majiayu000 Check Node Resource Usage
Preconditions
Before applying this skill, verify:
- Metrics server is available
- Need to understand cluster resource status
Actions
1. Get Resource Usage for All Nodes
Retrieve CPU and memory usage metrics for all nodes.
mcp_tool: kubernetes-mcp-server/nodes_top
params: {}
timeout: 30s
Success Criteria
The skill succeeds when:
- Node resource metrics retrieved successfully
- Resource pressure identified if present
Failure Handling
If metrics unavailable:
- Check if metrics-server is deployed
- Request prometheus-mcp-server deployment
Examples
Input Context:
{
"reason": "investigating scheduling issues",
"suspected_node": "worker-01"
}
Expected Outcome: Node metrics retrieved showing CPU and memory usage for each node, with any nodes under pressure clearly identified.