[CASSANDRA-19755] Coordinator read latency metrics are inflated for some queries

• Type: Bug
• Status: Resolved
• Priority: Normal
• Resolution: Fixed
• Fix Version/s: 4.1.x, 5.0-rc1, 5.0
• Component/s: Observability/Metrics
• Bug Category: Correctness
• Severity: Normal
• Complexity: Normal
• Discovered By: User Report
• Platform: All
• Since Version: NA

Source Control Link

https://github.com/apache/cassandra/commit/1cd0b382143ec56118105a6ed991c0803f400b18

Test and Documentation Plan

new and existing tests

• Assignee: Sam Tunnicliffe
• Reporter: Sam Tunnicliffe

Authors

Sam Tunnicliffe

Reviewers

Marcus Eriksson

• Created: 05/Jul/24 17:26
• Updated: 18/Jul/24 21:11
• Resolved: 10/Jul/24 08:30

Description

When a partition read is decomposed on the coordinator into multiple single partition read queries, the latency metric captured in StorageProxy can be artificially increased.
This primarily affects reads where paging and aggregates are used or where an IN clause selects multiple partition keys.