Add AcknowledgementSet support to DocumentDB/MongoDB streams #4379
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,45 @@ | ||
| package org.opensearch.dataprepper.plugins.mongo.model; | ||
|
|
||
| public class CheckpointStatus { | ||
| private final String resumeToken; | ||
| private final long recordCount; | ||
| private boolean acknowledged; | ||
| private final long createTimestamp; | ||
| private Long acknowledgedTimestamp; | ||
|
|
||
| public CheckpointStatus(final String resumeToken, final long recordCount, final long createTimestamp) { | ||
| this.resumeToken = resumeToken; | ||
| this.recordCount = recordCount; | ||
| this.acknowledged = false; | ||
| this.createTimestamp = createTimestamp; | ||
| } | ||
|
|
||
| public void setAcknowledgedTimestamp(final Long acknowledgedTimestamp) { | ||
| this.acknowledgedTimestamp = acknowledgedTimestamp; | ||
| } | ||
|
|
||
| public void setAcknowledged(boolean acknowledged) { | ||
| this.acknowledged = acknowledged; | ||
| } | ||
|
|
||
| public String getResumeToken() { | ||
| return resumeToken; | ||
| } | ||
| public long getRecordCount() { | ||
| return recordCount; | ||
| } | ||
|
|
||
| public boolean isAcknowledged() { | ||
| return acknowledged; | ||
| } | ||
|
|
||
| public long getCreateTimestamp() { | ||
| return createTimestamp; | ||
| } | ||
|
|
||
| public long getAcknowledgedTimestamp() { | ||
| return acknowledgedTimestamp; | ||
| } | ||
|
|
||
|
|
||
| } | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,127 @@ | ||
| package org.opensearch.dataprepper.plugins.mongo.stream; | ||
|
|
||
| import com.google.common.annotations.VisibleForTesting; | ||
| import org.opensearch.dataprepper.common.concurrent.BackgroundThreadFactory; | ||
| import org.opensearch.dataprepper.model.acknowledgements.AcknowledgementSet; | ||
| import org.opensearch.dataprepper.model.acknowledgements.AcknowledgementSetManager; | ||
| import org.opensearch.dataprepper.plugins.mongo.model.CheckpointStatus; | ||
| import org.slf4j.Logger; | ||
| import org.slf4j.LoggerFactory; | ||
|
|
||
| import java.time.Duration; | ||
| import java.time.Instant; | ||
| import java.util.Optional; | ||
| import java.util.concurrent.ConcurrentHashMap; | ||
| import java.util.concurrent.ConcurrentLinkedQueue; | ||
| import java.util.concurrent.ExecutorService; | ||
| import java.util.concurrent.Executors; | ||
| import java.util.function.Consumer; | ||
|
|
||
| public class StreamAcknowledgementManager { | ||
| private static final Logger LOG = LoggerFactory.getLogger(StreamAcknowledgementManager.class); | ||
| private final ConcurrentLinkedQueue<CheckpointStatus> checkpoints = new ConcurrentLinkedQueue<>(); | ||
| private final ConcurrentHashMap<String, CheckpointStatus> ackStatus = new ConcurrentHashMap<>(); | ||
| private final AcknowledgementSetManager acknowledgementSetManager; | ||
| private final DataStreamPartitionCheckpoint partitionCheckpoint; | ||
|
|
||
| private final Duration partitionAcknowledgmentTimeout; | ||
| private final int acknowledgementMonitorWaitTimeInMs; | ||
| private final int checkPointIntervalInMs; | ||
| private final ExecutorService executorService; | ||
|
|
||
| private boolean enableAcknowledgement = false; | ||
|
|
||
| public StreamAcknowledgementManager(final AcknowledgementSetManager acknowledgementSetManager, | ||
| final DataStreamPartitionCheckpoint partitionCheckpoint, | ||
| final Duration partitionAcknowledgmentTimeout, | ||
| final int acknowledgementMonitorWaitTimeInMs, | ||
| final int checkPointIntervalInMs) { | ||
| this.acknowledgementSetManager = acknowledgementSetManager; | ||
| this.partitionCheckpoint = partitionCheckpoint; | ||
| this.partitionAcknowledgmentTimeout = partitionAcknowledgmentTimeout; | ||
| this.acknowledgementMonitorWaitTimeInMs = acknowledgementMonitorWaitTimeInMs; | ||
| this.checkPointIntervalInMs = checkPointIntervalInMs; | ||
| executorService = Executors.newSingleThreadExecutor(BackgroundThreadFactory.defaultExecutorThreadFactory("mongodb-stream-ack-monitor")); | ||
| } | ||
|
|
||
| void init(final Consumer<Void> stopWorkerConsumer) { | ||
| enableAcknowledgement = true; | ||
| executorService.submit(() -> monitorAcknowledgment(executorService, stopWorkerConsumer)); | ||
| } | ||
|
|
||
| private void monitorAcknowledgment(final ExecutorService executorService, final Consumer<Void> stopWorkerConsumer) { | ||
| long lastCheckpointTime = System.currentTimeMillis(); | ||
| CheckpointStatus lastCheckpointStatus = null; | ||
| while (!Thread.currentThread().isInterrupted()) { | ||
| final CheckpointStatus checkpointStatus = checkpoints.peek(); | ||
| if (checkpointStatus != null) { | ||
| if (checkpointStatus.isAcknowledged()) { | ||
| lastCheckpointStatus = checkpoints.poll(); | ||
| ackStatus.remove(checkpointStatus.getResumeToken()); | ||
| if (System.currentTimeMillis() - lastCheckpointTime >= checkPointIntervalInMs) { | ||
| LOG.debug("Perform regular checkpointing for resume token {} at record count {}", checkpointStatus.getResumeToken(), checkpointStatus.getRecordCount()); | ||
| partitionCheckpoint.checkpoint(checkpointStatus.getResumeToken(), checkpointStatus.getRecordCount()); | ||
| lastCheckpointTime = System.currentTimeMillis(); | ||
| } | ||
| } else { | ||
| LOG.debug("Checkpoint not complete for resume token {}", checkpointStatus.getResumeToken()); | ||
| final Duration ackWaitDuration = Duration.between(Instant.ofEpochMilli(checkpointStatus.getCreateTimestamp()), Instant.now()); | ||
| // Acknowledgement not received for the checkpoint after twice ack wait time | ||
| if (ackWaitDuration.getSeconds() >= partitionAcknowledgmentTimeout.getSeconds() * 2) { | ||
| // Give up partition and should interrupt parent thread to stop processing stream | ||
| if (lastCheckpointStatus != null && lastCheckpointStatus.isAcknowledged()) { | ||
| partitionCheckpoint.checkpoint(lastCheckpointStatus.getResumeToken(), lastCheckpointStatus.getRecordCount()); | ||
| } | ||
| LOG.warn("Acknowledgement not received for the checkpoint {} past wait time. Giving up partition.", checkpointStatus.getResumeToken()); | ||
| partitionCheckpoint.giveUpPartition(); | ||
| break; | ||
|
There was a problem hiding this comment. I don't understand this. This will end the thread and then also the stream thread. But, would these ever start again? It seems that when this happens, this node is now done working and unable to contribute. There was a problem hiding this comment. Yes, this will end the stream. New node or same node can acquire the partition and work from the checkpoint state. There was a problem hiding this comment. OK. I see. The |
||
| } | ||
| } | ||
| } | ||
|
|
||
| try { | ||
| Thread.sleep(acknowledgementMonitorWaitTimeInMs); | ||
| } catch (InterruptedException ex) { | ||
| break; | ||
| } | ||
| } | ||
| stopWorkerConsumer.accept(null); | ||
| executorService.shutdown(); | ||
|
There was a problem hiding this comment. Should L88-L89 be in init method instead of the callable? There was a problem hiding this comment. No, once this monitor exists because of ack wait timeout, it should stop the stream worker. |
||
| } | ||
|
|
||
| Optional<AcknowledgementSet> createAcknowledgementSet(final String resumeToken, final long recordNumber) { | ||
| if (!enableAcknowledgement) { | ||
| return Optional.empty(); | ||
| } | ||
|
|
||
| final CheckpointStatus checkpointStatus = new CheckpointStatus(resumeToken, recordNumber, Instant.now().toEpochMilli()); | ||
| checkpoints.add(checkpointStatus); | ||
| ackStatus.put(resumeToken, checkpointStatus); | ||
| return Optional.of(acknowledgementSetManager.create((result) -> { | ||
| if (result) { | ||
| final CheckpointStatus ackCheckpointStatus = ackStatus.get(resumeToken); | ||
| ackCheckpointStatus.setAcknowledgedTimestamp(Instant.now().toEpochMilli()); | ||
| ackCheckpointStatus.setAcknowledged(true); | ||
| LOG.debug("Received acknowledgment of completion from sink for checkpoint {}", resumeToken); | ||
| } else { | ||
| LOG.warn("Negative acknowledgment received for checkpoint {}, resetting checkpoint", resumeToken); | ||
| // default CheckpointStatus acknowledged value is false. The monitorCheckpoints method will time out | ||
| // and reprocess stream from last successful checkpoint in the order. | ||
| } | ||
| }, partitionAcknowledgmentTimeout)); | ||
| } | ||
|
|
||
| void shutdown() { | ||
| executorService.shutdown(); | ||
| } | ||
|
|
||
| @VisibleForTesting | ||
| ConcurrentHashMap<String, CheckpointStatus> getAcknowledgementStatus() { | ||
| return ackStatus; | ||
| } | ||
|
|
||
| @VisibleForTesting | ||
| ConcurrentLinkedQueue<CheckpointStatus> getCheckpoints() { | ||
| return checkpoints; | ||
| } | ||
| } | ||
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This class is only used by
StreamAcknowledgementManager. So you can move this into thestreampackage and make it package-private to help keep it encapsulated.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sure. will do.