https://prow.ci.openshift.org/view/gs/origin-ci-test/logs/aggregated-aws-sdn-upgrade-4.14-micro-release-openshift-release-analysis-aggregator/1670859888839888896

This is a brand new disruption backend that went live around June 10th. We did not intend to start testing on it, but the aggregated tests are going against all backends.

This test uses days 3 through 10 prior to now for it's comparisons.

It is a bug that it started enforcing before we had at least that number of days, or a minimum number of runs.

However we now have our 10 days of data, the test should be valid. The P75 has moved up to 4s so presumably this test has a good chance of passing now. https://grafana-loki.ci.openshift.org/d/ISnBj4LVk/disruption?orgId=1&var-platform=aws&var-percentile=P75&var-backend=kube-api-http2-external-lb-reused-connections&var-releases=4.14&var-upgrade_type=minor&var-upgrade_type=micro&var-networks=sdn&var-networks=ovn&var-topologies=ha&var-architectures=amd64&var-min_job_runs=10&var-lookback=1

Also we should NEVER test ci-cluster-network-liveness, both aggreated or per job. This backend has no bearing on the code in the cluster and is purely checking the network. Disable testing for this everywhere.

What should we do here?

• fix to not test if lacking job runs or a window of job runs
• stop enforcing on these tests even though we should now have enough data?