Re-read the disruptions page this morning. The doc is very clear that PDBs only apply to *voluntary* disruptions — drains, evictions via the API.
Yet in almost every postmortem I've seen, someone says "but we had a PDB" about a node that died. How do you communicate this to teams that treat PDBs as an availability guarantee?
We stopped calling it an availability setting internally. We call it a "drain throttle". The renaming did more than any doc link.
Stealing this.
The other common mistake: minAvailable equal to replicas. Then no drain can ever proceed and cluster upgrades stall.
We had a cluster upgrade blocked for two days because a team set minAvailable: 3 on a 3-replica Deployment. The node drain just loops on "Cannot evict pod as it would violate the pod's disruption budget".
Our admission policy now rejects PDBs where minAvailable >= replicas, or maxUnavailable: 0. Teams complain for a week, then forget about it.
