* Custom Device Fingerprinting on CPPM:
* To create a custom fingerprint, you first need to connect a known device of that type to the network.
* CPPM will discover the device in its Endpoints Repository, allowing you to analyze its attributes (e.g., MAC OUI, DHCP options) and create custom profiling rules.
* Option Analysis:
* Option A: Correct. Discovering a known device in the Endpoints Repository is a prerequisite for creating accurate custom fingerprint rules.
* Option B: Incorrect. CPDI integration is not required for custom fingerprints on CPPM.
* Option C: Incorrect. XML rules are not pre-defined; they are created dynamically based on observed attributes.
* Option D: Incorrect. The "Automatically download Endpoint Profiler Fingerprints" setting is unrelated to custom profiling.

For centralized tracking of administrator command activity, TACACS+ with ClearPass is the correct solution.
TACACS+ is designed for network device administration because it separates authentication, authorization, and accounting. When ClearPass acts as a TACACS+ server, AOS-CX switches can send administrative login and command-related information to a centralized policy and audit platform. RADIUS start-stop accounting with the port-access option is for network access sessions, not command auditing. SSH public key authentication improves login security, but it does not create a centralized record of entered commands. Basic syslog at error severity records system events and errors, not a reliable command audit trail. TACACS+ command authorization with CPPM is the correct administrative control.

The Network Analytics Engine (NAE) in AOS-CX switches provides intelligent monitoring, troubleshooting, and performance analysis through predefined or custom scripts. Here's an analysis of the guidelines for NAE:
A: Each switch model has a maximum number of supported monitors, and one agent might have multiple monitors.
* Correct:
* Each AOS-CX switch model has hardware and software limitations, including the number of agents and monitors it supports.
* Monitors are data collection points for tracking specific metrics like interface statistics, CPU usage, or custom-defined parameters.
* Agents are scripts that use monitors to evaluate data, trigger actions, or generate alerts.
* Since one agent can have multiple monitors, the total number of monitors might impact the scalability of agents.
B: You can install multiple scripts on a switch, but you can deploy only one agent per script.
* Incorrect:
* Multiple agents can be deployed from the same script if they monitor different parameters or have different configurations.
* The limitation is usually related to the total number of agents and monitors supported by the switch model, not the script itself.
C: The switch will permit you to deploy as many NAE agents as you want, but they might degrade the switch functionality.
* Incorrect:
* AOS-CX enforces hardware and software limits on the number of agents and monitors. These limits are designed to prevent degradation of switch performance.
* You cannot deploy an unlimited number of agents, as the system enforces these restrictions.
D: When you use custom scripts, you can create as many agents from each script as you want.
* Incorrect:
* While you can use custom scripts to create agents, the total number of agents is subject to the switch's maximum supported limits.
* The scalability of agents is still bound by hardware and software constraints, even with custom scripts.
References
* HPE Aruba AOS-CX Network Analytics Engine Configuration Guide.
* Aruba AOS-CX Switch Series Technical Specifications.
* Best Practices for NAE Deployment in AOS-CX Networks.