Skip to content

[client][backport] Skip blocked bucket instead of stopping drain loop #2953

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
fresh-borzoni wants to merge 1 commit into
base: main
Choose a base branch
from
+46 −1
Open

[client][backport] Skip blocked bucket instead of stopping drain loop #2953

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion fluss-client/src/main/java/org/apache/fluss/client/write/RecordAccumulator.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -729,7 +729,8 @@ private List<ReadyWriteBatch> drainBatchesForOneNode(Cluster cluster, Integer no
break;
} else {
if (shouldStopDrainBatchesForBucket(first, tableBucket)) {
break;
// Buckets are independent — skip this one, keep draining others.
continue;
}
}

Expand Down
44 changes: 44 additions & 0 deletions fluss-client/src/test/java/org/apache/fluss/client/write/RecordAccumulatorTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -602,6 +602,50 @@ private void verifyTableBucketInBatches(
assertThat(tableBucketsInBatch).containsExactlyInAnyOrder(tb);
}

@Test
void testDrainContinuesWhenBucketAtMaxInflight() throws Exception {
int batchSize = 1024;
conf.set(ConfigOptions.CLIENT_WRITER_BATCH_TIMEOUT, Duration.ofMillis(0));
conf.set(ConfigOptions.CLIENT_WRITER_BUFFER_MEMORY_SIZE, new MemorySize(10L * batchSize));
conf.set(ConfigOptions.CLIENT_WRITER_BUFFER_PAGE_SIZE, new MemorySize(256));
conf.set(ConfigOptions.CLIENT_WRITER_BATCH_SIZE, new MemorySize(batchSize));

IdempotenceManager idempotenceManager =
new IdempotenceManager(true, /* maxInflightPerBucket */ 1, null, null);
idempotenceManager.setWriterId(1L);

RecordAccumulator accum =
new RecordAccumulator(
conf, idempotenceManager, TestingWriterMetricGroup.newInstance(), clock);

cluster = updateCluster(Arrays.asList(bucket1, bucket2));
IndexedRow row = indexedRow(DATA1_ROW_TYPE, new Object[] {1, "a"});

// Drain both buckets so each has 1 in-flight batch.
accum.append(createRecord(row), writeCallback, cluster, tb1.getBucket(), false);
accum.append(createRecord(row), writeCallback, cluster, tb2.getBucket(), false);

Map<Integer, List<ReadyWriteBatch>> firstDrain =
accum.drain(cluster, Collections.singleton(node1.id()), Integer.MAX_VALUE);
List<ReadyWriteBatch> firstBatches = firstDrain.get(node1.id());
assertThat(firstBatches).hasSize(2);

// Complete only tb2, leaving tb1 at max in-flight.
ReadyWriteBatch tb2Batch =
firstBatches.stream().filter(b -> b.tableBucket().equals(tb2)).findFirst().get();
idempotenceManager.handleCompletedBatch(tb2Batch);

// Append again to both. On drain, tb1 should be skipped but tb2 should still be drained.
accum.append(createRecord(row), writeCallback, cluster, tb1.getBucket(), false);
accum.append(createRecord(row), writeCallback, cluster, tb2.getBucket(), false);

Map<Integer, List<ReadyWriteBatch>> secondDrain =
accum.drain(cluster, Collections.singleton(node1.id()), Integer.MAX_VALUE);
List<ReadyWriteBatch> secondBatches = secondDrain.get(node1.id());
assertThat(secondBatches).hasSize(1);
assertThat(secondBatches.get(0).tableBucket()).isEqualTo(tb2);
}

/** Return the offset delta. */
private int expectedNumAppends(IndexedRow row, int batchSize) {
int size = recordBatchHeaderSize(CURRENT_LOG_MAGIC_VALUE);
Expand Down