It's highly recommended to use the embeded Sippy 'risk-analysis' to evaluate the results, instead of 'calling for flakes' as we are using to apply filters on the 'report' command.

The 'risk-analysis' should be called right after the openshift-tests have finished, considering all the metadata. It's important to evaluate the needed to collect the metadata for the execution to help on disconnected clusters OPCT-33 .

openshift-tests risk-analysis --junit-dir "${ARTIFACT_DIR}/junit" || true
 

References:

• https://github.com/mtulio/release/blob/master/ci-operator/step-registry/openshift/e2e/test/openshift-e2e-test-commands.sh#L43-L49