pvary commented on code in PR #10484: URL: https://github.com/apache/iceberg/pull/10484#discussion_r1700304518
########## flink/v1.19/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/TagBasedLockFactory.java: ########## @@ -0,0 +1,152 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.iceberg.flink.maintenance.operator; + +import java.io.IOException; +import java.util.Map; +import org.apache.flink.annotation.Internal; +import org.apache.iceberg.ManageSnapshots; +import org.apache.iceberg.SnapshotRef; +import org.apache.iceberg.Table; +import org.apache.iceberg.flink.TableLoader; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.util.Tasks; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** + * Iceberg table {@link ManageSnapshots#createTag(String, long)}/{@link + * ManageSnapshots#removeTag(String)} based lock implementation for {@link TriggerLockFactory}. + */ +@Internal +public class TagBasedLockFactory implements TriggerLockFactory { + private static final Logger LOG = LoggerFactory.getLogger(TagBasedLockFactory.class); + private static final String RUNNING_TAG = "__flink_maintenance_running"; + private static final String RECOVERING_TAG = "__flink_maintenance_recovering"; + private static final int CHANGE_ATTEMPTS = 3; + + private final TableLoader tableLoader; + private transient Table table; + + public TagBasedLockFactory(TableLoader tableLoader) { + this.tableLoader = tableLoader; + } + + @Override + public void open() { + tableLoader.open(); + this.table = tableLoader.loadTable(); + } + + @Override + public TriggerLockFactory.Lock createLock() { + return new Lock(table, RUNNING_TAG); + } + + @Override + public TriggerLockFactory.Lock createRecoveryLock() { + return new Lock(table, RECOVERING_TAG); + } + + @Override + public void close() throws IOException { + tableLoader.close(); + } + + public static class Lock implements TriggerLockFactory.Lock { + private final Table table; + private final String lockKey; + + public Lock(Table table, String lockKey) { + Preconditions.checkNotNull(table, "Table should not be null"); + Preconditions.checkNotNull(lockKey, "Lock key should not be null"); + this.table = table; + this.lockKey = lockKey; + } + + /** + * The lock will be acquired by jobs with creating a new tag. A new empty commit is added for a + * table without snapshots. + * + * @return <code>true</code> if the lock is acquired by this operator + */ + @Override + public boolean tryLock() { + if (isHeld()) { + LOG.info("Lock is already held"); + return false; + } + + if (table.currentSnapshot() == null) { + // Create an empty commit + table.newFastAppend().commit(); + LOG.info("Empty table, new empty commit added for using tags"); + } + + try { + Tasks.foreach(1) + .retry(CHANGE_ATTEMPTS) Review Comment: > > wondering if retry is safe here. Could the lock/unblock action initiated by multiple threads/operator subtasks? > > This is a valid issue. We need to write a more complicated locking with lockIds, and checks like we have in `MetastoreLock` and `LockRequest`. The main idea is to have something like `__flink_maitenance_<RND>` for the lock. > > * For checking the lock, we check if there is any tag starting with `__flink_maitenance_` > * For creating the lock, we create a tag with a `__flink_maitenance_<RND>` > * For retries for creation, we check if the lock with `RND` is created, and if not, try to create again > * For removal, we list the locks and remove the tag with `__flink_maitenance_<RND>` > * For retries for removal, we check if the lock with the `RND` is removed, and if not, try to remove again Fixed this. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
