[CASSANDRA-19693] Relax slow_query_log_timeout for MultiNodeSAITest

• Type: Improvement
• Status: Resolved
• Priority: Normal
• Resolution: Fixed
• Fix Version/s: 5.1
• Component/s: Feature/SAI, Test/fuzz
• Epic Link: Storage Attached Indexes (Phase 3)
• Change Category: Performance
• Complexity: Low Hanging Fruit
• Platform: All

Source Control Link

https://github.com/apache/cassandra/commit/ab1c3e4877f215380dc6881cd7793ab4890ccfb4

Test and Documentation Plan

This is a modification to an existing test.

• Assignee: Caleb Rackliffe
• Reporter: Caleb Rackliffe

Authors

Caleb Rackliffe

Reviewers

Alex Petrov

• Created: 07/Jun/24 18:44
• Updated: 11/Jun/24 16:20
• Resolved: 11/Jun/24 16:20

Description

To stress the paging subsystem, we intentionally use a comically low fetch size in MultiNodeSAITest. This can lead to some very slow queries when we get matches into the hundreds of rows. It looks like CASSANDRA-19534 has gotten a little more aggressive about how the slow query timeout is triggered, and there’s a lot of noise around this in the logs, even in local runs. I think bumping the default slow query timeout and perhaps the native protocol timeout a bit should clear this up.

Issue Links

• links to: GitHub Pull Request #3365