Skip to content

HDDS-17405. Intermittent failure in testFollowerReadOmProxyProviderFailoverOnConnectionFailure - #11728

Merged
peterxcli merged 1 commit into
apache:masterfrom
chihsuan:HDDS-17405
Oct 24, 2026
Merged

peterxcli merged 1 commit into
apache:masterfrom
chihsuan:HDDS-17405

Conversation

@chihsuan

@chihsuan chihsuan commented Oct 24, 2026 •

Copy link
Copy Markdown
Member

What changes were proposed in this pull request?

The test stops one OM and sends a read right away. If the stopped OM was the leader, the other two OMs need a moment to pick a new one. Sometimes the client gives up before that happens, and the test fails with "Leader is unknown".

This patch waits for the new leader before sending the read, same as HDDS-16843 did for the sibling test. The client still starts on the stopped OM, so the test still checks the failover.

What is the link to the Apache JIRA

https://issues.apache.org/jira/browse/HDDS-17405

How was this patch tested?

  • Made the failure happen every time locally by letting the client retry less. It failed without the fix and passed with it.
  • 10x10 flaky-test-check for the whole class: passed 100/100.
  • Fork CI: build-branch.

Generated-by: Claude Code (Opus 5.5)

…follower read test

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Copilot AI balanced review requested due to automatic review settings October 24, 2026 02:20

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟢 Approval recommended

The focused change removes the election-timing race without changing the client proxy or weakening failover assertions.

No open issues 🎉

What changed in this PR

This PR stabilizes the OM follower-read failover test by separating leader-election readiness from client connection failover.

Changes:

  • Waits for an OM leader after stopping the current proxy OM, before issuing the read.
  • Preserves failover coverage by leaving the client pointed at the stopped OM.
File Description
hadoop-ozone/​integration-test/​src/​test/​java/​org/​apache/​hadoop/​ozone/​om/​TestOzoneManagerHAFollowerReadWithStoppedNodes.java Adds the existing leader-readiness wait before the failover read.

Review effort: Balanced

@chihsuan chihsuan changed the title HDDS-17405. [Test] Wait for OM leader after stopping the proxy OM in follower read test HDDS-17405. Intermittent failure in testFollowerReadOmProxyProviderFailoverOnConnectionFailure Oct 24, 2026
@chihsuan
chihsuan requested a review from adoroszlai October 24, 2026 02:22
@smengcl
smengcl marked this pull request as ready for review October 24, 2026 05:50

@smengcl smengcl left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks good

@F64116045 F64116045 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks @chihsuan for the fix, lgtm

@Russole Russole left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks @chihsuan for the patch. LGTM!

@peterxcli peterxcli left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM +1

@peterxcli
peterxcli merged commit 1db07ca into apache:master Oct 24, 2026
41 checks passed
@peterxcli

Copy link
Copy Markdown
Member

Thanks @chihsuan @smengcl @F64116045 @Russole

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

7 participants