Stop Node Drains from Evicting Your Entire Kubernetes Workload
PodDisruptionBudgets limit voluntary evictions during maintenance, but only if the selector matches your pods and you know what they can't prevent.
ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
PodDisruptionBudgets limit voluntary evictions during maintenance, but only if the selector matches your pods and you know what they can't prevent.
A node drain hangs with no obvious error. The cause is often a PodDisruptionBudget returning HTTP 429 from the eviction API. Learn how minAvailable and maxUnavailable actually work, why percentages round down, and how to avoid undrainable nodes.
Learn how a PodDisruptionBudget keeps a minimum number of your application’s pods running while you drain or upgrade nodes, and see the trade‑offs of setting it too tightly.
Learn how a Pod Disruption Budget (PDB) guarantees a minimum number of replicas stay up while you drain nodes, plus a concrete example and the trade‑offs to watch for.
Learn how a PodDisruptionBudget limits voluntary pod evictions during node drains, keeping replica counts above a safe threshold while avoiding overly restrictive settings that can block maintenance.
When draining a node, the Kubernetes control plane evicts pods to honor Pod Disruption Budgets (PDBs). The goal is to verify whether the eviction sequence across multiple overlapping PDBs is deterministic enough to never temporarily violate a PDB’s minimum‑available count. Current documentation notes that the kube‑controller‑manager may evict pods from diffe
A StatefulSet scales down by deleting its highest‑ordinal pods, and a node drain may issue a direct DELETE request that bypasses the eviction subresource. PodDisruptionBudget evaluates minAvailable or maxUnavailable against the current count of ready pods, but the definition of ready can differ for pods that are terminating yet still counted as available, an