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->mDescription = "Clean up abandoned files in temporary uploaded file stash";
  41         }
  42
  43         public function execute() {
  44                 global $wgUploadStashMaxAge;
  45
  46                 $repo = RepoGroup::singleton()->getLocalRepo();
  47                 $tempRepo = $repo->getTempRepo();
  48
  49                 $dbr = $repo->getSlaveDb();
  50
  51                 // how far back should this look for files to delete?
  52                 $cutoff = time() - $wgUploadStashMaxAge;
  53
  54                 $this->output( "Getting list of files to clean up...\n" );
  55                 $res = $dbr->select(
  56                         'uploadstash',
  57                         'us_key',
  58                         'us_timestamp < ' . $dbr->addQuotes( $dbr->timestamp( $cutoff ) ),
  59                         __METHOD__
  60                 );
  61
  62                 // Delete all registered stash files...
  63                 if ( $res->numRows() == 0 ) {
  64                         $this->output( "No stashed files to cleanup according to the DB.\n" );
  65                 } else {
  66                         // finish the read before starting writes.
  67                         $keys = array();
  68                         foreach ( $res as $row ) {
  69                                 array_push( $keys, $row->us_key );
  70                         }
  71
  72                         $this->output( 'Removing ' . count( $keys ) . " file(s)...\n" );
  73                         // this could be done some other, more direct/efficient way, but using
  74                         // UploadStash's own methods means it's less likely to fall accidentally
  75                         // out-of-date someday
  76                         $stash = new UploadStash( $repo );
  77
  78                         $i = 0;
  79                         foreach ( $keys as $key ) {
  80                                 $i++;
  81                                 try {
  82                                         $stash->getFile( $key, true );
  83                                         $stash->removeFileNoAuth( $key );
  84                                 } catch ( UploadStashBadPathException $ex ) {
  85                                         $this->output( "Failed removing stashed upload with key: $key\n"  );
  86                                 } catch ( UploadStashZeroLengthFileException $ex ) {
  87                                         $this->output( "Failed removing stashed upload with key: $key\n"  );
  88                                 }
  89                                 if ( $i % 100 == 0 ) {
  90                                         $this->output( "$i\n" );
  91                                 }
  92                         }
  93                         $this->output( "$i done\n" );
  94                 }
  95
  96                 // Delete all the corresponding thumbnails...
  97                 $dir = $tempRepo->getZonePath( 'thumb' );
  98                 $iterator = $tempRepo->getBackend()->getFileList( array( 'dir' => $dir ) );
  99                 $this->output( "Deleting old thumbnails...\n" );
 100                 $i = 0;
 101                 foreach ( $iterator as $file ) {
 102                         if ( wfTimestamp( TS_UNIX, $tempRepo->getFileTimestamp( "$dir/$file" ) ) < $cutoff ) {
 103                                 $status = $tempRepo->quickPurge( "$dir/$file" );
 104                                 if ( !$status->isOK() ) {
 105                                         $this->error( print_r( $status->getErrorsArray(), true ) );
 106                                 }
 107                                 if ( ( ++$i % 100 ) == 0 ) {
 108                                         $this->output( "$i\n" );
 109                                 }
 110                         }
 111                 }
 112                 $this->output( "$i done\n" );
 113
 114                 // Apparently lots of stash files are not registered in the DB...
 115                 $dir = $tempRepo->getZonePath( 'public' );
 116                 $iterator = $tempRepo->getBackend()->getFileList( array( 'dir' => $dir ) );
 117                 $this->output( "Deleting orphaned temp files...\n" );
 118                 if ( strpos( $dir, '/local-temp' ) === false ) { // sanity check
 119                         $this->error( "Temp repo is not using the temp container.", 1 ); // die
 120                 }
 121                 $i = 0;
 122                 foreach ( $iterator as $file ) {
 123                         // Absolute sanity check for stashed files and file segments
 124                         if ( !preg_match( '#(^\d{14}!|\.\d+\.\w+\.\d+$)#', basename( $file ) ) ) {
 125                                 $this->output( "Skipped non-stash $file\n" );
 126                                 continue;
 127                         }
 128                         if ( wfTimestamp( TS_UNIX, $tempRepo->getFileTimestamp( "$dir/$file" ) ) < $cutoff ) {
 129                                 $status = $tempRepo->quickPurge( "$dir/$file" );
 130                                 if ( !$status->isOK() ) {
 131                                         $this->error( print_r( $status->getErrorsArray(), true ) );
 132                                 }
 133                                 if ( ( ++$i % 100 ) == 0 ) {
 134                                         $this->output( "$i\n" );
 135                                 }
 136                         }
 137                 }
 138                 $this->output( "$i done\n" );
 139         }
 140 }
 141
 142 $maintClass = "UploadStashCleanup";
 143 require_once( RUN_MAINTENANCE_IF_MAIN );