Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
8.9, 8.10
-
None
-
New
Description
The ML team at Elastic reported that a large scroll with an Elasticsearch nightly build that uses Lucene 9.0 snapshot returns fewer documents than expected. I looked into it and found that the competitive iterator can wrongly skip docs with a chunked bulk scorer.