[lennartkats-db]

Changes

• Apply added or changed configured MLflow experiment tags during bundle updates through the experiment tag API.
• Detect local tag changes while ignoring tags generated by the backend.
• Extend the experiment acceptance fixture to verify an added tag and a stable experiment ID. Sort its tag output so API ordering does not affect the snapshot.

Why

Issue #6521 reports that configured tags, including mlflow.workload_creation_policy_id, are ignored when an experiment already exists. Recreating the experiment to apply a tag would lose its existing runs or traces.

Tests

• go test ./bundle/direct/dresources -run TestExperimentDoUpdateSetsConfiguredTags -count=1 passed at 39dfd4f. The regression read old-policy after requesting new-policy at base 3aa5c96.
• go test ./acceptance -run 'TestAccept/bundle/resources/experiments/basic' -tail -test.v passed locally in both DMS modes.
• ./task check-changelog passed.
• The previous cloud failure was a tag-order snapshot mismatch. At head 39dfd4f, the Integration Tests check passed with all six cloud test jobs successful. Linux, macOS, and Windows direct/directdms test jobs, lint, validate-generated, and regression-report checks also passed.

This PR was written by Codex.

[github-actions]

Commit: 39dfd4f7a | Workflow run

Regression test report

Selected 1 changed tests (limit=20, 0 not selected)

2 tests: regression=2 unreleased=0 coverage=0 failing=0

test	branch	main (7adc80d)	latest (v1.24.0)
TestAccept/bundle/resources/experiments/basic/DMS=	✅	❌	❌
TestAccept/bundle/resources/experiments/basic/DMS=true	✅	❌	❌

TestAccept/bundle/resources/experiments/basic/DMS=: branch ✅ | main (7adc80d) ❌ | latest (v1.24.0) ❌

main (7adc80d):

acceptance_test.go:1345: Diff:
        --- bundle/resources/experiments/basic/output.txt
        +++ /tmp/TestAcceptbundleresourcesexperimentsbasicDMS=2966919059/002/output.txt
        @@ -150,10 +150,6 @@
               "value": "v1"
             },
             {
        -      "key": "k2",
        -      "value": "v2"
        -    },
        -    {
               "key": "mlflow.experiment.sourceName",
               "value": "/Users/[USERNAME]/new-name-[UNIQUE_NAME]"
             },
        
    acceptance_test.go:1345: 
        	Error Trace:	/home/runner/work/cli/cli/libs/testdiff/testdiff.go:22
        	            				/home/runner/work/cli/cli/acceptance/acceptance_test.go:1345
        	            				/home/runner/work/cli/cli/acceptance/acceptance_test.go:1098
        	            				/home/runner/work/cli/cli/acceptance/acceptance_test.go:660
        	Error:      	Not equal: 
        	            	expected: "update experiments.my_experiment\n\nPlan: 0 to add, 1 to change, 0 to delete, 0 unchanged\n"
        	            	actual  : "Plan: 0 to add, 0 to change, 0 to delete, 1 unchanged\n"
        	            	
        	            	Diff:
        	            	--- Expected
        	            	+++ Actual
        	            	@@ -1,4 +1,2 @@
        	            	-update experiments.my_experiment
        	            	+Plan: 0 to add, 0 to change, 0 to delete, 1 unchanged
        	            	 
        	            	-Plan: 0 to add, 1 to change, 0 to delete, 0 unchanged
        	            	-
        	Test:       	TestAccept/bundle/resources/experiments/basic/DMS=
        	Messages:   	bundle/resources/experiments/basic/out.plan.txt vs /tmp/TestAcceptbundleresourcesexperimentsbasicDMS=2966919059/001/out.plan.txt

latest (v1.24.0): same output as main (7adc80d)

TestAccept/bundle/resources/experiments/basic/DMS=true: branch ✅ | main (7adc80d) ❌ | latest (v1.24.0) ❌

main (7adc80d):

acceptance_test.go:1345: Diff:
        --- bundle/resources/experiments/basic/output.txt
        +++ /tmp/TestAcceptbundleresourcesexperimentsbasicDMS=true1808210503/002/output.txt
        @@ -150,10 +150,6 @@
               "value": "v1"
             },
             {
        -      "key": "k2",
        -      "value": "v2"
        -    },
        -    {
               "key": "mlflow.experiment.sourceName",
               "value": "/Users/[USERNAME]/new-name-[UNIQUE_NAME]"
             },
        
    acceptance_test.go:1345: 
        	Error Trace:	/home/runner/work/cli/cli/libs/testdiff/testdiff.go:22
        	            				/home/runner/work/cli/cli/acceptance/acceptance_test.go:1345
        	            				/home/runner/work/cli/cli/acceptance/acceptance_test.go:1098
        	            				/home/runner/work/cli/cli/acceptance/acceptance_test.go:660
        	Error:      	Not equal: 
        	            	expected: "update experiments.my_experiment\n\nPlan: 0 to add, 1 to change, 0 to delete, 0 unchanged\n"
        	            	actual  : "Plan: 0 to add, 0 to change, 0 to delete, 1 unchanged\n"
        	            	
        	            	Diff:
        	            	--- Expected
        	            	+++ Actual
        	            	@@ -1,4 +1,2 @@
        	            	-update experiments.my_experiment
        	            	+Plan: 0 to add, 0 to change, 0 to delete, 1 unchanged
        	            	 
        	            	-Plan: 0 to add, 1 to change, 0 to delete, 0 unchanged
        	            	-
        	Test:       	TestAccept/bundle/resources/experiments/basic/DMS=true
        	Messages:   	bundle/resources/experiments/basic/out.plan.txt vs /tmp/TestAcceptbundleresourcesexperimentsbasicDMS=true1808210503/001/out.plan.txt
FAIL

latest (v1.24.0): same output as main (7adc80d)

[eng-dev-ecosystem-bot]

Integration test report

Commit: 39dfd4f

Run: 41974864513

	Env	🔄​flaky	✅​pass	🙈​skip	Time
✅​	aws linux-2core-8gb		274	15	4:17
🔄​	aws-windows-latest-4core-16gb	11	265	13	14:51
✅​	azure linux-2core-8gb		273	15	4:17
✅​	azure-windows-latest-4core-16gb		275	13	4:15
✅​	gcp linux-2core-8gb		274	15	4:14
✅​	gcp-windows-latest-4core-16gb		276	13	4:29

11 interesting tests: 11 flaky

	Test Name	aws-windows-latest-4core-16gb
🔄​	TestFsCatOnADir	🔄​f
🔄​	TestFsCatOnADir/uc-volumes	🔄​f
🔄​	TestFsCatOnNonExistentFile	🔄​f
🔄​	TestFsCatOnNonExistentFile/uc-volumes	🔄​f
🔄​	TestFsCpDir/dbfs_to_uc-volumes	🔄​f
🔄​	TestFsCpDir/local_to_uc-volumes	🔄​f
🔄​	TestFsCpDir/uc-volumes_to_dbfs	🔄​f
🔄​	TestFsCpDir/uc-volumes_to_local	🔄​f
🔄​	TestFsCpDir/uc-volumes_to_uc-volumes	🔄​f
🔄​	TestFsCpFileToFile	🔄​f
🔄​	TestFsCpFileToFile/local_to_uc-volumes	🔄​f

Top 3 slowest tests (at least 2 minutes):

duration	env	testname
3:52	aws-windows-latest-4core-16gb	TestAccept
3:34	gcp-windows-latest-4core-16gb	TestAccept
3:22	azure-windows-latest-4core-16gb	TestAccept