[IMPALA-12960] Incorrect RowsPassedThrough Metric in Streaming Aggregation

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: Impala 4.4.0
• Component/s: Backend
• Epic Color: ghx-label-1
• Assignee: Yida Wu
• Reporter: Yida Wu
• Created: 31/Mar/24 03:31
• Updated: 04/Apr/24 21:18
• Resolved: 04/Apr/24 21:18

Description

The logic in grouping aggregation uses the row number of the output batch as the number of the current aggregator returning, however the output batch is shared with other aggregators in streaming aggregation and may not necessarily start at 0 initially, this can result in incorrect numbering and subsequently affect the accuracy of the RowsPassedThrough metric.

https://github.com/apache/impala/blob/f55077007bf68e6cbeaa15cf270c333af847a1f1/be/src/exec/grouping-aggregator.cc#L519