maintenance/cleanupUploadStash.php

   1 <?php
   2 /**
   3  * Remove old or broken uploads from temporary uploaded file storage,
   4  * clean up associated database records
   5  *
   6  * Copyright © 2011, Wikimedia Foundation
   7  *
   8  * This program is free software; you can redistribute it and/or modify
   9  * it under the terms of the GNU General Public License as published by
  10  * the Free Software Foundation; either version 2 of the License, or
  11  * (at your option) any later version.
  12  *
  13  * This program is distributed in the hope that it will be useful,
  14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  16  * GNU General Public License for more details.
  17  *
  18  * You should have received a copy of the GNU General Public License along
  19  * with this program; if not, write to the Free Software Foundation, Inc.,
  20  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  21  * http://www.gnu.org/copyleft/gpl.html
  22  *
  23  * @file
  24  * @author Ian Baker <ibaker@wikimedia.org>
  25  * @ingroup Maintenance
  26  */
  27
  28 require_once __DIR__ . '/Maintenance.php';
  29
  30 /**
  31  * Maintenance script to remove old or broken uploads from temporary uploaded
  32  * file storage and clean up associated database records.
  33  *
  34  * @ingroup Maintenance
  35  */
  36 class UploadStashCleanup extends Maintenance {
  37
  38         public function __construct() {
  39                 parent::__construct();
  40                 $this->addDescription( 'Clean up abandoned files in temporary uploaded file stash' );
  41                 $this->setBatchSize( 50 );
  42         }
  43
  44         public function execute() {
  45                 global $wgUploadStashMaxAge;
  46
  47                 $repo = RepoGroup::singleton()->getLocalRepo();
  48                 $tempRepo = $repo->getTempRepo();
  49
  50                 $dbr = $repo->getReplicaDB();
  51
  52                 // how far back should this look for files to delete?
  53                 $cutoff = time() - $wgUploadStashMaxAge;
  54
  55                 $this->output( "Getting list of files to clean up...\n" );
  56                 $res = $dbr->select(
  57                         'uploadstash',
  58                         'us_key',
  59                         'us_timestamp < ' . $dbr->addQuotes( $dbr->timestamp( $cutoff ) ),
  60                         __METHOD__
  61                 );
  62
  63                 // Delete all registered stash files...
  64                 if ( $res->numRows() == 0 ) {
  65                         $this->output( "No stashed files to cleanup according to the DB.\n" );
  66                 } else {
  67                         // finish the read before starting writes.
  68                         $keys = [];
  69                         foreach ( $res as $row ) {
  70                                 array_push( $keys, $row->us_key );
  71                         }
  72
  73                         $this->output( 'Removing ' . count( $keys ) . " file(s)...\n" );
  74                         // this could be done some other, more direct/efficient way, but using
  75                         // UploadStash's own methods means it's less likely to fall accidentally
  76                         // out-of-date someday
  77                         $stash = new UploadStash( $repo );
  78
  79                         $i = 0;
  80                         foreach ( $keys as $key ) {
  81                                 $i++;
  82                                 try {
  83                                         $stash->getFile( $key, true );
  84                                         $stash->removeFileNoAuth( $key );
  85                                 } catch ( UploadStashException $ex ) {
  86                                         $type = get_class( $ex );
  87                                         $this->output( "Failed removing stashed upload with key: $key ($type)\n" );
  88                                 }
  89                                 if ( $i % 100 == 0 ) {
  90                                         wfWaitForSlaves();
  91                                         $this->output( "$i\n" );
  92                                 }
  93                         }
  94                         $this->output( "$i done\n" );
  95                 }
  96
  97                 // Delete all the corresponding thumbnails...
  98                 $dir = $tempRepo->getZonePath( 'thumb' );
  99                 $iterator = $tempRepo->getBackend()->getFileList( [ 'dir' => $dir, 'adviseStat' => 1 ] );
 100                 $this->output( "Deleting old thumbnails...\n" );
 101                 $i = 0;
 102                 $batch = []; // operation batch
 103                 foreach ( $iterator as $file ) {
 104                         if ( wfTimestamp( TS_UNIX, $tempRepo->getFileTimestamp( "$dir/$file" ) ) < $cutoff ) {
 105                                 $batch[] = [ 'op' => 'delete', 'src' => "$dir/$file" ];
 106                                 if ( count( $batch ) >= $this->getBatchSize() ) {
 107                                         $this->doOperations( $tempRepo, $batch );
 108                                         $i += count( $batch );
 109                                         $batch = [];
 110                                         $this->output( "$i\n" );
 111                                 }
 112                         }
 113                 }
 114                 if ( count( $batch ) ) {
 115                         $this->doOperations( $tempRepo, $batch );
 116                         $i += count( $batch );
 117                 }
 118                 $this->output( "$i done\n" );
 119
 120                 // Apparently lots of stash files are not registered in the DB...
 121                 $dir = $tempRepo->getZonePath( 'public' );
 122                 $iterator = $tempRepo->getBackend()->getFileList( [ 'dir' => $dir, 'adviseStat' => 1 ] );
 123                 $this->output( "Deleting orphaned temp files...\n" );
 124                 if ( strpos( $dir, '/local-temp' ) === false ) { // sanity check
 125                         $this->fatalError( "Temp repo is not using the temp container." );
 126                 }
 127                 $i = 0;
 128                 $batch = []; // operation batch
 129                 foreach ( $iterator as $file ) {
 130                         if ( wfTimestamp( TS_UNIX, $tempRepo->getFileTimestamp( "$dir/$file" ) ) < $cutoff ) {
 131                                 $batch[] = [ 'op' => 'delete', 'src' => "$dir/$file" ];
 132                                 if ( count( $batch ) >= $this->getBatchSize() ) {
 133                                         $this->doOperations( $tempRepo, $batch );
 134                                         $i += count( $batch );
 135                                         $batch = [];
 136                                         $this->output( "$i\n" );
 137                                 }
 138                         }
 139                 }
 140                 if ( count( $batch ) ) {
 141                         $this->doOperations( $tempRepo, $batch );
 142                         $i += count( $batch );
 143                 }
 144                 $this->output( "$i done\n" );
 145         }
 146
 147         protected function doOperations( FileRepo $tempRepo, array $ops ) {
 148                 $status = $tempRepo->getBackend()->doQuickOperations( $ops );
 149                 if ( !$status->isOK() ) {
 150                         $this->error( print_r( $status->getErrorsArray(), true ) );
 151                 }
 152         }
 153 }
 154
 155 $maintClass = UploadStashCleanup::class;
 156 require_once RUN_MAINTENANCE_IF_MAIN;