store.class</name>
610 <value>org.apache.hadoop.hive.thrift.MemoryTokenStore</value>
611 <description>The delegation token store implementation. Set to org.apache.hadoop.hive.thrift.ZooKeeperTokenStore for load-balanced cluster.</description>
612 </property>
613 <property>
614 <name>hive.cluster.delegation.token.store.zookeeper.connectString</name>
615 <value/>
616 <description>
617 The ZooKeeper token store connect string. You can re-use the configuration value
618 set in hive.zookeeper.quorum, by leaving this parameter unset.
619 </description>
620 </property>
621 <property>
622 <name>hive.cluster.delegation.token.store.zookeeper.znode</name>
623 <value>/hivedelegation</value>
624 <description>
625 The root path for token store data. Note that this is used by both HiveServer2 and
626 MetaStore to store delegation Token. One directory gets created for each of them.
627 The final directory names would have the servername appended to it (HIVESERVER2,
628 METASTORE).
629 </description>
630 </property>
631 <property>
632 <name>hive.cluster.delegation.token.store.zookeeper.acl</name>
633 <value/>
634 <description>
635 ACL for token store entries. Comma separated list of ACL entries. For example:
636 sasl:hive/host1@MY.DOMAIN:cdrwa,sasl:hive/host2@MY.DOMAIN:cdrwa
637 Defaults to all permissions for the hiveserver2/metastore process user.
638 </description>
639 </property>
640 <property>
641 <name>hive.metastore.cache.pinobjtypes</name>
642 <value>Table,StorageDescriptor,SerDeInfo,Partition,Database,Type,FieldSchema,Order</value>
643 <description>List of comma separated metastore object types that should be pinned in the cache</description>
644 </property>
645 <property>
646 <name>datanucleus.connectionPoolingType</name>
647 <value>BONECP</value>
648 <description>Specify connection pool library for datanucleus</description>
649 </property>
650 <property>
651 <name>datanucleus.schema.validateTables</name>
652 <value>false</value>
653 <description>validates existing schema against code. turn this on if you want to verify existing schema</description>
654 </property>
655 <property>
656 <name>datanucleus.schema.validateColumns</name>
657 <value>false</value>
658 <description>validates existing schema against code. turn this on if you want to verify existing schema</description>
659 </property>
660 <property>
661 <name>datanucleus.schema.validateConstraints</name>
662 <value>false</value>
663 <description>validates existing schema against code. turn this on if you want to verify existing schema</description>
664 </property>
665 <property>
666 <name>datanucleus.storeManagerType</name>
667 <value>rdbms</value>
668 <description>metadata store type</description>
669 </property>
670 <property>
671 <name>datanucleus.schema.autoCreateAll</name>
672 <value>false</value>
673 <description>creates necessary schema on a startup if one doesn't exist. set this to false, after creating it once</description>
674 </property>
675 <property>
676 <name>hive.metastore.schema.verification</name>
677 <value>false</value>
678 <description>
679 Enforce metastore schema version consistency.
680 True: Verify that version information stored in metastore matches with one from Hive jars. Also disable automatic
681 schema migration attempt. Users are required to manually migrate schema after Hive upgrade which ensures
682 proper metastore schema migration. (Default)
683 False: Warn if the version information stored in metastore doesn't match with one from in Hive jars.
684 </description>
685 </property>
686 <property>
687 <name>hive.metastore.schema.verification.record.version</name>
688 <value>true</value>
689 <description>
690 When true the current MS version is recorded in the VERSION table. If this is disabled and verification is
691 enabled the MS will be unusable.
692 </description>
693 </property>
694 <property>
695 <name>datanucleus.autoStartMechanismMode</name>
696 <value>checked</value>
697 <description>throw exception if metadata tables are incorrect</description>
698 </property>
699 <property>
700 <name>datanucleus.transactionIsolation</name>
701 <value>read-committed</value>
702 <description>Default transaction isolation level for identity generation.</description>
703 </property>
704 <property>
705 <name>datanucleus.cache.level2</name>
706 <value>false</value>
707 <description>Use a level 2 cache. Turn this off if metadata is changed independently of Hive metastore server</description>
708 </property>
709 <property>
710 <name>datanucleus.cache.level2.type</name>
711 <value>none</value>
712 <description/>
713 </property>
714 <property>
715 <name>datanucleus.identifierFactory</name>
716 <value>datanucleus1</value>
717 <description>
718 Name of the identifier factory to use when generating table/column names etc.
719 'datanucleus1' is used for backward compatibility with DataNucleus v1
720 </description>
721 </property>
722 <property>
723 <name>datanucleus.rdbms.useLegacyNativeValueStrategy</name>
724 <value>true</value>
725 <description/>
726 </property>
727 <property>
728 <name>datanucleus.plugin.pluginRegistryBundleCheck</name>
729 <value>LOG</value>
730 <description>Defines what happens when plugin bundles are found and are duplicated [EXCEPTION|LOG|NONE]</description>
731 </property>
732 <property>
733 <name>hive.metastore.batch.retrieve.max</name>
734 <value>300</value>
735 <description>
736 Maximum number of objects (tables/partitions) can be retrieved from metastore in one batch.
737 The higher the number, the less the number of round trips is needed to the Hive metastore server,
738 but it may also cause higher memory requirement at the client side.
739 </description>
740 </property>
741 <property>
742 <name>hive.metastore.batch.retrieve.table.partition.max</name>
743 <value>1000</value>
744 <description>Maximum number of objects that metastore internally retrieves in one batch.</description>
745 </property>
746 <property>
747 <name>hive.metastore.init.hooks</name>
748 <value/>
749 <description>
750 A comma separated list of hooks to be invoked at the beginning of HMSHandler initialization.
751 An init hook is specified as the name of Java class which extends org.apache.hadoop.hive.metastore.MetaStoreInitListener.
752 </description>
753 </property>
754 <property>
755 <name>hive.metastore.pre.event.listeners</name>
756 <value/>
757 <description>List of comma separated listeners for metastore events.</description>
758 </property>
759 <property>
760 <name>hive.metastore.event.listeners</name>
761 <value/>
762 <description/>
763 </property>
764 <property>
765 <name>hive.metastore.event.db.listener.timetolive</name>
766 <value>86400s</value>
767 <description>
768 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
769 time after which events will be removed from the database listener queue
770 </description>
771 </property>
772 <property>
773 <name>hive.metastore.authorization.storage.checks</name>
774 <value>false</value>
775 <description>
776 Should the metastore do authorization checks against the underlying storage (usually hdfs)
777 for operations like drop-partition (disallow the drop-partition if the user in
778 question doesn't have permissions to delete the corresponding directory
779 on the storage).
780 </description>
781 </property>
782 <property>
783 <name>hive.metastore.event.clean.freq</name>
784 <value>0s</value>
785 <description>
786 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
787 Frequency at which timer task runs to purge expired events in metastore.
788 </description>
789 </property>
790 <property>
791 <name>hive.metastore.event.expiry.duration</name>
792 <value>0s</value>
793 <description>
794 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
795 Duration after which events expire from events table
796 </description>
797 </property>
798 <property>
799 <name>hive.metastore.execute.setugi</name>
800 <value>true</value>
801 <description>
802 In unsecure mode, setting this property to true will cause the metastore to execute DFS operations using
803 the client's reported user and group permissions. Note that this property must be set on
804 both the client and server sides. Further note that its best effort.
805 If client sets its to true and server sets it to false, client setting will be ignored.
806 </description>
807 </property>
808 <property>
809 <name>hive.metastore.partition.name.whitelist.pattern</name>
810 <value/>
811 <description>Partition names will be checked against this regex pattern and rejected if not matched.</description>
812 </property>
813 <property>
814 <name>hive.metastore.integral.jdo.pushdown</name>
815 <value>false</value>
816 <description>
817 Allow JDO query pushdown for integral partition columns in metastore. Off by default. This
818 improves metastore perf for integral columns, especially if there's a large number of partitions.
819 However, it doesn't work correctly with integral values that are not normalized (e.g. have
820 leading zeroes, like 0012). If metastore direct SQL is enabled and works, this optimization
821 is also irrelevant.
822 </description>
823 </property>
824 <property>
825 <name>hive.metastore.try.direct.sql</name>
826 <value>true</value>
827 <description>
828 Whether the Hive metastore should try to use direct SQL queries instead of the
829 DataNucleus for certain read paths. This can improve metastore performance when
830 fetching many partitions or column statistics by orders of magnitude; however, it
831 is not guaranteed to work on all RDBMS-es and all versions. In case of SQL failures,
832 the metastore will fall back to the DataNucleus, so it's safe even if SQL doesn't
833 work for all queries on your datastore. If all SQL queries fail (for example, your
834 metastore is backed by MongoDB), you might want to disable this to save the
835 try-and-fall-back cost.
836 </description>
837 </property>
838 <property>
839 <name>hive.metastore.direct.sql.batch.size</name>
840 <value>0</value>
841 <description>
842 Batch size for partition and other object retrieval from the underlying DB in direct
843 SQL. For some DBs like Oracle and MSSQL, there are hardcoded or perf-based limitations
844 that necessitate this. For DBs that can handle the queries, this isn't necessary and
845 may impede performance. -1 means no batching, 0 means automatic batching.
846 </description>
847 </property>
848 <property>
849 <name>hive.metastore.try.direct.sql.ddl</name>
850 <value>true</value>
851 <description>
852 Same as hive.metastore.try.direct.sql, for read statements within a transaction that
853 modifies metastore data. Due to non-standard behavior in Postgres, if a direct SQL
854 select query has incorrect syntax or something similar inside a transaction, the
855 entire transaction will fail and fall-back to DataNucleus will not be possible. You
856 should disable the usage of direct SQL inside transactions if that happens in your case.
857 </description>
858 </property>
859 <property>
860 <name>hive.direct.sql.max.query.length</name>
861 <value>100</value>
862 <description>
863 The maximum
864 size of a query string (in KB).
865 </description>
866 </property>
867 <property>
868 <name>hive.direct.sql.max.elements.in.clause</name>
869 <value>1000</value>
870 <description>
871 The maximum number of values in a IN clause. Once exceeded, it will be broken into
872 multiple OR separated IN clauses.
873 </description>
874 </property>
875 <property>
876 <name>hive.direct.sql.max.elements.values.clause</name>
877 <value>1000</value>
878 <description>The maximum number of values in a VALUES clause for INSERT statement.</description>
879 </property>
880 <property>
881 <name>hive.metastore.orm.retrieveMapNullsAsEmptyStrings</name>
882 <value>false</value>
883 <description>Thrift does not support nulls in maps, so any nulls present in maps retrieved from ORM must either be pruned or converted to empty strings. Some backing dbs such as Oracle persist empty strings as nulls, so we should set this parameter if we wish to reverse that behaviour. For others, pruning is the correct behaviour</description>
884 </property>
885 <property>
886 <name>hive.metastore.disallow.incompatible.col.type.changes</name>
887 <value>true</value>
888 <description>
889 If true (default is false), ALTER TABLE operations which change the type of a
890 column (say STRING) to an incompatible type (say MAP) are disallowed.
891 RCFile default SerDe (ColumnarSerDe) serializes the values in such a way that the
892 datatypes can be converted from string to any type. The map is also serialized as
893 a string, which can be read as a string as well. However, with any binary
894 serialization, this is not true. Blocking the ALTER TABLE prevents ClassCastExceptions
895 when subsequently trying to access old partitions.
896
897 Primitive types like INT, STRING, BIGINT, etc., are compatible with each other and are
898 not blocked.
899
900 See HIVE-4409 for more details.
901 </description>
902 </property>
903 <property>
904 <name>hive.table.parameters.default</name>
905 <value/>
906 <description>Default property values for newly created tables</description>
907 </property>
908 <property>
909 <name>hive.ddl.createtablelike.properties.whitelist</name>
910 <value/>
911 <description>Table Properties to copy over when executing a Create Table Like.</description>
912 </property>
913 <property>
914 <name>hive.metastore.rawstore.impl</name>
915 <value>org.apache.hadoop.hive.metastore.ObjectStore</value>
916 <description>
917 Name of the class that implements org.apache.hadoop.hive.metastore.rawstore interface.
918 This class is used to store and retrieval of raw metadata objects such as table, database
919 </description>
920 </property>
921 <property>
922 <name>hive.metastore.txn.store.impl</name>
923 <value>org.apache.hadoop.hive.metastore.txn.CompactionTxnHandler</value>
924 <description>Name of class that implements org.apache.hadoop.hive.metastore.txn.TxnStore. This class is used to store and retrieve transactions and locks</description>
925 </property>
926 <property>
927 <name>javax.jdo.option.ConnectionDriverName</name>
928 <value>org.apache.derby.jdbc.EmbeddedDriver</value>
929 <description>Driver class name for a JDBC metastore</description>
930 </property>
931 <property>
932 <name>javax.jdo.PersistenceManagerFactoryClass</name>
933 <value>org.datanucleus.api.jdo.JDOPersistenceManagerFactory</value>
934 <description>class implementing the jdo persistence</description>
935 </property>
936 <property>
937 <name>hive.metastore.expression.proxy</name>
938 <value>org.apache.hadoop.hive.ql.optimizer.ppr.PartitionExpressionForMetastore</value>
939 <description/>
940 </property>
941 <property>
942 <name>javax.jdo.option.DetachAllOnCommit</name>
943 <value>true</value>
944 <description>Detaches all objects from session so that they can be used after transaction is committed</description>
945 </property>
946 <property>
947 <name>javax.jdo.option.NonTransactionalRead</name>
948 <value>true</value>
949 <description>Reads outside of transactions</description>
950 </property>
951 <property>
952 <name>javax.jdo.option.ConnectionUserName</name>
953 <value>APP</value>
954 <description>Username to use against metastore database</description>
955 </property>
956 <property>
957 <name>hive.metastore.end.function.listeners</name>
958 <value/>
959 <description>List of comma separated listeners for the end of metastore functions.</description>
960 </property>
961 <property>
962 <name>hive.metastore.partition.inherit.table.properties</name>
963 <value/>
964 <description>
965 List of comma separated keys occurring in table properties which will get inherited to newly created partitions.
966 * implies all the keys will get inherited.
967 </description>
968 </property>
969 <property>
970 <name>hive.metastore.filter.hook</name>
971 <value>org.apache.hadoop.hive.metastore.DefaultMetaStoreFilterHookImpl</value>
972 <description>Metastore hook class for filtering the metadata read results. If hive.security.authorization.manageris set to instance of HiveAuthorizerFactory, then this value is ignored.</description>
973 </property>
974 <property>
975 <name>hive.metastore.dml.events</name>
976 <value>false</value>
977 <description>If true, the metastore will be asked to fire events for DML operations</description>
978 </property>
979 <property>
980 <name>hive.metastore.client.drop.partitions.using.expressions</name>
981 <value>true</value>
982 <description>Choose whether dropping partitions with HCatClient pushes the partition-predicate to the metastore, or drops partitions iteratively</description>
983 </property>
984 <property>
985 <name>hive.metastore.aggregate.stats.cache.enabled</name>
986 <value>true</value>
987 <description>Whether aggregate stats caching is enabled or not.</description>
988 </property>
989 <property>
990 <name>hive.metastore.aggregate.stats.cache.size</name>
991 <value>10000</value>
992 <description>Maximum number of aggregate stats nodes that we will place in the metastore aggregate stats cache.</description>
993 </property>
994 <property>
995 <name>hive.metastore.aggregate.stats.cache.max.partitions</name>
996 <value>10000</value>
997 <description>Maximum number of partitions that are aggregated per cache node.</description>
998 </property>
999 <property>
1000 <name>hive.metastore.aggregate.stats.cache.fpp</name>
1001 <value>0.01</value>
1002 <description>Maximum false positive probability for the Bloom Filter used in each aggregate stats cache node (default 1%).</description>
1003 </property>
1004 <property>
1005 <name>hive.metastore.aggregate.stats.cache.max.variance</name>
1006 <value>0.01</value>
1007 <description>Maximum tolerable variance in number of partitions between a cached node and our request (default 1%).</description>
1008 </property>
1009 <property>
1010 <name>hive.metastore.aggregate.stats.cache.ttl</name>
1011 <value>600s</value>
1012 <description>
1013 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
1014 Number of seconds for a cached node to be active in the cache before they become stale.
1015 </description>
1016 </property>
1017 <property>
1018 <name>hive.metastore.aggregate.stats.cache.max.writer.wait</name>
1019 <value>5000ms</value>
1020 <description>
1021 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
1022 Number of milliseconds a writer will wait to acquire the writelock before giving up.
1023 </description>
1024 </property>
1025 <property>
1026 <name>hive.metastore.aggregate.stats.cache.max.reader.wait</name>
1027 <value>1000ms</value>
1028 <description>
1029 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
1030 Number of milliseconds a reader will wait to acquire the readlock before giving up.
1031 </description>
1032 </property>
1033 <property>
1034 <name>hive.metastore.aggregate.stats.cache.max.full</name>
1035 <value>0.9</value>
1036 <description>Maximum cache full % after which the cache cleaner thread kicks in.</description>
1037 </property>
1038 <property>
1039 <name>hive.metastore.aggregate.stats.cache.clean.until</name>
1040 <value>0.8</value>
1041 <description>The cleaner thread cleans until cache reaches this % full size.</description>
1042 </property>
1043 <property>
1044 <name>hive.metastore.metrics.enabled</name>
1045 <value>false</value>
1046 <description>Enable metrics on the metastore.</description>
1047 </property>
1048 <property>
1049 <name>hive.metastore.initial.metadata.count.enabled</name>
1050 <value>true</value>
1051 <description>Enable a metadata count at metastore startup for metrics.</description>
1052 </property>
1053 <property>
1054 <name>hive.metadata.export.location</name>
1055 <value/>
1056 <description>
1057 When used in conjunction with the org.apache.hadoop.hive.ql.parse.MetaDataExportListener pre event listener,
1058 it is the location to which the metadata will be exported. The default is an empty string, which results in the
1059 metadata being exported to the current user's home directory on HDFS.
1060 </description>
1061 </property>
1062 <property>
1063 <name>hive.metadata.move.exported.metadata.to.trash</name>
1064 <value>true</value>
1065 <description>
1066 When used in conjunction with the org.apache.hadoop.hive.ql.parse.MetaDataExportListener pre event listener,
1067 this setting determines if the metadata that is exported will subsequently be moved to the user's trash directory
1068 alongside the dropped table data. This ensures that the metadata will be cleaned up along with the dropped table data.
1069 </description>
1070 </property>
1071 <property>
1072 <name>hive.cli.errors.ignore</name>
1073 <value>false</value>
1074 <description/>
1075 </property>
1076 <property>
1077 <name>hive.cli.print.current.db</name>
1078 <value>false</value>
1079 <description>Whether to include the current database in the Hive prompt.</description>
1080 </property>
1081 <property>
1082 <name>hive.cli.prompt</name>
1083 <value>hive</value>
1084 <description>
1085 Command line prompt configuration value. Other hiveconf can be used in this configuration value.
1086 Variable substitution will only be invoked at the Hive CLI startup.
1087 </description>
1088 </property>
1089 <property>
1090 <name>hive.cli.pretty.output.num.cols</name>
1091 <value>-1</value>
1092 <description>
1093 The number of columns to use when formatting output generated by the DESCRIBE PRETTY table_name command.
1094 If the value of this property is -1, then Hive will use the auto-detected terminal width.
1095 </description>
1096 </property>
1097 <property>
1098 <name>hive.metastore.fs.handler.class</name>
1099 <value>org.apache.hadoop.hive.metastore.HiveMetaStoreFsImpl</value>
1100 <description/>
1101 </property>
1102 <property>
1103 <name>hive.session.id</name>
1104 <value/>
1105 <description/>
1106 </property>
1107 <property>
1108 <name>hive.session.silent</name>
1109 <value>false</value>
1110 <description/>
1111 </property>
1112 <property>
1113 <name>hive.session.history.enabled</name>
1114 <value>false</value>
1115 <description>Whether to log Hive query, query plan, runtime statistics etc.</description>
1116 </property>
1117 <property>
1118 <name>hive.query.string</name>
1119 <value/>
1120 <description>Query being executed (might be multiple per a session)</description>
1121 </property>
1122 <property>
1123 <name>hive.query.id</name>
1124 <value/>
1125 <description>ID for query being executed (might be multiple per a session)</description>
1126 </property>
1127 <property>
1128 <name>hive.jobname.length</name>
1129 <value>50</value>
1130 <description>max jobname length</description>
1131 </property>
1132 <property>
1133 <name>hive.jar.path</name>
1134 <value/>
1135 <description>The location of hive_cli.jar that is used when submitting jobs in a separate jvm.</description>
1136 </property>
1137 <property>
1138 <name>hive.aux.jars.path</name>
1139 <value/>
1140 <description>The location of the plugin jars that contain implementations of user defined functions and serdes.</description>
1141 </property>
1142 <property>
1143 <name>hive.reloadable.aux.jars.path</name>
1144 <value/>
1145 <description>Jars can be renewed by executing reload command. And these jars can be used as the auxiliary classes like creating a UDF or SerDe.</description>
1146 </property>
1147 <property>
1148 <name>hive.added.files.path</name>
1149 <value/>
1150 <description>This an internal parameter.</description>
1151 </property>
1152 <property>
1153 <name>hive.added.jars.path</name>
1154 <value/>
1155 <description>This an internal parameter.</description>
1156 </property>
1157 <property>
1158 <name>hive.added.archives.path</name>
1159 <value/>
1160 <description>This an internal parameter.</description>
1161 </property>
1162 <property>
1163 <name>hive.auto.progress.timeout</name>
1164 <value>0s</value>
1165 <description>
1166 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
1167 How long to run autoprogressor for the script/UDTF operators.
1168 Set to 0 for forever.
1169 </description>
1170 </property>
1171 <property>
1172 <name>hive.script.auto.progress</name>
1173 <value>false</value>
1174 <description>
1175 Whether Hive Transform/Map/Reduce Clause should automatically send progress information to TaskTracker
1176 to avoid the task getting killed because of inactivity. Hive sends progress information when the script is
1177 outputting to stderr. This option removes the need of periodically producing stderr messages,
1178 but users should be cautious because this may prevent infinite loops in the scripts to be killed by TaskTracker.
1179 </description>
1180 </property>
1181 <property>
1182 <name>hive.script.operator.id.env.var</name>
1183 <value>HIVE_SCRIPT_OPERATOR_ID</value>
1184 <description>
1185 Name of the environment variable that holds the unique script operator ID in the user's
1186 transform function (the custom mapper/reducer that the user has specified in the query)
1187 </description>
1188 </property>
1189 <property>
1190 <name>hive.script.operator.truncate.env</name>
1191 <value>false</value>
1192 <description>Truncate each environment variable for external script in scripts operator to 20KB (to fit system limits)</description>
1193 </property>
1194 <property>
1195 <name>hive.script.operator.env.blacklist</name>
1196 <value>hive.txn.valid.txns,hive.script.operator.env.blacklist</value>
1197 <description>Comma separated list of keys from the configuration file not to convert to environment variables when envoking the script operator</description>
1198 </property>
1199 <property>
1200 <name>hive.strict.checks.large.query</name>
1201 <value>false</value>
1202 <description>
1203 Enabling strict large query checks disallows the following:
1204 Orderby without limit.
1205 No partition being picked up for a query against partitioned table.
1206 Note that these checks currently do not consider data size, only the query pattern.
1207 </description>
1208 </property>
1209 <property>
1210 <name>hive.strict.checks.type.safety</name>
1211 <value>true</value>
1212 <description>
1213 Enabling strict type safety checks disallows the following:
1214 Comparing bigints and strings.
1215 Comparing bigints and doubles.
1216 </description>
1217 </property>
1218 <property>
1219 <name>hive.strict.checks.cartesian.product</name>
1220 <value>true</value>
1221 <description>
1222 Enabling strict large query checks disallows the following:
1223 Cartesian product (cross join).
1224 </description>
1225 </property>
1226 <property>
1227 <name>hive.mapred.mode</name>
1228 <value>nonstrict</value>
1229 <description>Deprecated; use hive.strict.checks.* settings instead.</description>
1230 </property>
1231 <property>
1232 <name>hive.alias</name>
1233 <value/>
1234 <description/>
1235 </property>
1236 <property>
1237 <name>hive.map.aggr</name>
1238 <value>true</value>
1239 <description>Whether to use map-side aggregation in Hive Group By queries</description>
1240 </property>
1241 <property>
1242 <name>hive.groupby.skewindata</name>
1243 <value>false</value>
1244 <description>Whether there is skew in data to optimize group by queries</description>
1245 </property>
1246 <property>
1247 <name>hive.join.emit.interval</name>
1248 <value>1000</value>
1249 <description>How many rows in the right-most join operand Hive should buffer before emitting the join result.</description>
1250 </property>
1251 <property>
1252 <name>hive.join.cache.size</name>
1253 <value>25000</value>
1254 <description>How many rows in the joining tables (except the streaming table) should be cached in memory.</description>
1255 </property>
1256 <property>
1257 <name>hive.cbo.enable</name>
1258 <value>true</value>
1259 <description>Flag to control enabling Cost Based Optimizations using Calcite framework.</description>
1260 </property>
1261 <property>
1262 <name>hive.cbo.returnpath.hiveop</name>
1263 <value>false</value>
1264 <description>Flag to control calcite plan to hive operator conversion</description>
1265 </property>
1266 <property>
1267 <name>hive.cbo.costmodel.extended</name>
1268 <value>false</value>
1269 <description>Flag to control enabling the extended cost model based onCPU, IO and cardinality. Otherwise, the cost model is based on cardinality.</description>
1270 </property>
1271 <property>
1272 <name>hive.cbo.costmodel.cpu</name>
1273 <value>0.000001</value>
1274 <description>Default cost of a comparison</description>
1275 </property>
1276 <property>
1277 <name>hive.cbo.costmodel.network</name>
1278 <value>150.0</value>
1279 <description>Default cost of a transfering a byte over network; expressed as multiple of CPU cost</description>
1280 </property>
1281 <property>
1282 <name>hive.cbo.costmodel.local.fs.write</name>
1283 <value>4.0</value>
1284 <description>Default cost of writing a byte to local FS; expressed as multiple of NETWORK cost</description>
1285 </property>
1286 <property>
1287 <name>hive.cbo.costmodel.local.fs.read</name>
1288 <value>4.0</value>
1289 <description>Default cost of reading a byte from local FS; expressed as multiple of NETWORK cost</description>
1290 </property>
1291 <property>
1292 <name>hive.cbo.costmodel.hdfs.write</name>
1293 <value>10.0</value>
1294 <description>Default cost of writing a byte to HDFS; expressed as multiple of Local FS write cost</description>
1295 </property>
1296 <property>
1297 <name>hive.cbo.costmodel.hdfs.read</name>
1298 <value>1.5</value>
1299 <description>Default cost of reading a byte from HDFS; expressed as multiple of Local FS read cost</description>
1300 </property>
1301 <property>
1302 <name>hive.transpose.aggr.join</name>
1303 <value>false</value>
1304 <description>push aggregates through join</description>
1305 </property>
1306 <property>
1307 <name>hive.mapjoin.bucket.cache.size</name>
1308 <value>100</value>
1309 <description/>
1310 </property>
1311 <property>
1312 <name>hive.mapjoin.optimized.hashtable</name>
1313 <value>true</value>
1314 <description>
1315 Whether Hive should use memory-optimized hash table for MapJoin.
1316 Only works on Tez and Spark, because memory-optimized hashtable cannot be serialized.
1317 </description>
1318 </property>
1319 <property>
1320 <name>hive.mapjoin.optimized.hashtable.probe.percent</name>
1321 <value>0.5</value>
1322 <description>Probing space percentage of the optimized hashtable</description>
1323 </property>
1324 <property>
1325 <name>hive.mapjoin.hybridgrace.hashtable</name>
1326 <value>true</value>
1327 <description>Whether to use hybridgrace hash join as the join method for mapjoin. Tez only.</description>
1328 </property>
1329 <property>
1330 <name>hive.mapjoin.hybridgrace.memcheckfrequency</name>
1331 <value>1024</value>
1332 <description>For hybrid grace hash join, how often (how many rows apart) we check if memory is full. This number should be power of 2.</description>
1333 </property>
1334 <property>
1335 <name>hive.mapjoin.hybridgrace.minwbsize</name>
1336 <value>524288</value>
1337 <description>For hybrid graceHash join, the minimum write buffer size used by optimized hashtable. Default is 512 KB.</description>
1338 </property>
1339 <property>
1340 <name>hive.mapjoin.hybridgrace.minnumpartitions</name>
1341 <value>16</value>
1342 <description>ForHybrid grace hash join, the minimum number of partitions to create.</description>
1343 </property>
1344 <property>
1345 <name>hive.mapjoin.optimized.hashtable.wbsize</name>
1346 <value>8388608</value>
1347 <description>
1348 Optimized hashtable (see hive.mapjoin.optimized.hashtable) uses a chain of buffers to
1349 store data. This is one buffer size. HT may be slightly faster if this is larger, but for small
1350 joins unnecessary memory will be allocated and then trimmed.
1351 </description>
1352 </property>
1353 <property>
1354 <name>hive.smbjoin.cache.rows</name>
1355 <value>10000</value>
1356 <description>How many rows with the same key value should be cached in memory per smb joined table.</description>
1357 </property>
1358 <property>
1359 <name>hive.groupby.mapaggr.checkinterval</name>
1360 <value>100000</value>
1361 <description>Number of rows after which size of the grouping keys/aggregation classes is performed</description>
1362 </property>
1363 <property>
1364 <name>hive.map.aggr.hash.percentmemory</name>
1365 <value>0.5</value>
1366 <description>Portion of total memory to be used by map-side group aggregation hash table</description>
1367 </property>
1368 <property>
1369 <name>hive.mapjoin.followby.map.aggr.hash.percentmemory</name>
1370 <value>0.3</value>
1371 <description>Portion of total memory to be used by map-side group aggregation hash table, when this group by is followed by map join</description>
1372 </property>
1373 <property>
1374 <name>hive.map.aggr.hash.force.flush.memory.threshold</name>
1375 <value>0.9</value>
1376 <description>
1377 The max memory to be used by map-side group aggregation hash table.
1378 If the memory usage is higher than this number, force to flush data
1379 </description>
1380 </property>
1381 <property>
1382 <name>hive.map.aggr.hash.min.reduction</name>
1383 <value>0.5</value>
1384 <description>
1385 Hash aggregation will be turned off if the ratio between hash table size and input rows is bigger than this number.
1386 Set to 1 to make sure hash aggregation is never turned off.
1387 </description>
1388 </property>
1389 <property>
1390 <name>hive.multigroupby.singlereducer</name>
1391 <value>true</value>
1392 <description>
1393 Whether to optimize multi group by query to generate single M/R job plan. If the multi group by query has
1394 common group by keys, it will be optimized to generate single M/R job.
1395 </description>
1396 </property>
1397 <property>
1398 <name>hive.map.groupby.sorted</name>
1399 <value>true</value>
1400 <description>
1401 If the bucketing/sorting properties of the table exactly match the grouping key, whether to perform
1402 the group by in the mapper by using BucketizedHiveInputFormat. The only downside to this
1403 is that it limits the number of mappers to the number of files.
1404 </description>
1405 </property>
1406 <property>
1407 <name>hive.groupby.orderby.position.alias</name>
1408 <value>false</value>
1409 <description>Whether to enable using Column Position Alias in Group By or Order By</description>
1410 </property>
1411 <property>
1412 <name>hive.new.job.grouping.set.cardinality</name>
1413 <value>30</value>
1414 <description>
1415 Whether a new map-reduce job should be launched for grouping sets/rollups/cubes.
1416 For a query like: select a, b, c, count(1) from T group by a, b, c with rollup;
1417 4 rows are created per row: (a, b, c), (a, b, null), (a, null, null), (null, null, null).
1418 This can lead to explosion across map-reduce boundary if the cardinality of T is very high,
1419 and map-side aggregation does not do a very good job.
1420
1421 This parameter decides if Hive should add an additional map-reduce job. If the grouping set
1422 cardinality (4 in the example above), is more than this value, a new MR job is added under the
1423 assumption that the original group by will reduce the data size.
1424 </description>
1425 </property>
1426 <property>
1427 <name>hive.groupby.limit.extrastep</name>
1428 <value>true</value>
1429 <description>
1430 This parameter decides if Hive should
1431 create new MR job for sorting final output
1432 </description>
1433 </property>
1434 <property>
1435 <name>hive.exec.copyfile.maxsize</name>
1436 <value>33554432</value>
1437 <description>Maximum file size (in Mb) that Hive uses to do single HDFS copies between directories.Distributed copies (distcp) will be used instead for bigger files so that copies can be done faster.</description>
1438 </property>
1439 <property>
1440 <name>hive.udtf.auto.progress</name>
1441 <value>false</value>
1442 <description>
1443 Whether Hive should automatically send progress information to TaskTracker
1444 when using UDTF's to prevent the task getting killed because of inactivity. Users should be cautious
1445 because this may prevent TaskTracker from killing tasks with infinite loops.
1446 </description>
1447 </property>
1448 <property>
1449 <name>hive.default.fileformat</name>
1450 <value>TextFile</value>
1451 <description>
1452 Expects one of [textfile, sequencefile, rcfile, orc].
1453 Default file format for CREATE TABLE statement. Users can explicitly override it by CREATE TABLE ... STORED AS [FORMAT]
1454 </description>
1455 </property>
1456 <property>
1457 <name>hive.default.fileformat.managed</name>
1458 <value>none</value>
1459 <description>
1460 Expects one of [none, textfile, sequencefile, rcfile, orc].
1461 Default file format for CREATE TABLE statement applied to managed tables only. External tables will be
1462 created with format specified by hive.default.fileformat. Leaving this null will result in using hive.default.fileformat
1463 for all tables.
1464 </description>
1465 </property>
1466 <property>
1467 <name>hive.query.result.fileformat</name>
1468 <value>SequenceFile</value>
1469 <description>
1470 Expects one of [textfile, sequencefile, rcfile, llap].
1471 Default file format for storing result of the query.
1472 </description>
1473 </property>
1474 <property>
1475 <name>hive.fileformat.check</name>
1476 <value>true</value>
1477 <description>Whether to check file format or not when loading data files</description>
1478 </property>
1479 <property>
1480 <name>hive.default.rcfile.serde</name>
1481 <value>org.apache.hadoop.hive.serde2.columnar.LazyBinaryColumnarSerDe</value>
1482 <description>The default SerDe Hive will use for the RCFile format</description>
1483 </property>
1484 <property>
1485 <name>hive.default.serde</name>
1486 <value>org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe</value>
1487 <description>The default SerDe Hive will use for storage formats that do not specify a SerDe.</description>
1488 </property>
1489 <property>
1490 <name>hive.serdes.using.metastore.for.schema</name>
1491 <value>org.apache.hadoop.hive.ql.io.orc.OrcSerde,org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe,org.apache.hadoop.hive.serde2.columnar.ColumnarSerDe,org.apache.hadoop.hive.serde2.dynamic_type.DynamicSerDe,org.apache.hadoop.hive.serde2.MetadataTypedColumnsetSerDe,org.apache.hadoop.hive.serde2.columnar.LazyBinaryColumnarSerDe,org.apache.hadoop.hive.ql.io.parquet.serde.ParquetHiveSerDe,org.apache.hadoop.hive.serde2.lazybinary.LazyBinarySerDe</value>
1492 <description>SerDes retrieving schema from metastore. This is an internal parameter.</description>
1493 </property>
1494
1495 <property>
1496 <name>system:java.io.tmpdir</name>
1497 <value>/tmp/hive/java</value>
1498 </property>
1499 <property>
1500 <name>system:user.name</name>
1501 <value>${user.name}</value>
1502 </property>
1503
1504 <property>
1505 <name>hive.querylog.location</name>
1506 <value>${system:java.io.tmpdir}/${system:user.name}</value>
1507 <description>Location of Hive run time structured log file</description>
1508 </property>
1509 <property>
1510 <name>hive.querylog.enable.plan.progress</name>
1511 <value>true</value>
1512 <description>
1513 Whether to log the plan's progress every time a job's progress is checked.
1514 These logs are written to the location specified by hive.querylog.location
1515 </description>
1516 </property>
1517 <property>
1518 <name>hive.querylog.plan.progress.interval</name>
1519 <value>60000ms</value>
1520 <description>
1521 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
1522 The interval to wait between logging the plan's progress.
1523 If there is a whole number percentage change in the progress of the mappers or the reducers,
1524 the progress is logged regardless of this value.
1525 The actual interval will be the ceiling of (this value divided by the value of
1526 hive.exec.counters.pull.interval) multiplied by the value of hive.exec.counters.pull.interval
1527 I.e. if it is not divide evenly by the value of hive.exec.counters.pull.interval it will be
1528 logged less frequently than specified.
1529 This only has an effect if hive.querylog.enable.plan.progress is set to true.
1530 </description>
1531 </property>
1532 <property>
1533 <name>hive.script.serde</name>
1534 <value>org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe</value>
1535 <description>The default SerDe for transmitting input data to and reading output data from the user scripts. </description>
1536 </property>
1537 <property>
1538 <name>hive.script.recordreader</name>
1539 <value>org.apache.hadoop.hive.ql.exec.TextRecordReader</value>
1540 <description>The default record reader for reading data from the user scripts. </description>
1541 </property>
1542 <property>
1543 <name>hive.script.recordwriter</name>
1544 <value>org.apache.hadoop.hive.ql.exec.TextRecordWriter</value>
1545 <description>The default record writer for writing data to the user scripts. </description>
1546 </property>
1547 <property>
1548 <name>hive.transform.escape.input</name>
1549 <value>false</value>
1550 <description>
1551 This adds an option to escape special chars (newlines, carriage returns and
1552 tabs) when they are passed to the user script. This is useful if the Hive tables
1553 can contain data that contains special characters.
1554 </description>
1555 </property>
1556 <property>
1557 <name>hive.binary.record.max.length</name>
1558 <value>1000</value>
1559 <description>
1560 Read from a binary stream and treat each hive.binary.record.max.length bytes as a record.
1561 The last record before the end of stream can have less than hive.binary.record.max.length bytes
1562 </description>
1563 </property>
1564 <property>
1565 <name>hive.hwi.listen.host</name>
1566 <value>0.0.0.0</value>
1567 <description>This is the host address the Hive Web Interface will listen on</description>
1568 </property>
1569 <property>
1570 <name>hive.hwi.listen.port</name>
1571 <value>9999</value>
1572 <description>This is the port the Hive Web Interface will listen on</description>
1573 </property>
1574 <property>
1575 <name>hive.hwi.war.file</name>
1576 <value>${env:HWI_WAR_FILE}</value>
1577 <description>This sets the path to the HWI war file, relative to ${HIVE_HOME}. </description>
1578 </property>
1579 <property>
1580 <name>hive.mapred.local.mem</name>
1581 <value>0</value>
1582 <description>mapper/reducer memory in local mode</description>
1583 </property>
1584 <property>
1585 <name>hive.mapjoin.smalltable.filesize</name>
1586 <value>25000000</value>
1587 <description>
1588 The threshold for the input file size of the small tables; if the file size is smaller
1589 than this threshold, it will try to convert the common join into map join
1590 </description>
1591 </property>
1592 <property>
1593 <name>hive.exec.schema.evolution</name>
1594 <value>true</value>
1595 <description>Use schema evolution to convert self-describing file format's data to the schema desired by the reader.</description>
1596 </property>
1597 <property>
1598 <name>hive.sample.seednumber</name>
1599 <value>0</value>
1600 <description>A number used to percentage sampling. By changing this number, user will change the subsets of data sampled.</description>
1601 </property>
1602 <property>
1603 <name>hive.test.mode</name>
1604 <value>false</value>
1605 <description>Whether Hive is running in test mode. If yes, it turns on sampling and prefixes the output tablename.</description>
1606 </property>
1607 <property>
1608 <name>hive.test.mode.prefix</name>
1609 <value>test_</value>
1610 <description>In test mode, specfies prefixes for the output table</description>
1611 </property>
1612 <property>
1613 <name>hive.test.mode.samplefreq</name>
1614 <value>32</value>
1615 <description>
1616 In test mode, specfies sampling frequency for table, which is not bucketed,
1617 For example, the following query:
1618 INSERT OVERWRITE TABLE dest SELECT col1 from src
1619 would be converted to
1620 INSERT OVERWRITE TABLE test_dest
1621 SELECT col1 from src TABLESAMPLE (BUCKET 1 out of 32 on rand(1))
1622 </description>
1623 </property>
1624 <property>
1625 <name>hive.test.mode.nosamplelist</name>
1626 <value/>
1627 <description>In test mode, specifies comma separated table names which would not apply sampling</description>
1628 </property>
1629 <property>
1630 <name>hive.test.dummystats.aggregator</name>
1631 <value/>
1632 <description>internal variable for test</description>
1633 </property>
1634 <property>
1635 <name>hive.test.dummystats.publisher</name>
1636 <value/>
1637 <description>internal variable for test</description>
1638 </property>
1639 <property>
1640 <name>hive.test.currenttimestamp</name>
1641 <value/>
1642 <description>current timestamp for test</description>
1643 </property>
1644 <property>
1645 <name>hive.test.rollbacktxn</name>
1646 <value>false</value>
1647 <description>For testing only. Will mark every ACID transaction aborted</description>
1648 </property>
1649 <property>
1650 <name>hive.test.fail.compaction</name>
1651 <value>false</value>
1652 <description>For testing only. Will cause CompactorMR to fail.</description>
1653 </property>
1654 <property>
1655 <name>hive.test.fail.heartbeater</name>
1656 <value>false</value>
1657 <description>For testing only. Will cause Heartbeater to fail.</description>
1658 </property>
1659 <property>
1660 <name>hive.merge.mapfiles</name>
1661 <value>true</value>
1662 <description>Merge small files at the end of a map-only job</description>
1663 </property>
1664 <property>
1665 <name>hive.merge.mapredfiles</name>
1666 <value>false</value>
1667 <description>Merge small files at the end of a map-reduce job</description>
1668 </property>
1669 <property>
1670 <name>hive.merge.tezfiles</name>
1671 <value>false</value>
1672 <description>Merge small files at the end of a Tez DAG</description>
1673 </property>
1674 <property>
1675 <name>hive.merge.sparkfiles</name>
1676 <value>false</value>
1677 <description>Merge small files at the end of a Spark DAG Transformation</description>
1678 </property>
1679 <property>
1680 <name>hive.merge.size.per.task</name>
1681 <value>256000000</value>
1682 <description>Size of merged files at the end of the job</description>
1683 </property>
1684 <property>
1685 <name>hive.merge.smallfiles.avgsize</name>
1686 <value>16000000</value>
1687 <description>
1688 When the average output file size of a job is less than this number, Hive will start an additional
1689 map-reduce job to merge the output files into bigger files. This is only done for map-only jobs
1690 if hive.merge.mapfiles is true, and for map-reduce jobs if hive.merge.mapredfiles is true.
1691 </description>
1692 </property>
1693 <property>
1694 <name>hive.merge.rcfile.block.level</name>
1695 <value>true</value>
1696 <description/>
1697 </property>
1698 <property>
1699 <name>hive.merge.orcfile.stripe.level</name>
1700 <value>true</value>
1701 <description>
1702 When hive.merge.mapfiles, hive.merge.mapredfiles or hive.merge.tezfiles is enabled
1703 while writing a table with ORC file format, enabling this config will do stripe-level
1704 fast merge for small ORC files. Note that enabling this config will not honor the
1705 padding tolerance config (hive.exec.orc.block.padding.tolerance).
1706 </description>
1707 </property>
1708 <property>
1709 <name>hive.exec.rcfile.use.explicit.header</name>
1710 <value>true</value>
1711 <description>
1712 If this is set the header for RCFiles will simply be RCF. If this is not
1713 set the header will be that borrowed from sequence files, e.g. SEQ- followed
1714 by the input and output RCFile formats.
1715 </description>
1716 </property>
1717 <property>
1718 <name>hive.exec.rcfile.use.sync.cache</name>
1719 <value>true</value>
1720 <description/>
1721 </property>
1722 <property>
1723 <name>hive.io.rcfile.record.interval</name>
1724 <value>2147483647</value>
1725 <description/>
1726 </property>
1727 <property>
1728 <name>hive.io.rcfile.column.number.conf</name>
1729 <value>0</value>
1730 <description/>
1731 </property>
1732 <property>
1733 <name>hive.io.rcfile.tolerate.corruptions</name>
1734 <value>false</value>
1735 <description/>
1736 </property>
1737 <property>
1738 <name>hive.io.rcfile.record.buffer.size</name>
1739 <value>4194304</value>
1740 <description/>
1741 </property>
1742 <property>
1743 <name>parquet.memory.pool.ratio</name>
1744 <value>0.5</value>
1745 <description>
1746 Maximum fraction of heap that can be used by Parquet file writers in one task.
1747 It is for avoiding OutOfMemory error in tasks. Work with Parquet 1.6.0 and above.
1748 This config parameter is defined in Parquet, so that it does not start with 'hive.'.
1749 </description>
1750 </property>
1751 <property>
1752 <name>hive.parquet.timestamp.skip.conversion</name>
1753 <value>true</value>
1754 <description>Current Hive implementation of parquet stores timestamps to UTC, this flag allows skipping of the conversionon reading parquet files from other tools</description>
1755 </property>
1756 <property>
1757 <name>hive.int.timestamp.conversion.in.seconds</name>
1758 <value>false</value>
1759 <description>
1760 Boolean/tinyint/smallint/int/bigint value is interpreted as milliseconds during the timestamp conversion.
1761 Set this flag to true to interpret the value as seconds to be consistent with float/double.
1762 </description>
1763 </property>
1764 <property>
1765 <name>hive.exec.orc.memory.pool</name>
1766 <value>0.5</value>
1767 <description>Maximum fraction of heap that can be used by ORC file writers</description>
1768 </property>
1769 <property>
1770 <name>hive.exec.orc.write.format</name>
1771 <value/>
1772 <description>
1773 Define the version of the file to write. Possible values are 0.11 and 0.12.
1774 If this parameter is not defined, ORC will use the run length encoding (RLE)
1775 introduced in Hive 0.12. Any value other than 0.11 results in the 0.12 encoding.
1776 </description>
1777 </property>
1778 <property>
1779 <name>hive.exec.orc.default.stripe.size</name>
1780 <value>67108864</value>
1781 <description>Define the default ORC stripe size, in bytes.</description>
1782 </property>
1783 <property>
1784 <name>hive.exec.orc.default.block.size</name>
1785 <value>268435456</value>
1786 <description>Define the default file system block size for ORC files.</description>
1787 </property>
1788 <property>
1789 <name>hive.exec.orc.dictionary.key.size.threshold</name>
1790 <value>0.8</value>
1791 <description>
1792 If the number of keys in a dictionary is greater than this fraction of the total number of
1793 non-null rows, turn off dictionary encoding. Use 1 to always use dictionary encoding.
1794 </description>
1795 </property>
1796 <property>
1797 <name>hive.exec.orc.default.row.index.stride</name>
1798 <value>10000</value>
1799 <description>
1800 Define the default ORC index stride in number of rows. (Stride is the number of rows
1801 an index entry represents.)
1802 </description>
1803 </property>
1804 <property>
1805 <name>hive.orc.row.index.stride.dictionary.check</name>
1806 <value>true</value>
1807 <description>
1808 If enabled dictionary check will happen after first row index stride (default 10000 rows)
1809 else dictionary check will happen before writing first stripe. In both cases, the decision
1810 to use dictionary or not will be retained thereafter.
1811 </description>
1812 </property>
1813 <property>
1814 <name>hive.exec.orc.default.buffer.size</name>
1815 <value>262144</value>
1816 <description>Define the default ORC buffer size, in bytes.</description>
1817 </property>
1818 <property>
1819 <name>hive.exec.orc.base.delta.ratio</name>
1820 <value>8</value>
1821 <description>
1822 The ratio of base writer and
1823 delta writer in terms of STRIPE_SIZE and BUFFER_SIZE.
1824 </description>
1825 </property>
1826 <property>
1827 <name>hive.exec.orc.default.block.padding</name>
1828 <value>true</value>
1829 <description>Define the default block padding, which pads stripes to the HDFS block boundaries.</description>
1830 </property>
1831 <property>
1832 <name>hive.exec.orc.block.padding.tolerance</name>
1833 <value>0.05</value>
1834 <description>
1835 Define the tolerance for block padding as a decimal fraction of stripe size (for
1836 example, the default value 0.05 is 5% of the stripe size). For the defaults of 64Mb
1837 ORC stripe and 256Mb HDFS blocks, the default block padding tolerance of 5% will
1838 reserve a maximum of 3.2Mb for padding within the 256Mb block. In that case, if the
1839 available size within the block is more than 3.2Mb, a new smaller stripe will be
1840 inserted to fit within that space. This will make sure that no stripe written will
1841 cross block boundaries and cause remote reads within a node local task.
1842 </description>
1843 </property>
1844 <property>
1845 <name>hive.exec.orc.default.compress</name>
1846 <value>ZLIB</value>
1847 <description>Define the default compression codec for ORC file</description>
1848 </property>
1849 <property>
1850 <name>hive.exec.orc.encoding.strategy</name>
1851 <value>SPEED</value>
1852 <description>
1853 Expects one of [speed, compression].
1854 Define the encoding strategy to use while writing data. Changing this will
1855 only affect the light weight encoding for integers. This flag will not
1856 change the compression level of higher level compression codec (like ZLIB).
1857 </description>
1858 </property>
1859 <property>
1860 <name>hive.exec.orc.compression.strategy</name>
1861 <value>SPEED</value>
1862 <description>
1863 Expects one of [speed, compression].
1864 Define the compression strategy to use while writing data.
1865 This changes the compression level of higher level compression codec (like ZLIB).
1866 </description>
1867 </property>
1868 <property>
1869 <name>hive.exec.orc.split.strategy</name>
1870 <value>HYBRID</value>
1871 <description>
1872 Expects one of [hybrid, bi, etl].
1873 This is not a user level config. BI strategy is used when the requirement is to spend less time in split generation as opposed to query execution (split generation does not read or cache file footers). ETL strategy is used when spending little more time in split generation is acceptable (split generation reads and caches file footers). HYBRID chooses between the above strategies based on heuristics.
1874 </description>
1875 </property>
1876 <property>
1877 <name>hive.orc.splits.ms.footer.cache.enabled</name>
1878 <value>false</value>
1879 <description>Whether to enable using file metadata cache in metastore for ORC file footers.</description>
1880 </property>
1881 <property>
1882 <name>hive.orc.splits.ms.footer.cache.ppd.enabled</name>
1883 <value>true</value>
1884 <description>
1885 Whether to enable file footer cache PPD (hive.orc.splits.ms.footer.cache.enabled
1886 must also be set to true for this to work).
1887 </description>
1888 </property>
1889 <property>
1890 <name>hive.orc.splits.include.file.footer</name>
1891 <value>false</value>
1892 <description>
1893 If turned on splits generated by orc will include metadata about the stripes in the file. This
1894 data is read remotely (from the client or HS2 machine) and sent to all the tasks.
1895 </description>
1896 </property>
1897 <property>
1898 <name>hive.orc.splits.directory.batch.ms</name>
1899 <value>0</value>
1900 <description>
1901 How long, in ms, to wait to batch input directories for processing during ORC split
1902 generation. 0 means process directories individually. This can increase the number of
1903 metastore calls if metastore metadata cache is used.
1904 </description>
1905 </property>
1906 <property>
1907 <name>hive.orc.splits.include.fileid</name>
1908 <value>true</value>
1909 <description>Include file ID in splits on file systems that support it.</description>
1910 </property>
1911 <property>
1912 <name>hive.orc.splits.allow.synthetic.fileid</name>
1913 <value>true</value>
1914 <description>Allow synthetic file ID in splits on file systems that don't have a native one.</description>
1915 </property>
1916 <property>
1917 <name>hive.orc.cache.stripe.details.size</name>
1918 <value>10000</value>
1919 <description>Max cache size for keeping meta info about orc splits cached in the client.</description>
1920 </property>
1921 <property>
1922 <name>hive.orc.compute.splits.num.threads</name>
1923 <value>10</value>
1924 <description>How many threads orc should use to create splits in parallel.</description>
1925 </property>
1926 <property>
1927 <name>hive.exec.orc.skip.corrupt.data</name>
1928 <value>false</value>
1929 <description>
1930 If ORC reader encounters corrupt data, this value will be used to determine
1931 whether to skip the corrupt data or throw exception. The default behavior is to throw exception.
1932 </description>
1933 </property>
1934 <property>
1935 <name>hive.exec.orc.zerocopy</name>
1936 <value>false</value>
1937 <description>Use zerocopy reads with ORC. (This requires Hadoop 2.3 or later.)</description>
1938 </property>
1939 <property>
1940 <name>hive.lazysimple.extended_boolean_literal</name>
1941 <value>false</value>
1942 <description>
1943 LazySimpleSerde uses this property to determine if it treats 'T', 't', 'F', 'f',
1944 '1', and '0' as extened, legal boolean literal, in addition to 'TRUE' and 'FALSE'.
1945 The default is false, which means only 'TRUE' and 'FALSE' are treated as legal
1946 boolean literal.
1947 </description>
1948 </property>
1949 <property>
1950 <name>hive.optimize.skewjoin</name>
1951 <value>false</value>
1952 <description>
1953 Whether to enable skew join optimization.
1954 The algorithm is as follows: At runtime, detect the keys with a large skew. Instead of
1955 processing those keys, store them temporarily in an HDFS directory. In a follow-up map-reduce
1956 job, process those skewed keys. The same key need not be skewed for all the tables, and so,
1957 the follow-up map-reduce job (for the skewed keys) would be much faster, since it would be a
1958 map-join.
1959 </description>
1960 </property>
1961 <property>
1962 <name>hive.optimize.dynamic.partition.hashjoin</name>
1963 <value>false</value>
1964 <description>
1965 Whether to enable dynamically partitioned hash join optimization.
1966 This setting is also dependent on enabling hive.auto.convert.join
1967 </description>
1968 </property>
1969 <property>
1970 <name>hive.auto.convert.join</name>
1971 <value>true</value>
1972 <description>Whether Hive enables the optimization about converting common join into mapjoin based on the input file size</description>
1973 </property>
1974 <property>
1975 <name>hive.auto.convert.join.noconditionaltask</name>
1976 <value>true</value>
1977 <description>
1978 Whether Hive enables the optimization about converting common join into mapjoin based on the input file size.
1979 If this parameter is on, and the sum of size for n-1 of the tables/partitions for a n-way join is smaller than the
1980 specified size, the join is directly converted to a mapjoin (there is no conditional task).
1981 </description>
1982 </property>
1983 <property>
1984 <name>hive.auto.convert.join.noconditionaltask.size</name>
1985 <value>10000000</value>
1986 <description>
1987 If hive.auto.convert.join.noconditionaltask is off, this parameter does not take affect.
1988 However, if it is on, and the sum of size for n-1 of the tables/partitions for a n-way join is smaller than this size,
1989 the join is directly converted to a mapjoin(there is no conditional task). The default is 10MB
1990 </description>
1991 </property>
1992 <property>
1993 <name>hive.auto.convert.join.use.nonstaged</name>
1994 <value>false</value>
1995 <description>
1996 For conditional joins, if input stream from a small alias can be directly applied to join operator without
1997 filtering or projection, the alias need not to be pre-staged in distributed cache via mapred local task.
1998 Currently, this is not working with vectorization or tez execution engine.
1999 </description>
2000 </property>
2001 <property>
2002 <name>hive.skewjoin.key</name>
2003 <value>100000</value>
2004 <description>
2005 Determine if we get a skew key in join. If we see more than the specified number of rows with the same key in join operator,
2006 we think the key as a skew join key.
2007 </description>
2008 </property>
2009 <property>
2010 <name>hive.skewjoin.mapjoin.map.tasks</name>
2011 <value>10000</value>
2012 <description>
2013 Determine the number of map task used in the follow up map join job for a skew join.
2014 It should be used together with hive.skewjoin.mapjoin.min.split to perform a fine grained control.
2015 </description>
2016 </property>
2017 <property>
2018 <name>hive.skewjoin.mapjoin.min.split</name>
2019 <value>33554432</value>
2020 <description>
2021 Determine the number of map task at most used in the follow up map join job for a skew join by specifying
2022 the minimum split size. It should be used together with hive.skewjoin.mapjoin.map.tasks to perform a fine grained control.
2023 </description>
2024 </property>
2025 <property>
2026 <name>hive.heartbeat.interval</name>
2027 <value>1000</value>
2028 <description>Send a heartbeat after this interval - used by mapjoin and filter operators</description>
2029 </property>
2030 <property>
2031 <name>hive.limit.row.max.size</name>
2032 <value>100000</value>
2033 <description>When trying a smaller subset of data for simple LIMIT, how much size we need to guarantee each row to have at least.</description>
2034 </property>
2035 <property>
2036 <name>hive.limit.optimize.limit.file</name>
2037 <value>10</value>
2038 <description>When trying a smaller subset of data for simple LIMIT, maximum number of files we can sample.</description>
2039 </property>
2040 <property>
2041 <name>hive.limit.optimize.enable</name>
2042 <value>false</value>
2043 <description>Whether to enable to optimization to trying a smaller subset of data for simple LIMIT first.</description>
2044 </property>
2045 <property>
2046 <name>hive.limit.optimize.fetch.max</name>
2047 <value>50000</value>
2048 <description>
2049 Maximum number of rows allowed for a smaller subset of data for simple LIMIT, if it is a fetch query.
2050 Insert queries are not restricted by this limit.
2051 </description>
2052 </property>
2053 <property>
2054 <name>hive.limit.pushdown.memory.usage</name>
2055 <value>0.1</value>
2056 <description>
2057 Expects value between 0.0f and 1.0f.
2058 The fraction of available memory to be used for buffering rows in Reducesink operator for limit pushdown optimization.
2059 </description>
2060 </property>
2061 <property>
2062 <name>hive.limit.query.max.table.partition</name>
2063 <value>-1</value>
2064 <description>
2065 This controls how many partitions can be scanned for each partitioned table.
2066 The default value "-1" means no limit.
2067 </description>
2068 </property>
2069 <property>
2070 <name>hive.hashtable.key.count.adjustment</name>
2071 <value>1.0</value>
2072 <description>Adjustment to mapjoin hashtable size derived from table and column statistics; the estimate of the number of keys is divided by this value. If the value is 0, statistics are not usedand hive.hashtable.initialCapacity is used instead.</description>
2073 </property>
2074 <property>
2075 <name>hive.hashtable.initialCapacity</name>
2076 <value>100000</value>
2077 <description>Initial capacity of mapjoin hashtable if statistics are absent, or if hive.hashtable.key.count.adjustment is set to 0</description>
2078 </property>
2079 <property>
2080 <name>hive.hashtable.loadfactor</name>
2081 <value>0.75</value>
2082 <description/>
2083 </property>
2084 <property>
2085 <name>hive.mapjoin.followby.gby.localtask.max.memory.usage</name>
2086 <value>0.55</value>
2087 <description>
2088 This number means how much memory the local task can take to hold the key/value into an in-memory hash table
2089 when this map join is followed by a group by. If the local task's memory usage is more than this number,
2090 the local task will abort by itself. It means the data of the small table is too large to be held in memory.
2091 </description>
2092 </property>
2093 <property>
2094 <name>hive.mapjoin.localtask.max.memory.usage</name>
2095 <value>0.9</value>
2096 <description>
2097 This number means how much memory the local task can take to hold the key/value into an in-memory hash table.
2098 If the local task's memory usage is more than this number, the local task will abort by itself.
2099 It means the data of the small table is too large to be held in memory.
2100 </description>
2101 </property>
2102 <property>
2103 <name>hive.mapjoin.check.memory.rows</name>
2104 <value>100000</value>
2105 <description>The number means after how many rows processed it needs to check the memory usage</description>
2106 </property>
2107 <property>
2108 <name>hive.debug.localtask</name>
2109 <value>false</value>
2110 <description/>
2111 </property>
2112 <property>
2113 <name>hive.input.format</name>
2114 <value>org.apache.hadoop.hive.ql.io.CombineHiveInputFormat</value>
2115 <description>The default input format. Set this to HiveInputFormat if you encounter problems with CombineHiveInputFormat.</description>
2116 </property>
2117 <property>
2118 <name>hive.tez.input.format</name>
2119 <value>org.apache.hadoop.hive.ql.io.HiveInputFormat</value>
2120 <description>The default input format for tez. Tez groups splits in the AM.</description>
2121 </property>
2122 <property>
2123 <name>hive.tez.container.size</name>
2124 <value>-1</value>
2125 <description>By default Tez will spawn containers of the size of a mapper. This can be used to overwrite.</description>
2126 </property>
2127 <property>
2128 <name>hive.tez.cpu.vcores</name>
2129 <value>-1</value>
2130 <description>
2131 By default Tez will ask for however many cpus map-reduce is configured to use per container.
2132 This can be used to overwrite.
2133 </description>
2134 </property>
2135 <property>
2136 <name>hive.tez.java.opts</name>
2137 <value/>
2138 <description>By default Tez will use the Java options from map tasks. This can be used to overwrite.</description>
2139 </property>
2140 <property>
2141 <name>hive.tez.log.level</name>
2142 <value>INFO</value>
2143 <description>
2144 The log level to use for tasks executing as part of the DAG.
2145 Used only if hive.tez.java.opts is used to configure Java options.
2146 </description>
2147 </property>
2148 <property>
2149 <name>hive.query.name</name>
2150 <value/>
2151 <description>
2152 This named is used by Tez to set the dag name. This name in turn will appear on
2153 the Tez UI representing the work that was done.
2154 </description>
2155 </property>
2156 <property>
2157 <name>hive.optimize.bucketingsorting</name>
2158 <value>true</value>
2159 <description>
2160 Don't create a reducer for enforcing
2161 bucketing/sorting for queries of the form:
2162 insert overwrite table T2 select * from T1;
2163 where T1 and T2 are bucketed/sorted by the same keys into the same number of buckets.
2164 </description>
2165 </property>
2166 <property>
2167 <name>hive.mapred.partitioner</name>
2168 <value>org.apache.hadoop.hive.ql.io.DefaultHivePartitioner</value>
2169 <description/>
2170 </property>
2171 <property>
2172 <name>hive.enforce.sortmergebucketmapjoin</name>
2173 <value>false</value>
2174 <description>If the user asked for sort-merge bucketed map-side join, and it cannot be performed, should the query fail or not ?</description>
2175 </property>
2176 <property>
2177 <name>hive.enforce.bucketmapjoin</name>
2178 <value>false</value>
2179 <description>
2180 If the user asked for bucketed map-side join, and it cannot be performed,
2181 should the query fail or not ? For example, if the buckets in the tables being joined are
2182 not a multiple of each other, bucketed map-side join cannot be performed, and the
2183 query will fail if hive.enforce.bucketmapjoin is set to true.
2184 </description>
2185 </property>
2186 <property>
2187 <name>hive.auto.convert.sortmerge.join</name>
2188 <value>false</value>
2189 <description>Will the join be automatically converted to a sort-merge join, if the joined tables pass the criteria for sort-merge join.</description>
2190 </property>
2191 <property>
2192 <name>hive.auto.convert.sortmerge.join.bigtable.selection.policy</name>
2193 <value>org.apache.hadoop.hive.ql.optimizer.AvgPartitionSizeBasedBigTableSelectorForAutoSMJ</value>
2194 <description>
2195 The policy to choose the big table for automatic conversion to sort-merge join.
2196 By default, the table with the largest partitions is assigned the big table. All policies are:
2197 . based on position of the table - the leftmost table is selected
2198 org.apache.hadoop.hive.ql.optimizer.LeftmostBigTableSMJ.
2199 . based on total size (all the partitions selected in the query) of the table
2200 org.apache.hadoop.hive.ql.optimizer.TableSizeBasedBigTableSelectorForAutoSMJ.
2201 . based on average size (all the partitions selected in the query) of the table
2202 org.apache.hadoop.hive.ql.optimizer.AvgPartitionSizeBasedBigTableSelectorForAutoSMJ.
2203 New policies can be added in future.
2204 </description>
2205 </property>
2206 <property>
2207 <name>hive.auto.convert.sortmerge.join.to.mapjoin</name>
2208 <value>false</value>
2209 <description>
2210 If hive.auto.convert.sortmerge.join is set to true, and a join was converted to a sort-merge join,
2211 this parameter decides whether each table should be tried as a big table, and effectively a map-join should be
2212 tried. That would create a conditional task with n+1 children for a n-way join (1 child for each table as the
2213 big table), and the backup task will be the sort-merge join. In some cases, a map-join would be faster than a
2214 sort-merge join, if there is no advantage of having the output bucketed and sorted. For example, if a very big sorted
2215 and bucketed table with few files (say 10 files) are being joined with a very small sorter and bucketed table
2216 with few files (10 files), the sort-merge join will only use 10 mappers, and a simple map-only join might be faster
2217 if the complete small table can fit in memory, and a map-join can be performed.
2218 </description>
2219 </property>
2220 <property>
2221 <name>hive.exec.script.trust</name>
2222 <value>false</value>
2223 <description/>
2224 </property>
2225 <property>
2226 <name>hive.exec.rowoffset</name>
2227 <value>false</value>
2228 <description>Whether to provide the row offset virtual column</description>
2229 </property>
2230 <property>
2231 <name>hive.optimize.index.filter</name>
2232 <value>false</value>
2233 <description>Whether to enable automatic use of indexes</description>
2234 </property>
2235 <property>
2236 <name>hive.optimize.index.autoupdate</name>
2237 <value>false</value>
2238 <description>Whether to update stale indexes automatically</description>
2239 </property>
2240 <property>
2241 <name>hive.optimize.ppd</name>
2242 <value>true</value>
2243 <description>Whether to enable predicate pushdown</description>
2244 </property>
2245 <property>
2246 <name>hive.optimize.ppd.windowing</name>
2247 <value>true</value>
2248 <description>Whether to enable predicate pushdown through windowing</description>
2249 </property>
2250 <property>
2251 <name>hive.ppd.recognizetransivity</name>
2252 <value>true</value>
2253 <description>Whether to transitively replicate predicate filters over equijoin conditions.</description>
2254 </property>
2255 <property>
2256 <name>hive.ppd.remove.duplicatefilters</name>
2257 <value>true</value>
2258 <description>
2259 During query optimization, filters may be pushed down in the operator tree.
2260 If this config is true only pushed down filters remain in the operator tree,
2261 and the original filter is removed. If this config is false, the original filter
2262 is also left in the operator tree at the original place.
2263 </description>
2264 </property>
2265 <property>
2266 <name>hive.optimize.point.lookup</name>
2267 <value>true</value>
2268 <description>Whether to transform OR clauses in Filter operators into IN clauses</description>
2269 </property>
2270 <property>
2271 <name>hive.optimize.point.lookup.min</name>
2272 <value>31</value>
2273 <description>Minimum number of OR clauses needed to transform into IN clauses</description>
2274 </property>
2275 <property>
2276 <name>hive.optimize.partition.columns.separate</name>
2277 <value>true</value>
2278 <description>Extract partition columns from IN clauses</description>
2279 </property>
2280 <property>
2281 <name>hive.optimize.constant.propagation</name>
2282 <value>true</value>
2283 <description>Whether to enable constant propagation optimizer</description>
2284 </property>
2285 <property>
2286 <name>hive.optimize.remove.identity.project</name>
2287 <value>true</value>
2288 <description>Removes identity project from operator tree</description>
2289 </property>
2290 <property>
2291 <name>hive.optimize.metadataonly</name>
2292 <value>true</value>
2293 <description/>
2294 </property>
2295 <property>
2296 <name>hive.optimize.null.scan</name>
2297 <value>true</value>
2298 <description>Dont scan relations which are guaranteed to not generate any rows</description>
2299 </property>
2300 <property>
2301 <name>hive.optimize.ppd.storage</name>
2302 <value>true</value>
2303 <description>Whether to push predicates down to storage handlers</description>
2304 </property>
2305 <property>
2306 <name>hive.optimize.groupby</name>
2307 <value>true</value>
2308 <description>Whether to enable the bucketed group by from bucketed partitions/tables.</description>
2309 </property>
2310 <property>
2311 <name>hive.optimize.bucketmapjoin</name>
2312 <value>false</value>
2313 <description>Whether to try bucket mapjoin</description>
2314 </property>
2315 <property>
2316 <name>hive.optimize.bucketmapjoin.sortedmerge</name>
2317 <value>false</value>
2318 <description>Whether to try sorted bucket merge map join</description>
2319 </property>
2320 <property>
2321 <name>hive.optimize.reducededuplication</name>
2322 <value>true</value>
2323 <description>
2324 Remove extra map-reduce jobs if the data is already clustered by the same key which needs to be used again.
2325 This should always be set to true. Since it is a new feature, it has been made configurable.
2326 </description>
2327 </property>
2328 <property>
2329 <name>hive.optimize.reducededuplication.min.reducer</name>
2330 <value>4</value>
2331 <description>
2332 Reduce deduplication merges two RSs by moving key/parts/reducer-num of the child RS to parent RS.
2333 That means if reducer-num of the child RS is fixed (order by or forced bucketing) and small, it can make very slow, single MR.
2334 The optimization will be automatically disabled if number of reducers would be less than specified value.
2335 </description>
2336 </property>
2337 <property>
2338 <name>hive.optimize.sort.dynamic.partition</name>
2339 <value>false</value>
2340 <description>
2341 When enabled dynamic partitioning column will be globally sorted.
2342 This way we can keep only one record writer open for each partition value
2343 in the reducer thereby reducing the memory pressure on reducers.
2344 </description>
2345 </property>
2346 <property>
2347 <name>hive.optimize.sampling.orderby</name>
2348 <value>false</value>
2349 <description>Uses sampling on order-by clause for parallel execution.</description>
2350 </property>
2351 <property>
2352 <name>hive.optimize.sampling.orderby.number</name>
2353 <value>1000</value>
2354 <description>Total number of samples to be obtained.</description>
2355 </property>
2356 <property>
2357 <name>hive.optimize.sampling.orderby.percent</name>
2358 <value>0.1</value>
2359 <description>
2360 Expects value between 0.0f and 1.0f.
2361 Probability with which a row will be chosen.
2362 </description>
2363 </property>
2364 <property>
2365 <name>hive.optimize.distinct.rewrite</name>
2366 <value>true</value>
2367 <description>When applicable this optimization rewrites distinct aggregates from a single stage to multi-stage aggregation. This may not be optimal in all cases. Ideally, whether to trigger it or not should be cost based decision. Until Hive formalizes cost model for this, this is config driven.</description>
2368 </property>
2369 <property>
2370 <name>hive.optimize.union.remove</name>
2371 <value>false</value>
2372 <description>
2373 Whether to remove the union and push the operators between union and the filesink above union.
2374 This avoids an extra scan of the output by union. This is independently useful for union
2375 queries, and specially useful when hive.optimize.skewjoin.compiletime is set to true, since an
2376 extra union is inserted.
2377
2378 The merge is triggered if either of hive.merge.mapfiles or hive.merge.mapredfiles is set to true.
2379 If the user has set hive.merge.mapfiles to true and hive.merge.mapredfiles to false, the idea was the
2380 number of reducers are few, so the number of files anyway are small. However, with this optimization,
2381 we are increasing the number of files possibly by a big margin. So, we merge aggressively.
2382 </description>
2383 </property>
2384 <property>
2385 <name>hive.optimize.correlation</name>
2386 <value>false</value>
2387 <description>exploit intra-query correlations.</description>
2388 </property>
2389 <property>
2390 <name>hive.optimize.limittranspose</name>
2391 <value>false</value>
2392 <description>
2393 Whether to push a limit through left/right outer join or union. If the value is true and the size of the outer
2394 input is reduced enough (as specified in hive.optimize.limittranspose.reduction), the limit is pushed
2395 to the outer input or union; to remain semantically correct, the limit is kept on top of the join or the union too.
2396 </description>
2397 </property>
2398 <property>
2399 <name>hive.optimize.limittranspose.reductionpercentage</name>
2400 <value>1.0</value>
2401 <description>
2402 When hive.optimize.limittranspose is true, this variable specifies the minimal reduction of the
2403 size of the outer input of the join or input of the union that we should get in order to apply the rule.
2404 </description>
2405 </property>
2406 <property>
2407 <name>hive.optimize.limittranspose.reductiontuples</name>
2408 <value>0</value>
2409 <description>
2410 When hive.optimize.limittranspose is true, this variable specifies the minimal reduction in the
2411 number of tuples of the outer input of the join or the input of the union that you should get in order to apply the rule.
2412 </description>
2413 </property>
2414 <property>
2415 <name>hive.optimize.filter.stats.reduction</name>
2416 <value>false</value>
2417 <description>
2418 Whether to simplify comparison
2419 expressions in filter operators using column stats
2420 </description>
2421 </property>
2422 <property>
2423 <name>hive.optimize.skewjoin.compiletime</name>
2424 <value>false</value>
2425 <description>
2426 Whether to create a separate plan for skewed keys for the tables in the join.
2427 This is based on the skewed keys stored in the metadata. At compile time, the plan is broken
2428 into different joins: one for the skewed keys, and the other for the remaining keys. And then,
2429 a union is performed for the 2 joins generated above. So unless the same skewed key is present
2430 in both the joined tables, the join for the skewed key will be performed as a map-side join.
2431
2432 The main difference between this parameter and hive.optimize.skewjoin is that this parameter
2433 uses the skew information stored in the metastore to optimize the plan at compile time itself.
2434 If there is no skew information in the metadata, this parameter will not have any affect.
2435 Both hive.optimize.skewjoin.compiletime and hive.optimize.skewjoin should be set to true.
2436 Ideally, hive.optimize.skewjoin should be renamed as hive.optimize.skewjoin.runtime, but not doing
2437 so for backward compatibility.
2438
2439 If the skew information is correctly stored in the metadata, hive.optimize.skewjoin.compiletime
2440 would change the query plan to take care of it, and hive.optimize.skewjoin will be a no-op.
2441 </description>
2442 </property>
2443 <property>
2444 <name>hive.optimize.cte.materialize.threshold</name>
2445 <value>-1</value>
2446 <description>
2447 If the number of references to a CTE clause exceeds this threshold, Hive will materialize it
2448 before executing the main query block. -1 will disable this feature.
2449 </description>
2450 </property>
2451 <property>
2452 <name>hive.optimize.index.filter.compact.minsize</name>
2453 <value>5368709120</value>
2454 <description>Minimum size (in bytes) of the inputs on which a compact index is automatically used.</description>
2455 </property>
2456 <property>
2457 <name>hive.optimize.index.filter.compact.maxsize</name>
2458 <value>-1</value>
2459 <description>Maximum size (in bytes) of the inputs on which a compact index is automatically used. A negative number is equivalent to infinity.</description>
2460 </property>
2461 <property>
2462 <name>hive.index.compact.query.max.entries</name>
2463 <value>10000000</value>
2464 <description>The maximum number of index entries to read during a query that uses the compact index. Negative value is equivalent to infinity.</description>
2465 </property>
2466 <property>
2467 <name>hive.index.compact.query.max.size</name>
2468 <value>10737418240</value>
2469 <description>The maximum number of bytes that a query using the compact index can read. Negative value is equivalent to infinity.</description>
2470 </property>
2471 <property>
2472 <name>hive.index.compact.binary.search</name>
2473 <value>true</value>
2474 <description>Whether or not to use a binary search to find the entries in an index table that match the filter, where possible</description>
2475 </property>
2476 <property>
2477 <name>hive.stats.autogather</name>
2478 <value>true</value>
2479 <description>A flag to gather statistics (only basic) automatically during the INSERT OVERWRITE command.</description>
2480 </property>
2481 <property>
2482 <name>hive.stats.column.autogather</name>
2483 <value>false</value>
2484 <description>A flag to gather column statistics automatically.</description>
2485 </property>
2486 <property>
2487 <name>hive.stats.dbclass</name>
2488 <value>fs</value>
2489 <description>
2490 Expects one of the pattern in [custom, fs].
2491 The storage that stores temporary Hive statistics. In filesystem based statistics collection ('fs'),
2492 each task writes statistics it has collected in a file on the filesystem, which will be aggregated
2493 after the job has finished. Supported values are fs (filesystem) and custom as defined in StatsSetupConst.java.
2494 </description>
2495 </property>
2496 <property>
2497 <name>hive.stats.default.publisher</name>
2498 <value/>
2499 <description>The Java class (implementing the StatsPublisher interface) that is used by default if hive.stats.dbclass is custom type.</description>
2500 </property>
2501 <property>
2502 <name>hive.stats.default.aggregator</name>
2503 <value/>
2504 <description>The Java class (implementing the StatsAggregator interface) that is used by default if hive.stats.dbclass is custom type.</description>
2505 </property>
2506 <property>
2507 <name>hive.stats.atomic</name>
2508 <value>false</value>
2509 <description>whether to update metastore stats only if all stats are available</description>
2510 </property>
2511 <property>
2512 <name>hive.client.stats.counters</name>
2513 <value/>
2514 <description>
2515 Subset of counters that should be of interest for hive.client.stats.publishers (when one wants to limit their publishing).
2516 Non-display names should be used
2517 </description>
2518 </property>
2519 <property>
2520 <name>hive.stats.reliable</name>
2521 <value>false</value>
2522 <description>
2523 Whether queries will fail because stats cannot be collected completely accurately.
2524 If this is set to true, reading/writing from/into a partition may fail because the stats
2525 could not be computed accurately.
2526 </description>
2527 </property>
2528 <property>
2529 <name>hive.analyze.stmt.collect.partlevel.stats</name>
2530 <value>true</value>
2531 <description>analyze table T compute statistics for columns. Queries like these should compute partitionlevel stats for partitioned table even when no part spec is specified.</description>
2532 </property>
2533 <property>
2534 <name>hive.stats.gather.num.threads</name>
2535 <value>10</value>
2536 <description>
2537 Number of threads used by partialscan/noscan analyze command for partitioned tables.
2538 This is applicable only for file formats that implement StatsProvidingRecordReader (like ORC).
2539 </description>
2540 </property>
2541 <property>
2542 <name>hive.stats.collect.tablekeys</name>
2543 <value>false</value>
2544 <description>
2545 Whether join and group by keys on tables are derived and maintained in the QueryPlan.
2546 This is useful to identify how tables are accessed and to determine if they should be bucketed.
2547 </description>
2548 </property>
2549 <property>
2550 <name>hive.stats.collect.scancols</name>
2551 <value>false</value>
2552 <description>
2553 Whether column accesses are tracked in the QueryPlan.
2554 This is useful to identify how tables are accessed and to determine if there are wasted columns that can be trimmed.
2555 </description>
2556 </property>
2557 <property>
2558 <name>hive.stats.ndv.error</name>
2559 <value>20.0</value>
2560 <description>
2561 Standard error expressed in percentage. Provides a tradeoff between accuracy and compute cost.
2562 A lower value for error indicates higher accuracy and a higher compute cost.
2563 </description>
2564 </property>
2565 <property>
2566 <name>hive.metastore.stats.ndv.densityfunction</name>
2567 <value>false</value>
2568 <description>Whether to use density function to estimate the NDV for the whole table based on the NDV of partitions</description>
2569 </property>
2570 <property>
2571 <name>hive.stats.max.variable.length</name>
2572 <value>100</value>
2573 <description>
2574 To estimate the size of data flowing through operators in Hive/Tez(for reducer estimation etc.),
2575 average row size is multiplied with the total number of rows coming out of each operator.
2576 Average row size is computed from average column size of all columns in the row. In the absence
2577 of column statistics, for variable length columns (like string, bytes etc.), this value will be
2578 used. For fixed length columns their corresponding Java equivalent sizes are used
2579 (float - 4 bytes, double - 8 bytes etc.).
2580 </description>
2581 </property>
2582 <property>
2583 <name>hive.stats.list.num.entries</name>
2584 <value>10</value>
2585 <description>
2586 To estimate the size of data flowing through operators in Hive/Tez(for reducer estimation etc.),
2587 average row size is multiplied with the total number of rows coming out of each operator.
2588 Average row size is computed from average column size of all columns in the row. In the absence
2589 of column statistics and for variable length complex columns like list, the average number of
2590 entries/values can be specified using this config.
2591 </description>
2592 </property>
2593 <property>
2594 <name>hive.stats.map.num.entries</name>
2595 <value>10</value>
2596 <description>
2597 To estimate the size of data flowing through operators in Hive/Tez(for reducer estimation etc.),
2598 average row size is multiplied with the total number of rows coming out of each operator.
2599 Average row size is computed from average column size of all columns in the row. In the absence
2600 of column statistics and for variable length complex columns like map, the average number of
2601 entries/values can be specified using this config.
2602 </description>
2603 </property>
2604 <property>
2605 <name>hive.stats.fetch.partition.stats</name>
2606 <value>true</value>
2607 <description>
2608 Annotation of operator tree with statistics information requires partition level basic
2609 statistics like number of rows, data size and file size. Partition statistics are fetched from
2610 metastore. Fetching partition statistics for each needed partition can be expensive when the
2611 number of partitions is high. This flag can be used to disable fetching of partition statistics
2612 from metastore. When this flag is disabled, Hive will make calls to filesystem to get file sizes
2613 and will estimate the number of rows from row schema.
2614 </description>
2615 </property>
2616 <property>
2617 <name>hive.stats.fetch.column.stats</name>
2618 <value>false</value>
2619 <description>
2620 Annotation of operator tree with statistics information requires column statistics.
2621 Column statistics are fetched from metastore. Fetching column statistics for each needed column
2622 can be expensive when the number of columns is high. This flag can be used to disable fetching
2623 of column statistics from metastore.
2624 </description>
2625 </property>
2626 <property>
2627 <name>hive.stats.join.factor</name>
2628 <value>1.1</value>
2629 <description>
2630 Hive/Tez optimizer estimates the data size flowing through each of the operators. JOIN operator
2631 uses column statistics to estimate the number of rows flowing out of it and hence the data size.
2632 In the absence of column statistics, this factor determines the amount of rows that flows out
2633 of JOIN operator.
2634 </description>
2635 </property>
2636 <property>
2637 <name>hive.stats.deserialization.factor</name>
2638 <value>1.0</value>
2639 <description>
2640 Hive/Tez optimizer estimates the data size flowing through each of the operators. In the absence
2641 of basic statistics like number of rows and data size, file size is used to estimate the number
2642 of rows and data size. Since files in tables/partitions are serialized (and optionally
2643 compressed) the estimates of number of rows and data size cannot be reliably determined.
2644 This factor is multiplied with the file size to account for serialization and compression.
2645 </description>
2646 </property>
2647 <property>
2648 <name>hive.support.concurrency</name>
2649 <value>false</value>
2650 <description>
2651 Whether Hive supports concurrency control or not.
2652 A ZooKeeper instance must be up and running when using zookeeper Hive lock manager
2653 </description>
2654 </property>
2655 <property>
2656 <name>hive.lock.manager</name>
2657 <value>org.apache.hadoop.hive.ql.lockmgr.zookeeper.ZooKeeperHiveLockManager</value>
2658 <description/>
2659 </property>
2660 <property>
2661 <name>hive.lock.numretries</name>
2662 <value>100</value>
2663 <description>The number of times you want to try to get all the locks</description>
2664 </property>
2665 <property>
2666 <name>hive.unlock.numretries</name>
2667 <value>10</value>
2668 <description>The number of times you want to retry to do one unlock</description>
2669 </property>
2670 <property>
2671 <name>hive.lock.sleep.between.retries</name>
2672 <value>60s</value>
2673 <description>
2674 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
2675 The time should be in between 0 sec (exclusive) and 9223372036854775807 sec (exclusive).
2676 The maximum sleep time between various retries
2677 </description>
2678 </property>
2679 <property>
2680 <name>hive.lock.mapred.only.operation</name>
2681 <value>false</value>
2682 <description>
2683 This param is to control whether or not only do lock on queries
2684 that need to execute at least one mapred job.
2685 </description>
2686 </property>
2687 <property>
2688 <name>hive.zookeeper.quorum</name>
2689 <value/>
2690 <description>
2691 List of ZooKeeper servers to talk to. This is needed for:
2692 1. Read/write locks - when hive.lock.manager is set to
2693 org.apache.hadoop.hive.ql.lockmgr.zookeeper.ZooKeeperHiveLockManager,
2694 2. When HiveServer2 supports service discovery via Zookeeper.
2695 3. For delegation token storage if zookeeper store is used, if
2696 hive.cluster.delegation.token.store.zookeeper.connectString is not set
2697 4. LLAP daemon registry service
2698 </description>
2699 </property>
2700 <property>
2701 <name>hive.zookeeper.client.port</name>
2702 <value>2181</value>
2703 <description>
2704 The port of ZooKeeper servers to talk to.
2705 If the list of Zookeeper servers specified in hive.zookeeper.quorum
2706 does not contain port numbers, this value is used.
2707 </description>
2708 </property>
2709 <property>
2710 <name>hive.zookeeper.session.timeout</name>
2711 <value>1200000ms</value>
2712 <description>
2713 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
2714 ZooKeeper client's session timeout (in milliseconds). The client is disconnected, and as a result, all locks released,
2715 if a heartbeat is not sent in the timeout.
2716 </description>
2717 </property>
2718 <property>
2719 <name>hive.zookeeper.namespace</name>
2720 <value>hive_zookeeper_namespace</value>
2721 <description>The parent node under which all ZooKeeper nodes are created.</description>
2722 </property>
2723 <property>
2724 <name>hive.zookeeper.clean.extra.nodes</name>
2725 <value>false</value>
2726 <description>Clean extra nodes at the end of the session.</description>
2727 </property>
2728 <property>
2729 <name>hive.zookeeper.connection.max.retries</name>
2730 <value>3</value>
2731 <description>Max number of times to retry when connecting to the ZooKeeper server.</description>
2732 </property>
2733 <property>
2734 <name>hive.zookeeper.connection.basesleeptime</name>
2735 <value>1000ms</value>
2736 <description>
2737 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
2738 Initial amount of time (in milliseconds) to wait between retries
2739 when connecting to the ZooKeeper server when using ExponentialBackoffRetry policy.
2740 </description>
2741 </property>
2742 <property>
2743 <name>hive.txn.manager</name>
2744 <value>org.apache.hadoop.hive.ql.lockmgr.DummyTxnManager</value>
2745 <description>
2746 Set to org.apache.hadoop.hive.ql.lockmgr.DbTxnManager as part of turning on Hive
2747 transactions, which also requires appropriate settings for hive.compactor.initiator.on,
2748 hive.compactor.worker.threads, hive.support.concurrency (true), hive.enforce.bucketing
2749 (true), and hive.exec.dynamic.partition.mode (nonstrict).
2750 The default DummyTxnManager replicates pre-Hive-0.13 behavior and provides
2751 no transactions.
2752 </description>
2753 </property>
2754 <property>
2755 <name>hive.txn.timeout</name>
2756 <value>300s</value>
2757 <description>
2758 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
2759 time after which transactions are declared aborted if the client has not sent a heartbeat.
2760 </description>
2761 </property>
2762 <property>
2763 <name>hive.txn.heartbeat.threadpool.size</name>
2764 <value>5</value>
2765 <description>The number of threads to use for heartbeating. For Hive CLI, 1 is enough. For HiveServer2, we need a few</description>
2766 </property>
2767 <property>
2768 <name>hive.txn.manager.dump.lock.state.on.acquire.timeout</name>
2769 <value>false</value>
2770 <description>Set this to true so that when attempt to acquire a lock on resource times out, the current state of the lock manager is dumped to log file. This is for debugging. See also hive.lock.numretries and hive.lock.sleep.between.retries.</description>
2771 </property>
2772 <property>
2773 <name>hive.max.open.txns</name>
2774 <value>100000</value>
2775 <description>
2776 Maximum number of open transactions. If
2777 current open transactions reach this limit, future open transaction requests will be
2778 rejected, until this number goes below the limit.
2779 </description>
2780 </property>
2781 <property>
2782 <name>hive.count.open.txns.interval</name>
2783 <value>1s</value>
2784 <description>
2785 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
2786 Time in seconds between checks to count open transactions.
2787 </description>
2788 </property>
2789 <property>
2790 <name>hive.txn.max.open.batch</name>
2791 <value>1000</value>
2792 <description>
2793 Maximum number of transactions that can be fetched in one call to open_txns().
2794 This controls how many transactions streaming agents such as Flume or Storm open
2795 simultaneously. The streaming agent then writes that number of entries into a single
2796 file (per Flume agent or Storm bolt). Thus increasing this value decreases the number
2797 of delta files created by streaming agents. But it also increases the number of open
2798 transactions that Hive has to track at any given time, which may negatively affect
2799 read performance.
2800 </description>
2801 </property>
2802 <property>
2803 <name>hive.txn.retryable.sqlex.regex</name>
2804 <value/>
2805 <description>
2806 Comma separated list
2807 of regular expression patterns for SQL state, error code, and error message of
2808 retryable SQLExceptions, that's suitable for the metastore DB.
2809 For example: Can't serialize.*,40001$,^Deadlock,.*ORA-08176.*
2810 The string that the regex will be matched against is of the following form, where ex is a SQLException:
2811 ex.getMessage() + " (SQLState=" + ex.getSQLState() + ", ErrorCode=" + ex.getErrorCode() + ")"
2812 </description>
2813 </property>
2814 <property>
2815 <name>hive.compactor.initiator.on</name>
2816 <value>false</value>
2817 <description>
2818 Whether to run the initiator and cleaner threads on this metastore instance or not.
2819 Set this to true on one instance of the Thrift metastore service as part of turning
2820 on Hive transactions. For a complete list of parameters required for turning on
2821 transactions, see hive.txn.manager.
2822 </description>
2823 </property>
2824 <property>
2825 <name>hive.compactor.worker.threads</name>
2826 <value>0</value>
2827 <description>
2828 How many compactor worker threads to run on this metastore instance. Set this to a
2829 positive number on one or more instances of the Thrift metastore service as part of
2830 turning on Hive transactions. For a complete list of parameters required for turning
2831 on transactions, see hive.txn.manager.
2832 Worker threads spawn MapReduce jobs to do compactions. They do not do the compactions
2833 themselves. Increasing the number of worker threads will decrease the time it takes
2834 tables or partitions to be compacted once they are determined to need compaction.
2835 It will also increase the background load on the Hadoop cluster as more MapReduce jobs
2836 will be running in the background.
2837 </description>
2838 </property>
2839 <property>
2840 <name>hive.compactor.worker.timeout</name>
2841 <value>86400s</value>
2842 <description>
2843 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
2844 Time in seconds after which a compaction job will be declared failed and the
2845 compaction re-queued.
2846 </description>
2847 </property>
2848 <property>
2849 <name>hive.compactor.check.interval</name>
2850 <value>300s</value>
2851 <description>
2852 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
2853 Time in seconds between checks to see if any tables or partitions need to be
2854 compacted. This should be kept high because each check for compaction requires
2855 many calls against the NameNode.
2856 Decreasing this value will reduce the time it takes for compaction to be started
2857 for a table or partition that requires compaction. However, checking if compaction
2858 is needed requires several calls to the NameNode for each table or partition that
2859 has had a transaction done on it since the last major compaction. So decreasing this
2860 value will increase the load on the NameNode.
2861 </description>
2862 </property>
2863 <property>
2864 <name>hive.compactor.delta.num.threshold</name>
2865 <value>10</value>
2866 <description>
2867 Number of delta directories in a table or partition that will trigger a minor
2868 compaction.
2869 </description>
2870 </property>
2871 <property>
2872 <name>hive.compactor.delta.pct.threshold</name>
2873 <value>0.1</value>
2874 <description>
2875 Percentage (fractional) size of the delta files relative to the base that will trigger
2876 a major compaction. (1.0 = 100%, so the default 0.1 = 10%.)
2877 </description>
2878 </property>
2879 <property>
2880 <name>hive.compactor.max.num.delta</name>
2881 <value>500</value>
2882 <description>Maximum number of delta files that the compactor will attempt to handle in a single job.</description>
2883 </property>
2884 <property>
2885 <name>hive.compactor.abortedtxn.threshold</name>
2886 <value>1000</value>
2887 <description>
2888 Number of aborted transactions involving a given table or partition that will trigger
2889 a major compaction.
2890 </description>
2891 </property>
2892 <property>
2893 <name>hive.compactor.initiator.failed.compacts.threshold</name>
2894 <value>2</value>
2895 <description>
2896 Expects value between 1 and 20.
2897 Number of consecutive compaction failures (per table/partition) after which automatic compactions will not be scheduled any more. Note that this must be less than hive.compactor.history.retention.failed.
2898 </description>
2899 </property>
2900 <property>
2901 <name>hive.compactor.cleaner.run.interval</name>
2902 <value>5000ms</value>
2903 <description>
2904 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
2905 Time between runs of the cleaner thread
2906 </description>
2907 </property>
2908 <property>
2909 <name>hive.compactor.job.queue</name>
2910 <value/>
2911 <description>
2912 Used to specify name of Hadoop queue to which
2913 Compaction jobs will be submitted. Set to empty string to let Hadoop choose the queue.
2914 </description>
2915 </property>
2916 <property>
2917 <name>hive.compactor.history.retention.succeeded</name>
2918 <value>3</value>
2919 <description>
2920 Expects value between 0 and 100.
2921 Determines how many successful compaction records will be retained in compaction history for a given table/partition.
2922 </description>
2923 </property>
2924 <property>
2925 <name>hive.compactor.history.retention.failed</name>
2926 <value>3</value>
2927 <description>
2928 Expects value between 0 and 100.
2929 Determines how many failed compaction records will be retained in compaction history for a given table/partition.
2930 </description>
2931 </property>
2932 <property>
2933 <name>hive.compactor.history.retention.attempted</name>
2934 <value>2</value>
2935 <description>
2936 Expects value between 0 and 100.
2937 Determines how many attempted compaction records will be retained in compaction history for a given table/partition.
2938 </description>
2939 </property>
2940 <property>
2941 <name>hive.compactor.history.reaper.interval</name>
2942 <value>2m</value>
2943 <description>
2944 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
2945 Determines how often compaction history reaper runs
2946 </description>
2947 </property>
2948 <property>
2949 <name>hive.timedout.txn.reaper.start</name>
2950 <value>100s</value>
2951 <description>
2952 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
2953 Time delay of 1st reaper run after metastore start
2954 </description>
2955 </property>
2956 <property>
2957 <name>hive.timedout.txn.reaper.interval</name>
2958 <value>180s</value>
2959 <description>
2960 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
2961 Time interval describing how often the reaper runs
2962 </description>
2963 </property>
2964 <property>
2965 <name>hive.writeset.reaper.interval</name>
2966 <value>60s</value>
2967 <description>
2968 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
2969 Frequency of WriteSet reaper runs
2970 </description>
2971 </property>
2972 <property>
2973 <name>hive.hbase.wal.enabled</name>
2974 <value>true</value>
2975 <description>
2976 Whether writes to HBase should be forced to the write-ahead log.
2977 Disabling this improves HBase write performance at the risk of lost writes in case of a crash.
2978 </description>
2979 </property>
2980 <property>
2981 <name>hive.hbase.generatehfiles</name>
2982 <value>false</value>
2983 <description>True when HBaseStorageHandler should generate hfiles instead of operate against the online table.</description>
2984 </property>
2985 <property>
2986 <name>hive.hbase.snapshot.name</name>
2987 <value/>
2988 <description>The HBase table snapshot name to use.</description>
2989 </property>
2990 <property>
2991 <name>hive.hbase.snapshot.restoredir</name>
2992 <value>/tmp</value>
2993 <description>The directory in which to restore the HBase table snapshot.</description>
2994 </property>
2995 <property>
2996 <name>hive.archive.enabled</name>
2997 <value>false</value>
2998 <description>Whether archiving operations are permitted</description>
2999 </property>
3000 <property>
3001 <name>hive.optimize.index.groupby</name>
3002 <value>false</value>
3003 <description>Whether to enable optimization of group-by queries using Aggregate indexes.</description>
3004 </property>
3005 <property>
3006 <name>hive.outerjoin.supports.filters</name>
3007 <value>true</value>
3008 <description/>
3009 </property>
3010 <property>
3011 <name>hive.fetch.task.conversion</name>
3012 <value>more</value>
3013 <description>
3014 Expects one of [none, minimal, more].
3015 Some select queries can be converted to single FETCH task minimizing latency.
3016 Currently the query should be single sourced not having any subquery and should not have
3017 any aggregations or distincts (which incurs RS), lateral views and joins.
3018 0. none : disable hive.fetch.task.conversion
3019 1. minimal : SELECT STAR, FILTER on partition columns, LIMIT only
3020 2. more : SELECT, FILTER, LIMIT only (support TABLESAMPLE and virtual columns)
3021 </description>
3022 </property>
3023 <property>
3024 <name>hive.fetch.task.conversion.threshold</name>
3025 <value>1073741824</value>
3026 <description>
3027 Input threshold for applying hive.fetch.task.conversion. If target table is native, input length
3028 is calculated by summation of file lengths. If it's not native, storage handler for the table
3029 can optionally implement org.apache.hadoop.hive.ql.metadata.InputEstimator interface.
3030 </description>
3031 </property>
3032 <property>
3033 <name>hive.fetch.task.aggr</name>
3034 <value>false</value>
3035 <description>
3036 Aggregation queries with no group-by clause (for example, select count(*) from src) execute
3037 final aggregations in single reduce task. If this is set true, Hive delegates final aggregation
3038 stage to fetch task, possibly decreasing the query time.
3039 </description>
3040 </property>
3041 <property>
3042 <name>hive.compute.query.using.stats</name>
3043 <value>false</value>
3044 <description>
3045 When set to true Hive will answer a few queries like count(1) purely using stats
3046 stored in metastore. For basic stats collection turn on the config hive.stats.autogather to true.
3047 For more advanced stats collection need to run analyze table queries.
3048 </description>
3049 </property>
3050 <property>
3051 <name>hive.fetch.output.serde</name>
3052 <value>org.apache.hadoop.hive.serde2.DelimitedJSONSerDe</value>
3053 <description>The SerDe used by FetchTask to serialize the fetch output.</description>
3054 </property>
3055 <property>
3056 <name>hive.cache.expr.evaluation</name>
3057 <value>true</value>
3058 <description>
3059 If true, the evaluation result of a deterministic expression referenced twice or more
3060 will be cached.
3061 For example, in a filter condition like '.. where key + 10 = 100 or key + 10 = 0'
3062 the expression 'key + 10' will be evaluated/cached once and reused for the following
3063 expression ('key + 10 = 0'). Currently, this is applied only to expressions in select
3064 or filter operators.
3065 </description>
3066 </property>
3067 <property>
3068 <name>hive.variable.substitute</name>
3069 <value>true</value>
3070 <description>This enables substitution using syntax like ${var} ${system:var} and ${env:var}.</description>
3071 </property>
3072 <property>
3073 <name>hive.variable.substitute.depth</name>
3074 <value>40</value>
3075 <description>The maximum replacements the substitution engine will do.</description>
3076 </property>
3077 <property>
3078 <name>hive.conf.validation</name>
3079 <value>true</value>
3080 <description>Enables type checking for registered Hive configurations</description>
3081 </property>
3082 <property>
3083 <name>hive.semantic.analyzer.hook</name>
3084 <value/>
3085 <description/>
3086 </property>
3087 <property>
3088 <name>hive.security.authorization.enabled</name>
3089 <value>false</value>
3090 <description>enable or disable the Hive client authorization</description>
3091 </property>
3092 <property>
3093 <name>hive.security.authorization.manager</name>
3094 <value>org.apache.hadoop.hive.ql.security.authorization.plugin.sqlstd.SQLStdHiveAuthorizerFactory</value>
3095 <description>
3096 The Hive client authorization manager class name. The user defined authorization class should implement
3097 interface org.apache.hadoop.hive.ql.security.authorization.HiveAuthorizationProvider.
3098 </description>
3099 </property>
3100 <property>
3101 <name>hive.security.authenticator.manager</name>
3102 <value>org.apache.hadoop.hive.ql.security.HadoopDefaultAuthenticator</value>
3103 <description>
3104 hive client authenticator manager class name. The user defined authenticator should implement
3105 interface org.apache.hadoop.hive.ql.security.HiveAuthenticationProvider.
3106 </description>
3107 </property>
3108 <property>
3109 <name>hive.security.metastore.authorization.manager</name>
3110 <value>org.apache.hadoop.hive.ql.security.authorization.DefaultHiveMetastoreAuthorizationProvider</value>
3111 <description>
3112 Names of authorization manager classes (comma separated) to be used in the metastore
3113 for authorization. The user defined authorization class should implement interface
3114 org.apache.hadoop.hive.ql.security.authorization.HiveMetastoreAuthorizationProvider.
3115 All authorization manager classes have to successfully authorize the metastore API
3116 call for the command execution to be allowed.
3117 </description>
3118 </property>
3119 <property>
3120 <name>hive.security.metastore.authorization.auth.reads</name>
3121 <value>true</value>
3122 <description>If this is true, metastore authorizer authorizes read actions on database, table</description>
3123 </property>
3124 <property>
3125 <name>hive.security.metastore.authenticator.manager</name>
3126 <value>org.apache.hadoop.hive.ql.security.HadoopDefaultMetastoreAuthenticator</value>
3127 <description>
3128 authenticator manager class name to be used in the metastore for authentication.
3129 The user defined authenticator should implement interface org.apache.hadoop.hive.ql.security.HiveAuthenticationProvider.
3130 </description>
3131 </property>
3132 <property>
3133 <name>hive.security.authorization.createtable.user.grants</name>
3134 <value/>
3135 <description>
3136 the privileges automatically granted to some users whenever a table gets created.
3137 An example like "userX,userY:select;userZ:create" will grant select privilege to userX and userY,
3138 and grant create privilege to userZ whenever a new table created.
3139 </description>
3140 </property>
3141 <property>
3142 <name>hive.security.authorization.createtable.group.grants</name>
3143 <value/>
3144 <description>
3145 the privileges automatically granted to some groups whenever a table gets created.
3146 An example like "groupX,groupY:select;groupZ:create" will grant select privilege to groupX and groupY,
3147 and grant create privilege to groupZ whenever a new table created.
3148 </description>
3149 </property>
3150 <property>
3151 <name>hive.security.authorization.createtable.role.grants</name>
3152 <value/>
3153 <description>
3154 the privileges automatically granted to some roles whenever a table gets created.
3155 An example like "roleX,roleY:select;roleZ:create" will grant select privilege to roleX and roleY,
3156 and grant create privilege to roleZ whenever a new table created.
3157 </description>
3158 </property>
3159 <property>
3160 <name>hive.security.authorization.createtable.owner.grants</name>
3161 <value/>
3162 <description>
3163 The privileges automatically granted to the owner whenever a table gets created.
3164 An example like "select,drop" will grant select and drop privilege to the owner
3165 of the table. Note that the default gives the creator of a table no access to the
3166 table (but see HIVE-8067).
3167 </description>
3168 </property>
3169 <property>
3170 <name>hive.security.authorization.task.factory</name>
3171 <value>org.apache.hadoop.hive.ql.parse.authorization.HiveAuthorizationTaskFactoryImpl</value>
3172 <description>Authorization DDL task factory implementation</description>
3173 </property>
3174 <property>
3175 <name>hive.security.authorization.sqlstd.confwhitelist</name>
3176 <value/>
3177 <description>
3178 List of comma separated Java regexes. Configurations parameters that match these
3179 regexes can be modified by user when SQL standard authorization is enabled.
3180 To get the default value, use the 'set <param>' command.
3181 Note that the hive.conf.restricted.list checks are still enforced after the white list
3182 check
3183 </description>
3184 </property>
3185 <property>
3186 <name>hive.security.authorization.sqlstd.confwhitelist.append</name>
3187 <value/>
3188 <description>
3189 List of comma separated Java regexes, to be appended to list set in
3190 hive.security.authorization.sqlstd.confwhitelist. Using this list instead
3191 of updating the original list means that you can append to the defaults
3192 set by SQL standard authorization instead of replacing it entirely.
3193 </description>
3194 </property>
3195 <property>
3196 <name>hive.cli.print.header</name>
3197 <value>false</value>
3198 <description>Whether to print the names of the columns in query output.</description>
3199 </property>
3200 <property>
3201 <name>hive.cli.tez.session.async</name>
3202 <value>true</value>
3203 <description>
3204 Whether to start Tez
3205 session in background when running CLI with Tez, allowing CLI to be available earlier.
3206 </description>
3207 </property>
3208 <property>
3209 <name>hive.error.on.empty.partition</name>
3210 <value>false</value>
3211 <description>Whether to throw an exception if dynamic partition insert generates empty results.</description>
3212 </property>
3213 <property>
3214 <name>hive.index.compact.file</name>
3215 <value/>
3216 <description>internal variable</description>
3217 </property>
3218 <property>
3219 <name>hive.index.blockfilter.file</name>
3220 <value/>
3221 <description>internal variable</description>
3222 </property>
3223 <property>
3224 <name>hive.index.compact.file.ignore.hdfs</name>
3225 <value>false</value>
3226 <description>
3227 When true the HDFS location stored in the index file will be ignored at runtime.
3228 If the data got moved or the name of the cluster got changed, the index data should still be usable.
3229 </description>
3230 </property>
3231 <property>
3232 <name>hive.exim.uri.scheme.whitelist</name>
3233 <value>hdfs,pfile</value>
3234 <description>A comma separated list of acceptable URI schemes for import and export.</description>
3235 </property>
3236 <property>
3237 <name>hive.exim.strict.repl.tables</name>
3238 <value>true</value>
3239 <description>
3240 Parameter that determines if 'regular' (non-replication) export dumps can be
3241 imported on to tables that are the target of replication. If this parameter is
3242 set, regular imports will check if the destination table(if it exists) has a 'repl.last.id' set on it. If so, it will fail.
3243 </description>
3244 </property>
3245 <property>
3246 <name>hive.repl.task.factory</name>
3247 <value>org.apache.hive.hcatalog.api.repl.exim.EximReplicationTaskFactory</value>
3248 <description>
3249 Parameter that can be used to override which ReplicationTaskFactory will be
3250 used to instantiate ReplicationTask events. Override for third party repl plugins
3251 </description>
3252 </property>
3253 <property>
3254 <name>hive.mapper.cannot.span.multiple.partitions</name>
3255 <value>false</value>
3256 <description/>
3257 </property>
3258 <property>
3259 <name>hive.rework.mapredwork</name>
3260 <value>false</value>
3261 <description>
3262 should rework the mapred work or not.
3263 This is first introduced by SymlinkTextInputFormat to replace symlink files with real paths at compile time.
3264 </description>
3265 </property>
3266 <property>
3267 <name>hive.exec.concatenate.check.index</name>
3268 <value>true</value>
3269 <description>
3270 If this is set to true, Hive will throw error when doing
3271 'alter table tbl_name [partSpec] concatenate' on a table/partition
3272 that has indexes on it. The reason the user want to set this to true
3273 is because it can help user to avoid handling all index drop, recreation,
3274 rebuild work. This is very helpful for tables with thousands of partitions.
3275 </description>
3276 </property>
3277 <property>
3278 <name>hive.io.exception.handlers</name>
3279 <value/>
3280 <description>
3281 A list of io exception handler class names. This is used
3282 to construct a list exception handlers to handle exceptions thrown
3283 by record readers
3284 </description>
3285 </property>
3286 <property>
3287 <name>hive.log4j.file</name>
3288 <value/>
3289 <description>
3290 Hive log4j configuration file.
3291 If the property is not set, then logging will be initialized using hive-log4j2.properties found on the classpath.
3292 If the property is set, the value must be a valid URI (java.net.URI, e.g. "file:///tmp/my-logging.xml"),
3293 which you can then extract a URL from and pass to PropertyConfigurator.configure(URL).
3294 </description>
3295 </property>
3296 <property>
3297 <name>hive.exec.log4j.file</name>
3298 <value/>
3299 <description>
3300 Hive log4j configuration file for execution mode(sub command).
3301 If the property is not set, then logging will be initialized using hive-exec-log4j2.properties found on the classpath.
3302 If the property is set, the value must be a valid URI (java.net.URI, e.g. "file:///tmp/my-logging.xml"),
3303 which you can then extract a URL from and pass to PropertyConfigurator.configure(URL).
3304 </description>
3305 </property>
3306 <property>
3307 <name>hive.async.log.enabled</name>
3308 <value>true</value>
3309 <description>
3310 Whether to enable Log4j2's asynchronous logging. Asynchronous logging can give
3311 significant performance improvement as logging will be handled in separate thread
3312 that uses LMAX disruptor queue for buffering log messages.
3313 Refer https://logging.apache.org/log4j/2.x/manual/async.html for benefits and
3314 drawbacks.
3315 </description>
3316 </property>
3317 <property>
3318 <name>hive.log.explain.output</name>
3319 <value>false</value>
3320 <description>
3321 Whether to log explain output for every query.
3322 When enabled, will log EXPLAIN EXTENDED output for the query at INFO log4j log level.
3323 </description>
3324 </property>
3325 <property>
3326 <name>hive.explain.user</name>
3327 <value>true</value>
3328 <description>
3329 Whether to show explain result at user level.
3330 When enabled, will log EXPLAIN output for the query at user level.
3331 </description>
3332 </property>
3333 <property>
3334 <name>hive.autogen.columnalias.prefix.label</name>
3335 <value>_c</value>
3336 <description>
3337 String used as a prefix when auto generating column alias.
3338 By default the prefix label will be appended with a column position number to form the column alias.
3339 Auto generation would happen if an aggregate function is used in a select clause without an explicit alias.
3340 </description>
3341 </property>
3342 <property>
3343 <name>hive.autogen.columnalias.prefix.includefuncname</name>
3344 <value>false</value>
3345 <description>Whether to include function name in the column alias auto generated by Hive.</description>
3346 </property>
3347 <property>
3348 <name>hive.service.metrics.class</name>
3349 <value>org.apache.hadoop.hive.common.metrics.metrics2.CodahaleMetrics</value>
3350 <description>
3351 Expects one of [org.apache.hadoop.hive.common.metrics.metrics2.codahalemetrics, org.apache.hadoop.hive.common.metrics.legacymetrics].
3352 Hive metrics subsystem implementation class.
3353 </description>
3354 </property>
3355 <property>
3356 <name>hive.service.metrics.reporter</name>
3357 <value>JSON_FILE, JMX</value>
3358 <description>Reporter type for metric class org.apache.hadoop.hive.common.metrics.metrics2.CodahaleMetrics, comma separated list of JMX, CONSOLE, JSON_FILE, HADOOP2</description>
3359 </property>
3360 <property>
3361 <name>hive.service.metrics.file.location</name>
3362 <value>/tmp/report.json</value>
3363 <description>For metric class org.apache.hadoop.hive.common.metrics.metrics2.CodahaleMetrics JSON_FILE reporter, the location of local JSON metrics file. This file will get overwritten at every interval.</description>
3364 </property>
3365 <property>
3366 <name>hive.service.metrics.file.frequency</name>
3367 <value>5s</value>
3368 <description>
3369 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
3370 For metric class org.apache.hadoop.hive.common.metrics.metrics2.CodahaleMetrics JSON_FILE reporter, the frequency of updating JSON metrics file.
3371 </description>
3372 </property>
3373 <property>
3374 <name>hive.service.metrics.hadoop2.frequency</name>
3375 <value>30s</value>
3376 <description>
3377 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
3378 For metric class org.apache.hadoop.hive.common.metrics.metrics2.CodahaleMetrics HADOOP2 reporter, the frequency of updating the HADOOP2 metrics system.
3379 </description>
3380 </property>
3381 <property>
3382 <name>hive.service.metrics.hadoop2.component</name>
3383 <value>hive</value>
3384 <description>Component name to provide to Hadoop2 Metrics system. Ideally 'hivemetastore' for the MetaStore and and 'hiveserver2' for HiveServer2.</description>
3385 </property>
3386 <property>
3387 <name>hive.exec.perf.logger</name>
3388 <value>org.apache.hadoop.hive.ql.log.PerfLogger</value>
3389 <description>
3390 The class responsible for logging client side performance metrics.
3391 Must be a subclass of org.apache.hadoop.hive.ql.log.PerfLogger
3392 </description>
3393 </property>
3394 <property>
3395 <name>hive.start.cleanup.scratchdir</name>
3396 <value>false</value>
3397 <description>To cleanup the Hive scratchdir when starting the Hive Server</description>
3398 </property>
3399 <property>
3400 <name>hive.scratchdir.lock</name>
3401 <value>false</value>
3402 <description>To hold a lock file in scratchdir to prevent to be removed by cleardanglingscratchdir</description>
3403 </property>
3404 <property>
3405 <name>hive.insert.into.multilevel.dirs</name>
3406 <value>false</value>
3407 <description>
3408 Where to insert into multilevel directories like
3409 "insert directory '/HIVEFT25686/chinna/' from table"
3410 </description>
3411 </property>
3412 <property>
3413 <name>hive.warehouse.subdir.inherit.perms</name>
3414 <value>true</value>
3415 <description>
3416 Set this to false if the table directories should be created
3417 with the permissions derived from dfs umask instead of
3418 inheriting the permission of the warehouse or database directory.
3419 </description>
3420 </property>
3421 <property>
3422 <name>hive.insert.into.external.tables</name>
3423 <value>true</value>
3424 <description>whether insert into external tables is allowed</description>
3425 </property>
3426 <property>
3427 <name>hive.exec.temporary.table.storage</name>
3428 <value>default</value>
3429 <description>
3430 Expects one of [memory, ssd, default].
3431 Define the storage policy for temporary tables.Choices between memory, ssd and default
3432 </description>
3433 </property>
3434 <property>
3435 <name>hive.exec.driver.run.hooks</name>
3436 <value/>
3437 <description>A comma separated list of hooks which implement HiveDriverRunHook. Will be run at the beginning and end of Driver.run, these will be run in the order specified.</description>
3438 </property>
3439 <property>
3440 <name>hive.ddl.output.format</name>
3441 <value/>
3442 <description>
3443 The data format to use for DDL output. One of "text" (for human
3444 readable text) or "json" (for a json object).
3445 </description>
3446 </property>
3447 <property>
3448 <name>hive.entity.separator</name>
3449 <value>@</value>
3450 <description>Separator used to construct names of tables and partitions. For example, dbname@tablename@partitionname</description>
3451 </property>
3452 <property>
3453 <name>hive.entity.capture.transform</name>
3454 <value>false</value>
3455 <description>Compiler to capture transform URI referred in the query</description>
3456 </property>
3457 <property>
3458 <name>hive.display.partition.cols.separately</name>
3459 <value>true</value>
3460 <description>
3461 In older Hive version (0.10 and earlier) no distinction was made between
3462 partition columns or non-partition columns while displaying columns in describe
3463 table. From 0.12 onwards, they are displayed separately. This flag will let you
3464 get old behavior, if desired. See, test-case in patch for HIVE-6689.
3465 </description>
3466 </property>
3467 <property>
3468 <name>hive.ssl.protocol.blacklist</name>
3469 <value>SSLv2,SSLv3</value>
3470 <description>SSL Versions to disable for all Hive Servers</description>
3471 </property>
3472 <property>
3473 <name>hive.server2.max.start.attempts</name>
3474 <value>30</value>
3475 <description>
3476 Expects value bigger than 0.
3477 Number of times HiveServer2 will attempt to start before exiting, sleeping 60 seconds between retries.
3478 The default of 30 will keep trying for 30 minutes.
3479 </description>
3480 </property>
3481 <property>
3482 <name>hive.server2.support.dynamic.service.discovery</name>
3483 <value>false</value>
3484 <description>Whether HiveServer2 supports dynamic service discovery for its clients. To support this, each instance of HiveServer2 currently uses ZooKeeper to register itself, when it is brought up. JDBC/ODBC clients should use the ZooKeeper ensemble: hive.zookeeper.quorum in their connection string.</description>
3485 </property>
3486 <property>
3487 <name>hive.server2.zookeeper.namespace</name>
3488 <value>hiveserver2</value>
3489 <description>The parent node in ZooKeeper used by HiveServer2 when supporting dynamic service discovery.</description>
3490 </property>
3491 <property>
3492 <name>hive.server2.zookeeper.publish.configs</name>
3493 <value>true</value>
3494 <description>Whether we should publish HiveServer2's configs to ZooKeeper.</description>
3495 </property>
3496 <property>
3497 <name>hive.server2.global.init.file.location</name>
3498 <value>${env:HIVE_CONF_DIR}</value>
3499 <description>
3500 Either the location of a HS2 global init file or a directory containing a .hiverc file. If the
3501 property is set, the value must be a valid path to an init file or directory where the init file is located.
3502 </description>
3503 </property>
3504 <property>
3505 <name>hive.server2.transport.mode</name>
3506 <value>binary</value>
3507 <description>
3508 Expects one of [binary, http].
3509 Transport mode of HiveServer2.
3510 </description>
3511 </property>
3512 <property>
3513 <name>hive.server2.thrift.bind.host</name>
3514 <value/>
3515 <description>Bind host on which to run the HiveServer2 Thrift service.</description>
3516 </property>
3517 <property>
3518 <name>hive.driver.parallel.compilation</name>
3519 <value>false</value>
3520 <description>
3521 Whether to
3522 enable parallel compilation between sessions on HiveServer2. The default is false.
3523 </description>
3524 </property>
3525 <property>
3526 <name>hive.server2.compile.lock.timeout</name>
3527 <value>0s</value>
3528 <description>
3529 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
3530 Number of seconds a request will wait to acquire the compile lock before giving up. Setting it to 0s disables the timeout.
3531 </description>
3532 </property>
3533 <property>
3534 <name>hive.server2.webui.host</name>
3535 <value>0.0.0.0</value>
3536 <description>The host address the HiveServer2 WebUI will listen on</description>
3537 </property>
3538 <property>
3539 <name>hive.server2.webui.port</name>
3540 <value>10002</value>
3541 <description>The port the HiveServer2 WebUI will listen on. This can beset to 0 or a negative integer to disable the web UI</description>
3542 </property>
3543 <property>
3544 <name>hive.server2.webui.max.threads</name>
3545 <value>50</value>
3546 <description>The max HiveServer2 WebUI threads</description>
3547 </property>
3548 <property>
3549 <name>hive.server2.webui.use.ssl</name>
3550 <value>false</value>
3551 <description>Set this to true for using SSL encryption for HiveServer2 WebUI.</description>
3552 </property>
3553 <property>
3554 <name>hive.server2.webui.keystore.path</name>
3555 <value/>
3556 <description>SSL certificate keystore location for HiveServer2 WebUI.</description>
3557 </property>
3558 <property>
3559 <name>hive.server2.webui.keystore.password</name>
3560 <value/>
3561 <description>SSL certificate keystore password for HiveServer2 WebUI.</description>
3562 </property>
3563 <property>
3564 <name>hive.server2.webui.use.spnego</name>
3565 <value>false</value>
3566 <description>If true, the HiveServer2 WebUI will be secured with SPNEGO. Clients must authenticate with Kerberos.</description>
3567 </property>
3568 <property>
3569 <name>hive.server2.webui.spnego.keytab</name>
3570 <value/>
3571 <description>The path to the Kerberos Keytab file containing the HiveServer2 WebUI SPNEGO service principal.</description>
3572 </property>
3573 <property>
3574 <name>hive.server2.webui.spnego.principal</name>
3575 <value>HTTP/_HOST@EXAMPLE.COM</value>
3576 <description>
3577 The HiveServer2 WebUI SPNEGO service principal.
3578 The special string _HOST will be replaced automatically with
3579 the value of hive.server2.webui.host or the correct host name.
3580 </description>
3581 </property>
3582 <property>
3583 <name>hive.server2.webui.max.historic.queries</name>
3584 <value>25</value>
3585 <description>The maximum number of past queries to show in HiverSever2 WebUI.</description>
3586 </property>
3587 <property>
3588 <name>hive.server2.tez.default.queues</name>
3589 <value/>
3590 <description>
3591 A list of comma separated values corresponding to YARN queues of the same name.
3592 When HiveServer2 is launched in Tez mode, this configuration needs to be set
3593 for multiple Tez sessions to run in parallel on the cluster.
3594 </description>
3595 </property>
3596 <property>
3597 <name>hive.server2.tez.sessions.per.default.queue</name>
3598 <value>1</value>
3599 <description>
3600 A positive integer that determines the number of Tez sessions that should be
3601 launched on each of the queues specified by "hive.server2.tez.default.queues".
3602 Determines the parallelism on each queue.
3603 </description>
3604 </property>
3605 <property>
3606 <name>hive.server2.tez.initialize.default.sessions</name>
3607 <value>false</value>
3608 <description>
3609 This flag is used in HiveServer2 to enable a user to use HiveServer2 without
3610 turning on Tez for HiveServer2. The user could potentially want to run queries
3611 over Tez without the pool of sessions.
3612 </description>
3613 </property>
3614 <property>
3615 <name>hive.server2.tez.session.lifetime</name>
3616 <value>162h</value>
3617 <description>
3618 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is hour if not specified.
3619 The lifetime of the Tez sessions launched by HS2 when default sessions are enabled.
3620 Set to 0 to disable session expiration.
3621 </description>
3622 </property>
3623 <property>
3624 <name>hive.server2.tez.session.lifetime.jitter</name>
3625 <value>3h</value>
3626 <description>
3627 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is hour if not specified.
3628 The jitter for Tez session lifetime; prevents all the sessions from restarting at once.
3629 </description>
3630 </property>
3631 <property>
3632 <name>hive.server2.tez.sessions.init.threads</name>
3633 <value>16</value>
3634 <description>
3635 If hive.server2.tez.initialize.default.sessions is enabled, the maximum number of
3636 threads to use to initialize the default sessions.
3637 </description>
3638 </property>
3639 <property>
3640 <name>hive.server2.logging.operation.enabled</name>
3641 <value>true</value>
3642 <description>When true, HS2 will save operation logs and make them available for clients</description>
3643 </property>
3644 <property>
3645 <name>hive.server2.logging.operation.log.location</name>
3646 <value>${system:java.io.tmpdir}/${system:user.name}/operation_logs</value>
3647 <description>Top level directory where operation logs are stored if logging functionality is enabled</description>
3648 </property>
3649 <property>
3650 <name>hive.server2.logging.operation.level</name>
3651 <value>EXECUTION</value>
3652 <description>
3653 Expects one of [none, execution, performance, verbose].
3654 HS2 operation logging mode available to clients to be set at session level.
3655 For this to work, hive.server2.logging.operation.enabled should be set to true.
3656 NONE: Ignore any logging
3657 EXECUTION: Log completion of tasks
3658 PERFORMANCE: Execution + Performance logs
3659 VERBOSE: All logs
3660 </description>
3661 </property>
3662 <property>
3663 <name>hive.server2.metrics.enabled</name>
3664 <value>false</value>
3665 <description>Enable metrics on the HiveServer2.</description>
3666 </property>
3667 <property>
3668 <name>hive.server2.thrift.http.port</name>
3669 <value>10001</value>
3670 <description>Port number of HiveServer2 Thrift interface when hive.server2.transport.mode is 'http'.</description>
3671 </property>
3672 <property>
3673 <name>hive.server2.thrift.http.path</name>
3674 <value>cliservice</value>
3675 <description>Path component of URL endpoint when in HTTP mode.</description>
3676 </property>
3677 <property>
3678 <name>hive.server2.thrift.max.message.size</name>
3679 <value>104857600</value>
3680 <description>Maximum message size in bytes a HS2 server will accept.</description>
3681 </property>
3682 <property>
3683 <name>hive.server2.thrift.http.max.idle.time</name>
3684 <value>1800s</value>
3685 <description>
3686 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
3687 Maximum idle time for a connection on the server when in HTTP mode.
3688 </description>
3689 </property>
3690 <property>
3691 <name>hive.server2.thrift.http.worker.keepalive.time</name>
3692 <value>60s</value>
3693 <description>
3694 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
3695 Keepalive time for an idle http worker thread. When the number of workers exceeds min workers, excessive threads are killed after this time interval.
3696 </description>
3697 </property>
3698 <property>
3699 <name>hive.server2.thrift.http.request.header.size</name>
3700 <value>6144</value>
3701 <description>Request header size in bytes, when using HTTP transport mode. Jetty defaults used.</description>
3702 </property>
3703 <property>
3704 <name>hive.server2.thrift.http.response.header.size</name>
3705 <value>6144</value>
3706 <description>Response header size in bytes, when using HTTP transport mode. Jetty defaults used.</description>
3707 </property>
3708 <property>
3709 <name>hive.server2.thrift.http.cookie.auth.enabled</name>
3710 <value>true</value>
3711 <description>When true, HiveServer2 in HTTP transport mode, will use cookie based authentication mechanism.</description>
3712 </property>
3713 <property>
3714 <name>hive.server2.thrift.http.cookie.max.age</name>
3715 <value>86400s</value>
3716 <description>
3717 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
3718 Maximum age in seconds for server side cookie used by HS2 in HTTP mode.
3719 </description>
3720 </property>
3721 <property>
3722 <name>hive.server2.thrift.http.cookie.domain</name>
3723 <value/>
3724 <description>Domain for the HS2 generated cookies</description>
3725 </property>
3726 <property>
3727 <name>hive.server2.thrift.http.cookie.path</name>
3728 <value/>
3729 <description>Path for the HS2 generated cookies</description>
3730 </property>
3731 <property>
3732 <name>hive.server2.thrift.http.cookie.is.secure</name>
3733 <value>true</value>
3734 <description>Secure attribute of the HS2 generated cookie.</description>
3735 </property>
3736 <property>
3737 <name>hive.server2.thrift.http.cookie.is.httponly</name>
3738 <value>true</value>
3739 <description>HttpOnly attribute of the HS2 generated cookie.</description>
3740 </property>
3741 <property>
3742 <name>hive.server2.thrift.port</name>
3743 <value>10000</value>
3744 <description>Port number of HiveServer2 Thrift interface when hive.server2.transport.mode is 'binary'.</description>
3745 </property>
3746 <property>
3747 <name>hive.server2.thrift.sasl.qop</name>
3748 <value>auth</value>
3749 <description>
3750 Expects one of [auth, auth-int, auth-conf].
3751 Sasl QOP value; set it to one of following values to enable higher levels of
3752 protection for HiveServer2 communication with clients.
3753 Setting hadoop.rpc.protection to a higher level than HiveServer2 does not
3754 make sense in most situations. HiveServer2 ignores hadoop.rpc.protection in favor
3755 of hive.server2.thrift.sasl.qop.
3756 "auth" - authentication only (default)
3757 "auth-int" - authentication plus integrity protection
3758 "auth-conf" - authentication plus integrity and confidentiality protection
3759 This is applicable only if HiveServer2 is configured to use Kerberos authentication.
3760 </description>
3761 </property>
3762 <property>
3763 <name>hive.server2.thrift.min.worker.threads</name>
3764 <value>5</value>
3765 <description>Minimum number of Thrift worker threads</description>
3766 </property>
3767 <property>
3768 <name>hive.server2.thrift.max.worker.threads</name>
3769 <value>500</value>
3770 <description>Maximum number of Thrift worker threads</description>
3771 </property>
3772 <property>
3773 <name>hive.server2.thrift.exponential.backoff.slot.length</name>
3774 <value>100ms</value>
3775 <description>
3776 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
3777 Binary exponential backoff slot time for Thrift clients during login to HiveServer2,
3778 for retries until hitting Thrift client timeout
3779 </description>
3780 </property>
3781 <property>
3782 <name>hive.server2.thrift.login.timeout</name>
3783 <value>20s</value>
3784 <description>
3785 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
3786 Timeout for Thrift clients during login to HiveServer2
3787 </description>
3788 </property>
3789 <property>
3790 <name>hive.server2.thrift.worker.keepalive.time</name>
3791 <value>60s</value>
3792 <description>
3793 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
3794 Keepalive time (in seconds) for an idle worker thread. When the number of workers exceeds min workers, excessive threads are killed after this time interval.
3795 </description>
3796 </property>
3797 <property>
3798 <name>hive.server2.async.exec.threads</name>
3799 <value>100</value>
3800 <description>Number of threads in the async thread pool for HiveServer2</description>
3801 </property>
3802 <property>
3803 <name>hive.server2.async.exec.shutdown.timeout</name>
3804 <value>10s</value>
3805 <description>
3806 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
3807 How long HiveServer2 shutdown will wait for async threads to terminate.
3808 </description>
3809 </property>
3810 <property>
3811 <name>hive.server2.async.exec.wait.queue.size</name>
3812 <value>100</value>
3813 <description>
3814 Size of the wait queue for async thread pool in HiveServer2.
3815 After hitting this limit, the async thread pool will reject new requests.
3816 </description>
3817 </property>
3818 <property>
3819 <name>hive.server2.async.exec.keepalive.time</name>
3820 <value>10s</value>
3821 <description>
3822 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
3823 Time that an idle HiveServer2 async thread (from the thread pool) will wait for a new task
3824 to arrive before terminating
3825 </description>
3826 </property>
3827 <property>
3828 <name>hive.server2.async.exec.async.compile</name>
3829 <value>false</value>
3830 <description>Whether to enable compiling async query asynchronously. If enabled, it is unknown if the query will have any resultset before compilation completed.</description>
3831 </property>
3832 <property>
3833 <name>hive.server2.long.polling.timeout</name>
3834 <value>5000ms</value>
3835 <description>
3836 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
3837 Time that HiveServer2 will wait before responding to asynchronous calls that use long polling
3838 </description>
3839 </property>
3840 <property>
3841 <name>hive.session.impl.classname</name>
3842 <value/>
3843 <description>Classname for custom implementation of hive session</description>
3844 </property>
3845 <property>
3846 <name>hive.session.impl.withugi.classname</name>
3847 <value/>
3848 <description>Classname for custom implementation of hive session with UGI</description>
3849 </property>
3850 <property>
3851 <name>hive.server2.authentication</name>
3852 <value>NONE</value>
3853 <description>
3854 Expects one of [nosasl, none, ldap, kerberos, pam, custom].
3855 Client authentication types.
3856 NONE: no authentication check
3857 LDAP: LDAP/AD based authentication
3858 KERBEROS: Kerberos/GSSAPI authentication
3859 CUSTOM: Custom authentication provider
3860 (Use with property hive.server2.custom.authentication.class)
3861 PAM: Pluggable authentication module
3862 NOSASL: Raw transport
3863 </description>
3864 </property>
3865 <property>
3866 <name>hive.server2.allow.user.substitution</name>
3867 <value>true</value>
3868 <description>Allow alternate user to be specified as part of HiveServer2 open connection request.</description>
3869 </property>
3870 <property>
3871 <name>hive.server2.authentication.kerberos.keytab</name>
3872 <value/>
3873 <description>Kerberos keytab file for server principal</description>
3874 </property>
3875 <property>
3876 <name>hive.server2.authentication.kerberos.principal</name>
3877 <value/>
3878 <description>Kerberos server principal</description>
3879 </property>
3880 <property>
3881 <name>hive.server2.authentication.spnego.keytab</name>
3882 <value/>
3883 <description>
3884 keytab file for SPNego principal, optional,
3885 typical value would look like /etc/security/keytabs/spnego.service.keytab,
3886 This keytab would be used by HiveServer2 when Kerberos security is enabled and
3887 HTTP transport mode is used.
3888 This needs to be set only if SPNEGO is to be used in authentication.
3889 SPNego authentication would be honored only if valid
3890 hive.server2.authentication.spnego.principal
3891 and
3892 hive.server2.authentication.spnego.keytab
3893 are specified.
3894 </description>
3895 </property>
3896 <property>
3897 <name>hive.server2.authentication.spnego.principal</name>
3898 <value/>
3899 <description>
3900 SPNego service principal, optional,
3901 typical value would look like HTTP/_HOST@EXAMPLE.COM
3902 SPNego service principal would be used by HiveServer2 when Kerberos security is enabled
3903 and HTTP transport mode is used.
3904 This needs to be set only if SPNEGO is to be used in authentication.
3905 </description>
3906 </property>
3907 <property>
3908 <name>hive.server2.authentication.ldap.url</name>
3909 <value/>
3910 <description>
3911 LDAP connection URL(s),
3912 this value could contain URLs to mutiple LDAP servers instances for HA,
3913 each LDAP URL is separated by a SPACE character. URLs are used in the
3914 order specified until a connection is successful.
3915 </description>
3916 </property>
3917 <property>
3918 <name>hive.server2.authentication.ldap.baseDN</name>
3919 <value/>
3920 <description>LDAP base DN</description>
3921 </property>
3922 <property>
3923 <name>hive.server2.authentication.ldap.Domain</name>
3924 <value/>
3925 <description/>
3926 </property>
3927 <property>
3928 <name>hive.server2.authentication.ldap.groupDNPattern</name>
3929 <value/>
3930 <description>
3931 COLON-separated list of patterns to use to find DNs for group entities in this directory.
3932 Use %s where the actual group name is to be substituted for.
3933 For example: CN=%s,CN=Groups,DC=subdomain,DC=domain,DC=com.
3934 </description>
3935 </property>
3936 <property>
3937 <name>hive.server2.authentication.ldap.groupFilter</name>
3938 <value/>
3939 <description>
3940 COMMA-separated list of LDAP Group names (short name not full DNs).
3941 For example: HiveAdmins,HadoopAdmins,Administrators
3942 </description>
3943 </property>
3944 <property>
3945 <name>hive.server2.authentication.ldap.userDNPattern</name>
3946 <value/>
3947 <description>
3948 COLON-separated list of patterns to use to find DNs for users in this directory.
3949 Use %s where the actual group name is to be substituted for.
3950 For example: CN=%s,CN=Users,DC=subdomain,DC=domain,DC=com.
3951 </description>
3952 </property>
3953 <property>
3954 <name>hive.server2.authentication.ldap.userFilter</name>
3955 <value/>
3956 <description>
3957 COMMA-separated list of LDAP usernames (just short names, not full DNs).
3958 For example: hiveuser,impalauser,hiveadmin,hadoopadmin
3959 </description>
3960 </property>
3961 <property>
3962 <name>hive.server2.authentication.ldap.guidKey</name>
3963 <value>uid</value>
3964 <description>
3965 LDAP attribute name whose values are unique in this LDAP server.
3966 For example: uid or CN.
3967 </description>
3968 </property>
3969 <property>
3970 <name>hive.server2.authentication.ldap.groupMembershipKey</name>
3971 <value>member</value>
3972 <description>
3973 LDAP attribute name on the user entry that references a group, the user belongs to.
3974 For example: member, uniqueMember or memberUid
3975 </description>
3976 </property>
3977 <property>
3978 <name>hive.server2.authentication.ldap.groupClassKey</name>
3979 <value>groupOfNames</value>
3980 <description>
3981 LDAP attribute name on the group entry that is to be used in LDAP group searches.
3982 For example: group, groupOfNames or groupOfUniqueNames.
3983 </description>
3984 </property>
3985 <property>
3986 <name>hive.server2.authentication.ldap.customLDAPQuery</name>
3987 <value/>
3988 <description>
3989 A full LDAP query that LDAP Atn provider uses to execute against LDAP Server.
3990 If this query returns a null resultset, the LDAP Provider fails the Authentication
3991 request, succeeds if the user is part of the resultset.For example: (&(objectClass=group)(objectClass=top)(instanceType=4)(cn=Domain*))
3992 (&(objectClass=person)(|(sAMAccountName=admin)(|(memberOf=CN=Domain Admins,CN=Users,DC=domain,DC=com)(memberOf=CN=Administrators,CN=Builtin,DC=domain,DC=com))))
3993 </description>
3994 </property>
3995 <property>
3996 <name>hive.server2.custom.authentication.class</name>
3997 <value/>
3998 <description>
3999 Custom authentication class. Used when property
4000 'hive.server2.authentication' is set to 'CUSTOM'. Provided class
4001 must be a proper implementation of the interface
4002 org.apache.hive.service.auth.PasswdAuthenticationProvider. HiveServer2
4003 will call its Authenticate(user, passed) method to authenticate requests.
4004 The implementation may optionally implement Hadoop's
4005 org.apache.hadoop.conf.Configurable class to grab Hive's Configuration object.
4006 </description>
4007 </property>
4008 <property>
4009 <name>hive.server2.authentication.pam.services</name>
4010 <value/>
4011 <description>
4012 List of the underlying pam services that should be used when auth type is PAM
4013 A file with the same name must exist in /etc/pam.d
4014 </description>
4015 </property>
4016 <property>
4017 <name>hive.server2.enable.doAs</name>
4018 <value>true</value>
4019 <description>
4020 Setting this property to true will have HiveServer2 execute
4021 Hive operations as the user making the calls to it.
4022 </description>
4023 </property>
4024 <property>
4025 <name>hive.server2.table.type.mapping</name>
4026 <value>CLASSIC</value>
4027 <description>
4028 Expects one of [classic, hive].
4029 This setting reflects how HiveServer2 will report the table types for JDBC and other
4030 client implementations that retrieve the available tables and supported table types
4031 HIVE : Exposes Hive's native table types like MANAGED_TABLE, EXTERNAL_TABLE, VIRTUAL_VIEW
4032 CLASSIC : More generic types like TABLE and VIEW
4033 </description>
4034 </property>
4035 <property>
4036 <name>hive.server2.session.hook</name>
4037 <value/>
4038 <description/>
4039 </property>
4040 <property>
4041 <name>hive.server2.use.SSL</name>
4042 <value>false</value>
4043 <description>Set this to true for using SSL encryption in HiveServer2.</description>
4044 </property>
4045 <property>
4046 <name>hive.server2.keystore.path</name>
4047 <value/>
4048 <description>SSL certificate keystore location.</description>
4049 </property>
4050 <property>
4051 <name>hive.server2.keystore.password</name>
4052 <value/>
4053 <description>SSL certificate keystore password.</description>
4054 </property>
4055 <property>
4056 <name>hive.server2.map.fair.scheduler.queue</name>
4057 <value>true</value>
4058 <description>
4059 If the YARN fair scheduler is configured and HiveServer2 is running in non-impersonation mode,
4060 this setting determines the user for fair scheduler queue mapping.
4061 If set to true (default), the logged-in user determines the fair scheduler queue
4062 for submitted jobs, so that map reduce resource usage can be tracked by user.
4063 If set to false, all Hive jobs go to the 'hive' user's queue.
4064 </description>
4065 </property>
4066 <property>
4067 <name>hive.server2.builtin.udf.whitelist</name>
4068 <value/>
4069 <description>
4070 Comma separated list of builtin udf names allowed in queries.
4071 An empty whitelist allows all builtin udfs to be executed. The udf black list takes precedence over udf white list
4072 </description>
4073 </property>
4074 <property>
4075 <name>hive.server2.builtin.udf.blacklist</name>
4076 <value/>
4077 <description>Comma separated list of udfs names. These udfs will not be allowed in queries. The udf black list takes precedence over udf white list</description>
4078 </property>
4079 <property>
4080 <name>hive.allow.udf.load.on.demand</name>
4081 <value>false</value>
4082 <description>
4083 Whether enable loading UDFs from metastore on demand; this is mostly relevant for
4084 HS2 and was the default behavior before Hive 1.2. Off by default.
4085 </description>
4086 </property>
4087 <property>
4088 <name>hive.server2.session.check.interval</name>
4089 <value>6h</value>
4090 <description>
4091 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
4092 The time should be bigger than or equal to 3000 msec.
4093 The check interval for session/operation timeout, which can be disabled by setting to zero or negative value.
4094 </description>
4095 </property>
4096 <property>
4097 <name>hive.server2.close.session.on.disconnect</name>
4098 <value>true</value>
4099 <description>Session will be closed when connection is closed. Set this to false to have session outlive its parent connection.</description>
4100 </property>
4101 <property>
4102 <name>hive.server2.idle.session.timeout</name>
4103 <value>7d</value>
4104 <description>
4105 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
4106 Session will be closed when it's not accessed for this duration, which can be disabled by setting to zero or negative value.
4107 </description>
4108 </property>
4109 <property>
4110 <name>hive.server2.idle.operation.timeout</name>
4111 <value>5d</value>
4112 <description>
4113 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
4114 Operation will be closed when it's not accessed for this duration of time, which can be disabled by setting to zero value.
4115 With positive value, it's checked for operations in terminal state only (FINISHED, CANCELED, CLOSED, ERROR).
4116 With negative value, it's checked for all of the operations regardless of state.
4117 </description>
4118 </property>
4119 <property>
4120 <name>hive.server2.idle.session.check.operation</name>
4121 <value>true</value>
4122 <description>
4123 Session will be considered to be idle only if there is no activity, and there is no pending operation.
4124 This setting takes effect only if session idle timeout (hive.server2.idle.session.timeout) and checking
4125 (hive.server2.session.check.interval) are enabled.
4126 </description>
4127 </property>
4128 <property>
4129 <name>hive.server2.thrift.client.retry.limit</name>
4130 <value>1</value>
4131 <description>Number of retries upon failure of Thrift HiveServer2 calls</description>
4132 </property>
4133 <property>
4134 <name>hive.server2.thrift.client.connect.retry.limit</name>
4135 <value>1</value>
4136 <description>Number of retries while opening a connection to HiveServe2</description>
4137 </property>
4138 <property>
4139 <name>hive.server2.thrift.client.retry.delay.seconds</name>
4140 <value>1s</value>
4141 <description>
4142 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
4143 Number of seconds for the HiveServer2 thrift client to wait between consecutive connection attempts. Also specifies the time to wait between retrying thrift calls upon failures
4144 </description>
4145 </property>
4146 <property>
4147 <name>hive.server2.thrift.client.user</name>
4148 <value>anonymous</value>
4149 <description>Username to use against thrift client</description>
4150 </property>
4151 <property>
4152 <name>hive.server2.thrift.client.password</name>
4153 <value>anonymous</value>
4154 <description>Password to use against thrift client</description>
4155 </property>
4156 <property>
4157 <name>hive.server2.thrift.resultset.serialize.in.tasks</name>
4158 <value>false</value>
4159 <description>
4160 Whether we should serialize the Thrift structures used in JDBC ResultSet RPC in task nodes.
4161 We use SequenceFile and ThriftJDBCBinarySerDe to read and write the final results if this is true.
4162 </description>
4163 </property>
4164 <property>
4165 <name>hive.server2.thrift.resultset.max.fetch.size</name>
4166 <value>1000</value>
4167 <description>Max number of rows sent in one Fetch RPC call by the server to the client.</description>
4168 </property>
4169 <property>
4170 <name>hive.server2.xsrf.filter.enabled</name>
4171 <value>false</value>
4172 <description>If enabled, HiveServer2 will block any requests made to it over http if an X-XSRF-HEADER header is not present</description>
4173 </property>
4174 <property>
4175 <name>hive.security.command.whitelist</name>
4176 <value>set,reset,dfs,add,list,delete,reload,compile</value>
4177 <description>Comma separated list of non-SQL Hive commands users are authorized to execute</description>
4178 </property>
4179 <property>
4180 <name>hive.mv.files.thread</name>
4181 <value>25</value>
4182 <description>
4183 Expects a byte size value with unit (blank for bytes, kb, mb, gb, tb, pb).
4184 The size should be in between 0Pb (inclusive) and 1Kb (inclusive).
4185 Number of threads used to move files in move task. Set it to 0 to disable multi-threaded file moves. This parameter is also used by MSCK to check tables.
4186 </description>
4187 </property>
4188 <property>
4189 <name>hive.multi.insert.move.tasks.share.dependencies</name>
4190 <value>false</value>
4191 <description>
4192 If this is set all move tasks for tables/partitions (not directories) at the end of a
4193 multi-insert query will only begin once the dependencies for all these move tasks have been
4194 met.
4195 Advantages: If concurrency is enabled, the locks will only be released once the query has
4196 finished, so with this config enabled, the time when the table/partition is
4197 generated will be much closer to when the lock on it is released.
4198 Disadvantages: If concurrency is not enabled, with this disabled, the tables/partitions which
4199 are produced by this query and finish earlier will be available for querying
4200 much earlier. Since the locks are only released once the query finishes, this
4201 does not apply if concurrency is enabled.
4202 </description>
4203 </property>
4204 <property>
4205 <name>hive.exec.infer.bucket.sort</name>
4206 <value>false</value>
4207 <description>
4208 If this is set, when writing partitions, the metadata will include the bucketing/sorting
4209 properties with which the data was written if any (this will not overwrite the metadata
4210 inherited from the table if the table is bucketed/sorted)
4211 </description>
4212 </property>
4213 <property>
4214 <name>hive.exec.infer.bucket.sort.num.buckets.power.two</name>
4215 <value>false</value>
4216 <description>
4217 If this is set, when setting the number of reducers for the map reduce task which writes the
4218 final output files, it will choose a number which is a power of two, unless the user specifies
4219 the number of reducers to use using mapred.reduce.tasks. The number of reducers
4220 may be set to a power of two, only to be followed by a merge task meaning preventing
4221 anything from being inferred.
4222 With hive.exec.infer.bucket.sort set to true:
4223 Advantages: If this is not set, the number of buckets for partitions will seem arbitrary,
4224 which means that the number of mappers used for optimized joins, for example, will
4225 be very low. With this set, since the number of buckets used for any partition is
4226 a power of two, the number of mappers used for optimized joins will be the least
4227 number of buckets used by any partition being joined.
4228 Disadvantages: This may mean a much larger or much smaller number of reducers being used in the
4229 final map reduce job, e.g. if a job was originally going to take 257 reducers,
4230 it will now take 512 reducers, similarly if the max number of reducers is 511,
4231 and a job was going to use this many, it will now use 256 reducers.
4232 </description>
4233 </property>
4234 <property>
4235 <name>hive.optimize.listbucketing</name>
4236 <value>false</value>
4237 <description>Enable list bucketing optimizer. Default value is false so that we disable it by default.</description>
4238 </property>
4239 <property>
4240 <name>hive.server.read.socket.timeout</name>
4241 <value>10s</value>
4242 <description>
4243 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
4244 Timeout for the HiveServer to close the connection if no response from the client. By default, 10 seconds.
4245 </description>
4246 </property>
4247 <property>
4248 <name>hive.server.tcp.keepalive</name>
4249 <value>true</value>
4250 <description>Whether to enable TCP keepalive for the Hive Server. Keepalive will prevent accumulation of half-open connections.</description>
4251 </property>
4252 <property>
4253 <name>hive.decode.partition.name</name>
4254 <value>false</value>
4255 <description>Whether to show the unquoted partition names in query results.</description>
4256 </property>
4257 <property>
4258 <name>hive.execution.engine</name>
4259 <value>mr</value>
4260 <description>
4261 Expects one of [mr, tez, spark].
4262 Chooses execution engine. Options are: mr (Map reduce, default), tez, spark. While MR
4263 remains the default engine for historical reasons, it is itself a historical engine
4264 and is deprecated in Hive 2 line. It may be removed without further warning.
4265 </description>
4266 </property>
4267 <property>
4268 <name>hive.execution.mode</name>
4269 <value>container</value>
4270 <description>
4271 Expects one of [container, llap].
4272 Chooses whether query fragments will run in container or in llap
4273 </description>
4274 </property>
4275 <property>
4276 <name>hive.jar.directory</name>
4277 <value/>
4278 <description>
4279 This is the location hive in tez mode will look for to find a site wide
4280 installed hive instance.
4281 </description>
4282 </property>
4283 <property>
4284 <name>hive.user.install.directory</name>
4285 <value>/user/</value>
4286 <description>
4287 If hive (in tez mode only) cannot find a usable hive jar in "hive.jar.directory",
4288 it will upload the hive jar to "hive.user.install.directory/user.name"
4289 and use it to run queries.
4290 </description>
4291 </property>
4292 <property>
4293 <name>hive.vectorized.execution.enabled</name>
4294 <value>false</value>
4295 <description>
4296 This flag should be set to true to enable vectorized mode of query execution.
4297 The default value is false.
4298 </description>
4299 </property>
4300 <property>
4301 <name>hive.vectorized.execution.reduce.enabled</name>
4302 <value>true</value>
4303 <description>
4304 This flag should be set to true to enable vectorized mode of the reduce-side of query execution.
4305 The default value is true.
4306 </description>
4307 </property>
4308 <property>
4309 <name>hive.vectorized.execution.reduce.groupby.enabled</name>
4310 <value>true</value>
4311 <description>
4312 This flag should be set to true to enable vectorized mode of the reduce-side GROUP BY query execution.
4313 The default value is true.
4314 </description>
4315 </property>
4316 <property>
4317 <name>hive.vectorized.execution.mapjoin.native.enabled</name>
4318 <value>true</value>
4319 <description>
4320 This flag should be set to true to enable native (i.e. non-pass through) vectorization
4321 of queries using MapJoin.
4322 The default value is true.
4323 </description>
4324 </property>
4325 <property>
4326 <name>hive.vectorized.execution.mapjoin.native.multikey.only.enabled</name>
4327 <value>false</value>
4328 <description>
4329 This flag should be set to true to restrict use of native vector map join hash tables to
4330 the MultiKey in queries using MapJoin.
4331 The default value is false.
4332 </description>
4333 </property>
4334 <property>
4335 <name>hive.vectorized.execution.mapjoin.minmax.enabled</name>
4336 <value>false</value>
4337 <description>
4338 This flag should be set to true to enable vector map join hash tables to
4339 use max / max filtering for integer join queries using MapJoin.
4340 The default value is false.
4341 </description>
4342 </property>
4343 <property>
4344 <name>hive.vectorized.execution.mapjoin.overflow.repeated.threshold</name>
4345 <value>-1</value>
4346 <description>
4347 The number of small table rows for a match in vector map join hash tables
4348 where we use the repeated field optimization in overflow vectorized row batch for join queries using MapJoin.
4349 A value of -1 means do use the join result optimization. Otherwise, threshold value can be 0 to maximum integer.
4350 </description>
4351 </property>
4352 <property>
4353 <name>hive.vectorized.execution.mapjoin.native.fast.hashtable.enabled</name>
4354 <value>false</value>
4355 <description>
4356 This flag should be set to true to enable use of native fast vector map join hash tables in
4357 queries using MapJoin.
4358 The default value is false.
4359 </description>
4360 </property>
4361 <property>
4362 <name>hive.vectorized.groupby.checkinterval</name>
4363 <value>100000</value>
4364 <description>Number of entries added to the group by aggregation hash before a recomputation of average entry size is performed.</description>
4365 </property>
4366 <property>
4367 <name>hive.vectorized.groupby.maxentries</name>
4368 <value>1000000</value>
4369 <description>
4370 Max number of entries in the vector group by aggregation hashtables.
4371 Exceeding this will trigger a flush irrelevant of memory pressure condition.
4372 </description>
4373 </property>
4374 <property>
4375 <name>hive.vectorized.groupby.flush.percent</name>
4376 <value>0.1</value>
4377 <description>Percent of entries in the group by aggregation hash flushed when the memory threshold is exceeded.</description>
4378 </property>
4379 <property>
4380 <name>hive.vectorized.execution.reducesink.new.enabled</name>
4381 <value>true</value>
4382 <description>
4383 This flag should be set to true to enable the new vectorization
4384 of queries using ReduceSink.
4385 iThe default value is true.
4386 </description>
4387 </property>
4388 <property>
4389 <name>hive.vectorized.use.vectorized.input.format</name>
4390 <value>true</value>
4391 <description>
4392 This flag should be set to true to enable vectorizing with vectorized input file format capable SerDe.
4393 The default value is true.
4394 </description>
4395 </property>
4396 <property>
4397 <name>hive.vectorized.use.vector.serde.deserialize</name>
4398 <value>false</value>
4399 <description>
4400 This flag should be set to true to enable vectorizing rows using vector deserialize.
4401 The default value is false.
4402 </description>
4403 </property>
4404 <property>
4405 <name>hive.vectorized.use.row.serde.deserialize</name>
4406 <value>false</value>
4407 <description>
4408 This flag should be set to true to enable vectorizing using row deserialize.
4409 The default value is false.
4410 </description>
4411 </property>
4412 <property>
4413 <name>hive.typecheck.on.insert</name>
4414 <value>true</value>
4415 <description>This property has been extended to control whether to check, convert, and normalize partition value to conform to its column type in partition operations including but not limited to insert, such as alter, describe etc.</description>
4416 </property>
4417 <property>
4418 <name>hive.hadoop.classpath</name>
4419 <value/>
4420 <description>
4421 For Windows OS, we need to pass HIVE_HADOOP_CLASSPATH Java parameter while starting HiveServer2
4422 using "-hiveconf hive.hadoop.classpath=%HIVE_LIB%".
4423 </description>
4424 </property>
4425 <property>
4426 <name>hive.rpc.query.plan</name>
4427 <value>false</value>
4428 <description>Whether to send the query plan via local resource or RPC</description>
4429 </property>
4430 <property>
4431 <name>hive.compute.splits.in.am</name>
4432 <value>true</value>
4433 <description>Whether to generate the splits locally or in the AM (tez only)</description>
4434 </property>
4435 <property>
4436 <name>hive.tez.input.generate.consistent.splits</name>
4437 <value>true</value>
4438 <description>Whether to generate consistent split locations when generating splits in the AM</description>
4439 </property>
4440 <property>
4441 <name>hive.prewarm.enabled</name>
4442 <value>false</value>
4443 <description>Enables container prewarm for Tez/Spark (Hadoop 2 only)</description>
4444 </property>
4445 <property>
4446 <name>hive.prewarm.numcontainers</name>
4447 <value>10</value>
4448 <description>Controls the number of containers to prewarm for Tez/Spark (Hadoop 2 only)</description>
4449 </property>
4450 <property>
4451 <name>hive.stageid.rearrange</name>
4452 <value>none</value>
4453 <description>
4454 Expects one of [none, idonly, traverse, execution].
4455 </description>
4456 </property>
4457 <property>
4458 <name>hive.explain.dependency.append.tasktype</name>
4459 <value>false</value>
4460 <description/>
4461 </property>
4462 <property>
4463 <name>hive.counters.group.name</name>
4464 <value>HIVE</value>
4465 <description>The name of counter group for internal Hive variables (CREATED_FILE, FATAL_ERROR, etc.)</description>
4466 </property>
4467 <property>
4468 <name>hive.support.quoted.identifiers</name>
4469 <value>column</value>
4470 <description>
4471 Expects one of [none, column].
4472 Whether to use quoted identifier. 'none' or 'column' can be used.
4473 none: default(past) behavior. Implies only alphaNumeric and underscore are valid characters in identifiers.
4474 column: implies column names can contain any character.
4475 </description>
4476 </property>
4477 <property>
4478 <name>hive.support.sql11.reserved.keywords</name>
4479 <value>true</value>
4480 <description>
4481 This flag should be set to true to enable support for SQL2011 reserved keywords.
4482 The default value is true.
4483 </description>
4484 </property>
4485 <property>
4486 <name>hive.support.special.characters.tablename</name>
4487 <value>true</value>
4488 <description>
4489 This flag should be set to true to enable support for special characters in table names.
4490 When it is set to false, only [a-zA-Z_0-9]+ are supported.
4491 The only supported special character right now is '/'. This flag applies only to quoted table names.
4492 The default value is true.
4493 </description>
4494 </property>
4495 <property>
4496 <name>hive.users.in.admin.role</name>
4497 <value/>
4498 <description>
4499 Comma separated list of users who are in admin role for bootstrapping.
4500 More users can be added in ADMIN role later.
4501 </description>
4502 </property>
4503 <property>
4504 <name>hive.compat</name>
4505 <value>0.12</value>
4506 <description>
4507 Enable (configurable) deprecated behaviors by setting desired level of backward compatibility.
4508 Setting to 0.12:
4509 Maintains division behavior: int / int = double
4510 </description>
4511 </property>
4512 <property>
4513 <name>hive.convert.join.bucket.mapjoin.tez</name>
4514 <value>false</value>
4515 <description>
4516 Whether joins can be automatically converted to bucket map joins in hive
4517 when tez is used as the execution engine.
4518 </description>
4519 </property>
4520 <property>
4521 <name>hive.exec.check.crossproducts</name>
4522 <value>true</value>
4523 <description>Check if a plan contains a Cross Product. If there is one, output a warning to the Session's console.</description>
4524 </property>
4525 <property>
4526 <name>hive.localize.resource.wait.interval</name>
4527 <value>5000ms</value>
4528 <description>
4529 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
4530 Time to wait for another thread to localize the same resource for hive-tez.
4531 </description>
4532 </property>
4533 <property>
4534 <name>hive.localize.resource.num.wait.attempts</name>
4535 <value>5</value>
4536 <description>The number of attempts waiting for localizing a resource in hive-tez.</description>
4537 </property>
4538 <property>
4539 <name>hive.tez.auto.reducer.parallelism</name>
4540 <value>false</value>
4541 <description>
4542 Turn on Tez' auto reducer parallelism feature. When enabled, Hive will still estimate data sizes
4543 and set parallelism estimates. Tez will sample source vertices' output sizes and adjust the estimates at runtime as
4544 necessary.
4545 </description>
4546 </property>
4547 <property>
4548 <name>hive.tez.max.partition.factor</name>
4549 <value>2.0</value>
4550 <description>When auto reducer parallelism is enabled this factor will be used to over-partition data in shuffle edges.</description>
4551 </property>
4552 <property>
4553 <name>hive.tez.min.partition.factor</name>
4554 <value>0.25</value>
4555 <description>
4556 When auto reducer parallelism is enabled this factor will be used to put a lower limit to the number
4557 of reducers that tez specifies.
4558 </description>
4559 </property>
4560 <property>
4561 <name>hive.tez.bucket.pruning</name>
4562 <value>false</value>
4563 <description>
4564 When pruning is enabled, filters on bucket columns will be processed by
4565 filtering the splits against a bitset of included buckets. This needs predicates
4566 produced by hive.optimize.ppd and hive.optimize.index.filters.
4567 </description>
4568 </property>
4569 <property>
4570 <name>hive.tez.bucket.pruning.compat</name>
4571 <value>true</value>
4572 <description>
4573 When pruning is enabled, handle possibly broken inserts due to negative hashcodes.
4574 This occasionally doubles the data scan cost, but is default enabled for safety
4575 </description>
4576 </property>
4577 <property>
4578 <name>hive.tez.dynamic.partition.pruning</name>
4579 <value>true</value>
4580 <description>
4581 When dynamic pruning is enabled, joins on partition keys will be processed by sending
4582 events from the processing vertices to the Tez application master. These events will be
4583 used to prune unnecessary partitions.
4584 </description>
4585 </property>
4586 <property>
4587 <name>hive.tez.dynamic.partition.pruning.max.event.size</name>
4588 <value>1048576</value>
4589 <description>Maximum size of events sent by processors in dynamic pruning. If this size is crossed no pruning will take place.</description>
4590 </property>
4591 <property>
4592 <name>hive.tez.dynamic.partition.pruning.max.data.size</name>
4593 <value>104857600</value>
4594 <description>Maximum total data size of events in dynamic pruning.</description>
4595 </property>
4596 <property>
4597 <name>hive.tez.smb.number.waves</name>
4598 <value>0.5</value>
4599 <description>The number of waves in which to run the SMB join. Account for cluster being occupied. Ideally should be 1 wave.</description>
4600 </property>
4601 <property>
4602 <name>hive.tez.exec.print.summary</name>
4603 <value>false</value>
4604 <description>Display breakdown of execution steps, for every query executed by the shell.</description>
4605 </property>
4606 <property>
4607 <name>hive.tez.exec.inplace.progress</name>
4608 <value>true</value>
4609 <description>Updates tez job execution progress in-place in the terminal.</description>
4610 </property>
4611 <property>
4612 <name>hive.llap.io.enabled</name>
4613 <value/>
4614 <description>Whether the LLAP IO layer is enabled.</description>
4615 </property>
4616 <property>
4617 <name>hive.llap.io.memory.mode</name>
4618 <value>cache</value>
4619 <description>
4620 Expects one of [cache, allocator, none].
4621 LLAP IO memory usage; 'cache' (the default) uses data and metadata cache with a
4622 custom off-heap allocator, 'allocator' uses the custom allocator without the caches,
4623 'none' doesn't use either (this mode may result in significant performance degradation)
4624 </description>
4625 </property>
4626 <property>
4627 <name>hive.llap.io.allocator.alloc.min</name>
4628 <value>16Kb</value>
4629 <description>
4630 Expects a byte size value with unit (blank for bytes, kb, mb, gb, tb, pb).
4631 Minimum allocation possible from LLAP buddy allocator. Allocations below that are
4632 padded to minimum allocation. For ORC, should generally be the same as the expected
4633 compression buffer size, or next lowest power of 2. Must be a power of 2.
4634 </description>
4635 </property>
4636 <property>
4637 <name>hive.llap.io.allocator.alloc.max</name>
4638 <value>16Mb</value>
4639 <description>
4640 Expects a byte size value with unit (blank for bytes, kb, mb, gb, tb, pb).
4641 Maximum allocation possible from LLAP buddy allocator. For ORC, should be as large as
4642 the largest expected ORC compression buffer size. Must be a power of 2.
4643 </description>
4644 </property>
4645 <property>
4646 <name>hive.llap.io.allocator.arena.count</name>
4647 <value>8</value>
4648 <description>
4649 Arena count for LLAP low-level cache; cache will be allocated in the steps of
4650 (size/arena_count) bytes. This size must be <= 1Gb and >= max allocation; if it is
4651 not the case, an adjusted size will be used. Using powers of 2 is recommended.
4652 </description>
4653 </property>
4654 <property>
4655 <name>hive.llap.io.memory.size</name>
4656 <value>1Gb</value>
4657 <description>
4658 Expects a byte size value with unit (blank for bytes, kb, mb, gb, tb, pb).
4659 Maximum size for IO allocator or ORC low-level cache.
4660 </description>
4661 </property>
4662 <property>
4663 <name>hive.llap.io.allocator.direct</name>
4664 <value>true</value>
4665 <description>Whether ORC low-level cache should use direct allocation.</description>
4666 </property>
4667 <property>
4668 <name>hive.llap.io.allocator.mmap</name>
4669 <value>false</value>
4670 <description>
4671 Whether ORC low-level cache should use memory mapped allocation (direct I/O).
4672 This is recommended to be used along-side NVDIMM (DAX) or NVMe flash storage.
4673 </description>
4674 </property>
4675 <property>
4676 <name>hive.llap.io.allocator.mmap.path</name>
4677 <value>/tmp</value>
4678 <description>
4679 Expects a writable directory on the local filesystem.
4680 The directory location for mapping NVDIMM/NVMe flash storage into the ORC low-level cache.
4681 </description>
4682 </property>
4683 <property>
4684 <name>hive.llap.io.use.lrfu</name>
4685 <value>true</value>
4686 <description>Whether ORC low-level cache should use LRFU cache policy instead of default (FIFO).</description>
4687 </property>
4688 <property>
4689 <name>hive.llap.io.lrfu.lambda</name>
4690 <value>0.01</value>
4691 <description>
4692 Lambda for ORC low-level cache LRFU cache policy. Must be in [0, 1]. 0 makes LRFU
4693 behave like LFU, 1 makes it behave like LRU, values in between balance accordingly.
4694 </description>
4695 </property>
4696 <property>
4697 <name>hive.llap.cache.allow.synthetic.fileid</name>
4698 <value>false</value>
4699 <description>
4700 Whether LLAP cache should use synthetic file ID if real one is not available. Systems
4701 like HDFS, Isilon, etc. provide a unique file/inode ID. On other FSes (e.g. local
4702 FS), the cache would not work by default because LLAP is unable to uniquely track the
4703 files; enabling this setting allows LLAP to generate file ID from the path, size and
4704 modification time, which is almost certain to identify file uniquely. However, if you
4705 use a FS without file IDs and rewrite files a lot (or are paranoid), you might want
4706 to avoid this setting.
4707 </description>
4708 </property>
4709 <property>
4710 <name>hive.llap.orc.gap.cache</name>
4711 <value>true</value>
4712 <description>
4713 Whether LLAP cache for ORC should remember gaps in ORC compression buffer read
4714 estimates, to avoid re-reading the data that was read once and discarded because it
4715 is unneeded. This is only necessary for ORC files written before HIVE-9660.
4716 </description>
4717 </property>
4718 <property>
4719 <name>hive.llap.io.use.fileid.path</name>
4720 <value>true</value>
4721 <description>
4722 Whether LLAP should use fileId (inode)-based path to ensure better consistency for the
4723 cases of file overwrites. This is supported on HDFS.
4724 </description>
4725 </property>
4726 <property>
4727 <name>hive.llap.io.orc.time.counters</name>
4728 <value>true</value>
4729 <description>Whether to enable time counters for LLAP IO layer (time spent in HDFS, etc.)</description>
4730 </property>
4731 <property>
4732 <name>hive.llap.auto.allow.uber</name>
4733 <value>false</value>
4734 <description>Whether or not to allow the planner to run vertices in the AM.</description>
4735 </property>
4736 <property>
4737 <name>hive.llap.auto.enforce.tree</name>
4738 <value>true</value>
4739 <description>Enforce that all parents are in llap, before considering vertex</description>
4740 </property>
4741 <property>
4742 <name>hive.llap.auto.enforce.vectorized</name>
4743 <value>true</value>
4744 <description>Enforce that inputs are vectorized, before considering vertex</description>
4745 </property>
4746 <property>
4747 <name>hive.llap.auto.enforce.stats</name>
4748 <value>true</value>
4749 <description>Enforce that col stats are available, before considering vertex</description>
4750 </property>
4751 <property>
4752 <name>hive.llap.auto.max.input.size</name>
4753 <value>10737418240</value>
4754 <description>Check input size, before considering vertex (-1 disables check)</description>
4755 </property>
4756 <property>
4757 <name>hive.llap.auto.max.output.size</name>
4758 <value>1073741824</value>
4759 <description>Check output size, before considering vertex (-1 disables check)</description>
4760 </property>
4761 <property>
4762 <name>hive.llap.skip.compile.udf.check</name>
4763 <value>false</value>
4764 <description>
4765 Whether to skip the compile-time check for non-built-in UDFs when deciding whether to
4766 execute tasks in LLAP. Skipping the check allows executing UDFs from pre-localized
4767 jars in LLAP; if the jars are not pre-localized, the UDFs will simply fail to load.
4768 </description>
4769 </property>
4770 <property>
4771 <name>hive.llap.allow.permanent.fns</name>
4772 <value>true</value>
4773 <description>Whether LLAP decider should allow permanent UDFs.</description>
4774 </property>
4775 <property>
4776 <name>hive.llap.execution.mode</name>
4777 <value>none</value>
4778 <description>
4779 Expects one of [auto, none, all, map].
4780 Chooses whether query fragments will run in container or in llap
4781 </description>
4782 </property>
4783 <property>
4784 <name>hive.llap.object.cache.enabled</name>
4785 <value>true</value>
4786 <description>Cache objects (plans, hashtables, etc) in llap</description>
4787 </property>
4788 <property>
4789 <name>hive.llap.io.decoding.metrics.percentiles.intervals</name>
4790 <value>30</value>
4791 <description>
4792 Comma-delimited set of integers denoting the desired rollover intervals (in seconds)
4793 for percentile latency metrics on the LLAP daemon IO decoding time.
4794 hive.llap.queue.metrics.percentiles.intervals
4795 </description>
4796 </property>
4797 <property>
4798 <name>hive.llap.io.threadpool.size</name>
4799 <value>10</value>
4800 <description>Specify the number of threads to use for low-level IO thread pool.</description>
4801 </property>
4802 <property>
4803 <name>hive.llap.daemon.service.principal</name>
4804 <value/>
4805 <description>The name of the LLAP daemon's service principal.</description>
4806 </property>
4807 <property>
4808 <name>hive.llap.daemon.keytab.file</name>
4809 <value/>
4810 <description>The path to the Kerberos Keytab file containing the LLAP daemon's service principal.</description>
4811 </property>
4812 <property>
4813 <name>hive.llap.zk.sm.principal</name>
4814 <value/>
4815 <description>The name of the principal to use to talk to ZooKeeper for ZooKeeper SecretManager.</description>
4816 </property>
4817 <property>
4818 <name>hive.llap.zk.sm.keytab.file</name>
4819 <value/>
4820 <description>
4821 The path to the Kerberos Keytab file containing the principal to use to talk to
4822 ZooKeeper for ZooKeeper SecretManager.
4823 </description>
4824 </property>
4825 <property>
4826 <name>hive.llap.zk.sm.connectionString</name>
4827 <value/>
4828 <description>ZooKeeper connection string for ZooKeeper SecretManager.</description>
4829 </property>
4830 <property>
4831 <name>hive.llap.zk.registry.user</name>
4832 <value/>
4833 <description>
4834 In the LLAP ZooKeeper-based registry, specifies the username in the Zookeeper path.
4835 This should be the hive user or whichever user is running the LLAP daemon.
4836 </description>
4837 </property>
4838 <property>
4839 <name>hive.llap.daemon.acl</name>
4840 <value>*</value>
4841 <description>The ACL for LLAP daemon.</description>
4842 </property>
4843 <property>
4844 <name>hive.llap.daemon.acl.blocked</name>
4845 <value/>
4846 <description>The deny ACL for LLAP daemon.</description>
4847 </property>
4848 <property>
4849 <name>hive.llap.management.acl</name>
4850 <value>*</value>
4851 <description>The ACL for LLAP daemon management.</description>
4852 </property>
4853 <property>
4854 <name>hive.llap.management.acl.blocked</name>
4855 <value/>
4856 <description>The deny ACL for LLAP daemon management.</description>
4857 </property>
4858 <property>
4859 <name>hive.llap.remote.token.requires.signing</name>
4860 <value>true</value>
4861 <description>
4862 Expects one of [false, except_llap_owner, true].
4863 Whether the token returned from LLAP management API should require fragment signing.
4864 True by default; can be disabled to allow CLI to get tokens from LLAP in a secure
4865 cluster by setting it to true or 'except_llap_owner' (the latter returns such tokens
4866 to everyone except the user LLAP cluster is authenticating under).
4867 </description>
4868 </property>
4869 <property>
4870 <name>hive.llap.daemon.delegation.token.lifetime</name>
4871 <value>14d</value>
4872 <description>
4873 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
4874 LLAP delegation token lifetime, in seconds if specified without a unit.
4875 </description>
4876 </property>
4877 <property>
4878 <name>hive.llap.management.rpc.port</name>
4879 <value>15004</value>
4880 <description>RPC port for LLAP daemon management service.</description>
4881 </property>
4882 <property>
4883 <name>hive.llap.auto.auth</name>
4884 <value>false</value>
4885 <description>Whether or not to set Hadoop configs to enable auth in LLAP web app.</description>
4886 </property>
4887 <property>
4888 <name>hive.llap.daemon.rpc.num.handlers</name>
4889 <value>5</value>
4890 <description>Number of RPC handlers for LLAP daemon.</description>
4891 </property>
4892 <property>
4893 <name>hive.llap.daemon.work.dirs</name>
4894 <value/>
4895 <description>
4896 Working directories for the daemon. Needs to be set for a secure cluster, since LLAP may
4897 not have access to the default YARN working directories. yarn.nodemanager.local-dirs is
4898 used if this is not set
4899 </description>
4900 </property>
4901 <property>
4902 <name>hive.llap.daemon.yarn.shuffle.port</name>
4903 <value>15551</value>
4904 <description>YARN shuffle port for LLAP-daemon-hosted shuffle.</description>
4905 </property>
4906 <property>
4907 <name>hive.llap.daemon.yarn.container.mb</name>
4908 <value>-1</value>
4909 <description>llap server yarn container size in MB. Used in LlapServiceDriver and package.py</description>
4910 </property>
4911 <property>
4912 <name>hive.llap.daemon.queue.name</name>
4913 <value/>
4914 <description>Queue name within which the llap slider application will run. Used in LlapServiceDriver and package.py</description>
4915 </property>
4916 <property>
4917 <name>hive.llap.daemon.container.id</name>
4918 <value/>
4919 <description>ContainerId of a running LlapDaemon. Used to publish to the registry</description>
4920 </property>
4921 <property>
4922 <name>hive.llap.daemon.shuffle.dir.watcher.enabled</name>
4923 <value>false</value>
4924 <description>TODO doc</description>
4925 </property>
4926 <property>
4927 <name>hive.llap.daemon.am.liveness.heartbeat.interval.ms</name>
4928 <value>10000ms</value>
4929 <description>
4930 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
4931 Tez AM-LLAP heartbeat interval (milliseconds). This needs to be below the task timeout
4932 interval, but otherwise as high as possible to avoid unnecessary traffic.
4933 </description>
4934 </property>
4935 <property>
4936 <name>hive.llap.am.liveness.connection.timeout.ms</name>
4937 <value>10000ms</value>
4938 <description>
4939 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
4940 Amount of time to wait on connection failures to the AM from an LLAP daemon before
4941 considering the AM to be dead.
4942 </description>
4943 </property>
4944 <property>
4945 <name>hive.llap.am.liveness.connection.sleep.between.retries.ms</name>
4946 <value>2000ms</value>
4947 <description>
4948 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
4949 Sleep duration while waiting to retry connection failures to the AM from the daemon for
4950 the general keep-alive thread (milliseconds).
4951 </description>
4952 </property>
4953 <property>
4954 <name>hive.llap.task.scheduler.timeout.seconds</name>
4955 <value>60s</value>
4956 <description>
4957 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
4958 Amount of time to wait before failing the query when there are no llap daemons running
4959 (alive) in the cluster.
4960 </description>
4961 </property>
4962 <property>
4963 <name>hive.llap.daemon.num.executors</name>
4964 <value>4</value>
4965 <description>
4966 Number of executors to use in LLAP daemon; essentially, the number of tasks that can be
4967 executed in parallel.
4968 </description>
4969 </property>
4970 <property>
4971 <name>hive.llap.daemon.rpc.port</name>
4972 <value>15001</value>
4973 <description>The LLAP daemon RPC port.</description>
4974 </property>
4975 <property>
4976 <name>hive.llap.daemon.memory.per.instance.mb</name>
4977 <value>4096</value>
4978 <description>The total amount of memory to use for the executors inside LLAP (in megabytes).</description>
4979 </property>
4980 <property>
4981 <name>hive.llap.daemon.vcpus.per.instance</name>
4982 <value>4</value>
4983 <description>The total number of vcpus to use for the executors inside LLAP.</description>
4984 </property>
4985 <property>
4986 <name>hive.llap.daemon.num.file.cleaner.threads</name>
4987 <value>1</value>
4988 <description>Number of file cleaner threads in LLAP.</description>
4989 </property>
4990 <property>
4991 <name>hive.llap.file.cleanup.delay.seconds</name>
4992 <value>300s</value>
4993 <description>
4994 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
4995 How long to delay before cleaning up query files in LLAP (in seconds, for debugging).
4996 </description>
4997 </property>
4998 <property>
4999 <name>hive.llap.daemon.service.hosts</name>
5000 <value/>
5001 <description>
5002 Explicitly specified hosts to use for LLAP scheduling. Useful for testing. By default,
5003 YARN registry is used.
5004 </description>
5005 </property>
5006 <property>
5007 <name>hive.llap.daemon.service.refresh.interval.sec</name>
5008 <value>60s</value>
5009 <description>
5010 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
5011 LLAP YARN registry service list refresh delay, in seconds.
5012 </description>
5013 </property>
5014 <property>
5015 <name>hive.llap.daemon.communicator.num.threads</name>
5016 <value>10</value>
5017 <description>Number of threads to use in LLAP task communicator in Tez AM.</description>
5018 </property>
5019 <property>
5020 <name>hive.llap.daemon.download.permanent.fns</name>
5021 <value>false</value>
5022 <description>Whether LLAP daemon should localize the resources for permanent UDFs.</description>
5023 </property>
5024 <property>
5025 <name>hive.llap.task.scheduler.node.reenable.min.timeout.ms</name>
5026 <value>200ms</value>
5027 <description>
5028 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
5029 Minimum time after which a previously disabled node will be re-enabled for scheduling,
5030 in milliseconds. This may be modified by an exponential back-off if failures persist.
5031 </description>
5032 </property>
5033 <property>
5034 <name>hive.llap.task.scheduler.node.reenable.max.timeout.ms</name>
5035 <value>10000ms</value>
5036 <description>
5037 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
5038 Maximum time after which a previously disabled node will be re-enabled for scheduling,
5039 in milliseconds. This may be modified by an exponential back-off if failures persist.
5040 </description>
5041 </property>
5042 <property>
5043 <name>hive.llap.task.scheduler.node.disable.backoff.factor</name>
5044 <value>1.5</value>
5045 <description>
5046 Backoff factor on successive blacklists of a node due to some failures. Blacklist times
5047 start at the min timeout and go up to the max timeout based on this backoff factor.
5048 </description>
5049 </property>
5050 <property>
5051 <name>hive.llap.task.scheduler.num.schedulable.tasks.per.node</name>
5052 <value>0</value>
5053 <description>
5054 The number of tasks the AM TaskScheduler will try allocating per node. 0 indicates that
5055 this should be picked up from the Registry. -1 indicates unlimited capacity; positive
5056 values indicate a specific bound.
5057 </description>
5058 </property>
5059 <property>
5060 <name>hive.llap.task.scheduler.locality.delay</name>
5061 <value>0ms</value>
5062 <description>
5063 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
5064 The time should be in between -1 msec (inclusive) and 9223372036854775807 msec (inclusive).
5065 Amount of time to wait before allocating a request which contains location information, to a location other than the ones requested. Set to -1 for an infinite delay, 0for no delay.
5066 </description>
5067 </property>
5068 <property>
5069 <name>hive.llap.daemon.task.preemption.metrics.intervals</name>
5070 <value>30,60,300</value>
5071 <description>
5072 Comma-delimited set of integers denoting the desired rollover intervals (in seconds)
5073 for percentile latency metrics. Used by LLAP daemon task scheduler metrics for
5074 time taken to kill task (due to pre-emption) and useful time wasted by the task that
5075 is about to be preempted.
5076 </description>
5077 </property>
5078 <property>
5079 <name>hive.llap.daemon.task.scheduler.wait.queue.size</name>
5080 <value>10</value>
5081 <description>LLAP scheduler maximum queue size.</description>
5082 </property>
5083 <property>
5084 <name>hive.llap.daemon.wait.queue.comparator.class.name</name>
5085 <value>org.apache.hadoop.hive.llap.daemon.impl.comparator.ShortestJobFirstComparator</value>
5086 <description>
5087 The priority comparator to use for LLAP scheduler prioroty queue. The built-in options
5088 are org.apache.hadoop.hive.llap.daemon.impl.comparator.ShortestJobFirstComparator and
5089 .....FirstInFirstOutComparator
5090 </description>
5091 </property>
5092 <property>
5093 <name>hive.llap.daemon.task.scheduler.enable.preemption</name>
5094 <value>true</value>
5095 <description>
5096 Whether non-finishable running tasks (e.g. a reducer waiting for inputs) should be
5097 preempted by finishable tasks inside LLAP scheduler.
5098 </description>
5099 </property>
5100 <property>
5101 <name>hive.llap.task.communicator.connection.timeout.ms</name>
5102 <value>16000ms</value>
5103 <description>
5104 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
5105 Connection timeout (in milliseconds) before a failure to an LLAP daemon from Tez AM.
5106 </description>
5107 </property>
5108 <property>
5109 <name>hive.llap.task.communicator.connection.sleep.between.retries.ms</name>
5110 <value>2000ms</value>
5111 <description>
5112 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
5113 Sleep duration (in milliseconds) to wait before retrying on error when obtaining a
5114 connection to LLAP daemon from Tez AM.
5115 </description>
5116 </property>
5117 <property>
5118 <name>hive.llap.daemon.web.port</name>
5119 <value>15002</value>
5120 <description>LLAP daemon web UI port.</description>
5121 </property>
5122 <property>
5123 <name>hive.llap.daemon.web.ssl</name>
5124 <value>false</value>
5125 <description>Whether LLAP daemon web UI should use SSL.</description>
5126 </property>
5127 <property>
5128 <name>hive.llap.client.consistent.splits</name>
5129 <value>false</value>
5130 <description>Whether to setup split locations to match nodes on which llap daemons are running, instead of using the locations provided by the split itself</description>
5131 </property>
5132 <property>
5133 <name>hive.llap.validate.acls</name>
5134 <value>true</value>
5135 <description>
5136 Whether LLAP should reject permissive ACLs in some cases (e.g. its own management
5137 protocol or ZK paths), similar to how ssh refuses a key with bad access permissions.
5138 </description>
5139 </property>
5140 <property>
5141 <name>hive.llap.daemon.output.service.port</name>
5142 <value>15003</value>
5143 <description>LLAP daemon output service port</description>
5144 </property>
5145 <property>
5146 <name>hive.llap.daemon.output.service.send.buffer.size</name>
5147 <value>131072</value>
5148 <description>Send buffer size to be used by LLAP daemon output service</description>
5149 </property>
5150 <property>
5151 <name>hive.llap.enable.grace.join.in.llap</name>
5152 <value>false</value>
5153 <description>Override if grace join should be allowed to run in llap.</description>
5154 </property>
5155 <property>
5156 <name>hive.spark.client.future.timeout</name>
5157 <value>60s</value>
5158 <description>
5159 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
5160 Timeout for requests from Hive client to remote Spark driver.
5161 </description>
5162 </property>
5163 <property>
5164 <name>hive.spark.job.monitor.timeout</name>
5165 <value>60s</value>
5166 <description>
5167 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is sec if not specified.
5168 Timeout for job monitor to get Spark job state.
5169 </description>
5170 </property>
5171 <property>
5172 <name>hive.spark.client.connect.timeout</name>
5173 <value>1000ms</value>
5174 <description>
5175 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
5176 Timeout for remote Spark driver in connecting back to Hive client.
5177 </description>
5178 </property>
5179 <property>
5180 <name>hive.spark.client.server.connect.timeout</name>
5181 <value>90000ms</value>
5182 <description>
5183 Expects a time value with unit (d/day, h/hour, m/min, s/sec, ms/msec, us/usec, ns/nsec), which is msec if not specified.
5184 Timeout for handshake between Hive client and remote Spark driver. Checked by both processes.
5185 </description>
5186 </property>
5187 <property>
5188 <name>hive.spark.client.secret.bits</name>
5189 <value>256</value>
5190 <description>Number of bits of randomness in the generated secret for communication between Hive client and remote Spark driver. Rounded down to the nearest multiple of 8.</description>
5191 </property>
5192 <property>
5193 <name>hive.spark.client.rpc.threads</name>
5194 <value>8</value>
5195 <description>Maximum number of threads for remote Spark driver's RPC event loop.</description>
5196 </property>
5197 <property>
5198 <name>hive.spark.client.rpc.max.size</name>
5199 <value>52428800</value>
5200 <description>Maximum message size in bytes for communication between Hive client and remote Spark driver. Default is 50MB.</description>
5201 </property>
5202 <property>
5203 <name>hive.spark.client.channel.log.level</name>
5204 <value/>
5205 <description>Channel logging level for remote Spark driver. One of {DEBUG, ERROR, INFO, TRACE, WARN}.</description>
5206 </property>
5207 <property>
5208 <name>hive.spark.client.rpc.sasl.mechanisms</name>
5209 <value>DIGEST-MD5</value>
5210 <description>Name of the SASL mechanism to use for authentication.</description>
5211 </property>
5212 <property>
5213 <name>hive.spark.client.rpc.server.address</name>
5214 <value/>
5215 <description>The server address of HiverServer2 host to be used for communication between Hive client and remote Spark driver. Default is empty, which means the address will be determined in the same way as for hive.server2.thrift.bind.host.This is only necessary if the host has mutiple network addresses and if a different network address other than hive.server2.thrift.bind.host is to be used.</description>
5216 </property>
5217 <property>
5218 <name>hive.spark.dynamic.partition.pruning</name>
5219 <value>false</value>
5220 <description>
5221 When dynamic pruning is enabled, joins on partition keys will be processed by writing
5222 to a temporary HDFS file, and read later for removing unnecessary partitions.
5223 </description>
5224 </property>
5225 <property>
5226 <name>hive.spark.dynamic.partition.pruning.max.data.size</name>
5227 <value>104857600</value>
5228 <description>Maximum total data size in dynamic pruning.</description>
5229 </property>
5230 <property>
5231 <name>hive.reorder.nway.joins</name>
5232 <value>true</value>
5233 <description>Runs reordering of tables within single n-way join (i.e.: picks streamtable)</description>
5234 </property>
5235 <property>
5236 <name>hive.log.every.n.records</name>
5237 <value>0</value>
5238 <description>
5239 Expects value bigger than 0.
5240 If value is greater than 0 logs in fixed intervals of size n rather than exponentially.
5241 </description>
5242 </property>
5243 <property>
5244 <name>hive.msck.path.validation</name>
5245 <value>throw</value>
5246 <description>
5247 Expects one of [throw, skip, ignore].
5248 The approach msck should take with HDFS directories that are partition-like but contain unsupported characters. 'throw' (an exception) is the default; 'skip' will skip the invalid directories and still repair the others; 'ignore' will skip the validation (legacy behavior, causes bugs in many cases)
5249 </description>
5250 </property>
5251 <property>
5252 <name>hive.server2.llap.concurrent.queries</name>
5253 <value>-1</value>
5254 <description>The number of queries allowed in parallel via llap. Negative number implies 'infinite'.</description>
5255 </property>
5256 <property>
5257 <name>hive.tez.enable.memory.manager</name>
5258 <value>true</value>
5259 <description>Enable memory manager for tez</description>
5260 </property>
5261 <property>
5262 <name>hive.hash.table.inflation.factor</name>
5263 <value>2.0</value>
5264 <description>Expected inflation factor between disk/in memory representation of hash tables</description>
5265 </property>
5266 <property>
5267 <name>hive.log.trace.id</name>
5268 <value/>
5269 <description>Log tracing id that can be used by upstream clients for tracking respective logs. Truncated to 64 characters. Defaults to use auto-generated session id.</description>
5270 </property>
5271 <property>
5272 <name>hive.conf.restricted.list</name>
5273 <value>hive.security.authenticator.manager,hive.security.authorization.manager,hive.users.in.admin.role,hive.server2.xsrf.filter.enabled</value>
5274 <description>Comma separated list of configuration options which are immutable at runtime</description>
5275 </property>
5276 <property>
5277 <name>hive.conf.hidden.list</name>
5278 <value>javax.jdo.option.ConnectionPassword,hive.server2.keystore.password</value>
5279 <description>Comma separated list of configuration options which should not be read by normal user like passwords</description>
5280 </property>
5281 <property>
5282 <name>hive.conf.internal.variable.list</name>
5283 <value>hive.added.files.path,hive.added.jars.path,hive.added.archives.path</value>
5284 <description>Comma separated list of variables which are used internally and should not be configurable.</description>
5285 </property>
5286</configuration>