Apache Solr Reference Guide Covering Apache Solr 6.0

More documents

Recommendations

Info

socketTimeout 0 (use OS default) The amount of time in ms that a socket is allowed to wait. connTimeout 0 (use OS default) The amount of time in ms that is accepted for binding / connecting a socket maxConnectionsPerHost 20 The maximum number of concurrent connections that is made to each individual shard in a distributed search. maxConnections 10000 The total maximum number of concurrent connections in distributed searches. corePoolSize 0 The retained lowest limit on the number of threads used in coordinating distributed search. maximumPoolSize Integer.MAX_VALUE The maximum number of threads used for coordinating distributed search. maxThreadIdleTime 5 seconds The amount of time to wait for before threads are scaled back in response to a reduction in load. sizeOfQueue -1 If specified, the thread pool will use a backing queue instead of a direct handoff buffer. High throughput systems will want to configure this to be a direct hand off (with -1). Systems that desire better latency will want to configure a reasonable size of queue to handle variations in requests. fairnessPolicy false Chooses the JVM specifics dealing with fair policy queuing, if enabled distributed searches will be handled in a First in First out fashion at a cost to throughput. If disabled throughput will be favored over latency. Configuring statsCache (Distributed IDF) Document and term statistics are needed in order to calculate relevancy. Solr provides four implementations out of the box when it comes to document stats calculation: LocalStatsCache: This only uses local term and document statistics to compute relevance. In cases with uniform term distribution across shards, this works reasonably well. This option is the default if no is configured. ExactStatsCache: This implementation uses global values (across the collection) for document frequency. ExactSharedStatsCache: This is exactly like the exact stats cache in it's functionality but the global stats are reused for subsequent requests with the same terms. LRUStatsCache: This implementation uses an LRU cache to hold global stats, which are shared between requests. The implementation can be selected by setting in solrconfig.xml. For example, the following line makes Solr use the ExactStatsCache implementation: Avoiding Distributed Deadlock Apache Solr Reference Guide 6.0 552
Each shard serves top-level query requests and then makes sub-requests to all of the other shards. Care should be taken to ensure that the max number of threads serving HTTP requests is greater than the possible number of requests from both top-level clients and other shards. If this is not the case, the configuration may result in a distributed deadlock. For example, a deadlock might occur in the case of two shards, each with just a single thread to service HTTP requests. Both threads could receive a top-level request concurrently, and make sub-requests to each other. Because there are no more remaining threads to service requests, the incoming requests will be blocked until the other pending requests are finished, but they will not finish since they are waiting for the sub-requests. By ensuring that Solr is configured to handle a sufficient number of threads, you can avoid deadlock situations like this. Prefer Local Shards Solr allows you to pass an optional boolean parameter named preferLocalShards to indicate that a distributed query should prefer local replicas of a shard when available. In other words, if a query includes prefe rLocalShards=true, then the query controller will look for local replicas to service the query instead of selecting replicas at random from across the cluster. This is useful when a query requests many fields or large fields to be returned per document because it avoids moving large amounts of data over the network when it is available locally. In addition, this feature can be useful for minimizing the impact of a problematic replica with degraded performance, as it reduces the likelihood that the degraded replica will be hit by other healthy replicas. Lastly, it follows that the value of this feature diminishes as the number of shards in a collection increases because the query controller will have to direct the query to non-local replicas for most of the shards. In other words, this feature is mostly useful for optimizing queries directed towards collections with a small number of shards and many replicas. Also, this option should only be used if you are load balancing requests across all nodes that host replicas for the collection you are querying, as Solr's CloudSolrClient will do. If not load-balancing, this feature can introduce a hotspot in the cluster since queries won't be evenly distributed across the cluster. Read and Write Side Fault Tolerance SolrCloud supports elasticity, high availability, and fault tolerance in reads and writes. What this means, basically, is that when you have a large cluster, you can always make requests to the cluster: Reads will return results whenever possible, even if some nodes are down, and Writes will be acknowledged only if they are durable; i.e., you won't lose data. Read Side Fault Tolerance In a SolrCloud cluster each individual node load balances read requests across all the replicas in collection. You still need a load balancer on the 'outside' that talks to the cluster, or you need a smart client which understands how to read and interact with Solr's metadata in ZooKeeper and only requests the ZooKeeper ensemble's address to start discovering to which nodes it should send requests. (Solr provides a smart Java SolrJ client called CloudSolrClient.) Even if some nodes in the cluster are offline or unreachable, a Solr node will be able to correctly respond to a search request as long as it can communicate with at least one replica of every shard, or one replica of every rel evant shard if the user limited the search via the ' shards ' or '_route_' parameters. The more replicas there are of every shard, the more likely that the Solr cluster will be able to handle search results in the event of node failures. zkConnected A Solr node will return the results of a search request as long as it can communicate with at least one replica of Apache Solr Reference Guide 6.0 553
Page 1 and 2:
TM Apache Solr Reference Guide Cove
Page 3 and 4:
About This Guide This guide describ
Page 5 and 6:
Getting Started Solr makes it easy
Page 7 and 8:
Start Solr in the Foreground Since
Page 9 and 10:
documents. Do not worry too much ab
Page 11 and 12:
,price Faceted browsing is one of S
Page 13 and 14:
for the simplest keyword searching
Page 15 and 16:
Stop Informational Version Status H
Page 17 and 18:
-noprompt Start Solr and suppress a
Page 19 and 20:
The stop command sends a STOP reque
Page 21 and 22:
$ bin/solr healthcheck -c gettingst
Page 23 and 24:
collection. For example, the follow
Page 25 and 26:
An example of this command with the
Page 27 and 28:
with the fact that DocValues do not
Page 29 and 30:
Accessing the URL http://hostname:8
Page 31 and 32:
For an explanation of the various l
Page 33 and 34:
Java Properties The Java Properties
Page 35 and 36:
The collection-specific UI screens
Page 37 and 38:
This screen also lets you adjust va
Page 39 and 40:
Related Topics Uploading Data with
Page 41 and 42:
Request-handler (qt) Specifies the
Page 43 and 44:
about the core, and a secondary men
Page 45 and 46:
0 13 {!lucene}*:* false _text_ 10
Page 47 and 48:
This information may be useful for
Page 49 and 50:
In the Solr universe, documents are
Page 51 and 52:
The first line in the
Page 53 and 54:
omitNorms If true, omits the norms
Page 55 and 56:
TrieField TrieFloatField TrieIntFie
Page 57 and 58:
Working with Dates Date Formatting
Page 59 and 60:
UTC: http://localhost:8983/solr/my_
Page 61 and 62:
Working with External Files and Pro
Page 63 and 64:
str Stored string value of a field.
Page 65 and 66:
1 one two three version: 1 stored:
Page 67 and 68:
faceting 5 true 7 true 7 add multip
Page 69 and 70:
useDocValuesAsStored If the field h
Page 71 and 72:
Similarity Similarity is a Lucene c
Page 73 and 74:
Multiple Commands in a Single POST
Page 75 and 76:
Add a Dynamic Field Rule The add-dy
Page 77 and 78:
The replace-field-type command repl
Page 79 and 80:
Finally, repeated commands can be s
Page 81 and 82:
81 Apache Solr Reference Guide 6.0
Page 83 and 84:
id ... List Fields GET
Page 85 and 86:
INPUT Path Parameters Key Descripti
Page 87 and 88:
The query parameters can be added t
Page 89 and 90:
Key collection Description The coll
Page 91 and 92:
curl http://localhost:8983/solr/get
Page 93 and 94:
OUTPUT Output Content The output wi
Page 95 and 96:
Handling text properly will make yo
Page 97 and 98:
Retrieving DocValues During Search
Page 99 and 100:
true managed-schema Define an Upda
Page 101 and 102:
Javadocs for update processor facto
Page 103 and 104:
400 7 ERROR: [doc=19F] Error add
Page 105 and 106:
Analyzers An analyzer examines the
Page 107 and 108:
About Tokenizers The
Page 109 and 110:
You configure the tokenizer for a t
Page 111 and 112:
In: "Please, email john.doe@foo.c
Page 113 and 114:
In: "babaloo" Out:"ba", "bab", "b
Page 115 and 116:
Extract simple, capitalized words.
Page 117 and 118:
lement. For example: The foll
Page 119 and 120:
Example: Classic Filter This f
Page 121 and 122:
Tokenizer to Filter: "four"(1), "sc
Page 123 and 124:
In: "the quick brown fox jumped
Page 125 and 126:
name: (string) The name of the norm
Page 127 and 128:
appropriate for English language te
Page 129 and 130:
N-Gram Filter Generates n-gram toke
Page 131 and 132:
In: "cat concatenate catycat" To
Page 133 and 134:
In: "jump jumping jumped" Tokenizer
Page 135 and 136:
In: "To be, or not to be." Token
Page 137 and 138:
enablePositionIncrements: if lucene
Page 139 and 140:
couch,sofa,divan teh => the huge,gi
Page 141 and 142:
useWhitelist: If true, the file def
Page 143 and 144:
Tokenizer to Filter: "XL-4000/ES"(1
Page 145 and 146:
The table below presents examples o
Page 147 and 148:
KeywordRepeatFilterFactory Emits ea
Page 149 and 150:
Using a Tailored ruleset: custom: (
Page 151 and 152:
get the default rules for Germany /
Page 153 and 154:
Language-Specific Factories Thes
Page 155 and 156:
Arguments: language: (required) ste
Page 157 and 158:
Out: "preziden", "preziden", "prezi
Page 159 and 160:
Examples: In: "le chat,
Page 161 and 162:
Example: Indonesian Solr incl
Page 163 and 164:
Japanese Tokenizer Tokenizer for Ja
Page 165 and 166:
In: "tirgiem tirgus" Tokenize
Page 167 and 168:
Polish Solr provides support fo
Page 169 and 170:
Factory class: solr.RussianLightSte
Page 171 and 172:
Spanish Solr includes two stemmers
Page 173 and 174:
Algorithms discussed in this sectio
Page 175 and 176:
Caverphone To use this encoding in
Page 177 and 178:
The " text_general" field type is d
Page 179 and 180:
Indexing and Basic Data Operations
Page 181 and 182:
$ bin/post -h Usage: post -c [OPTI
Page 183 and 184:
The bin/post script currently deleg
Page 185 and 186:
XML Update Commands Commit and Opti
Page 187 and 188:

Page 189 and 190:
In general, the JSON update syntax
Page 191 and 192:
if split=/ (i.e., you want your JSO
Page 193 and 194:
curl 'http://localhost:8983/solr/my
Page 195 and 196:
So, if no params are passed, the en
Page 197 and 198:
Path /update/csv Default Parameters
Page 199 and 200:
Topics covered in this section: Key
Page 201 and 202:
defaultField extractOnly extractFor
Page 203 and 204:
EEEE, dd-MMM-yy HH:mm:ss zzz EEE MM
Page 205 and 206:
curl "http://localhost:8983/solr/te
Page 207 and 208:
Configuration Configuring solrconfi
Page 209 and 210:
Datasources can still be specifie
Page 211 and 212:
dateFormat A java.text.SimpleDateFo
Page 213 and 214:
The URLDataSource type accepts thes
Page 215 and 216:
deletedPkQuery deltaImportQuery SQL
Page 217 and 218:
The entity attributes unique to the
Page 219 and 220:
onError By default, the TikaEntityP
Page 221 and 222:
While there are use cases where you
Page 223 and 224:
The HTMLStripTransformer You can us
Page 225 and 226:
The emailids field in the table can
Page 227 and 228:
set Modifier Usage Set or replace t
Page 229 and 230:
$ curl -X POST -H 'Content-Type: ap
Page 231 and 232:
Method MD5Signature Lookup3Signatur
Page 233 and 234:
language identification and a field
Page 235 and 236:
langid.fallbackFields string none n
Page 237 and 238:
1. Copy solr-uima-VERSION.jar (unde
Page 239 and 240:
VALID_ALCHEMYAPI_KEY is your Alch
Page 241 and 242:
Overview of Searching in Solr Solr
Page 243 and 244:
Velocity Search UI Solr includes a
Page 245 and 246:
The table below summarizes Solr's c
Page 247 and 248:
Setting the start parameter to some
Page 249 and 250:
The debug Parameter The debug param
Page 251 and 252:
The echoParams Parameter The echoPa
Page 253 and 254:
The Standard Query Parser's Respons
Page 255 and 256:
Multiple characters (matches zero o
Page 257 and 258:
field. For example, suppose an inde
Page 259 and 260:
Grouping Terms to Form Sub-Queries
Page 261 and 262:
In addition to the common request p
Page 263 and 264:
If the calculations based on the pa
Page 265 and 266:
http://localhost:8983/solr/techprod
Page 267 and 268:
Examples of Queries Submitted to th
Page 269 and 270:
it includes parentheses. For exampl
Page 271 and 272:
dist docfreq(field,val) exists fiel
Page 273 and 274:
max maxdoc min ms norm( field) not
Page 275 and 276:
ecip rord scale sqedist Performs a
Page 277 and 278:
tf top totaltermfreq xor() Term fre
Page 279 and 280:
is equivalent to: q={!type=dismax q
Page 281 and 282:
This parser takes a query that matc
Page 283 and 284:
ds.txt for your collection, and ind
Page 285 and 286:
One way to model this graph as Solr
Page 287 and 288:
With this alternative document mode
Page 289 and 290:
y wrapping all SHOULD clauses in a
Page 291 and 292:
Re-Ranking Query Parser The ReRankQ
Page 293 and 294:
example configuration below, client
Page 295 and 296:
TermsQueryBuilder UserInputQueryB
Page 297 and 298:
The facet.field parameter identifie
Page 299 and 300:
The default value is fc (except for
Page 301 and 302:
The facet.range.hardend parameter i
Page 303 and 304:
Using the " bin/solr -e techproduct
Page 305 and 306:
"price":{ "min":479.95001220703125,
Page 307 and 308:
307 Apache Solr Reference Guide 6.0
Page 309 and 310:
... "start":"2006-01-01T00:00:00Z",
Page 311 and 312:
and exclude those filters when face
Page 313 and 314:
document sample 1 parent 11 Red
Page 315 and 316:
Standard Highlighter The standard h
Page 317 and 318:
hl.usePhraseHighlighter true If set
Page 319 and 320:
hl.maxMultiValuedToExamine integer.
Page 321 and 322:
Related Content HighlightingParamet
Page 323 and 324:
hl.score.b 0.75 Specifies BM25 leng
Page 325 and 326:
Levenshtein metric, which is the sa
Page 327 and 328:
spellcheck Turns on or off SpellChe
Page 329 and 330:
The spellcheck.alternativeTermCount
Page 331 and 332:
1 0 5 0 dell 1 1 6 17 0 u
Page 333 and 334:
Transforming Result Documents Docum
Page 335 and 336:
childFilter - query to filter which
Page 337 and 338:
name lookupImpl dictionaryImpl A sy
Page 339 and 340:
allTermsRequired: Boolean option fo
Page 341 and 342:
This dictionary implementation take
Page 343 and 344:
Example query: http://localhost:898
Page 345 and 346:
solrconfig.xml mySuggester Analyz
Page 347 and 348:
paging, and filtering using common
Page 349 and 350:
Fetching A Large Number of Sorted R
Page 351 and 352:
$ curl '...&rows=10&sort=id+asc&cur
Page 353 and 354:
The psuedo-code for tailing a curso
Page 355 and 356:
group.facet Boolean Determines whet
Page 357 and 358:
Page 359 and 360:
The Collapsing query parser and the
Page 361 and 362:
q=foo&fq={!collapse field=ISBN}&exp
Page 363 and 364:
To enable the clustering component
Page 365 and 366:
There were a few clusters discover
Page 367 and 368:
clustering When true, clustering co
Page 369 and 370:
The default language can also be se
Page 371 and 372:
See the section SpatialRecursivePre
Page 373 and 374:
sphere; and sqedist, to calculate t
Page 375 and 376:
maxDistErr worldBounds Defines the
Page 377 and 378:
interface being developed allows an
Page 379 and 380:
The parameters below allow you to c
Page 381 and 382:
0 2 5 3 3 3 3 3 3 3 3 3 Get
Page 383 and 384:
The TermVectorComponent is a search
Page 385 and 386:
Boolean Parameters Description Type
Page 387 and 388:
0.0 3.0 32 0 10.0 22.0 0.3125 0.7
Page 389 and 390:
Additional "Expert" local params ar
Page 391 and 392:
config-file Path to the file that d
Page 393 and 394:
csv json php phps python ruby smile
Page 395 and 396:
text/plain Python Response Writer
Page 397 and 398:
Page 399 and 400:
v.layout.enabled v.contentType v.js
Page 401 and 402:
maxDocs maxTime Integer. Defines th
Page 403 and 404:
${solr.ulog.dir:} Real Time Get re
Page 405 and 406:
To export the full sorted result se
Page 407 and 408:
Some streaming functions act as str
Page 409 and 410:
expr=search(collection1, zkHost="lo
Page 411 and 412:
facet(collection1, q="*:*", buckets
Page 413 and 414:
Parameters Syntax StreamExpression
Page 415 and 416:
-1, // checkpoint every X tuples, i
Page 417 and 418:
The order of the streams does not m
Page 419 and 420:
intersect( search(collection1, q=a_
Page 421 and 422:
outerHashJoin( search(people, q=*:*
Page 423 and 424:
Parameters Syntax StreamExpression
Page 425 and 426:
unique( search(collection1, q="*:*"
Page 427 and 428:
worker nodes. It involves sorting a
Page 429 and 430:
{"result-set":{"docs":[ {"count(*)"
Page 431 and 432:
engine and are generated by the Sta
Page 433 and 434:
the Data Tables. The worker nodes e
Page 435 and 436:
The Well-Configured Solr Instance T
Page 437 and 438:
#core.properties name=collection2 m
Page 439 and 440:
Managed Schema Default When a is n
Page 441 and 442:
Merging Index Segments mergePolicyF
Page 443 and 444:
lockType The LockFactory options sp
Page 445 and 446:
own request handler, you should mak
Page 447 and 448:
used instead of the default. First-
Page 449 and 450:
We've defined three paths with this
Page 451 and 452:
1000 commitWithin The commitWithin
Page 453 and 454:
accessed frequently tend to stay in
Page 455 and 456:
the filterCache will be checked for
Page 457 and 458:
equestParsers Element The sub-elem
Page 459 and 460:
2. required between requests. The f
Page 461 and 462:
updateRequestProcessorChains and up
Page 463 and 464:
Constructing a chain at request tim
Page 465 and 466:
stripping index.html), the domain a
Page 467 and 468:
In Solr, the term core is used to r
Page 469 and 470:
shareSchema transientCacheSize conf
Page 471 and 472:
name=my_core_name Placement of core
Page 473 and 474:
CREATE RELOAD RENAME SWAP UNLOAD ME
Page 475 and 476:
RELOAD The RELOAD action loads a ne
Page 477 and 478:
In this example, we use the indexDi
Page 479 and 480:
Parameter Type Required Default Des
Page 481 and 482:
{ } "response":{"numFound":1,"start
Page 483 and 484:
Commands for Common Properties The
Page 485 and 486:
The Config API does not let you cre
Page 487 and 488:
Examples Creating and Updating Comm
Page 489 and 490:
{"overlay":{ "znodeVersion":5, "use
Page 491 and 492:
curl http://localhost:8983/solr/tec
Page 493 and 494:
Managed Resources Managed resources
Page 495 and 496:
Synonyms For the most part, the API
Page 497 and 498:
{ "responseHeader":{ "status":0, "Q
Page 499 and 500:
curl http://localhost:8983/solr/tec
Page 501 and 502: the signature of the jar that you g
Page 503 and 504: Managing Solr This section describe
Page 505 and 506: and runs Solr as the solr user. Con
Page 507 and 508: Solr process PID running on port 89
Page 509 and 510: _JMX_OPTS property in the include f
Page 511 and 512: security.json { } "authentication"
Page 513 and 514: during initialization. The authoriz
Page 515 and 516: curl --user solr:SolrRocks http://l
Page 517 and 518: Consult Your Kerberos Admins! Befor
Page 519 and 520: If you already have a /security.jso
Page 521 and 522: in/solr -c -z server1:2181,server2:
Page 523 and 524: st Parameters API, and other APIs w
Page 525 and 526: curl --user solr:SolrRocks -H 'Cont
Page 527 and 528: in/solr.in.sh example SOLR_SSL_* co
Page 529 and 530: working through the previous sectio
Page 531 and 532: { "responseHeader":{ "status":0, "Q
Page 533 and 534: SolrCloud Instances In SolrCloud mo
Page 535 and 536: authentication from Solr, you need
Page 537 and 538: The backup command is an asynchrono
Page 539 and 540: In addition to the logging options
Page 541 and 542: log4j.appender.file.MaxFileSize=100
Page 543 and 544: MBean Request Handler The MBean Req
Page 545 and 546: SolrCloud Example Interactive Start
Page 547 and 548: startup a SolrCloud cluster using t
Page 549 and 550: Then at query time, you include the
Page 551: http://localhost:8983/solr/gettings
Page 555 and 556: Write Side Fault Tolerance SolrClou
Page 557 and 558: Create the instance Creating the in
Page 559 and 560: Finally, create your myid files in
Page 561 and 562: Changing ACL Schemes Example Usages
Page 563 and 564: When solr wants to create a new zno
Page 565 and 566: set SOLR_ZK_CREDS_AND_ACLS=-DzkDige
Page 567 and 568: maxShardsPerNode integer No 1 When
Page 569 and 570: It's possible to edit multiple attr
Page 571 and 572: Query Parameters Key Type Required
Page 573 and 574: Create a Shard Shards can only crea
Page 575 and 576: The CREATEALIAS action will create
Page 577 and 578: 0 603 0 19 0 67 Delete
Page 579 and 580: http://localhost:8983/solr/admin/co
Page 581 and 582: The response will include the statu
Page 583 and 584: 0 5007 0 8 0 1 test2_shard
Page 585 and 586: Key Type Required Description role
Page 587 and 588: Cluster Status /admin/collections?a
Page 589 and 590: } } "127.0.1.1:8900_solr"] Request
Page 595 and 596: 0 9 Examining the clusterstate a
Page 597 and 598: 0 123 success Already leader 19
Page 601 and 602: -z -zkhost -c -collection -d -con
Page 603 and 604:
... ${solr.data.dir:} ... 3. The
Page 605 and 606:
will explain why the request failed
Page 607 and 608:
The nodes are sorted first and the
Page 609 and 610:
Defining Rules Rules are specified
Page 611 and 612:
Updates and deletes are first writt
Page 613 and 614:
Rely on two storages: an ephemeral
Page 615 and 616:
Target Configuration Here is a typi
Page 617 and 618:
core/cdcr?action=QUEUES: Fetches st
Page 619 and 620:
Output { } "responseHeader": { "sta
Page 621 and 622:
{ } responseHeader={ status=0, QTim
Page 623 and 624:
http://localhost:898 ${TargetZk}
Page 625 and 626:
Legacy Scaling and Distribution Thi
Page 627 and 628:
Limitations to Distributed Search D
Page 629 and 630:
0 8 *:* localhost:8983/solr/core1
Page 631 and 632:
Master and Slave Update Optimizatio
Page 633 and 634:
http://remote_host:port/solr/core
Page 635 and 636:
A commit command is issued on the s
Page 637 and 638:
is a large expense, but not nearly
Page 639 and 640:
Apache Solr Reference Guide 6.0 639
Page 641 and 642:
The Solr Wiki contains a list of cl
Page 643 and 644:
pSolrClient , or CloudSolrClient .
Page 645 and 646:
upload using binary format, which i
Page 647 and 648:
}} 'manufacturedate_dt'=>'2005-10-1
Page 649 and 650:
Solr Glossary Where possible, terms
Page 651 and 652:
Literally, N data about data. Metad
Page 653 and 654:
See also #Inverse document frequenc
Page 655 and 656:
A new graph query parser makes it p
Page 657 and 658:
Upgrading a Solr Cluster This page
Page 659 and 660:
proceeding to upgrade the next node
show all

Apache Solr Reference Guide Covering Apache Solr 6.0

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?