Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[AUTOCUT] Gradle Check Flaky Test Report for RemoteStoreClusterStateRestoreIT #14326

Open
opensearch-ci-bot opened this issue Jun 13, 2024 · 13 comments
Labels
autocut ClusterManager:RemoteState flaky-test Random test failure that succeeds on second run >test-failure Test failure from CI, local build, etc.

Comments

@opensearch-ci-bot
Copy link
Collaborator

opensearch-ci-bot commented Jun 13, 2024

Flaky Test Report for RemoteStoreClusterStateRestoreIT

Noticed the RemoteStoreClusterStateRestoreIT has some flaky, failing tests that failed during post-merge actions.

Details

Git Reference Merged Pull Request Build Details Test Name
ccf5289 14221 40884 org.opensearch.remotestore.RemoteStoreClusterStateRestoreIT.testFullClusterRestoreGlobalMetadata

The other pull requests, besides those involved in post-merge actions, that contain failing tests with the RemoteStoreClusterStateRestoreIT class are:

For more details on the failed tests refer to OpenSearch Gradle Check Metrics dashboard.

@shiv0408
Copy link
Member

This test should not be flaky now, the fix was merged in #14230

@shiv0408
Copy link
Member

@prudhvigodithi Wanted to confirm if a test becomes flaky again after the issue has been resolved, does our workflow create a new issue or open the previous closed issue?

@prudhvigodithi
Copy link
Member

Hey @shiv0408 yes the automation will re-open the issue if it was closed within 3 days (which is configurable) else it will re-create a new issue. In both cases the issue body will have the latest flaky test information.
Thanks
@getsaurabh02

@shiv0408
Copy link
Member

@andrross Do you know how this issue got re-opened? I closed the issue as the fix was made, do we need to keep the issue open?

@andrross
Copy link
Member

andrross commented Jun 17, 2024

@shiv0408 This is the PR that caused it to reopen: #14345

I think @prudhvigodithi is looking into a similar case. It might be that a PR is open that hasn't rebased with your fix, and that causes it to reopen if the test fails.

@shiv0408
Copy link
Member

This PR was merged 3 days ago, but the CI bot opened is couple of hours ago.

@prudhvigodithi
Copy link
Member

The automation looks for last 30 days build data in post merge action and if found RemoteStoreClusterStateRestoreIT will re-open/create the issue. So even though the fix was pushed, the RemoteStoreClusterStateRestoreIT was found failing in past 30 days hence it re-opened the existing issue.

How about we allow the automation to close the issue? Since now we have all the issues created for the flaky tests, we can reduce to identify the flaky tests in last 15 days and auto-close the issue if not failing in last 15 days, this way the user need not worry about closing the issue ? WDYT @andrross @shiv0408
Adding @dblock @reta @getsaurabh02

@shiv0408
Copy link
Member

How about we allow the automation to close the issue?

This might present incorrect information that test is still flaky, while it might have been resolved. Anyway, we are going to reopen the issue if test turns to be flaky again.

@andrross
Copy link
Member

@prudhvigodithi If an issue exists and it is closed, and none of the failures are newer than the close date, then can we keep it closed?

@prudhvigodithi
Copy link
Member

Sure, what we can do it the following:

  • The automation will go back 30 days and check for failures in post merge action, flag the tests and create a detailed issue report. The issue body will be updated with the latest information as and when it has the latest information which will be indexed through the Gradle Check build. This is the current state.

  • If the issue is closed (considering the flaky test is fixed by the user) the automation should not re-open unless the data is different from what shown in the issue body, if anything (the issue body) is different after closed then it should re-open the issue. Here the data to compare is the markdown table and not the linked PR's as during the PR creation the failures sometimes could be genuine. So re-open when seen a new failure (with a different post merge commit) after the issue is closed. This should also solve the problem where sometimes we think the Flaky test is fixed but would re-occur and with new reoccurrence the issue should re-open with new data.

@peterzhuamazon
Copy link
Member

Can we compare the base of the PRs to determine if such issue is legit or we inform user to rebase?

It is possible that an issue can re-surface due to regression.

At least if we see the PR has a updated base, we can re-open the old issue if needed. Else, inform user to rebase their branch. Thanks.

@shiv0408
Copy link
Member

shiv0408 commented Jun 18, 2024

@prudhvigodithi we should create an issue to add your second point as an enhancement on our current state.

@dblock
Copy link
Member

dblock commented Jul 1, 2024

[Catch All Triage - Attendees 1, 2, 3, 4, 5]

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
autocut ClusterManager:RemoteState flaky-test Random test failure that succeeds on second run >test-failure Test failure from CI, local build, etc.
Projects
None yet
Development

No branches or pull requests

6 participants