HDDS-11063. TestSnapshotDiffManager#testThreadPoolIsFull is flaky without wait between batches #10581
+88
−87
There are no files selected for viewing
This file contains hidden or 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 hidden or 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 hidden or 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 |
|---|---|---|
|
|
@@ -95,11 +95,8 @@ | |
| import java.util.Optional; | ||
| import java.util.Set; | ||
| import java.util.UUID; | ||
| import java.util.concurrent.CountDownLatch; | ||
| import java.util.concurrent.ExecutionException; | ||
| import java.util.concurrent.TimeUnit; | ||
| import java.util.concurrent.atomic.AtomicInteger; | ||
| import java.util.function.BiFunction; | ||
|
|
@@ -1170,105 +1167,110 @@ private void uploadSnapshotDiffJobToDb(SnapshotInfo fromSnapshot, | |
|
|
||
| private static Stream<Arguments> threadPoolFullScenarios() { | ||
| return Stream.of( | ||
| Arguments.of("When the pool drains between job batches", | ||
| true, 45, 0), | ||
| // 10 running + 10 queued = 20 accepted, remaining 25 rejected | ||
| Arguments.of("When the pool does not drain between job batches", | ||
| false, 20, 25) | ||
| ); | ||
| } | ||
|
|
||
| @ParameterizedTest(name = "{0}") | ||
| @MethodSource("threadPoolFullScenarios") | ||
| public void testThreadPoolIsFull(String description, | ||
| boolean drainBetweenBatches, | ||
| int expectInProgressJobsCount, | ||
| int expectRejectedJobsCount) | ||
| throws Exception { | ||
| List<SnapshotInfo> snapshotInfos = createTestSnapshots(10); | ||
| SnapshotDiffManager spy = spy(snapshotDiffManager); | ||
|
|
||
| CountDownLatch blockWorkers = new CountDownLatch(1); | ||
| AtomicInteger completedJobs = new AtomicInteger(0); | ||
| doAnswer(invocation -> { | ||
| blockWorkers.await(); | ||
| completedJobs.incrementAndGet(); | ||
| return null; | ||
| }).when(spy).generateSnapshotDiffReport(anyString(), anyString(), | ||
| eq(VOLUME_NAME), eq(BUCKET_NAME), anyString(), anyString(), | ||
| eq(false), eq(false)); | ||
|
|
||
| if (drainBetweenBatches) { | ||
| blockWorkers.countDown(); | ||
| } | ||
|
|
||
| try { | ||
| List<SnapshotDiffResponse> responses = new ArrayList<>(); | ||
| int totalSubmitted = 0; | ||
| for (int i = 0; i < snapshotInfos.size(); i++) { | ||
| for (int j = i + 1; j < snapshotInfos.size(); j++) { | ||
| String fromSnapshotName = snapshotInfos.get(i).getName(); | ||
| String toSnapshotName = snapshotInfos.get(j).getName(); | ||
| responses.add(submitJob(spy, fromSnapshotName, toSnapshotName)); | ||
| totalSubmitted++; | ||
| } | ||
| if (drainBetweenBatches) { | ||
| final int expected = totalSubmitted; | ||
|
Contributor
There was a problem hiding this comment. Not in scope for the current PR to fix flakiness but may be considered as a follow up test improvement:
Contributor
Author
There was a problem hiding this comment. Great idea, I opened HDDS-15648 as a follow-up. |
||
| attempt(() -> { | ||
| if (completedJobs.get() < expected) { | ||
| throw new IllegalStateException("Waiting for jobs to complete"); | ||
| } | ||
| return null; | ||
| }, 50, TimeDuration.valueOf(100, TimeUnit.MILLISECONDS), null, null); | ||
| } | ||
| } | ||
|
|
||
| int inProgressJobsCount = 0; | ||
| int rejectedJobsCount = 0; | ||
| for (SnapshotDiffResponse response : responses) { | ||
| if (response.getJobStatus() == IN_PROGRESS) { | ||
| inProgressJobsCount++; | ||
| } else if (response.getJobStatus() == REJECTED) { | ||
| rejectedJobsCount++; | ||
| } else { | ||
| throw new IllegalStateException("Unexpected job status."); | ||
| } | ||
| } | ||
|
|
||
| assertEquals(expectInProgressJobsCount, inProgressJobsCount); | ||
| assertEquals(expectRejectedJobsCount, rejectedJobsCount); | ||
|
|
||
| int notFoundJobs = 0; | ||
| for (int i = 0; i < snapshotInfos.size(); i++) { | ||
| for (int j = i + 1; j < snapshotInfos.size(); j++) { | ||
| SnapshotDiffJob diffJob = | ||
| getSnapshotDiffJobFromDb(snapshotInfos.get(i), | ||
| snapshotInfos.get(j)); | ||
| if (diffJob == null) { | ||
| notFoundJobs++; | ||
| } | ||
| } | ||
| } | ||
|
|
||
| // assert that rejected jobs were removed from the job table as well. | ||
| assertEquals(expectRejectedJobsCount, notFoundJobs); | ||
| } finally { | ||
| blockWorkers.countDown(); | ||
| } | ||
| } | ||
|
|
||
| private List<SnapshotInfo> createTestSnapshots(int count) throws IOException { | ||
| List<SnapshotInfo> snapshotInfos = new ArrayList<>(); | ||
| for (int i = 0; i < count; i++) { | ||
| UUID snapshotId = UUID.randomUUID(); | ||
| String snapshotName = "snap-" + snapshotId; | ||
| SnapshotInfo snapInfo = new SnapshotInfo.Builder() | ||
| .setSnapshotId(snapshotId) | ||
| .setVolumeName(VOLUME_NAME) | ||
| .setBucketName(BUCKET_NAME) | ||
| .setName(snapshotName) | ||
| .setSnapshotPath("fromSnapshotPath") | ||
| .build(); | ||
| snapshotInfos.add(snapInfo); | ||
| when(snapshotInfoTable.get(getTableKey(VOLUME_NAME, BUCKET_NAME, snapshotName))) | ||
| .thenReturn(snapInfo); | ||
| } | ||
| return snapshotInfos; | ||
| } | ||
|
|
||
| private SnapshotDiffResponse submitJob(SnapshotDiffManager diffManager, | ||
|
|
@@ -1641,7 +1643,7 @@ public void testGetSnapshotDiffReportWhenDone() throws Exception { | |
|
|
||
| SnapshotDiffManager spy = spy(snapshotDiffManager); | ||
| SnapshotDiffReportOzone dummyReport = new SnapshotDiffReportOzone( | ||
| spy.getSnapshotRootPath(ctx.volumeName, ctx.bucketName).toString(), | ||
| ctx.volumeName, ctx.bucketName, ctx.fromSnapshotName, ctx.toSnapshotName, | ||
| expectedEntries, null); | ||
| doReturn(dummyReport).when(spy).createPageResponse(any(SnapshotDiffJob.class), | ||
|
|
||
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.
Uh oh!
There was an error while loading. Please reload this page.
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.
Thanks for the patch @rhalm.
nit: Use
full_thread_pool_size = 2 * OZONE_OM_SNAPSHOT_DIFF_THREAD_POOL_SIZEand45 - full_thread_pool_sizefor the expected accepted and rejected jobs here.Otherwise, LGTM.
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.
Thanks for the review @SaketaChalamchala, done.