View source for Manual:Job queue/it

<languages/>
<span lang="en" dir="ltr" class="mw-content-ltr">In 2009 ([[Special:MyLanguage/Release notes/1.6|MediaWiki 1.6]]) the '''job queue''' was introduced to perform long-running tasks asynchronously.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">The job queue is designed to hold many short tasks using [[w:batch processing|batch processing]].</span>

<div lang="en" dir="ltr" class="mw-content-ltr">
== Set up ==
</div>
<span lang="en" dir="ltr" class="mw-content-ltr">It is recommended that you instead schedule the running of jobs completely in the background, via the command line.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">By default, jobs are run at the end of a web request. Disable this default behaviour by setting {{wg|JobRunRate}} to <code>0</code>.</span>

{{note|1=<span lang="en" dir="ltr" class="mw-content-ltr">You should run <code>runJobs.php</code> as the same user as the web server runs as, to ensure that permissions to the filesystem are correctly accounted for if jobs touch uploaded files.</span>}}

<div lang="en" dir="ltr" class="mw-content-ltr">
=== Cron ===
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
You could use [[w:Cron|cron]] to run the jobs every hour. Add the following to your crontab file:
</div>

<syntaxhighlight lang="bash">
0 * * * * /usr/bin/php /var/www/wiki/maintenance/runJobs.php --maxtime=3600 > /var/log/runJobs.log 2>&1
</syntaxhighlight>

<div lang="en" dir="ltr" class="mw-content-ltr">
Using Cron makes it easy to get started, but can make email notifications and cascading template feel slow (to wait up to an hour). Consider using one of the below approaches to set up a continuous job runner instead.
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
=== Continuous service ===
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
If you have shell access and the possibility to create init scripts, you can create a simple service to run jobs as they become available, and also throttle them to prevent the job runner to monopolise the CPU resources of the server:
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
Create a bash script, for example at <code>/usr/local/bin/mwjobrunner</code>:
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
==== Create script ====
</div>
<syntaxhighlight lang="bash">
#!/bin/bash
# Put the MediaWiki installation path on the line below
MW_INSTALL_PATH="/home/www/www.mywikisite.example/mediawiki"
RUN_JOBS="$MW_INSTALL_PATH/maintenance/runJobs.php --maxtime=3600"
echo Starting job service...
# Wait a minute after the server starts up to give other processes time to get started
sleep 60
echo Started.
while true; do
	# Job types that need to be run ASAP no matter how many of them are in the queue
	# Those jobs should be very "cheap" to run
	php $RUN_JOBS --type="enotifNotify"
	# Everything else, limit the number of jobs on each batch
	# The --wait parameter will pause the execution here until new jobs are added,
	# to avoid running the loop without anything to do
	php $RUN_JOBS --wait --maxjobs=20
	# Wait some seconds to let the CPU do other things, like handling web requests, etc
	echo Waiting for 10 seconds...
	sleep 10
done
</syntaxhighlight>

<div lang="en" dir="ltr" class="mw-content-ltr">
Depending on how fast the server is and the load it handles, you can adapt the number of jobs to run on each cycle and the number of seconds to wait on each cycle.
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
Make the script executable (<code>chmod 755</code>).
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
==== Create service ====
</div>
<span lang="en" dir="ltr" class="mw-content-ltr">If using systemd, create a new service unit by creating the file <code>/etc/systemd/system/mw-jobqueue.service</code>.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">Change the <code>User</code> parameter to the user that runs PHP on your web server:</span>

<syntaxhighlight lang="ini">
[Unit]
Description=MediaWiki Job runner

[Service]
ExecStart=/usr/local/bin/mwjobrunner
Nice=10
ProtectSystem=full
User=php-fpm
OOMScoreAdjust=200
StandardOutput=journal

[Install]
WantedBy=multi-user.target
</syntaxhighlight>

<span lang="en" dir="ltr" class="mw-content-ltr">Enable it and start it with those commands:</span>

<syntaxhighlight lang="bash">
sudo systemctl enable mw-jobqueue
sudo systemctl start mw-jobqueue
sudo systemctl status mw-jobqueue
</syntaxhighlight>

<div lang="en" dir="ltr" class="mw-content-ltr">
== Job execution on page requests ==
</div>
<span lang="en" dir="ltr" class="mw-content-ltr">By default, at the end of each web request, one job is taken from the job queue and executed.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">This behaviour is controlled by the {{ll|Manual:$wgJobRunRate|$wgJobRunRate}} configuration variable.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">Setting this variable to <code>1</code>, will run a job on each request.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">Setting this variable to <code>0</code> will disable the execution of jobs during web requests completely, so that you can instead run {{ll|Manual:runJobs.php|runJobs.php}} manually or periodically from the command line.</span>
{{MW 1.23|+}}
<span lang="en" dir="ltr" class="mw-content-ltr">When enabled, jobs will be executed by opening a socket and making an internal HTTP request to an unlisted [[Special:MyLanguage/Manual:Special pages|special page]]: Special:RunJobs.</span>
Vedere anche la [[#Asynchronous|sezione asincrona]].

<div lang="en" dir="ltr" class="mw-content-ltr">
=== Performance issue ===
</div>
<span lang="en" dir="ltr" class="mw-content-ltr">If the performance burden of running jobs on every web request is too great but you are unable to run jobs from the command line, you can reduce {{ll|Manual:$wgJobRunRate|$wgJobRunRate}} to a number between <code>1</code> and <code>0</code>.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">This means a job will execute ''on average'' every <code>1 / $wgJobRunRate</code> requests.</span>

<syntaxhighlight lang="php">
$wgJobRunRate = 0.01;
</syntaxhighlight>

<div lang="en" dir="ltr" class="mw-content-ltr">
=== Manual usage ===
</div>
<span lang="en" dir="ltr" class="mw-content-ltr">There is also a way to empty the job queue manually, for example after changing a template that's present on many pages.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">Simply run the <code>{{ll|Manual:runJobs.php|maintenance/runJobs.php}}</code> maintenance script.</span>
Per esempio:

<syntaxhighlight lang="bash">
/path-to-my-wiki/maintenance$ php ./runJobs.php
</syntaxhighlight>

<div lang="en" dir="ltr" class="mw-content-ltr">
== Abandoned jobs ==
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
A job can fail for some reasons. To understand why, you have to inspect the related log file.
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
In any case, if a job fails 3 times (so if the system has done that number of attempts), the job is then considered "abandoned" and it's not executed again.
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
Relevant source code:
</div>

https://doc.wikimedia.org/mediawiki-core/master/php/JobQueue_8php_source.html#l00085

<div lang="en" dir="ltr" class="mw-content-ltr">
An abandoned job is:
</div>

* <span lang="en" dir="ltr" class="mw-content-ltr">not executed anymore from {{ll|Manual:runJobs.php|runJobs.php}}</span>
* <span lang="en" dir="ltr" class="mw-content-ltr">not counted from {{ll|Manual:showJobs.php}}</span>
* <span lang="en" dir="ltr" class="mw-content-ltr">not automatically removed from the database</span>
* <span lang="en" dir="ltr" class="mw-content-ltr">but are included in the count of [[#Special:Statistics|Special:Statistics]]</span>

<div lang="en" dir="ltr" class="mw-content-ltr">
== History ==
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
=== Asynchronous ===
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
The configuration variable {{ll|Manual:$wgRunJobsAsync|$wgRunJobsAsync}} has been added to force the execution of jobs synchronously, in scenarios where making an internal HTTP request for job execution is not wanted.
</div>

<span lang="en" dir="ltr" class="mw-content-ltr">When running jobs asynchronously, it will open an internal HTTP connection for handling the execution of jobs, and will return the contents of the page immediately to the client without waiting for the job to complete.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">Otherwise, the job will be executed in the same process and the client will have to wait until the job is completed.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">When the job does not run asynchronously, if a fatal error occurs during job execution, it will propagate to the client, aborting the load of the page.</span>

<span lang="en" dir="ltr" class="mw-content-ltr">Note that even if $wgRunJobsAsync is set to true, if PHP can't open a socket to make the internal HTTP request, it will fall back to the synchronous job execution.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">However, there are a variety of situations where this internal request may fail, and jobs won't be run, without falling back to the synchronous job execution.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">Starting with MediaWiki 1.28.1 and 1.27.2, $wgRunJobsAsync now defaults to false.</span>

<div lang="en" dir="ltr" class="mw-content-ltr">
=== Deferred updates ===
</div>

<span lang="en" dir="ltr" class="mw-content-ltr">The ''deferred updates'' mechanism allows the execution of code to be scheduled for the end of the request, after all content has been sent to the browser.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">This is similar to queuing a job, except that it runs immediately instead of upto several minutes/hours in the future.</span>

<span lang="en" dir="ltr" class="mw-content-ltr"><code>DeferredUpdates</code> was introduced in MediaWiki 1.23 and received major changes during MediaWiki 1.27 and 1.28.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">The goal of this mechanism is speed up the web responses by doing less work, as well as to prioritise some work that would previously be a job to run as soon as possible after the end of the response.</span>

<span lang="en" dir="ltr" class="mw-content-ltr">A deferrable update can implement <code>EnqueueableDataUpdate</code> in order to be queueable as a Job as well.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">This is used by <code>RefreshSecondaryDataUpdate</code> in core, for example, which means if the update fails for any reason, MediaWiki will fallback to queuing as a job and try again later as to fulfil the contract in question.</span> 


<span id="Changes_in_MediaWiki_1.22"></span>
=== Modifiche introdotte con MediaWiki 1.22 ===
<span lang="en" dir="ltr" class="mw-content-ltr">In {{ll|MediaWiki 1.22|MediaWiki 1.22}}, the job queue execution on each page request was changed ({{gerrit|59797}}) so, instead of executing the job inside the same PHP process that's rendering the page, a new PHP CLI command is spawned to execute {{ll|Manual:runJobs.php|runJobs.php}} in the background.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">It will only work if {{ll|Manual:$wgPhpCli|$wgPhpCli}} is set to an actual path or safe mode is off, otherwise, the old method will be used.</span>

<div lang="en" dir="ltr" class="mw-content-ltr">
This new execution method could cause some problems:
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
* If {{ll|Manual:$wgPhpCli|$wgPhpCli}} is set to an incompatible version of PHP (e.g.: an outdated version) jobs may fail to run (fixed in 1.23).
</div>
<div lang="en" dir="ltr" class="mw-content-ltr">
* PHP <code>open_basedir</code> restrictions are in effect, and {{ll|Manual:$wgPhpCli|$wgPhpCli}} is disallowed ({{task|T62208}}, fixed in 1.23).
</div>
<div lang="en" dir="ltr" class="mw-content-ltr">
* Performance: even if the job queue is empty, the new PHP process is started anyway ({{task|T62210}}, fixed in 1.23).
</div>
<div lang="en" dir="ltr" class="mw-content-ltr">
* Sometimes the spawning PHP process cause the server or only the CLI process to hang due to stdout and stderr descriptors not properly redirected ({{task|T60719}}, fixed in 1.22)
</div>
<div lang="en" dir="ltr" class="mw-content-ltr">
* It does not work for shared code (wiki farms), because it doesn't pass additional required parameters to runJobs.php to identify the wiki that's running the job ({{task|T62698}}, fixed in 1.23)
</div>
<div lang="en" dir="ltr" class="mw-content-ltr">
* Normal shell limits like {{ll|Manual:$wgMaxShellMemory|$wgMaxShellMemory}}, {{ll|Manual:$wgMaxShellTime|$wgMaxShellTime}}, {{ll|Manual:$wgMaxShellFileSize|$wgMaxShellFileSize}} are enforced on the runJobs.php process that's being executed in the background.
</div>

<span lang="en" dir="ltr" class="mw-content-ltr">There's no way to revert to the old on-request job queue handling, besides setting {{ll|Manual:$wgPhpCli|$wgPhpCli}} to <code>false</code>, for example, which may cause other problems ({{task|T63387}}).</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">It can be disabled completely by setting {{phpi|$wgJobRunRate {{=}} 0;}}, but jobs will no longer run on page requests, and you must explicitly run runJobs.php to periodically run pending jobs.</span>


<span id="Changes_in_MediaWiki_1.23"></span>
=== Modifiche introdotte con MediaWiki 1.23 ===

<div lang="en" dir="ltr" class="mw-content-ltr">
In MediaWiki 1.23, the 1.22 execution method is abandoned, and jobs are triggered by MediaWiki making an HTTP connection to itself.
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
It was first designed as an [[Special:MyLanguage/Manual:API|API]] entry point ({{gerrit|113038}}) but later changed to be the unlisted special page [[Special:RunJobs]] ({{gerrit|118336}}).
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
While it solves various bugs introduced in 1.22, it still requires loading a lot of PHP classes in memory on a new process to execute a job, and also makes a new HTTP request that the server must handle.
</div>

<span id="Changes_in_MediaWiki_1.27"></span>
=== Modifiche introdotte con MediaWiki 1.27 ===

<div lang="en" dir="ltr" class="mw-content-ltr">
In MediaWiki 1.25 and MediaWiki 1.26, use of {{ll|Manual:$wgRunJobsAsync|$wgRunJobsAsync}} would sometimes cause jobs not to get run if the wiki has custom {{ll|Manual:$wgServerName|$wgServerName}}  configuration. This was fixed in MediaWiki 1.27.
</div>
{{phabricator|T107290}}

<span id="Changes_in_MediaWiki_1.28"></span>
=== Modifiche introdotte con MediaWiki 1.28 ===

<div lang="en" dir="ltr" class="mw-content-ltr">
Between MediaWiki 1.23 and MediaWiki 1.27, use of {{ll|Manual:$wgRunJobsAsync|$wgRunJobsAsync}} would cause jobs not to get run on if MediaWiki requests are for a server name or protocol that does not match the currently configured server name one (e.g. when supporting both HTTP and HTTPS, or when MediaWiki is behind a reverse proxy that redirects to HTTPS). This was fixed in MediaWiki 1.28.
</div>
{{phabricator|T68485}}

<span id="Changes_in_MediaWiki_1.29"></span>
=== Modifiche introdotte con MediaWiki 1.29 ===

<div lang="en" dir="ltr" class="mw-content-ltr">
In MediaWiki 1.27.0 to 1.27.3 and 1.28.0 to 1.28.2, when {{ll|Manual:$wgJobRunRate|$wgJobRunRate}} is set to a value greater than 0, an error like the one below may appear in error logs, or on the page:
</div>

<pre>
PHP Notice: JobQueueGroup::__destruct: 1 buffered job(s) never inserted
</pre>

<span lang="en" dir="ltr" class="mw-content-ltr">As a result of this error, certain updates may fail in some cases, like category members not being updated on category pages, or recent changes displaying edits of deleted pages - even if you manually run {{ll|Manual:runJobs.php|runJobs.php}} to clear the job queue.</span>
<span lang="en" dir="ltr" class="mw-content-ltr">It has been reported as a bug ({{phabricator|T100085}}) and was solved in 1.27.4 and 1.28.3.</span>

<span id="Job_examples"></span>
== Esempi di lavoro ==

<div lang="en" dir="ltr" class="mw-content-ltr">
=== Updating links tables when a template changes ===
</div>
<span lang="en" dir="ltr" class="mw-content-ltr">When a template changes, MediaWiki adds a job to the job queue for each article transcluding that template.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">Each job is a command to read an article, expand any templates, and update the link table accordingly.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">Previously, the host articles would remain outdated until either their parser cache expires or until a user edits the article.</span>

<div lang="en" dir="ltr" class="mw-content-ltr">
=== HTML cache invalidation ===
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
A wider class of operations can result in invalidation of the HTML cache for a large number of pages:
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
*Changing an image (all the thumbnails have to be re-rendered, and their sizes recalculated)
</div>
<div lang="en" dir="ltr" class="mw-content-ltr">
*Deleting a page (all the links to it from other pages need to change from blue to red)
</div>
<div lang="en" dir="ltr" class="mw-content-ltr">
*Creating or undeleting a page (like above, but from red to blue)
</div>
<div lang="en" dir="ltr" class="mw-content-ltr">
*Changing a template (all the pages that transclude the template need updating)
</div>

<span lang="en" dir="ltr" class="mw-content-ltr">Except for template changes, these operations do not invalidate the links tables, but they do invalidate the HTML cache of all pages linking to that page, or using that image.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">Invalidating the cache of a page is a short operation; it only requires updating a single database field and sending a multicast packet to clear the caches.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">But if there are more than about 1000 to do, it takes a long time. By default, one job is added per '''300''' operations (see {{ll|Manual:$wgUpdateRowsPerJob|$wgUpdateRowsPerJob}})</span>

<span lang="en" dir="ltr" class="mw-content-ltr">Note, however, that even if purging the cache of a page is a short operation, reparsing a complex page that is not in the cache may be expensive, especially if a highly used template is edited and causes lots of pages to be purged in a short period of time and your wiki has lots of concurrent visitors loading a wide spread of pages.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">This can be mitigated by reducing the number of pages purged in a short period of time, by reducing {{wg|UpdateRowsPerJob}} to a small number (20, for example) and also set {{wg|JobBackoffThrottling}} for <code>htmlCacheUpdate</code> to a low number (5, for example).</span>

<div lang="en" dir="ltr" class="mw-content-ltr">
=== Audio and video transcoding ===
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
When using {{ll|Extension:TimedMediaHandler|nsp=0}} to process local uploads of audio and video files, the job queue is used to run the potentially very slow creation of derivative transcodes at various resolutions/formats.
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
These are '''not suitable''' for running on web requests -- you will need a background runner.
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
It's recommended to set up separate runners for the <code>webVideoTranscode</code> and <code>webVideoTranscodePrioritized</code> job types if possible. These two queues process different subsets of files -- the first for high resolution HD videos, and the second for lower-resolution videos and audio files which process more quickly.
</div>

<div lang="en" dir="ltr" class="mw-content-ltr">
== Typical values ==
</div>
<span lang="en" dir="ltr" class="mw-content-ltr">During a period of low load, the job queue might be zero.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">At Wikimedia, the job queue is, in practice, almost never zero.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">In off-peak hours, it might be a few hundred to a thousand.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">During a busy day, it might be a few million, but it can quickly fluctuate by 10% or more.</span>
<sup>[https://brionv.com/log/2008/04/22/so-whats-in-the-job-queue-anyway/]</sup>

<span id="Special:Statistics"></span>
== Speciale:Statistiche ==
<span lang="en" dir="ltr" class="mw-content-ltr">Up to MediaWiki 1.16, the job queue value was shown on [[Special:Statistics]].</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">However, since 1.17 ([[rev:75272]]) it's been removed, and can be seen now with {{ll|API:Siteinfo}}:</span>

{{ApiEx
| desc=
| p1=action=query
| p2=meta=siteinfo
| p3=siprop=statistics
}}

<span lang="en" dir="ltr" class="mw-content-ltr">The number of jobs returned in the API result may be slightly inaccurate when using MySQL, which estimates the number of jobs in the database.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">This number can fluctuate based on the number of jobs that have recently been added or deleted.</span> 
<span lang="en" dir="ltr" class="mw-content-ltr">For other databases that do not support fast result-size estimation, the actual number of jobs is given.</span>

<span id="For_developers"></span>
== Per gli sviluppatori ==
* {{ll|Manual:Job queue/For developers|2=<span lang="en" dir="ltr" class="mw-content-ltr">Job queue for developers</span>}}

<div lang="en" dir="ltr" class="mw-content-ltr">
== Code stewardship ==
</div>
{{Component|mediawiki-jobqueue}}

<span id="See_also"></span>
== Vedi anche ==
* {{ll|Manual:runJobs.php}}
* {{ll|Manual:showJobs.php}}
* {{ll|Manual:manageJobs.php}}
* {{ll|Manual:$wgJobRunRate|$wgJobRunRate}}
* {{ll|Manual:Job table}}

[[Category:Job queue{{#translation:}}| ]]
[[Category:Installation{{#translation:}}]]
[[Category:MediaWiki concepts{{#translation:}}]]