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 use MediaWiki\MainConfigNames;
  29 use MediaWiki\Status\Status;
  30
  31 require_once __DIR__ . '/Maintenance.php';
  32
  33 /**
  34  * Maintenance script to remove old or broken uploads from temporary uploaded
  35  * file storage and clean up associated database records.
  36  *
  37  * @ingroup Maintenance
  38  */
  39 class CleanupUploadStash extends Maintenance {
  40
  41         public function __construct() {
  42                 parent::__construct();
  43                 $this->addDescription( 'Clean up abandoned files in temporary uploaded file stash' );
  44                 $this->setBatchSize( 50 );
  45         }
  46
  47         public function execute() {
  48                 $repo = $this->getServiceContainer()->getRepoGroup()->getLocalRepo();
  49                 $tempRepo = $repo->getTempRepo();
  50
  51                 $dbr = $repo->getReplicaDB();
  52
  53                 // how far back should this look for files to delete?
  54                 $cutoff = time() - (int)$this->getConfig()->get( MainConfigNames::UploadStashMaxAge );
  55
  56                 $this->output( "Getting list of files to clean up...\n" );
  57                 $res = $dbr->newSelectQueryBuilder()
  58                         ->select( 'us_key' )
  59                         ->from( 'uploadstash' )
  60                         ->where( $dbr->expr( 'us_timestamp', '<', $dbr->timestamp( $cutoff ) ) )
  61                         ->caller( __METHOD__ )
  62                         ->fetchResultSet();
  63
  64                 // Delete all registered stash files...
  65                 if ( $res->numRows() == 0 ) {
  66                         $this->output( "No stashed files to cleanup according to the DB.\n" );
  67                 } else {
  68                         // finish the read before starting writes.
  69                         $keys = [];
  70                         foreach ( $res as $row ) {
  71                                 $keys[] = $row->us_key;
  72                         }
  73
  74                         $this->output( 'Removing ' . count( $keys ) . " file(s)...\n" );
  75                         // this could be done some other, more direct/efficient way, but using
  76                         // UploadStash's own methods means it's less likely to fall accidentally
  77                         // out-of-date someday
  78                         $stash = new UploadStash( $repo );
  79
  80                         $i = 0;
  81                         foreach ( $keys as $key ) {
  82                                 $i++;
  83                                 try {
  84                                         $stash->getFile( $key, true );
  85                                         $stash->removeFileNoAuth( $key );
  86                                 } catch ( UploadStashException $ex ) {
  87                                         $type = get_class( $ex );
  88                                         $this->output( "Failed removing stashed upload with key: $key ($type)\n" );
  89                                 }
  90                                 if ( $i % 100 == 0 ) {
  91                                         $this->waitForReplication();
  92                                         $this->output( "$i\n" );
  93                                 }
  94                         }
  95                         $this->output( "$i done\n" );
  96                 }
  97
  98                 // Delete all the corresponding thumbnails...
  99                 $dir = $tempRepo->getZonePath( 'thumb' );
 100                 $iterator = $tempRepo->getBackend()->getFileList( [ 'dir' => $dir, 'adviseStat' => 1 ] );
 101                 if ( $iterator === null ) {
 102                         $this->fatalError( "Could not get file listing." );
 103                 }
 104                 $this->output( "Deleting old thumbnails...\n" );
 105                 $i = 0;
 106                 $batch = [];
 107                 foreach ( $iterator as $file ) {
 108                         if ( wfTimestamp( TS_UNIX, $tempRepo->getFileTimestamp( "$dir/$file" ) ) < $cutoff ) {
 109                                 $batch[] = [ 'op' => 'delete', 'src' => "$dir/$file" ];
 110                                 if ( count( $batch ) >= $this->getBatchSize() ) {
 111                                         $this->doOperations( $tempRepo, $batch );
 112                                         $i += count( $batch );
 113                                         $batch = [];
 114                                         $this->output( "$i\n" );
 115                                 }
 116                         }
 117                 }
 118                 if ( count( $batch ) ) {
 119                         $this->doOperations( $tempRepo, $batch );
 120                         $i += count( $batch );
 121                 }
 122                 $this->output( "$i done\n" );
 123
 124                 // Apparently lots of stash files are not registered in the DB...
 125                 $dir = $tempRepo->getZonePath( 'public' );
 126                 $iterator = $tempRepo->getBackend()->getFileList( [ 'dir' => $dir, 'adviseStat' => 1 ] );
 127                 if ( $iterator === null ) {
 128                         $this->fatalError( "Could not get file listing." );
 129                 }
 130                 $this->output( "Deleting orphaned temp files...\n" );
 131                 if ( strpos( $dir, '/local-temp' ) === false ) {
 132                         $this->output( "Temp repo might be misconfigured. It points to directory: '$dir' \n" );
 133                 }
 134
 135                 $i = 0;
 136                 $batch = [];
 137                 foreach ( $iterator as $file ) {
 138                         if ( wfTimestamp( TS_UNIX, $tempRepo->getFileTimestamp( "$dir/$file" ) ) < $cutoff ) {
 139                                 $batch[] = [ 'op' => 'delete', 'src' => "$dir/$file" ];
 140                                 if ( count( $batch ) >= $this->getBatchSize() ) {
 141                                         $this->doOperations( $tempRepo, $batch );
 142                                         $i += count( $batch );
 143                                         $batch = [];
 144                                         $this->output( "$i\n" );
 145                                 }
 146                         }
 147                 }
 148                 if ( count( $batch ) ) {
 149                         $this->doOperations( $tempRepo, $batch );
 150                         $i += count( $batch );
 151                 }
 152                 $this->output( "$i done\n" );
 153         }
 154
 155         protected function doOperations( FileRepo $tempRepo, array $ops ) {
 156                 $status = $tempRepo->getBackend()->doQuickOperations( $ops );
 157                 if ( !$status->isOK() ) {
 158                         $this->error( print_r( Status::wrap( $status )->getErrorsArray(), true ) );
 159                 }
 160         }
 161 }
 162
 163 $maintClass = CleanupUploadStash::class;
 164 require_once RUN_MAINTENANCE_IF_MAIN;