Skip to content

Fix [BUG] ClusterMaxMergesAtOnceIT is flaky #19647

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
wants to merge 3 commits into
base: main
Choose a base branch
from
Open

Fix [BUG] ClusterMaxMergesAtOnceIT is flaky #19647

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
569ee65
Fix flack test #18056
Oct 15, 2025
88f189f
Merge remote-tracking branch 'ApacheFork/main' into flacky/test1
Oct 15, 2025
17e5f99
remove AwaitsFix annotation
Oct 15, 2025
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
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
- Fix flaky test FieldDataLoadingIT.testIndicesFieldDataCacheSizeSetting ([#19571](https://github.com/opensearch-project/OpenSearch/pull/19571))
- Avoid primary shard failure caused by merged segment warmer exceptions ([#19436](https://github.com/opensearch-project/OpenSearch/pull/19436))
- Fix pull-based ingestion out-of-bounds offset scenarios and remove persisted offsets ([#19607](https://github.com/opensearch-project/OpenSearch/pull/19607))
- Fix flaky test ClusterMaxMergesAtOnceIT.testClusterLevelDefaultUpdatesMergePolicy ([#18056](https://github.com/opensearch-project/OpenSearch/issues/18056))

### Dependencies
- Update to Gradle 9.1 ([#19575](https://github.com/opensearch-project/OpenSearch/pull/19575))
Expand Down
7 changes: 5 additions & 2 deletions server/src/internalClusterTest/java/org/opensearch/index/ClusterMaxMergesAtOnceIT.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,7 +64,6 @@ public void setUp() throws Exception {
internalCluster().startClusterManagerOnlyNode();
}

@AwaitsFix(bugUrl = "https://github.com/opensearch-project/OpenSearch/issues/18056")
public void testClusterLevelDefaultUpdatesMergePolicy() throws ExecutionException, InterruptedException {
String clusterManagerName = internalCluster().getClusterManagerName();
List<String> dataNodes = new ArrayList<>(internalCluster().getDataNodeNames());
Expand Down Expand Up @@ -96,11 +95,15 @@ public void testClusterLevelDefaultUpdatesMergePolicy() throws ExecutionExceptio
assertEquals(20, ((OpenSearchTieredMergePolicy) indexService.getIndexSettings().getMergePolicy(true)).getMaxMergeAtOnce());
assertEquals(20, ((OpenSearchTieredMergePolicy) secondIndexService.getIndexSettings().getMergePolicy(true)).getMaxMergeAtOnce());

int replicas = randomIntBetween(1, Math.max(1, internalCluster().numDataNodes() - 1));
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

More for my own curiosity than anything -- how does this end up fixing the test?

By the default, the replica count would be 1, based on the implementation of indexSettings() on line 56. So I don't understand how setting the replica count to a random number between 1 and 1 (since it looks like numDataNodes = 2) changes anything.

I'm not particularly familiar with these tests, but I'm trying to understand more. Thanks!

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe generating a random integer here is enough to change the random result from randomFrom(dataNodes) on line 108 (before)/111 (after)?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@msfroh I think you're right. I confirmed that just adding a randomInt() call here and doing nothing with the result is enough to make the test pass.

// Create index with index level override in settings
indexName = "log-myindex-3";
createIndex(
indexName,
Settings.builder().put(TieredMergePolicyProvider.INDEX_MERGE_POLICY_MAX_MERGE_AT_ONCE_SETTING.getKey(), 15).build()
Settings.builder()
.put(TieredMergePolicyProvider.INDEX_MERGE_POLICY_MAX_MERGE_AT_ONCE_SETTING.getKey(), 15)
.put(IndexMetadata.SETTING_NUMBER_OF_REPLICAS, replicas)
.build()
);
ensureYellowAndNoInitializingShards(indexName);
ensureGreen(indexName);
Expand Down
Loading