BeaverDeck logo
BeaverDeck

Home / BeaverDeck / Docs / Insights Guide / Node Insights / Node Capacity

Node Capacity

BeaverDeck uses this check to identify a specific node condition that may need operator review.

Permissions: viewing checks requires insights: view. Opening a linked object or logs requires the corresponding resource permission, and the BeaverDeck ServiceAccount must be allowed to read the Kubernetes resources used by the check. Suppressing a finding requires insights: edit and affects all users.
Check typenode-capacity
Insights sectionNode Insights
Alert severityCritical

When It Reports A Finding

The combined CPU requests, CPU limits, memory requests, or memory limits of scheduled non-terminal pods exceed the node's allocatable value for that resource.

Why This Is A Problem

Request overcommit indicates a scheduling or accounting problem. Limit overcommit can create contention, CPU throttling, memory pressure, and correlated failures when workloads peak together.

Recommended Response

  1. Use the listed scheduled pods to identify the largest requests and limits.
  2. Validate values against observed usage and correct accidental or stale resource settings in the owning workload.
  3. Redistribute workloads or add appropriately sized nodes when demand is legitimate.
  4. Review disruption budgets and availability before moving workloads from the node.

Scope And Limitations

Kubernetes commonly permits limit overcommit, so a limits-only finding can be intentional. Evaluate workload burst behavior and memory risk instead of treating every overcommit as an immediate failure.

After remediation: refresh Node Insights and verify the underlying resource or metric. Suppress the finding only when the condition is intentional and its risk is accepted.