diff options
Diffstat (limited to 'apps/files_sharing/lib/DeleteOrphanedSharesJob.php')
| -rw-r--r-- | apps/files_sharing/lib/DeleteOrphanedSharesJob.php | 144 |
1 files changed, 106 insertions, 38 deletions
diff --git a/apps/files_sharing/lib/DeleteOrphanedSharesJob.php b/apps/files_sharing/lib/DeleteOrphanedSharesJob.php index 6a641734680..63f057e3bf4 100644 --- a/apps/files_sharing/lib/DeleteOrphanedSharesJob.php +++ b/apps/files_sharing/lib/DeleteOrphanedSharesJob.php @@ -1,48 +1,45 @@ <?php + +declare(strict_types=1); /** - * @copyright Copyright (c) 2016, ownCloud, Inc. - * - * @author Christoph Wurst <christoph@winzerhof-wurst.at> - * @author Joas Schilling <coding@schilljs.com> - * @author Morris Jobke <hey@morrisjobke.de> - * @author Vincent Petry <vincent@nextcloud.com> - * - * @license AGPL-3.0 - * - * This code is free software: you can redistribute it and/or modify - * it under the terms of the GNU Affero General Public License, version 3, - * as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU Affero General Public License for more details. - * - * You should have received a copy of the GNU Affero General Public License, version 3, - * along with this program. If not, see <http://www.gnu.org/licenses/> - * + * SPDX-FileCopyrightText: 2020-2024 Nextcloud GmbH and Nextcloud contributors + * SPDX-FileCopyrightText: 2016 ownCloud, Inc. + * SPDX-License-Identifier: AGPL-3.0-only */ namespace OCA\Files_Sharing; -use OC\BackgroundJob\TimedJob; +use OCP\AppFramework\Db\TTransactional; +use OCP\AppFramework\Utility\ITimeFactory; +use OCP\BackgroundJob\TimedJob; +use OCP\DB\QueryBuilder\IQueryBuilder; +use OCP\IDBConnection; +use PDO; +use Psr\Log\LoggerInterface; +use function array_map; /** * Delete all share entries that have no matching entries in the file cache table. */ class DeleteOrphanedSharesJob extends TimedJob { - /** - * Default interval in minutes - * - * @var int $defaultIntervalMin - **/ - protected $defaultIntervalMin = 15; + use TTransactional; + + private const CHUNK_SIZE = 1000; + + private const INTERVAL = 24 * 60 * 60; /** * sets the correct interval for this timed job */ - public function __construct() { - $this->interval = $this->defaultIntervalMin * 60; + public function __construct( + ITimeFactory $time, + private IDBConnection $db, + private LoggerInterface $logger, + ) { + parent::__construct($time); + + $this->setInterval(self::INTERVAL); // 1 day + $this->setTimeSensitivity(self::TIME_INSENSITIVE); } /** @@ -51,15 +48,86 @@ class DeleteOrphanedSharesJob extends TimedJob { * @param array $argument unused argument */ public function run($argument) { - $connection = \OC::$server->getDatabaseConnection(); - $logger = \OC::$server->getLogger(); + if ($this->db->getShardDefinition('filecache')) { + $this->shardingCleanup(); + return; + } - $sql = - 'DELETE FROM `*PREFIX*share` ' . - 'WHERE `item_type` in (\'file\', \'folder\') ' . - 'AND NOT EXISTS (SELECT `fileid` FROM `*PREFIX*filecache` WHERE `file_source` = `fileid`)'; + $qbSelect = $this->db->getQueryBuilder(); + $qbSelect->select('id') + ->from('share', 's') + ->leftJoin('s', 'filecache', 'fc', $qbSelect->expr()->eq('s.file_source', 'fc.fileid')) + ->where($qbSelect->expr()->isNull('fc.fileid')) + ->setMaxResults(self::CHUNK_SIZE); + $deleteQb = $this->db->getQueryBuilder(); + $deleteQb->delete('share') + ->where( + $deleteQb->expr()->in('id', $deleteQb->createParameter('ids'), IQueryBuilder::PARAM_INT_ARRAY) + ); + + /** + * Read a chunk of orphan rows and delete them. Continue as long as the + * chunk is filled and time before the next cron run does not run out. + * + * Note: With isolation level READ COMMITTED, the database will allow + * other transactions to delete rows between our SELECT and DELETE. In + * that (unlikely) case, our DELETE will have fewer affected rows than + * IDs passed for the WHERE IN. If this happens while processing a full + * chunk, the logic below will stop prematurely. + * Note: The queries below are optimized for low database locking. They + * could be combined into one single DELETE with join or sub query, but + * that has shown to (dead)lock often. + */ + $cutOff = $this->time->getTime() + self::INTERVAL; + do { + $deleted = $this->atomic(function () use ($qbSelect, $deleteQb) { + $result = $qbSelect->executeQuery(); + $ids = array_map('intval', $result->fetchAll(PDO::FETCH_COLUMN)); + $result->closeCursor(); + $deleteQb->setParameter('ids', $ids, IQueryBuilder::PARAM_INT_ARRAY); + $deleted = $deleteQb->executeStatement(); + $this->logger->debug('{deleted} orphaned share(s) deleted', [ + 'app' => 'DeleteOrphanedSharesJob', + 'deleted' => $deleted, + ]); + return $deleted; + }, $this->db); + } while ($deleted >= self::CHUNK_SIZE && $this->time->getTime() <= $cutOff); + } + + private function shardingCleanup(): void { + $qb = $this->db->getQueryBuilder(); + $qb->selectDistinct('file_source') + ->from('share', 's'); + $sourceFiles = $qb->executeQuery()->fetchAll(PDO::FETCH_COLUMN); + + $deleteQb = $this->db->getQueryBuilder(); + $deleteQb->delete('share') + ->where( + $deleteQb->expr()->in('file_source', $deleteQb->createParameter('ids'), IQueryBuilder::PARAM_INT_ARRAY) + ); + + $chunks = array_chunk($sourceFiles, self::CHUNK_SIZE); + foreach ($chunks as $chunk) { + $deletedFiles = $this->findMissingSources($chunk); + $this->atomic(function () use ($deletedFiles, $deleteQb) { + $deleteQb->setParameter('ids', $deletedFiles, IQueryBuilder::PARAM_INT_ARRAY); + $deleted = $deleteQb->executeStatement(); + $this->logger->debug('{deleted} orphaned share(s) deleted', [ + 'app' => 'DeleteOrphanedSharesJob', + 'deleted' => $deleted, + ]); + return $deleted; + }, $this->db); + } + } - $deletedEntries = $connection->executeUpdate($sql); - $logger->debug("$deletedEntries orphaned share(s) deleted", ['app' => 'DeleteOrphanedSharesJob']); + private function findMissingSources(array $ids): array { + $qb = $this->db->getQueryBuilder(); + $qb->select('fileid') + ->from('filecache') + ->where($qb->expr()->in('fileid', $qb->createNamedParameter($ids, IQueryBuilder::PARAM_INT_ARRAY))); + $found = $qb->executeQuery()->fetchAll(\PDO::FETCH_COLUMN); + return array_diff($ids, $found); } } |