aweisberg commented on code in PR #4569: URL: https://github.com/apache/cassandra/pull/4569#discussion_r2710021898
########## src/java/org/apache/cassandra/replication/MutationTrackingSyncCoordinator.java: ########## @@ -0,0 +1,229 @@ +/* + * 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.cassandra.replication; + +import java.util.ArrayList; +import java.util.List; +import java.util.Map; +import java.util.concurrent.ConcurrentHashMap; +import java.util.concurrent.TimeUnit; +import java.util.concurrent.atomic.AtomicBoolean; + +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import org.apache.cassandra.dht.Range; +import org.apache.cassandra.dht.Token; +import org.apache.cassandra.utils.concurrent.AsyncPromise; +import org.apache.cassandra.utils.concurrent.Future; + +public class MutationTrackingSyncCoordinator +{ + private static final Logger logger = LoggerFactory.getLogger(MutationTrackingSyncCoordinator.class); + + private final String keyspace; + private final Range<Token> range; + private final AsyncPromise<Void> completionFuture = new AsyncPromise<>(); + + // Per-shard state: tracks what each node has reported for that shard + private final Map<Range<Token>, ShardSyncState> shardStates = new ConcurrentHashMap<>(); + + private final AtomicBoolean started = new AtomicBoolean(false); + private final AtomicBoolean completed = new AtomicBoolean(false); + + public MutationTrackingSyncCoordinator(String keyspace, Range<Token> range) + { + this.keyspace = keyspace; + this.range = range; + } + + public void start() + { + if (!started.compareAndSet(false, true)) + throw new IllegalStateException("Sync coordinator already started"); + + List<Shard> overlappingShards; + + overlappingShards = new ArrayList<>(); + MutationTrackingService.instance.forEachShardInKeyspace(keyspace, shard -> { + if (shard.range.intersects(range)) + overlappingShards.add(shard); + }); + + if (overlappingShards.isEmpty()) + { + completionFuture.setSuccess(null); + return; + } + + // Register to receive offset updates + MutationTrackingService.instance.registerSyncCoordinator(this); + + // Initialize state for each shard and capture targets + for (Shard shard : overlappingShards) + { + ShardSyncState state = new ShardSyncState(shard); + state.captureTargets(); + shardStates.put(shard.range, state); + } + + if (checkIfComplete()) + { + complete(); + return; + } + + logger.info("Sync coordinator started for keyspace {} range {}, tracking {} shards", + keyspace, range, overlappingShards.size()); + } + + private void complete() + { + if (!completed.compareAndSet(false, true)) + return; + MutationTrackingService.instance.unregisterSyncCoordinator(this); + completionFuture.setSuccess(null); + } + + private boolean checkIfComplete() + { + for (ShardSyncState state : shardStates.values()) + { + if (!state.isComplete()) + return false; + } + return true; + } + + public void onOffsetsReceived() + { + if (completed.get()) + return; + + // The underlying CoordinatorLog already updates its reconciled offsets. + // We just need to re-check if we're now complete. + if (checkIfComplete()) + { + complete(); + } + } + + public String getKeyspace() + { + return keyspace; + } + + public Range<Token> getRange() + { + return range; + } + + public Future<Void> awaitCompletion() + { + return completionFuture; + } + + /** + * Blocks until sync completes or timeout is reached. + * + * @param timeout Maximum time to wait + * @param unit Time unit + * @return true if completed, false if timed out + */ + public boolean awaitCompletion(long timeout, TimeUnit unit) throws InterruptedException + { + try + { + completionFuture.get(timeout, unit); + return true; + } + catch (java.util.concurrent.TimeoutException e) + { + return false; + } + catch (java.util.concurrent.ExecutionException e) + { + throw new RuntimeException(e.getCause()); + } + } + + public void cancel() + { + if (completed.compareAndSet(false, true)) + { + MutationTrackingService.instance.unregisterSyncCoordinator(this); + completionFuture.setFailure(new RuntimeException("Sync cancelled")); + } + } + + /** + * Tracks sync state for a single shard. + */ + private static class ShardSyncState + { + private final Shard shard; + + // Target offsets: LogId -> the offsets we're waiting for all nodes to have + private final Map<CoordinatorLogId, Offsets.Immutable> targets = new ConcurrentHashMap<>(); + + ShardSyncState(Shard shard) + { + this.shard = shard; + } + + void captureTargets() + { + BroadcastLogOffsets current = shard.collectReplicatedOffsets(false); Review Comment: This gets the offsets to wait for from the local coordinator. That doesn't make for a complete happens before edge between this repair and any already acknowledged writes that may not have been visible to this coordinator. So this repair could claim everything is replicated at ALL, but there are acknowledged writes which may not have been replicated yet. The repair would need to contact the other replicas and find out the highest known offset for every coordinator they know about and then wait on that. I think this is called out in the JIRA description as: > that would be collecting mutation offsets from all replicas for relevant shards, compute their union and wait until all these replicas report having received them. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
