diff --git a/apps/descheduler/README.md b/apps/descheduler/README.md index ff5bf5e..c55756e 100644 --- a/apps/descheduler/README.md +++ b/apps/descheduler/README.md @@ -1,27 +1,30 @@ -# Odroid CPU rebalancing +# Worker CPU rebalancing The upstream Descheduler chart runs every five minutes. `values.yaml` is the policy; the bootstrap Application selects its chart version. -The policy uses Metrics Server CPU usage: nodes below 40% are possible -receivers and nodes above 75% are possible eviction sources. Kubernetes still -chooses replacement placement. It does not live-migrate processes or guarantee -that a replacement lands on a different node. +The policy uses Metrics Server CPU usage: worker nodes below 25% are possible +receivers and nodes above 50% are possible eviction sources. The control-plane +node is excluded from eviction. Kubernetes chooses replacement placement; a +replacement can land on any eligible worker, including the original node. +No workload is pinned to the Xeon or an Odroid by this policy. Each cycle permits at most two evictions overall, one per node and namespace. Pods must be at least one hour old, belong to a controller with at least two replicas, have a PodDisruptionBudget, have no PVC, and fit another eligible -node. Only the six Odroids are eligible. Storage, networking, monitoring, -certificate management, and GitOps namespaces are excluded. EmptyDir alone is -not a protection for these replicated stateless workloads. +node. Storage, networking, monitoring, certificate management, and GitOps +namespaces are excluded. EmptyDir alone is not a protection for these +replicated stateless workloads. -Start with `cmdOptions.dry-run: true`, inspect Job logs, then remove that option -to allow evictions. To stop balancing, set `suspend: true` in values and let Argo -sync. Existing running Jobs must finish before suspension takes full effect. +To test a policy change, set `cmdOptions.dry-run: true` and inspect Job logs +before enabling evictions. To stop balancing, set `suspend: true` in values and +let Argo sync. Existing running Jobs must finish before suspension takes full +effect. -Review after changes to workload mix; do not remove PVC and infrastructure -protections to force a more even CPU graph. CPU requests determine whether a -replacement can fit even when observed utilization is low. +Review after changes to workload mix. CPU requests determine whether a +replacement can fit even when observed utilization is low. The Descheduler +cannot guarantee that a particular worker gains CPU usage; scheduling rules on +the workload and available memory still govern placement. -Live evictions enabled on 2026-11-01 after successful dry-run cycles. -The review selected replicated dashboard/ingress pods and respected the caps. +Live evictions were enabled on 2026-11-01 after successful dry-run cycles. +The 2026-11-22 dry run selected one Traefik replica and respected the caps. diff --git a/apps/descheduler/values.yaml b/apps/descheduler/values.yaml index 6f59930..748420a 100644 --- a/apps/descheduler/values.yaml +++ b/apps/descheduler/values.yaml @@ -12,7 +12,6 @@ resources: memory: 256Mi cmdOptions: v: 3 - dry-run: true deschedulerPolicy: maxNoOfPodsToEvictPerNode: 1 maxNoOfPodsToEvictPerNamespace: 1