[AUTOCUT] Gradle Check Flaky Test Report for StreamStringTermsAggregatorTests

[opensearch-ci-bot]

Flaky Test Report for StreamStringTermsAggregatorTests

Noticed the StreamStringTermsAggregatorTests has some flaky, failing tests that failed during post-merge actions and timer trigerred run on main branch.

Details

Git Reference	Merged Pull Request	Build Details	Test Name
main	null	61913 61919 62405 62461 62625 62704 62716 62799 62882 63161 63378 63656 63855 63960	org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testBuildAggregationsBatchWithSize org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testBuildAggregationsBatchWithSingleValuedOrds org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testReduceSingleAggregation org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testReduceWithSizeLimit
ff854c3	19097	63938	org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testReduceWithSizeLimit
b856451	19261	63760	org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testReduceWithSizeLimit
26a4014	19203	63222	org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testBuildAggregationsBatchWithSize
981127d	18918	62920	org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testBuildAggregationsBatchWithSize
e37b2cd	19107	62620	org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testBuildAggregationsBatchWithSize
79d3e38	18961	61948	org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testBuildAggregationsBatchWithCountOrder
a23bbc4	18946	61900	org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testBuildAggregationsBatchWithSize

The other pull requests, besides those involved in post-merge actions, that contain failing tests with the StreamStringTermsAggregatorTests class are:

• 18942
• 19063
• 19358
• 18523
• 18568
• 18922
• 18964
• 19001
• 19057
• 19096
• 19159
• 19178
• 19188
• 19214
• 19229
• 19241
• 19244
• 19253
• 19307
• 19345
• 19347
• 19350
• 19356
• 19387
• 19397

For more details on the failed tests refer to OpenSearch Gradle Check Metrics dashboard.

[prudhvigodithi]

Able to reproduce with AF249FA66A40E125 seed.

./gradlew ':server:test' --tests 'org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testBuildAggregationsBatchWithSingleValuedOrds' -Dtests.seed=AF249FA66A40E125

[mkhludnev]

./gradlew ':server:test' --tests 'org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testReduceWithSizeLimit' -Dtests.seed=EB1F22CBC8D5825B

when it asserts cat_7 it got

java.lang.AssertionError: 
    Expected: <1L>
         but: was <3L>
        at __randomizedtesting.SeedInfo.seed([EB1F22CBC8D5825B:78E3C73A42C2AE3A]:0)
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:18)
        at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:6)
        at org.apache.lucene.tests.util.LuceneTestCase.assertThat(LuceneTestCase.java:2104)
        at org.opensearch.search.aggregations.bucket.terms.StreamStringTermsAggregatorTests.testReduceWithSizeLimit(StreamStringTermsAggregatorTests.java:1029)

The problem for this seed is that the second reader got three segments for seven docs (2+2+1) and https://github.com/opensearch-project/OpenSearch/pull/18874/files#diff-86c755a522dd8e362366b3070e18089926a7d938a9ed2d7d2556b9d1a2cd5002R118 seems like is unaware of segments. It collects persegment ordinals into the single docCount without attempting to remap them across segments.

I'm in doubts re #18874 fwiw @rishabhmaurya @bowenlan-amzn

It's broken before reduce, aggregating the second reader got cat_7=3 due to squashing segments .

[rishabhmaurya]

@mkhludnev thank you for investigation. Taking a look