A zone-label migration is rolled out and later stateful search replicas remain Pending even though enough nodes exist in the cluster.
An OpenSearch cluster on Kubernetes stays yellow because topology spread and anti-affinity now conflict after a zone label migration, leaving one replica perpetually unschedulable
The workload spec is valid, yet the scheduler cannot satisfy both the old anti-affinity rule and the new zone spread expectation at once.
Scenario
What to check first
- Identify the primary failure signal in the The Scheduler Was Given Two Placement Rules That No Longer Shared the Same Labels scenario.
- Separate visible symptoms from the underlying technical dependency.
- Describe the safest recovery path and the follow-up prevention work.
Checking checklist
- Summarize the current impact and the last known change.
- Collect direct evidence from logs, runtime state, and configuration before changing anything.
- Separate immediate recovery from permanent prevention work.
Recovery and prevention
Compare placement rule inputs after the label migration before scaling the cluster out.
Questions worth viewing together
Community-field Kubernetes problem inspired by OpenSearch and scheduler discussions where zone-label changes made topology spread and anti-affinity rule... Scheduler conflicts often appear only after label migrations make two formerly compatible rules diverge.
Teams often blame lack of capacity when the real issue is contradictory placement logic.
Label migrations should include a scheduler policy audit for StatefulSets with both spread and anti-affinity rules.
Similar cases seen in the field