-
Bug
-
Resolution: Obsolete
-
Major
-
None
-
4.20
-
Quality / Stability / Reliability
-
False
-
-
2
-
Important
-
None
-
None
-
Rejected
-
MCO Sprint 276
-
1
-
In Progress
-
Release Note Not Required
-
None
-
None
-
None
-
None
-
None
Description of problem:
The OnClusterBuild tests running in the MCO disruptive suite do not have a healthy pass rate. This dashboard shows the OCB test 7-day pass rates and clicking into the list of failed runs for each test shows that the tests have been failing for a while.
Version-Release number of selected component (if applicable):
4.20 - I've only monitored tests in 4.20, but it's possible the tests are unstable in other versions also.
How reproducible:
Tests fail often.
Steps to Reproduce:
See this example test run to see 3/5 of the OCB tests failing as they seem to fail consistently on a pod readiness timeout.
Actual results:
The `MCO ocb PolarionID:83140-A MachineOSConfig with custom containerfile definition can be successfully applied`, `MCO ocb PolarionID:77781-A successfully built MachineOSConfig can be re-build`, and `A valid MachineOSConfig leads to a successful MachineOSBuild and cleanup of its associated resources` tests consistently fail on pod readiness timeouts across multiple platforms.
Expected results:
The OCB tests should meet the general 95% pass rate criteria for component readiness.
Additional info:
At the time of creating this bug, the OCB tests have the following 7-day pass rate:
Test Name | Pass Rate |
---|---|
MCO ocb PolarionID:77781-A successfully built MachineOSConfig can be re-build | 0% |
MCO ocb PolarionID:83140-A MachineOSConfig with custom containerfile definition can be successfully applied | 0% |
MCO ocb PolarionID:83141-A valid MachineOSConfig leads to a successful MachineOSBuild and cleanup of its associated resources | 1.7% |
MCO ocb PolarionID:77782-A MachineOSConfig with an unfinished build can be re-build | 89.7% |
MCO ocb PolarionID:83138-A MachineOSConfig fails to apply or degrades if invalid inputs are given | 91.2% |