Struct SystemMonitor 

pub struct SystemMonitor { /* private fields */ }

A utility for monitoring system resource usage of a process.

SystemMonitor can be configured to track a specific process ID or the current process. It periodically samples CPU usage, memory usage, disk I/O, and network I/O, and writes the metrics to a specified output file or to the tracing log.

Example

use std::time::Duration;

use xet_runtime::logging::SystemMonitor;
use xet_runtime::utils::TemplatedPathBuf;

let monitor = SystemMonitor::follow_process(
    Duration::from_secs(5),
    Some(TemplatedPathBuf::from("monitor_{PID}_{TIMESTAMP}.log")),
)?;

// ... application logic ...

monitor.stop()?;

Implementations

impl SystemMonitor

pub fn follow_process( sample_interval: Duration, log_path: Option<TemplatedPathBuf>, ) -> Result<Self, SystemMonitorError>

Creates a new SystemMonitor that follows the current process.

Monitoring starts immediately upon creation. The background thread begins sampling system metrics at the specified interval.

Arguments

• sample_interval - The interval at which to sample system metrics.
• log_path - Optional path template for the output log file. If None, logs to tracing at INFO level.

pub fn with_pid( pid: Pid, sample_interval: Duration, log_path: Option<TemplatedPathBuf>, ) -> Result<Self, SystemMonitorError>

Creates a new SystemMonitor that follows a specific process ID.

Monitoring starts immediately upon creation. The background thread begins sampling system metrics at the specified interval.

Arguments

• pid - The process ID to monitor.
• sample_interval - The interval at which to sample system metrics.
• log_path - Optional path template for the output log file. If None, logs to tracing at INFO level.

pub fn start(&self) -> Result<(), SystemMonitorError>

Starts the monitoring thread.

This function is called automatically by follow_process() and with_pid(), so it typically doesn’t need to be called manually. If the monitor is already running, this is a no-op.

Errors

Returns an error if:

• The log path is invalid or cannot be written to
• The monitored process no longer exists
• Internal synchronization fails

pub fn stop(&self) -> Result<(), SystemMonitorError>

Stops the monitoring thread.

Signals the background thread to stop and waits for it to join.

Errors

Returns an error if there is an issue stopping the thread, such as if the thread panicked or if there are internal synchronization issues.

Trait Implementations

impl Debug for SystemMonitor

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Drop for SystemMonitor

fn drop(&mut self)

Executes the destructor for this type.