Description of problem:

For years, the TechPreviewNoUpgrade alert has used:

cluster_feature_set{name!="", namespace="openshift-kube-apiserver-operator"} == 0

But recently testing 4.12.19, I saw the alert pending with name="LatencySensitive". Alerting on that not-useful-since-4.6 feature set is fine (OCPBUGS-14497), but TechPreviewNoUpgrade isn't a great name when the actual feature set is LatencySensitive. And the summary and descripition don't apply to LatencySensitive either.

Version-Release number of selected component (if applicable):

The buggy expr / alertname pair shipped in 4.3.0.

How reproducible:

All the time.

Steps to Reproduce:

1. Install a cluster like 4.12.19.
2. Set the LatencySensitive feature set:

$ oc patch featuregate cluster --type=json --patch='[{"op":"add","path":"/spec/featureSet","value":"LatencySensitive"}]'

3. Check alerts with /monitoring/alerts?rowFilter-alert-source=platform&resource-list-text=TechPreviewNoUpgrade in the web console.

Actual results:

TechPreviewNoUpgrade is pending or firing.

Expected results:

Something appropriate to LatencySensitive, like a generic alert that covers all non-default feature sets, is pending or firing.