includes/poolcounter/PoolCounter.php

   1 <?php
   2 /**
   3  * Provides of semaphore semantics for restricting the number
   4  * of workers that may be concurrently performing the same task.
   5  *
   6  * This program is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (at your option) any later version.
  10  *
  11  * This program is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  14  * GNU General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License along
  17  * with this program; if not, write to the Free Software Foundation, Inc.,
  18  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  19  * http://www.gnu.org/copyleft/gpl.html
  20  *
  21  * @file
  22  */
  23
  24 /**
  25  * When you have many workers (threads/servers) giving service, and a
  26  * cached item expensive to produce expires, you may get several workers
  27  * doing the job at the same time.
  28  *
  29  * Given enough requests and the item expiring fast (non-cacheable,
  30  * lots of edits...) that single work can end up unfairly using most (all)
  31  * of the cpu of the pool. This is also known as 'Michael Jackson effect'
  32  * since this effect triggered on the english wikipedia on the day Michael
  33  * Jackson died, the biographical article got hit with several edits per
  34  * minutes and hundreds of read hits.
  35  *
  36  * The PoolCounter provides semaphore semantics for restricting the number
  37  * of workers that may be concurrently performing such single task. Only one
  38  * key can be locked by any PoolCounter instance of a process, except for keys
  39  * that start with "nowait:". However, only 0 timeouts (non-blocking requests)
  40  * can be used with "nowait:" keys.
  41  *
  42  * By default PoolCounterNull is used, which provides no locking. You
  43  * can get a useful one in the PoolCounter extension.
  44  */
  45 abstract class PoolCounter {
  46         /* Return codes */
  47         const LOCKED = 1; /* Lock acquired */
  48         const RELEASED = 2; /* Lock released */
  49         const DONE = 3; /* Another worker did the work for you */
  50
  51         const ERROR = -1; /* Indeterminate error */
  52         const NOT_LOCKED = -2; /* Called release() with no lock held */
  53         const QUEUE_FULL = -3; /* There are already maxqueue workers on this lock */
  54         const TIMEOUT = -4; /* Timeout exceeded */
  55         const LOCK_HELD = -5; /* Cannot acquire another lock while you have one lock held */
  56
  57         /** @var string All workers with the same key share the lock */
  58         protected $key;
  59         /** @var int Maximum number of workers working on tasks with the same key simultaneously */
  60         protected $workers;
  61         /**
  62          * Maximum number of workers working on this task type, regardless of key.
  63          * 0 means unlimited. Max allowed value is 65536.
  64          * The way the slot limit is enforced is overzealous - this option should be used with caution.
  65          * @var int
  66          */
  67         protected $slots = 0;
  68         /** @var int If this number of workers are already working/waiting, fail instead of wait */
  69         protected $maxqueue;
  70         /** @var float Maximum time in seconds to wait for the lock */
  71         protected $timeout;
  72
  73         /**
  74          * @var bool Whether the key is a "might wait" key
  75          */
  76         private $isMightWaitKey;
  77         /**
  78          * @var bool Whether this process holds a "might wait" lock key
  79          */
  80         private static $acquiredMightWaitKey = 0;
  81
  82         /**
  83          * @param array $conf
  84          * @param string $type The class of actions to limit concurrency for (task type)
  85          * @param string $key
  86          */
  87         protected function __construct( $conf, $type, $key ) {
  88                 $this->workers = $conf['workers'];
  89                 $this->maxqueue = $conf['maxqueue'];
  90                 $this->timeout = $conf['timeout'];
  91                 if ( isset( $conf['slots'] ) ) {
  92                         $this->slots = $conf['slots'];
  93                 }
  94
  95                 if ( $this->slots ) {
  96                         $key = $this->hashKeyIntoSlots( $type, $key, $this->slots );
  97                 }
  98
  99                 $this->key = $key;
 100                 $this->isMightWaitKey = !preg_match( '/^nowait:/', $this->key );
 101         }
 102
 103         /**
 104          * Create a Pool counter. This should only be called from the PoolWorks.
 105          *
 106          * @param string $type The class of actions to limit concurrency for (task type)
 107          * @param string $key
 108          *
 109          * @return PoolCounter
 110          */
 111         public static function factory( $type, $key ) {
 112                 global $wgPoolCounterConf;
 113                 if ( !isset( $wgPoolCounterConf[$type] ) ) {
 114                         return new PoolCounterNull;
 115                 }
 116                 $conf = $wgPoolCounterConf[$type];
 117                 $class = $conf['class'];
 118
 119                 return new $class( $conf, $type, $key );
 120         }
 121
 122         /**
 123          * @return string
 124          */
 125         public function getKey() {
 126                 return $this->key;
 127         }
 128
 129         /**
 130          * I want to do this task and I need to do it myself.
 131          *
 132          * @return Status Value is one of Locked/Error
 133          */
 134         abstract public function acquireForMe();
 135
 136         /**
 137          * I want to do this task, but if anyone else does it
 138          * instead, it's also fine for me. I will read its cached data.
 139          *
 140          * @return Status Value is one of Locked/Done/Error
 141          */
 142         abstract public function acquireForAnyone();
 143
 144         /**
 145          * I have successfully finished my task.
 146          * Lets another one grab the lock, and returns the workers
 147          * waiting on acquireForAnyone()
 148          *
 149          * @return Status Value is one of Released/NotLocked/Error
 150          */
 151         abstract public function release();
 152
 153         /**
 154          * Checks that the lock request is sane.
 155          * @return Status - good for sane requests fatal for insane
 156          * @since 1.25
 157          */
 158         final protected function precheckAcquire() {
 159                 if ( $this->isMightWaitKey ) {
 160                         if ( self::$acquiredMightWaitKey ) {
 161                                 /*
 162                                  * The poolcounter itself is quite happy to allow you to wait
 163                                  * on another lock while you have a lock you waited on already
 164                                  * but we think that it is unlikely to be a good idea.  So we
 165                                  * made it an error.  If you are _really_ _really_ sure it is a
 166                                  * good idea then feel free to implement an unsafe flag or
 167                                  * something.
 168                                  */
 169                                 return Status::newFatal( 'poolcounter-usage-error',
 170                                         'You may only aquire a single non-nowait lock.' );
 171                         }
 172                 } elseif ( $this->timeout !== 0 ) {
 173                         return Status::newFatal( 'poolcounter-usage-error',
 174                                 'Locks starting in nowait: must have 0 timeout.' );
 175                 }
 176                 return Status::newGood();
 177         }
 178
 179         /**
 180          * Update any lock tracking information when the lock is acquired
 181          * @since 1.25
 182          */
 183         final protected function onAcquire() {
 184                 self::$acquiredMightWaitKey |= $this->isMightWaitKey;
 185         }
 186
 187         /**
 188          * Update any lock tracking information when the lock is released
 189          * @since 1.25
 190          */
 191         final protected function onRelease() {
 192                 self::$acquiredMightWaitKey &= !$this->isMightWaitKey;
 193         }
 194
 195         /**
 196          * Given a key (any string) and the number of lots, returns a slot key (a prefix with a suffix
 197          * integer from the [0..($slots-1)] range). This is used for a global limit on the number of
 198          * instances of a given type that can acquire a lock. The hashing is deterministic so that
 199          * PoolCounter::$workers is always an upper limit of how many instances with the same key
 200          * can acquire a lock.
 201          *
 202          * @param string $type The class of actions to limit concurrency for (task type)
 203          * @param string $key PoolCounter instance key (any string)
 204          * @param int $slots The number of slots (max allowed value is 65536)
 205          * @return string Slot key with the type and slot number
 206          */
 207         protected function hashKeyIntoSlots( $type, $key, $slots ) {
 208                 return $type . ':' . ( hexdec( substr( sha1( $key ), 0, 4 ) ) % $slots );
 209         }
 210 }