Customizing Agent Monitoring Items

The Agent's functionality and behavior can be tailored to specific monitoring needs by configuring its runtime parameters.

---

Runtime Parameters

Configuration Parameters

If you installed the Agent using the one-click script, you can edit the /opt/nezha/agent/config.yml file to add or modify parameters.
Multiple Agents can run on the same server, and each Agent's configuration file is independent.

Below is a list of runtime parameters and their functions:

• client_secret:

  • The client secret used for secure communication with the Dashboard.
  • This parameter must match the configuration in the Dashboard to enable proper communication.

• debug:

  • When set to true, enables debug mode and logs detailed information.

• disable_auto_update:

  • When set to true, disables the automatic update feature, enhancing system stability and security.

• disable_command_execute:

  • When set to true, disables the execution of online terminals, file management, and scheduled tasks for increased security.

• disable_force_update:

  • When set to true, disables forced updates, allowing only manual updates.

• disable_nat:

  • When set to true, disables the NAT traversal feature to protect internal network data.

• disable_send_query:

  • When set to true, prevents the Agent from initiating TCP Ping, ICMP Ping, and HTTP GET requests.

• gpu:

  • When set to true, enables GPU monitoring.
  • Note: Additional dependencies may be required for GPU monitoring. See the GPU Monitoring Guide for details.

• insecure_tls:

  • When set to true, disables certificate validation, suitable for scenarios using self-signed certificates.

• ip_report_period:

  • Sets the interval (in seconds) for local IP updates. The default is 1800 seconds (30 minutes).
  • If this value is less than the report_delay setting, the report_delay value will take precedence.

• report_delay:

  • Sets the interval (in seconds) for system information reporting. The default is 1 second. To reduce resource usage, a value of 3 seconds is recommended (valid range: 1-4 seconds).

• server:

  • The domain or IP address of the Dashboard for communication, including the port number.

• skip_connection_count:

  • When set to true, disables monitoring of network connection counts, suitable for servers with high connection volumes or limited resources.

• skip_procs_count:

  • When set to true, disables monitoring of process counts to lower resource usage.

• temperature:

  • When set to true, enables hardware temperature monitoring (only supported for specific hardware; unavailable for some VPS instances).

• tls:

  • When set to true, enables SSL/TLS encryption for communication between the Agent and the Dashboard.
  • Enable this option if the Agent uses an SSL/TLS-enabled Nginx reverse proxy.

• hard_drive_partition_allowlist:

  • A string array specifying the list of hard drive partitions to monitor.
  • Monitoring will only include the listed partitions.

• nic_allowlist:

  • A map[string]bool type whitelist specifying the network interfaces to monitor.
  • The key is the network interface name, and the value is true for monitoring and false for exclusion.
  • If not specified, all available interfaces are monitored by default.

• use_gitee_to_upgrade:

  • When set to true, uses the Gitee repository as the automatic update source, better suited for servers in mainland China.

• use_ipv6_country_code:

  • When set to true, forces the use of IPv6 addresses for country code queries (default uses IPv4).

• uuid:

  • The unique identifier for the current Agent, used by the Dashboard to recognize the data source.
  • To replace an existing Agent in the Dashboard, manually set this parameter.

• dns:

  • A string array to set custom DNS server lists.
  • When specified, the Agent will prioritize these DNS servers for domain resolution.

• custom_ip_api:

  • A string array specifying custom IP query API endpoints.
  • The Agent will use these APIs to retrieve the server's public IP information.

---

Apply Changes

After modifying the configuration file, restart the Agent service to apply changes. Follow these steps:

1. Restart the Service
   Use the following command to restart the default Agent service:

   sudo systemctl restart nezha-agent.service

2. For Multiple Agent Services
   If running multiple Agent services on the same server, first list all Agent services:

   sudo systemctl list-units --type=service | grep nezha-agent

   Then restart each Agent service using the following command:

   sudo systemctl restart <service-name>

   Replace <service-name> with the actual service name, such as nezha-agent@2.service.