includes/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.
  38  *
  39  * By default PoolCounter_Stub is used, which provides no locking. You
  40  * can get a useful one in the PoolCounter extension.
  41  */
  42 abstract class PoolCounter {
  43         /* Return codes */
  44         const LOCKED = 1; /* Lock acquired */
  45         const RELEASED = 2; /* Lock released */
  46         const DONE = 3; /* Another worker did the work for you */
  47
  48         const ERROR = -1; /* Indeterminate error */
  49         const NOT_LOCKED = -2; /* Called release() with no lock held */
  50         const QUEUE_FULL = -3; /* There are already maxqueue workers on this lock */
  51         const TIMEOUT = -4; /* Timeout exceeded */
  52         const LOCK_HELD = -5; /* Cannot acquire another lock while you have one lock held */
  53
  54         /** @var string All workers with the same key share the lock */
  55         protected $key;
  56         /** @var integer Maximum number of workers doing the task simultaneously */
  57         protected $workers;
  58         /** @var integer If this number of workers are already working/waiting, fail instead of wait */
  59         protected $maxqueue;
  60         /** @var float Maximum time in seconds to wait for the lock */
  61         protected $timeout;
  62
  63         /**
  64          * @param array $conf
  65          * @param string $type
  66          * @param string $key
  67          */
  68         protected function __construct( $conf, $type, $key ) {
  69                 $this->key = $key;
  70                 $this->workers = $conf['workers'];
  71                 $this->maxqueue = $conf['maxqueue'];
  72                 $this->timeout = $conf['timeout'];
  73         }
  74
  75         /**
  76          * Create a Pool counter. This should only be called from the PoolWorks.
  77          *
  78          * @param $type
  79          * @param $key
  80          *
  81          * @return PoolCounter
  82          */
  83         public static function factory( $type, $key ) {
  84                 global $wgPoolCounterConf;
  85                 if ( !isset( $wgPoolCounterConf[$type] ) ) {
  86                         return new PoolCounter_Stub;
  87                 }
  88                 $conf = $wgPoolCounterConf[$type];
  89                 $class = $conf['class'];
  90
  91                 return new $class( $conf, $type, $key );
  92         }
  93
  94         /**
  95          * @return string
  96          */
  97         public function getKey() {
  98                 return $this->key;
  99         }
 100
 101         /**
 102          * I want to do this task and I need to do it myself.
 103          *
 104          * @return Status Value is one of Locked/Error
 105          */
 106         abstract public function acquireForMe();
 107
 108         /**
 109          * I want to do this task, but if anyone else does it
 110          * instead, it's also fine for me. I will read its cached data.
 111          *
 112          * @return Status Value is one of Locked/Done/Error
 113          */
 114         abstract public function acquireForAnyone();
 115
 116         /**
 117          * I have successfully finished my task.
 118          * Lets another one grab the lock, and returns the workers
 119          * waiting on acquireForAnyone()
 120          *
 121          * @return Status value is one of Released/NotLocked/Error
 122          */
 123         abstract public function release();
 124 }
 125
 126 class PoolCounter_Stub extends PoolCounter {
 127         public function __construct() {
 128                 /* No parameters needed */
 129         }
 130
 131         public function acquireForMe() {
 132                 return Status::newGood( PoolCounter::LOCKED );
 133         }
 134
 135         public function acquireForAnyone() {
 136                 return Status::newGood( PoolCounter::LOCKED );
 137         }
 138
 139         public function release() {
 140                 return Status::newGood( PoolCounter::RELEASED );
 141         }
 142 }
 143
 144 /**
 145  * Class for dealing with PoolCounters using class members
 146  */
 147 abstract class PoolCounterWork {
 148         protected $cacheable = false; //Does this override getCachedWork() ?
 149
 150         /**
 151          * @param string $type The type of PoolCounter to use
 152          * @param string $key Key that identifies the queue this work is placed on
 153          */
 154         public function __construct( $type, $key ) {
 155                 $this->poolCounter = PoolCounter::factory( $type, $key );
 156         }
 157
 158         /**
 159          * Actually perform the work, caching it if needed
 160          * @return mixed work result or false
 161          */
 162         abstract public function doWork();
 163
 164         /**
 165          * Retrieve the work from cache
 166          * @return mixed work result or false
 167          */
 168         public function getCachedWork() {
 169                 return false;
 170         }
 171
 172         /**
 173          * A work not so good (eg. expired one) but better than an error
 174          * message.
 175          * @return mixed work result or false
 176          */
 177         public function fallback() {
 178                 return false;
 179         }
 180
 181         /**
 182          * Do something with the error, like showing it to the user.
 183          * @return bool
 184          */
 185         public function error( $status ) {
 186                 return false;
 187         }
 188
 189         /**
 190          * Log an error
 191          *
 192          * @param $status Status
 193          * @return void
 194          */
 195         public function logError( $status ) {
 196                 $key = $this->poolCounter->getKey();
 197
 198                 wfDebugLog( 'poolcounter', "Pool key '$key': "
 199                         . $status->getMessage()->inLanguage( 'en' )->useDatabase( false )->text() );
 200         }
 201
 202         /**
 203          * Get the result of the work (whatever it is), or the result of the error() function.
 204          * This returns the result of the first applicable method that returns a non-false value,
 205          * where the methods are checked in the following order:
 206          *   - a) doWork()       : Applies if the work is exclusive or no another process
 207          *                         is doing it, and on the condition that either this process
 208          *                         successfully entered the pool or the pool counter is down.
 209          *   - b) doCachedWork() : Applies if the work is cacheable and this blocked on another
 210          *                         process which finished the work.
 211          *   - c) fallback()     : Applies for all remaining cases.
 212          * If these all fall through (by returning false), then the result of error() is returned.
 213          *
 214          * @param $skipcache bool
 215          * @return mixed
 216          */
 217         public function execute( $skipcache = false ) {
 218                 if ( $this->cacheable && !$skipcache ) {
 219                         $status = $this->poolCounter->acquireForAnyone();
 220                 } else {
 221                         $status = $this->poolCounter->acquireForMe();
 222                 }
 223
 224                 if ( !$status->isOK() ) {
 225                         // Respond gracefully to complete server breakage: just log it and do the work
 226                         $this->logError( $status );
 227                         return $this->doWork();
 228                 }
 229
 230                 switch ( $status->value ) {
 231                         case PoolCounter::LOCKED:
 232                                 $result = $this->doWork();
 233                                 $this->poolCounter->release();
 234                                 return $result;
 235
 236                         case PoolCounter::DONE:
 237                                 $result = $this->getCachedWork();
 238                                 if ( $result === false ) {
 239                                         /* That someone else work didn't serve us.
 240                                          * Acquire the lock for me
 241                                          */
 242                                         return $this->execute( true );
 243                                 }
 244                                 return $result;
 245
 246                         case PoolCounter::QUEUE_FULL:
 247                         case PoolCounter::TIMEOUT:
 248                                 $result = $this->fallback();
 249
 250                                 if ( $result !== false ) {
 251                                         return $result;
 252                                 }
 253                                 /* no break */
 254
 255                         /* These two cases should never be hit... */
 256                         case PoolCounter::ERROR:
 257                         default:
 258                                 $errors = array(
 259                                         PoolCounter::QUEUE_FULL => 'pool-queuefull',
 260                                         PoolCounter::TIMEOUT => 'pool-timeout' );
 261
 262                                 $status = Status::newFatal( isset( $errors[$status->value] )
 263                                         ? $errors[$status->value]
 264                                         : 'pool-errorunknown' );
 265                                 $this->logError( $status );
 266                                 return $this->error( $status );
 267                 }
 268         }
 269 }
 270
 271 /**
 272  * Convenience class for dealing with PoolCounters using callbacks
 273  * @since 1.22
 274  */
 275 class PoolCounterWorkViaCallback extends PoolCounterWork {
 276         /** @var callable */
 277         protected $doWork;
 278         /** @var callable|null */
 279         protected $doCachedWork;
 280         /** @var callable|null */
 281         protected $fallback;
 282         /** @var callable|null */
 283         protected $error;
 284
 285         /**
 286          * Build a PoolCounterWork class from a type, key, and callback map.
 287          *
 288          * The callback map must at least have a callback for the 'doWork' method.
 289          * Additionally, callbacks can be provided for the 'doCachedWork', 'fallback',
 290          * and 'error' methods. Methods without callbacks will be no-ops that return false.
 291          * If a 'doCachedWork' callback is provided, then execute() may wait for any prior
 292          * process in the pool to finish and reuse its cached result.
 293          *
 294          * @param string $type
 295          * @param string $key
 296          * @param array $callbacks Map of callbacks
 297          * @throws MWException
 298          */
 299         public function __construct( $type, $key, array $callbacks ) {
 300                 parent::__construct( $type, $key );
 301                 foreach ( array( 'doWork', 'doCachedWork', 'fallback', 'error' ) as $name ) {
 302                         if ( isset( $callbacks[$name] ) ) {
 303                                 if ( !is_callable( $callbacks[$name] ) ) {
 304                                         throw new MWException( "Invalid callback provided for '$name' function." );
 305                                 }
 306                                 $this->$name = $callbacks[$name];
 307                         }
 308                 }
 309                 if ( !isset( $this->doWork ) ) {
 310                         throw new MWException( "No callback provided for 'doWork' function." );
 311                 }
 312                 $this->cacheable = isset( $this->doCachedWork );
 313         }
 314
 315         public function doWork() {
 316                 return call_user_func_array( $this->doWork, array() );
 317         }
 318
 319         public function getCachedWork() {
 320                 if ( $this->doCachedWork ) {
 321                         return call_user_func_array( $this->doCachedWork, array() );
 322                 }
 323                 return false;
 324         }
 325
 326         public function fallback() {
 327                 if ( $this->fallback ) {
 328                         return call_user_func_array( $this->fallback, array() );
 329                 }
 330                 return false;
 331         }
 332
 333         public function error( $status ) {
 334                 if ( $this->error ) {
 335                         return call_user_func_array( $this->error, array( $status ) );
 336                 }
 337                 return false;
 338         }
 339 }