fix(task): inactive task runs when updated

[raffs]

Closes #21874

When a tasks is being updated, ensure is scheduled only when the tasks status is active. Otherwise, an already-inactivated tasks might be scheduled if the task status is not being changed.

• CHANGELOG.md updated with a link to the PR (not the Issue)
• Well-formatted commit messages
• Rebased/mergeable
• Tests pass
• Signed CLA (if not already signed)

[williamhbaker]

Great job tracking this down @raffs!

One other thing you'll need to do is add a test case to show that the bug is fixed. I think you could create a new case in the existing Test_Coordinator_Scheduler_Methods test that shows the expected scheduler calls for updating an inactive task without changing it's status.

[raffs]

Great job tracking this down @raffs!

One other thing you'll need to do is add a test case to show that the bug is fixed. I think you could create a new case in the existing Test_Coordinator_Scheduler_Methods test that shows the expected scheduler calls for updating an inactive task without changing it's status.

Thanks @wbaker85.

I have added the test case as requested. Previously, I wasn't entire sure how to create the test case for a tasks that shouldn't be scheduled. But now, I can see that we diff the config on every test and having a empty schedule for updating an already inactive tasks seems to do trick for testing this case.

[danxmoran]

LGTM! The CI failure is unrelated to this PR, we added a new job that apparently doesn't work on PRs from forked repos. We'll patch it up and then a rebase/merge from master should get this passing.

[williamhbaker]

@raffs - I merged #22232 just now, so if you can rebase onto the latest master & resolve the conflict in the changelog we should be able to get this merged!

[raffs]

LGTM! The CI failure is unrelated to this PR, we added a new job that apparently doesn't work on PRs from forked repos. We'll patch it up and then a rebase/merge from master should get this passing.

Thank you for the info @danxmoran

@raffs - I merged #22232 just now, so if you can rebase onto the latest master & resolve the conflict in the changelog we should be able to get this merged!

Great! thanks for the info.