KAFKA-19467: Add a metric for controller thread idleness

[MahsaSeifikar]

This change adds the metric ControllerEventManager::AvgIdleRatio which
measures the amount of time the controller spends blocked waiting for
events vs the amount of time spent processing events. A value of 1.0
means that the controller spent the entire interval blocked waiting for
events.

Reviewers: José Armando García Sancio [email protected], Kevin Wu
[email protected], Alyssa Huang [email protected], TengYao
Chi [email protected], Reviewers: Chia-Ping Tsai
[email protected]

[frankvicky]

@MahsaSeifikar: Please apply spotless

[kevin-wu24]

Thanks for the changes @MahsaSeifikar. I have a question about the wrapper classes being introduced:

[kevin-wu24]

Thanks for the updates @MahsaSeifikar

Two more comments:

[kevin-wu24]

Changes LGTM @MahsaSeifikar.

[MahsaSeifikar]

@chia7712 Could you please review this metric KIP when you get a chance?

[ahuang98]

what do you think about moving updateIdleStartTime into updateEventQueueProcessingTime? you would need to call it in one other place outside of that.

otherwise looks reasonable to me, thanks Mahsa!

[jsancio]

Thanks for the change @MahsaSeifikar . I left some high-level comments and observations.

[jsancio]

Thanks for the changes @MahsaSeifikar

[MahsaSeifikar]

This AclControlManagerTest.testDeleteExceedsMaxRecords() failure seems unrelated to this PR and has shown flaky behavior in the past 28 days, link.

Here is the ticket KAFKA-19513 to fix this.

[jsancio]

LGTM

[MahsaSeifikar]

This AdjustStreamThreadCountTest.shouldAddAndRemoveThreadsMultipleTimes failure seems unrelated to this PR and has shown flaky behavior in the past 2 days, link.

Here is the pr to fix this.

[jsancio]

LGTM

[jsancio]

Test failures seem unrelated to this change:

Found 1 test failures:
FAILED ❌ AdjustStreamThreadCountTest > shouldAddAndRemoveThreadsMultipleTimes()
Found 1 flaky test failures:
FLAKY ⚠️ DescribeStreamsGroupTest > testDescribeStreamsGroupWithShortTimeout()