This appears to be a pretty rare combo only running in presubmits. https://sippy.dptools.openshift.org/sippy-ng/jobs/Presubmits/analysis?filters=%7B%22items%22%3A%5B%7B%22columnField%22%3A%22name%22%2C%22operatorValue%22%3A%22equals%22%2C%22value%22%3A%22pull-ci-openshift-kubernetes-master-e2e-gcp-ovn-upgrade%22%7D%5D%7D

The disruption data for some lb backed services appears noticably worse than for minor upgrades, which is confusing to us.

Add a periodic and run relatively often, roughly 5x a day will get us the required 100 runs over 3 weeks to be included in the query_results.json file.