Difference between revisions of "Elasticsearch"

From Freephile Wiki
Jump to navigation Jump to search
m (→‎Video: replace badly encoded character)
m (Freephile moved page Elasticsearch to Elasticsearch: not camelcase)
(47 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
This site uses Elasticsearch for it's search functionality under the hood.
 
{{Feature
 
{{Feature
|image=Elasticsearch_logo.svg
+
|explains= Search
|imgdesc=Search
+
|description= This site uses Elasticsearch for the best possible search experience [[File:System-search.svg|link=Search|thumb|64px]]
|title=
+
|notes=
}}
+
|tests=
{{#set:feature description = This site uses Elasticsearch for an amazing search experience! }}
+
|examples=
{{#set:feature notes = [https://medium.com/@AIMDekTech/what-is-elasticsearch-why-elasticsearch-advantages-of-elasticsearch-47b81b549f4d The What, Why and Advantages of Elasticsearch] }}
 
{{#set:feature tests = Search for something in "files" which indicates that PDF index results are returned instead of just articles. E.g.[https://wiki.freephile.org/wiki/index.php?title=Special:Search&profile=advanced&profile=advanced&fulltext=Search&search=ssh-agent&ns6=1 Search for 'ssh-agent' in the File namespace] }}
 
{{#set:feature examples =  }}
 
This site uses Elasticsearch for it's search functionality under the hood.
 
 
 
 
 
==About==
 
Elasticsearch is a distributed RESTful search engine built for the cloud.  See https://www.elastic.co/about  I'd like to recommend the intro video <ref>https://www.elastic.co/webinars/getting-started-elasticsearch</ref> but you have to submit an email to view it. (joe@example.com probably works)
 
 
 
==Community==
 
There is a Discourse forum at https://discuss.elastic.co/
 
 
 
==Features==
 
See [[mw:Help:CirrusSearch]] for help on how to best use the search functionality (including regex searches).
 
 
 
===Search Tips===
 
This wiki supports Elasticsearch features.  So, for example, let's say you want to search for 'Ansible' in the wiki, but you know that there is an Ansible page, and you don't want to be taken directly to that page.  Instead, you prefer to actually see all the [Special:Search search results for 'Ansible'].  Just prefix your search term with the <code>~</code> character.  Now when you press enter, you'll go to the search results page with a full listing of results.
 
 
 
Use the <code>morelike:</code> special prefix [https://wiki.freephile.org/wiki/api.php?action=query&list=search&srsearch=morelike:Elasticsearch morelike:Elasticsearch]
 
 
 
 
 
Use the <code>cirrusdump</code> action to see the document as ElasticSearch sees it [{{fullurl:{{PAGENAMEE}}
 
|action=cirrusdump}}]. This is especially useful to test whether (new) documents are being indexed.  Additional MediaWiki API methods like <code>cirrus-config-dump</code> are listed at [[mw:Extension:CirrusSearch#API]]
 
 
 
{{Messagebox
 
| type = success
 
| text = Different indexes are created for the entire contents of the wiki.  Each index is weighted differently.  So, for example, "Lead-in" text is the wikitext between the top of the page and the first heading.  Words found here are deemed more relevant to a users search query than the same word if found in the body text of an article.  So, in this wiki, [{{fullurl:Special:Search|search=yaml|fulltext=Search}} searching for the word "YAML"] puts the [[Ansible]] article ahead of the [[Eclipse]] article in search results.
 
}}
 
 
 
{{Messagebox
 
| type = success
 
| text = Content as well as all files uploaded into the system are indexed. For example, [{{fullurl:Special:Search|search=fai|fulltext=Search|profile=all}} a search for "FAI"] lists both the [[Cloning]] article as well as the [[:File:Fai poster a4.pdf|PDF file]]  And the file is not listed only because of the file name, but also because of the (indexed) file content.  [{{fullurl:Special:Search|search=ed%20roman|fulltext=Search|profile=all}} A search for "Ed Roman"] will bring up the Enterprise Java Beans Design Patterns PDF file ([{{fullurl:File:Ejbdesignpatterns.pdf|page=13}} see p. 13 where Ed Roman is mentioned].)
 
}}
 
 
 
{{Messagebox
 
| type = warning
 
| text =
 
Elasticsearch performs poorly when JVM starts swapping: you should ensure that it ''never'' swaps.
 
 
 
Set this property to true to lock the memory: <code>bootstrap.mlockall: true</code> (in /etc/elasticsearch/elasticsearch.yml)
 
 
 
Make sure that the <code>ES_MIN_MEM</code> and <code>ES_MAX_MEM</code> environment variables are set  to the same value, and that the machine has enough memory to allocate for Elasticsearch, leaving enough memory for the operating system itself.
 
 
 
You should also make sure that the Elasticsearch process is allowed to lock the memory, eg. by using <code>ulimit -l unlimited</code>.
 
 
}}
 
}}
  
==Video==
 
 
*[https://vimeo.com/136326424 Building Elasticsearch: From Idea to {code} to Adoption] The back side of a napkin, a pen, and a few beverages are often the ingredients that yield good ideas. Elasticsearch had a different origin. It started with a need for a simple search box for a collection of recipes. '''Shay Banon''', creator of Elasticsearch and CTO at Elastic, shares the history behind pushing the code for his first open source project that led to the creation of Elasticsearch and it's rapid adoption by users worldwide. -- ''RISE | August 2015''
 
  
==Troubleshooting==
+
== About ==
Make sure that you use the official packages from Elasticsearch, and NOT the Ubuntu packages. See below for the installation guide. Note that I had to actually un-comment and specify the bind.host as 0.0.0.0 on an older setup (Version: 1.7.3, Build: NA/NA, JVM: 1.8.0_171 on Ubuntu 16.04).  Plus, make sure that your firewall is allowing the ports 9200-9400. You can run the startup shell script directly to see what's wrong if there's no log output (and read the source for options): <code>/usr/share/elasticsearch/bin/elasticsearch</code>
+
Elasticsearch is a distributed RESTful search engine built for the cloud. Features include:
 
 
<pre>
 
Usage: /usr/share/elasticsearch/bin/elasticsearch [-vdh] [-p pidfile] [-D prop] [-X prop]
 
Start elasticsearch.
 
    -d            daemonize (run in background)
 
    -p pidfile    write PID to <pidfile>
 
    -h
 
    --help        print command line options
 
    -v            print elasticsearch version, then exit
 
    -D prop      set JAVA system property
 
    -X prop      set non-standard JAVA system property
 
  --prop=val
 
  --prop val    set elasticsearch property (i.e. -Des.<prop>=<val>)
 
</pre>
 
 
 
For version 1.7 on Ubuntu 16.04, although the system uses SystemD, there is still a SysV init script that controls elasticsearch. I think this means that you can't get the logging info into the journalctl system... See /etc/init.d/elasticsearch According to the [https://www.elastic.co/guide/en/elasticsearch/reference/current/deb.html install guide], you should be able to edit the elasticsearch.service file, and take out the --quiet option to make it log to the journal.  When that is enabled, you can do <code>journalctl --unit elasticsearch</code> to quickly see the info being logged.
 
 
 
==Production Configuration==
 
ElasticSearch looks for a configuration file to include, and uses a search path for that include. You can specify it on the command-line; through an environment variable; or just make sure that your file is found in the search path. (My default was found at /usr/share/elasticsearch/bin/elasticsearch.in.sh)
 
<source lang="bash">
 
# If an include wasn't specified in the environment, then search for one...
 
if [ "x$ES_INCLUDE" = "x" ]; then
 
    # Locations (in order) to use when searching for an include file.
 
    for include in /usr/share/elasticsearch/elasticsearch.in.sh \
 
                  /usr/local/share/elasticsearch/elasticsearch.in.sh \
 
                  /opt/elasticsearch/elasticsearch.in.sh \
 
                  ~/.elasticsearch.in.sh \
 
                  $ES_HOME/bin/elasticsearch.in.sh \
 
                  "`dirname "$0"`"/elasticsearch.in.sh; do
 
        if [ -r "$include" ]; then
 
            . "$include"
 
            break
 
        fi
 
    done
 
# ...otherwise, source the specified include.
 
elif [ -r "$ES_INCLUDE" ]; then
 
    . "$ES_INCLUDE"
 
fi
 
</source>
 
  
 +
* Distributed and Highly Available Search Engine.
 +
** Each index is fully sharded with a configurable number of shards.
 +
** Each shard can have one or more replicas.
 +
** Read / Search operations performed on either one of the replica shard.
 +
* Multi Tenant with Multi Types.
 +
** Support for more than one index.
 +
** Support for more than one type per index.
 +
** Index level configuration (number of shards, index storage, ...).
 +
* Various set of APIs
 +
** HTTP RESTful API
 +
** Native Java API.
 +
** All APIs perform automatic node operation rerouting.
 +
* Document oriented
 +
** No need for upfront schema definition.
 +
** Schema can be defined per type for customization of the indexing process.
 +
* Reliable, Asynchronous Write Behind for long term persistency.
 +
* (Near) Real Time Search.
 +
* Built on top of Lucene
 +
** Each shard is a fully functional Lucene index
 +
** All the power of Lucene easily exposed through simple configuration / plugins.
 +
* Per operation consistency
 +
** Single document level operations are atomic, consistent, isolated and durable.
 +
* Open Source under the Apache License, version 2 ("ALv2")
  
==Elasticsearch for MediaWiki==
+
== Video ==
To improve the out-of-the-box search experience with MediaWiki, you should install the [[mw:Extension:CirrusSearch]].  CirrusSearch is just a connector to the Elasticsearch engine. Thus, to use CirrusSearch, first install the [[Elasticsearch]] system (you can use <code>yum</code> or <code>apt</code> repositories for that).
+
* [https://vimeo.com/136326424 Building Elasticsearch: From Idea to {code} to Adoption] The back side of a napkin, a pen, and a few beverages are often the ingredients that yield good ideas. Elasticsearch had a different origin. It started with a need for a simple search box for a collection of recipes. '''Shay Banon''', creator of Elasticsearch and CTO at Elastic, shares the history behind pushing the code for his first open source project that led to the creation of Elasticsearch
 +
* https://www.elastic.co/about
  
Wikitech gives some information about how WMF uses Elasticsearch at https://wikitech.wikimedia.org/wiki/Search
+
== Elasticsearch for MediaWiki ==
 +
To improve the out-of-the-box search experience with MediaWiki, you should install the [[mw:Extension:CirrusSearch]].  CirrusSearch is just a connector to the Elasticsearch engine.  Thus, to use CirrusSearch, first install the [[Elasticsearch]] system (you can use [https://www.elastic.co/guide/en/elasticsearch/reference/current/setup-repositories.html the repositories for that]). 
  
 
This system has three components: Elastica, CirrusSearch, and Elasticsearch.
 
This system has three components: Elastica, CirrusSearch, and Elasticsearch.
 +
; Elastica : Elastica is a MediaWiki extension that provides the library to interface with Elasticsearch. It wraps the [https://github.com/ruflin/Elastica Elastica] library. It has no configuration.
 +
; CirrusSearch : CirrusSearch is a MediaWiki extension that provides search support backed by Elasticsearch.
 +
; Elasticsearch : is a Java application, so you need [[Java]] installed as well.  At the time of this writing, there is a version mismatch.  Elasticsearch is at version 2.1.1 in the repositories, but CirrusSearch is only compatible with the older 1.7 version.
  
;Elastica :Elastica is a MediaWiki extension that provides the library to interface with Elasticsearch. It wraps the [https://github.com/ruflin/Elastica Elastica] library. It has no configuration.
+
== Installation ==
;CirrusSearch :CirrusSearch is a MediaWiki extension that provides search support backed by Elasticsearch.
 
;Elasticsearch :is a Java application, so you need [[Java]] installed as well. As all these pieces continue to be developed and released, you must be sure to take heed of the requirements for matching the right versions together to compose your full setup.
 
 
 
==Elasticsearch for QualityBox==
 
 
 
<source lang="haproxy">
 
# disallow PUT and DELETE methods through the web
 
# administrators will need to use local curl commands to bypass the load-balancer
 
# in the event that you want to delete indexes etc.
 
frontend elastic
 
        bind *:9201
 
        mode http
 
        acl is_delete method DELETE
 
        http-request deny if is_delete
 
        acl is_put method PUT
 
        http-request deny if is_put
 
        default_backend elastic
 
 
backend elastic
 
        mode http
 
        option forwardfor
 
        balance source
 
        option httpclose
 
      server es1 127.0.0.1:9200 weight 1 check inter 1000 rise 5 fall 1
 
 
 
</source>
 
 
 
You can add multi-wiki search like so using [https://github.com/wikimedia/mediawiki-extensions-CirrusSearch/blob/master/CirrusSearch.php $wgCirrusSearchEnableCrossProjectSearch],  $wgCirrusSearchWikiToNameMap and $wgCirrusSearchInterwikiSources:
 
 
 
 
 
 
 
 
 
<source lang="php">
 
if ( $wikiId !== 'commons' ) {
 
$wgCirrusSearchEnableCrossProjectSearch = true;
 
$wgCirrusSearchWikiToNameMap = [
 
'commons' => 'wiki_commons',
 
];
 
$wgCirrusSearchInterwikiSources = [
 
'commons' => 'wiki_commons_content_first',
 
];
 
}
 
 
 
 
 
</source>
 
 
 
==Where is my Elasticsearch?==
 
Maybe you installed elasticsearch, but have no idea where it resides on your system.  Try this:
 
<source lang="bash">
 
curl -XGET "http://localhost:9200/_nodes/settings?pretty=true"
 
</source>
 
Other direct commands
 
<source lang="bash">
 
curl 'localhost:9200/_tasks?pretty'
 
curl 'localhost:9200/_cat/nodes?pretty'
 
curl 'localhost:9200/_nodes?pretty'
 
curl 'localhost:9200/_nodes/settings?pretty=true'
 
curl 'localhost:9200/_cat/health?pretty'
 
curl 'localhost:9200/_cluster/health?pretty=true'
 
curl 'localhost:9200/_cluster/state?pretty'
 
 
 
curl 'localhost:9200/_cat/indices?v'
 
</source>
 
 
 
The configuration for Elasticsearch is normally held in two files: <code>/etc/elasticsearch/elasticsearch.yml</code> and <code>/etc/elasticsearch/logging.yml</code>
 
 
 
==Starting / Stopping==
 
Elasticsearch is (usually) run as a service, so you can start and stop it the way you would depending on whether you run SysV init or SystemD
 
 
 
==Upgrading==
 
QualityBox 34 runs MediaWiki 1.34.x and Elasticsearch 6.x
 
 
 
The best way to upgrade to QB 34 from QB 32.x (ES 5.x) is to [https://gerrit.wikimedia.org/r/plugins/gitiles/mediawiki/extensions/CirrusSearch/+/HEAD/README blow away the index and start over].
 
 
 
* DO NOT follow the [https://www.elastic.co/guide/en/elastic-stack/6.8/upgrading-elastic-stack.html upgrade instructions]. Migration Assistant does not work for 5.6 in testing.
 
* DO NOT prepare a 32.x host by [https://www.elastic.co/guide/en/elasticsearch/reference/5.6/installing-xpack-es.html installing X-Pack] and preparing the indexes using the Migration Assistant. [https://github.com/elastic/elasticsearch/issues/30085#issuecomment-685286211 It does not work!]. And you'll avoid the hassle of registering for a (free) limited license. Plus the hassle of installing and upgrading Kibana just to get the Migration Assistant.
 
* [https://www.elastic.co/guide/en/elasticsearch/reference/6.0/breaking_60_indices_changes.html Breaking changes in 6.x] are taken care of by the Elastica Extension
 
 
 
{| class="wikitable"
 
|+Version Dependencies
 
|-
 
!MediaWiki!!ElasticSearch!!Elastica!!CirrusSearch!!Cluster Restart?!!Reindex?
 
|-
 
|REL1_27||1.x||REL1_27||REL1_27||n/a||n/a
 
|-
 
|REL1_28||2.x||REL1_28||REL1_28||restart||yes <ref>For more information about upgrading from 1.x to 2.4, see [https://www.elastic.co/guide/en/elasticsearch/reference/2.4/setup-upgrade.html Upgrading Elasticsearch] in the Elasticsearch 2.4 Reference.</ref>
 
|-
 
|REL1_29||5.3.x or 5.4.x||REL1_29||REL1_29||restart||yes <ref>For more information about upgrading from 2.4 to 5.6, see [https://www.elastic.co/guide/en/elasticsearch/reference/5.6/setup-upgrade.html Upgrading Elasticsearch] in the Elasticsearch 5.6 Reference.</ref>
 
|-
 
|REL1_30||5.3.x or 5.4.x||REL1_30||REL1_30||no <ref>Elasticsearch 6.x support rolling upgrades from Elasticsearch 5.6 Upgrading from earlier versions requires a full cluster restart. See https://www.elastic.co/guide/en/elasticsearch/reference/6.0/restart-upgrade.html</ref>||Depends <ref>Elasticsearch can read indices created in the '''previous major version'''. Older indices must be reindexed or deleted.  Elasticsearch will fail to start if incompatible indices are present.</ref>
 
|-
 
|REL 1_31 and 1_32
 
|''5.5.x or 5.6.x''
 
|
 
|
 
|
 
|
 
|-
 
|REL 1_33 and !_34 and 1_35
 
|6.5.x (6.5.4 rec)
 
|
 
|
 
|
 
|
 
|}
 
 
 
 
 
Elastic Co. provides an [https://github.com/elastic/ansible-elasticsearch/tree/master ansible role to manage your installation] (including a [https://github.com/elastic/ansible-elasticsearch/tree/2.x 2.x branch] for older setups).  Their [https://www.elastic.co/guide/en/elasticsearch/reference/6.0/setup-upgrade.html guide to upgrading] covers the nitty gritty.
 
 
 
==Reindexing==
 
 
 
The most basic form of the reindex API just copies documents from one index into another.  You might [https://www.elastic.co/guide/en/elasticsearch/reference/6.0/docs-reindex.html reindex] to change the name of a field.  Usually though, you are reindexing because you are forced to during a major version upgrade.
 
 
 
To assist in the upgrade process there is a plugin that assists with the tasks.
 
 
 
Also, you can reindex from a remote (cluster) so that you can upgrade without downtime because once the new cluster is ready, you can just switch to it with minimal disruption.
 
<ref>https://www.elastic.co/guide/en/elasticsearch/reference/6.0/reindex-upgrade.html</ref>
 
 
 
If you are re-indexing your existing Meza installation, you can <code> sudo meza maint rebuild monolith --tags search-index</code>
 
 
 
 
 
==Monitoring==
 
With the upgrade to Elasticsearch 5.x and 6.x, plugins are deprecated. It's suggested to use Kibana as a monitoring and management interface to Elasticsearch.
 
<img src="http://meta.qualitybox.us:20000/api/v1/badge.svg?chart=elasticsearch_local.cluster_health_status&alarm=elasticsearch_last_collected&refresh=auto" />
 
 
 
http://meta.qualitybox.us:20000/api/v1/badge.svg?chart=elasticsearch_local.cluster_health_status&alarm=elasticsearch_last_collected&refresh=auto" type="image/svg+xml
 
 
 
 
 
<html>
 
<embed src="http://meta.qualitybox.us:20000/api/v1/badge.svg?chart=elasticsearch_local.cluster_health_status&alarm=elasticsearch_last_collected&refresh=auto" type="image/svg+xml" height="20"/>
 
</html>
 
 
 
*Monitoring the Elastic Stack https://www.elastic.co/guide/en/elastic-stack-overview/6.4/xpack-monitoring.html
 
*Monitoring Settings https://www.elastic.co/guide/en/elasticsearch/reference/6.4/monitoring-settings.html
 
*X-Pack monitoring https://www.elastic.co/guide/en/kibana/current/xpack-monitoring.html
 
*Configuring Monitoring in Kibana https://www.elastic.co/guide/en/kibana/6.4/monitoring-xpack-kibana.html
 
*Monitoring with [https://github.com/lmenezes/cerebro Cerebro]
 
*You can still use [https://mobz.github.io/elasticsearch-head/ elasticsearch-head] if you use an SSH reverse tunnel to access the remote server.
 
 
 
 
 
==Installation==
 
 
Here's a quick example of how we got all the parts installed on an [[Ubuntu]] server.
 
Here's a quick example of how we got all the parts installed on an [[Ubuntu]] server.
 
<source lang="bash">
 
<source lang="bash">
Line 290: Line 92:
 
Checking if elasticsearch is running
 
Checking if elasticsearch is running
 
<source lang="bash">
 
<source lang="bash">
curl http://localhost:9200/
+
curl -X GET http://localhost:9200/
 
</source>
 
</source>
 
<source lang="javascript">
 
<source lang="javascript">
Line 321: Line 123:
 
</source>
 
</source>
  
==Resources==
+
== Resources ==
 
+
* https://phabricator.wikimedia.org/diffusion/ECIR/browse/master/CirrusSearch.php
*See [[mw:Help:CirrusSearch]] for help on how to best use the search functionality (including regex searches).
+
* https://git.wikimedia.org/blob/mediawiki%2Fextensions%2FCirrusSearch.git/HEAD/README
*https://phabricator.wikimedia.org/diffusion/ECIR/browse/master/CirrusSearch.php
+
* https://wikitech.wikimedia.org/wiki/Search
*https://phabricator.wikimedia.org/diffusion/ECIR/browse/master/README
 
*https://wikitech.wikimedia.org/wiki/Search
 
 
 
==Problems==
 
We recently ran a 'rebuild-all' script to update Elasticsearch indexes
 
<pre>
 
[centos@ip-10-0-50-189 .deploy-meza]$ time sudo ./elastic-rebuild-all.sh demo
 
Rebuilding index for demo
 
  Output log:
 
    /opt/data-meza/logs/search-index/demo.2017-09-07_13:15:01.log
 
elastic-build-index completed for "demo" at 2017-09-07_13:15:03
 
 
 
real    0m1.653s
 
user    0m1.327s
 
sys    0m0.199s
 
[centos@ip-10-0-50-189 .deploy-meza]$ tail /opt/data-meza/logs/search-index/demo.2017-09-07_13:15:01.log
 
        Inferring index identifier...error
 
Looks like the index has more than one identifier. </pre><span style="background-color:#c1f749;">You should delete all
 
but the one of them currently active. Here is the list: wiki_demo_content,wiki_demo_content_first</span><pre>
 
 
 
Notice: Undefined index: HTTP_HOST in /opt/htdocs/mediawiki/LocalSettings.php on line 219
 
[          wiki_demo] index(es) do not exist. Did you forget to run updateSearchIndexConfig?
 
 
 
Notice: Undefined index: HTTP_HOST in /opt/htdocs/mediawiki/LocalSettings.php on line 219
 
[          wiki_demo] index(es) do not exist. Did you forget to run updateSearchIndexConfig?
 
******* Elastic Search build index complete! *******
 
</pre>
 
''Emphasis added''
 
 
 
===SOLVED===
 
You can delete the unwanted index like this with [[curl]]:
 
<source lang="bash">
 
curl -XDELETE "http://localhost:9200/wiki_cod_content"
 
</source>
 
 
 
See more about deleting wikis and all indexes at https://github.com/freephile/meza/blob/6658c795a4b5e5b1a5afcb05c62cf0bcc2d0203b/src/scripts/delete.wikis.sh
 
 
 
{{References}}
 
  
 
[[Category:Search]]
 
[[Category:Search]]
[[Category:Elasticsearch]]
 
[[Category:Wiki]]
 
[[Category:QualityBox]]
 

Revision as of 11:39, 30 January 2016