> This page is for Health Validation.

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/health-validation/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/health-validation/_mcp/server.

# CPU Checks

CPU covers host processor inventory and availability signals.

## System CPU Count

`cpu_count`

Verifies every logical CPU in the Linux kernel's possible mask is online and available for scheduling.

| Property | Value |
|---|---|
| Provenance | <nobr>`nvidia`</nobr> |
| Condition scope | <nobr>`NODE`</nobr> |
| Execution class | <nobr>`PASSIVE`</nobr> |
| Scheduling cadence | <nobr>`CONTINUOUS`</nobr> |
| Maturity | <nobr>`stable`</nobr> |
| Execution contexts | <nobr>`OPERATOR_INFRA`</nobr>, <nobr>`TENANT_GUEST`</nobr> |
| Required signals | <nobr>`possible_cpu_count`</nobr>, <nobr>`online_cpu_count`</nobr> |

**Rules**

- **CRITICAL** when `online_cpu_count != possible_cpu_count`
  - Message: Expected all $\{possible_cpu_count\} possible logical CPUs to be online; found $\{online_cpu_count\} online.

**Recommended action:** operational state `drain`, then `reset_gpu` → `restart_os` → `service_required`

## NUMA Node Count

`numa_node_count`

Verifies a node brought online the expected number of NUMA nodes for its architecture.

| Property | Value |
|---|---|
| Provenance | <nobr>`nvidia`</nobr> |
| Condition scope | <nobr>`NODE`</nobr> |
| Execution class | <nobr>`PASSIVE`</nobr> |
| Scheduling cadence | <nobr>`CONTINUOUS`</nobr> |
| Maturity | <nobr>`stable`</nobr> |
| Execution contexts | <nobr>`OPERATOR_INFRA`</nobr>, <nobr>`TENANT_GUEST`</nobr> |
| Architectures | <nobr>`GB200`</nobr>, <nobr>`GB300`</nobr> |
| Required signals | <nobr>`numa_node_count`</nobr> |

**Thresholds**

| Name | Value | Unit | Category | Description |
|---|---|---|---|---|
| <nobr>`expected_numa_node_count`</nobr> | GB200: 34, GB300: 34, other: 34 | count | <nobr>`DEFAULT`</nobr> | Expected number of NUMA nodes online. GB200 and GB300 at 34. Override per fleet. |

**Rules**

- **CRITICAL** when `numa_node_count != expected_numa_node_count`
  - Message: Expected $\{expected_numa_node_count\} NUMA nodes online, found $\{numa_node_count\}.

**Recommended action:** operational state `drain`, then `reset_gpu` → `restart_os` → `service_required`