Compare commits
25 Commits
Author | SHA1 | Date | |
---|---|---|---|
75494e2a71 | |||
8e6e291f6a | |||
bfeb43f38a | |||
2f5e2391f7 | |||
820952089a | |||
c745125f20 | |||
758fa7608f | |||
ee6b947f29 | |||
18dd1ef528 | |||
8fa7866cc2 | |||
238ad87119 | |||
d584e21fd9 | |||
ac79a2e0b7 | |||
1798c1270b | |||
97d1019b69 | |||
e30ceb75d9 | |||
4681e0c694 | |||
49f1af392d | |||
f78e48900a | |||
8c49a63a50 | |||
d994be757e | |||
d7a58e00da | |||
d2efc64487 | |||
1f23c303c1 | |||
1cc85bfa14 |
15
.github/workflows/build.yml
vendored
15
.github/workflows/build.yml
vendored
@@ -192,11 +192,11 @@ jobs:
|
||||
docker-build:
|
||||
name: Build and Push Docker Image
|
||||
needs: [detect-project, build, set-tag]
|
||||
if: |
|
||||
always() &&
|
||||
needs.detect-project.result == 'success' &&
|
||||
needs.build.result == 'success' &&
|
||||
github.event_name != 'pull_request'
|
||||
#if: |
|
||||
# always() &&
|
||||
# needs.detect-project.result == 'success' &&
|
||||
# needs.build.result == 'success' &&
|
||||
# github.event_name != 'pull_request'
|
||||
runs-on: ubuntu-latest
|
||||
environment: production
|
||||
steps:
|
||||
@@ -233,9 +233,6 @@ jobs:
|
||||
tag:
|
||||
name: Create Tag
|
||||
needs: [docker-build, build, set-tag]
|
||||
if: |
|
||||
github.event_name == 'push' &&
|
||||
needs.docker-build.result == 'success'
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- uses: actions/checkout@v4
|
||||
@@ -257,7 +254,7 @@ jobs:
|
||||
|
||||
summary:
|
||||
name: Workflow Summary
|
||||
needs: [test, audit, build, docker-build]
|
||||
needs: [test, build, docker-build]
|
||||
if: always()
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
|
@@ -21,9 +21,6 @@ anyhow = "1.0.98"
|
||||
|
||||
regex = "1.11.3"
|
||||
|
||||
# Docker .env loading
|
||||
# config = "0.13"
|
||||
|
||||
# Docker API access
|
||||
bollard = "0.19"
|
||||
futures-util = "0.3"
|
||||
|
@@ -1,8 +1,22 @@
|
||||
/// # API Module
|
||||
///
|
||||
/// This module provides all HTTP communication between WatcherAgent and the backend server.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **Registration:** Registers the agent with the backend and retrieves its server ID and IP address.
|
||||
/// - **Heartbeat:** Periodically sends heartbeat signals to indicate liveness.
|
||||
/// - **Metrics Reporting:** Sends collected hardware and network metrics to the backend.
|
||||
/// - **Command Listening:** Polls for and executes remote commands from the backend (e.g., update image, restart container).
|
||||
///
|
||||
/// ## Usage
|
||||
/// These functions are called from the main agent loop and background tasks. All network operations are asynchronous and robust to transient failures.
|
||||
use std::time::Duration;
|
||||
|
||||
use crate::docker::serverclientcomm::handle_server_message;
|
||||
use crate::hardware::HardwareInfo;
|
||||
use crate::models::{HeartbeatDto, IdResponse, MetricDto, RegistrationDto, ServerMessage, Acknowledgment};
|
||||
use crate::serverclientcomm::handle_server_message;
|
||||
use crate::models::{
|
||||
Acknowledgment, HeartbeatDto, IdResponse, MetricDto, RegistrationDto, ServerMessage,
|
||||
};
|
||||
|
||||
use anyhow::Result;
|
||||
use reqwest::{Client, StatusCode};
|
||||
@@ -11,6 +25,18 @@ use tokio::time::sleep;
|
||||
|
||||
use bollard::Docker;
|
||||
|
||||
/// Registers this agent with the backend server and retrieves its server ID and IP address.
|
||||
///
|
||||
/// This function collects local hardware information, prepares a registration payload, and sends it to the backend. It will retry registration until successful, handling network errors and server-side failures gracefully.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `base_url` - The base URL of the backend server (e.g., `https://server.example.com`).
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(i32, String), Box<dyn Error + Send + Sync>>` - Tuple of server ID and registered IP address on success.
|
||||
///
|
||||
/// # Errors
|
||||
/// Returns an error if unable to register after repeated attempts.
|
||||
pub async fn register_with_server(
|
||||
base_url: &str,
|
||||
) -> Result<(i32, String), Box<dyn Error + Send + Sync>> {
|
||||
@@ -31,18 +57,18 @@ pub async fn register_with_server(
|
||||
|
||||
// Prepare registration data
|
||||
let registration = RegistrationDto {
|
||||
id: server_id,
|
||||
server_id: server_id,
|
||||
ip_address: registered_ip.clone(),
|
||||
cpu_type: hardware.cpu.name.clone().unwrap_or_default(),
|
||||
cpu_cores: (hardware.cpu.cores).unwrap_or_default(),
|
||||
gpu_type: hardware.gpu.name.clone().unwrap_or_default(),
|
||||
ram_size: hardware.memory.total.unwrap_or_default(),
|
||||
ram_size: hardware.memory.total_size.unwrap_or_default(),
|
||||
};
|
||||
|
||||
// Try to register (will retry on failure)
|
||||
loop {
|
||||
println!("Attempting to register with server...");
|
||||
let url = format!("{}/monitoring/register-agent-by-id", base_url);
|
||||
let url = format!("{}/monitoring/hardware-info", base_url);
|
||||
match client.post(&url).json(®istration).send().await {
|
||||
Ok(resp) if resp.status().is_success() => {
|
||||
println!("✅ Successfully registered with server.");
|
||||
@@ -64,6 +90,16 @@ pub async fn register_with_server(
|
||||
}
|
||||
}
|
||||
|
||||
/// Looks up the server ID for the given IP address from the backend server.
|
||||
///
|
||||
/// This function will retry until a valid response is received, handling network errors and server-side failures.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `base_url` - The base URL of the backend server.
|
||||
/// * `ip` - The local IP address to look up.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(i32, String), Box<dyn Error + Send + Sync>>` - Tuple of server ID and registered IP address.
|
||||
async fn get_server_id_by_ip(
|
||||
base_url: &str,
|
||||
ip: &str,
|
||||
@@ -72,7 +108,7 @@ async fn get_server_id_by_ip(
|
||||
.danger_accept_invalid_certs(true)
|
||||
.build()?;
|
||||
|
||||
let url = format!("{}/monitoring/server-id-by-ip?ipAddress={}", base_url, ip);
|
||||
let url = format!("{}/monitoring/register?ipAddress={}", base_url, ip);
|
||||
|
||||
loop {
|
||||
println!("Attempting to fetch server ID for IP {}...", ip);
|
||||
@@ -115,6 +151,16 @@ async fn get_server_id_by_ip(
|
||||
}
|
||||
}
|
||||
|
||||
/// Periodically sends heartbeat signals to the backend server to indicate agent liveness.
|
||||
///
|
||||
/// This function runs in a background task and will retry on network errors.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `base_url` - The base URL of the backend server.
|
||||
/// * `ip` - The IP address of the agent.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(), Box<dyn Error + Send + Sync>>` - Ok if heartbeats are sent successfully.
|
||||
pub async fn heartbeat_loop(base_url: &str, ip: &str) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
let client = Client::builder()
|
||||
.danger_accept_invalid_certs(true)
|
||||
@@ -138,6 +184,16 @@ pub async fn heartbeat_loop(base_url: &str, ip: &str) -> Result<(), Box<dyn Erro
|
||||
}
|
||||
}
|
||||
|
||||
/// Sends collected hardware and network metrics to the backend server.
|
||||
///
|
||||
/// This function is called periodically from the metrics collection loop. It logs the result and retries on network errors.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `base_url` - The base URL of the backend server.
|
||||
/// * `metrics` - The metrics data to send (see [`MetricDto`]).
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(), Box<dyn Error + Send + Sync>>` - Ok if metrics are sent successfully.
|
||||
pub async fn send_metrics(
|
||||
base_url: &str,
|
||||
metrics: &MetricDto,
|
||||
@@ -158,10 +214,23 @@ pub async fn send_metrics(
|
||||
Ok(())
|
||||
}
|
||||
|
||||
pub async fn listening_to_server(docker: &Docker, base_url: &str) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
/// Polls the backend server for remote commands and executes them.
|
||||
///
|
||||
/// This function runs in a background task, polling the server for new messages. It acknowledges receipt and execution of each command, and handles errors gracefully.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `docker` - Reference to a Bollard Docker client.
|
||||
/// * `base_url` - The base URL of the backend server.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(), Box<dyn Error + Send + Sync>>` - Ok if commands are handled successfully.
|
||||
pub async fn listening_to_server(
|
||||
docker: &Docker,
|
||||
base_url: &str,
|
||||
) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
let url = format!("{}/api/message", base_url);
|
||||
let client = reqwest::Client::new();
|
||||
|
||||
|
||||
loop {
|
||||
// Get message from server
|
||||
let resp = client.get(&url).send().await;
|
||||
@@ -172,20 +241,36 @@ pub async fn listening_to_server(docker: &Docker, base_url: &str) -> Result<(),
|
||||
match response.json::<ServerMessage>().await {
|
||||
Ok(msg) => {
|
||||
// Acknowledge receipt immediately
|
||||
if let Err(e) = send_acknowledgment(&client, base_url, &msg.message_id, "received", "Message received successfully").await {
|
||||
if let Err(e) = send_acknowledgment(
|
||||
&client,
|
||||
base_url,
|
||||
&msg.message_id,
|
||||
"received",
|
||||
"Message received successfully",
|
||||
)
|
||||
.await
|
||||
{
|
||||
eprintln!("Failed to send receipt acknowledgment: {}", e);
|
||||
}
|
||||
|
||||
|
||||
// Handle the message
|
||||
let result = handle_server_message(docker, msg.clone()).await;
|
||||
|
||||
|
||||
// Send execution result acknowledgment
|
||||
let (status, details) = match result {
|
||||
Ok(_) => ("success", "Message executed successfully".to_string()),
|
||||
Err(e) => ("error", format!("Execution failed: {}", e)),
|
||||
};
|
||||
|
||||
if let Err(e) = send_acknowledgment(&client, base_url, &msg.message_id, status, &details).await {
|
||||
|
||||
if let Err(e) = send_acknowledgment(
|
||||
&client,
|
||||
base_url,
|
||||
&msg.message_id,
|
||||
status,
|
||||
&details,
|
||||
)
|
||||
.await
|
||||
{
|
||||
eprintln!("Failed to send execution acknowledgment: {}", e);
|
||||
}
|
||||
}
|
||||
@@ -210,6 +295,19 @@ pub async fn listening_to_server(docker: &Docker, base_url: &str) -> Result<(),
|
||||
}
|
||||
}
|
||||
|
||||
/// Sends an acknowledgment to the backend server for a received or executed command message.
|
||||
///
|
||||
/// This function is used internally by [`listening_to_server`] to confirm receipt and execution status of commands.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `client` - Reference to a reqwest HTTP client.
|
||||
/// * `base_url` - The base URL of the backend server.
|
||||
/// * `message_id` - The ID of the message being acknowledged.
|
||||
/// * `status` - Status string (e.g., "received", "success", "error").
|
||||
/// * `details` - Additional details about the acknowledgment.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(), Box<dyn Error + Send + Sync>>` - Ok if acknowledgment is sent successfully.
|
||||
async fn send_acknowledgment(
|
||||
client: &reqwest::Client,
|
||||
base_url: &str,
|
||||
@@ -218,24 +316,26 @@ async fn send_acknowledgment(
|
||||
details: &str,
|
||||
) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
let ack_url = format!("{}/api/acknowledge", base_url);
|
||||
|
||||
|
||||
let acknowledgment = Acknowledgment {
|
||||
message_id: message_id.to_string(),
|
||||
status: status.to_string(),
|
||||
details: details.to_string(),
|
||||
};
|
||||
|
||||
let response = client
|
||||
.post(&ack_url)
|
||||
.json(&acknowledgment)
|
||||
.send()
|
||||
.await?;
|
||||
|
||||
|
||||
let response = client.post(&ack_url).json(&acknowledgment).send().await?;
|
||||
|
||||
if response.status().is_success() {
|
||||
println!("Acknowledgment sent successfully for message {}", message_id);
|
||||
println!(
|
||||
"Acknowledgment sent successfully for message {}",
|
||||
message_id
|
||||
);
|
||||
} else {
|
||||
eprintln!("Server returned error for acknowledgment: {}", response.status());
|
||||
eprintln!(
|
||||
"Server returned error for acknowledgment: {}",
|
||||
response.status()
|
||||
);
|
||||
}
|
||||
|
||||
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
|
231
WatcherAgent/src/docker/container.rs
Normal file
231
WatcherAgent/src/docker/container.rs
Normal file
@@ -0,0 +1,231 @@
|
||||
//! Docker container utilities for WatcherAgent
|
||||
//!
|
||||
//! Provides functions to list and process Docker containers using the Bollard library.
|
||||
//!
|
||||
use crate::docker::stats;
|
||||
use crate::docker::stats::{ContainerCpuInfo, ContainerNetworkInfo};
|
||||
use crate::models::{DockerContainerDto, DockerContainerRegistrationDto};
|
||||
|
||||
use bollard::query_parameters::{
|
||||
CreateImageOptions, ListContainersOptions, RestartContainerOptions,
|
||||
};
|
||||
use bollard::Docker;
|
||||
use futures_util::StreamExt;
|
||||
use std::error::Error;
|
||||
|
||||
/// Returns a list of available Docker containers.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `docker` - Reference to a Bollard Docker client.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Vec<DockerContainer>` - Vector of Docker container info.
|
||||
pub async fn get_available_containers(docker: &Docker) -> Vec<DockerContainerDto> {
|
||||
println!("=== DOCKER CONTAINER LIST ===");
|
||||
|
||||
let options = Some(ListContainersOptions {
|
||||
all: true,
|
||||
..Default::default()
|
||||
});
|
||||
|
||||
let containers_list = match docker.list_containers(options).await {
|
||||
Ok(containers) => {
|
||||
println!("Available containers ({}):", containers.len());
|
||||
containers
|
||||
.into_iter()
|
||||
.filter_map(|container| {
|
||||
container.id.as_ref()?; // Skip if no ID
|
||||
|
||||
let id = container.id?;
|
||||
let short_id = if id.len() > 12 { &id[..12] } else { &id };
|
||||
|
||||
let name = container
|
||||
.names
|
||||
.and_then(|names| names.into_iter().next())
|
||||
.map(|name| name.trim_start_matches('/').to_string())
|
||||
.unwrap_or_else(|| "unknown".to_string());
|
||||
|
||||
let image = container
|
||||
.image
|
||||
.as_ref()
|
||||
.map(|img| img.to_string())
|
||||
.unwrap_or_else(|| "unknown".to_string());
|
||||
|
||||
/*let status = container
|
||||
.status
|
||||
.as_ref()
|
||||
.map(|s| match s.to_lowercase().as_str() {
|
||||
s if s.contains("up") || s.contains("running") => "running".to_string(),
|
||||
s if s.contains("exited") || s.contains("stopped") => {
|
||||
"stopped".to_string()
|
||||
}
|
||||
_ => s.to_string(),
|
||||
})
|
||||
.unwrap_or_else(|| "unknown".to_string());
|
||||
|
||||
println!(
|
||||
" - ID: {}, Image: {}, Name: {}",
|
||||
short_id,
|
||||
container.image.unwrap(),
|
||||
name
|
||||
);*/
|
||||
|
||||
Some(DockerContainerDto {
|
||||
id: short_id.to_string(),
|
||||
image,
|
||||
name: name,
|
||||
})
|
||||
})
|
||||
.collect()
|
||||
}
|
||||
Err(e) => {
|
||||
eprintln!("Failed to list containers: {}", e);
|
||||
Vec::new()
|
||||
}
|
||||
};
|
||||
|
||||
containers_list
|
||||
}
|
||||
|
||||
/// Pulls a new Docker image and restarts the current container.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `docker` - Reference to a Bollard Docker client.
|
||||
/// * `image` - The name of the Docker image to pull.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(), Box<dyn Error + Send + Sync>>` - Ok if updated successfully, error otherwise.
|
||||
pub async fn update_docker_image(
|
||||
docker: &Docker,
|
||||
image: &str,
|
||||
) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
println!("Updating to {}", image);
|
||||
|
||||
// 1. Pull new image
|
||||
let mut stream = docker.create_image(
|
||||
Some(CreateImageOptions {
|
||||
from_image: Some(image.to_string()),
|
||||
..Default::default()
|
||||
}),
|
||||
None,
|
||||
None,
|
||||
);
|
||||
|
||||
// Use the stream with proper trait bounds
|
||||
while let Some(result) = StreamExt::next(&mut stream).await {
|
||||
match result {
|
||||
Ok(progress) => {
|
||||
if let Some(status) = progress.status {
|
||||
println!("Pull status: {}", status);
|
||||
}
|
||||
}
|
||||
Err(e) => {
|
||||
eprintln!("Error pulling image: {}", e);
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// 2. Restart the current container
|
||||
let options = Some(ListContainersOptions {
|
||||
all: true,
|
||||
..Default::default()
|
||||
});
|
||||
let container_id = docker
|
||||
.list_containers(options)
|
||||
.await?
|
||||
.into_iter()
|
||||
.find_map(|c| {
|
||||
c.image
|
||||
.as_ref()
|
||||
.and_then(|img| if img == image { c.id } else { None })
|
||||
});
|
||||
let _ = restart_container(docker, &container_id.unwrap()).await;
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
/// Restarts the agent's own Docker container.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `docker` - Reference to a Bollard Docker client.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(), Box<dyn Error + Send + Sync>>` - Ok if restarted successfully, error otherwise.
|
||||
pub async fn restart_container(
|
||||
docker: &Docker,
|
||||
container_id: &str,
|
||||
) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
println!("Restarting container {}", container_id);
|
||||
if let Err(e) = docker
|
||||
.restart_container(
|
||||
&container_id.to_string(),
|
||||
Some(RestartContainerOptions {
|
||||
signal: None,
|
||||
t: Some(0),
|
||||
}),
|
||||
)
|
||||
.await
|
||||
{
|
||||
eprintln!("Failed to restart container: {}", e);
|
||||
}
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
/*
|
||||
/// Extracts a Docker container ID from a string line.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `line` - The input string containing a container ID or related info.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Option<String>` - The extracted container ID if found.
|
||||
pub fn extract_client_container_id(line: &str) -> Option<String> {
|
||||
// ...existing code...
|
||||
}
|
||||
*/
|
||||
|
||||
/// Gets network statistics for a specific container
|
||||
pub async fn get_network_stats(
|
||||
docker: &Docker,
|
||||
container_id: &str,
|
||||
) -> Result<ContainerNetworkInfo, Box<dyn Error + Send + Sync>> {
|
||||
let (_, net_info) = stats::get_single_container_stats(docker, container_id).await?;
|
||||
|
||||
if let Some(net_info) = net_info {
|
||||
Ok(net_info)
|
||||
} else {
|
||||
// Return default network info if not found
|
||||
Ok(ContainerNetworkInfo {
|
||||
container_id: container_id.to_string(),
|
||||
rx_bytes: 0,
|
||||
tx_bytes: 0,
|
||||
rx_packets: 0,
|
||||
tx_packets: 0,
|
||||
rx_errors: 0,
|
||||
tx_errors: 0,
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
/// Gets CPU statistics for a specific container
|
||||
pub async fn get_cpu_stats(
|
||||
docker: &Docker,
|
||||
container_id: &str,
|
||||
) -> Result<ContainerCpuInfo, Box<dyn Error + Send + Sync>> {
|
||||
let (cpu_info, _) = stats::get_single_container_stats(docker, container_id).await?;
|
||||
|
||||
if let Some(cpu_info) = cpu_info {
|
||||
Ok(cpu_info)
|
||||
} else {
|
||||
// Return default CPU info if not found
|
||||
Ok(ContainerCpuInfo {
|
||||
container_id: container_id.to_string(),
|
||||
cpu_usage_percent: 0.0,
|
||||
system_cpu_usage: 0,
|
||||
container_cpu_usage: 0,
|
||||
online_cpus: 1,
|
||||
})
|
||||
}
|
||||
}
|
203
WatcherAgent/src/docker/mod.rs
Normal file
203
WatcherAgent/src/docker/mod.rs
Normal file
@@ -0,0 +1,203 @@
|
||||
/// # Docker Module
|
||||
///
|
||||
/// This module provides Docker integration for WatcherAgent, including container enumeration, statistics, and lifecycle management.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **Container Management:** Lists, inspects, and manages Docker containers relevant to the agent.
|
||||
/// - **Statistics Aggregation:** Collects network and CPU statistics for all managed containers.
|
||||
/// - **Lifecycle Operations:** Supports container restart and ID lookup for agent self-management.
|
||||
///
|
||||
pub mod container;
|
||||
pub mod serverclientcomm;
|
||||
pub mod stats;
|
||||
|
||||
use crate::models::{DockerContainerDto, DockerContainerMetricDto};
|
||||
use bollard::{query_parameters::InspectContainerOptions, Docker};
|
||||
use std::error::Error;
|
||||
|
||||
/// Main Docker manager that holds the Docker client and provides all operations
|
||||
#[derive(Debug, Clone)]
|
||||
pub struct DockerManager {
|
||||
pub docker: Docker,
|
||||
}
|
||||
|
||||
impl Default for DockerManager {
|
||||
fn default() -> Self {
|
||||
Self {
|
||||
docker: Docker::connect_with_local_defaults()
|
||||
.unwrap_or_else(|e| panic!("Failed to create default Docker connection: {}", e)),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
impl DockerManager {
|
||||
/// Creates a new DockerManager instance
|
||||
pub fn new() -> Result<Self, Box<dyn Error + Send + Sync>> {
|
||||
let docker = Docker::connect_with_local_defaults()
|
||||
.map_err(|e| format!("Failed to connect to Docker: {}", e))?;
|
||||
|
||||
Ok(Self { docker })
|
||||
}
|
||||
|
||||
/// Creates a DockerManager instance with optional Docker connection
|
||||
pub fn new_optional() -> Option<Self> {
|
||||
Docker::connect_with_local_defaults()
|
||||
.map(|docker| Self { docker })
|
||||
.ok()
|
||||
}
|
||||
|
||||
/// Finds the Docker container running the agent by image name
|
||||
pub async fn get_client_container(
|
||||
&self,
|
||||
) -> Result<Option<DockerContainerDto>, Box<dyn Error + Send + Sync>> {
|
||||
let containers = container::get_available_containers(&self.docker).await;
|
||||
let client_image = "watcher-agent";
|
||||
|
||||
Ok(containers
|
||||
.into_iter()
|
||||
.find(|c| c.image == client_image)
|
||||
.map(|container| DockerContainerDto {
|
||||
id: container.id,
|
||||
image: container.image,
|
||||
name: container.name,
|
||||
}))
|
||||
}
|
||||
|
||||
/// Gets the current client version (image name) if running in Docker
|
||||
pub async fn get_client_version(&self) -> String {
|
||||
match self.get_client_container().await {
|
||||
Ok(Some(container)) => container.image,
|
||||
Ok(None) => {
|
||||
eprintln!("Warning: No WatcherAgent container found");
|
||||
"unknown".to_string()
|
||||
}
|
||||
Err(e) => {
|
||||
eprintln!("Warning: Could not get current image version: {}", e);
|
||||
"unknown".to_string()
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// Checks if Docker is available and the agent is running in a container
|
||||
pub async fn is_dockerized(&self) -> bool {
|
||||
self.get_client_container()
|
||||
.await
|
||||
.map(|c| c.is_some())
|
||||
.unwrap_or(false)
|
||||
}
|
||||
|
||||
/// Gets all available containers as DTOs for registration
|
||||
pub async fn get_containers_for_registration(
|
||||
&self,
|
||||
) -> Result<Vec<DockerContainerDto>, Box<dyn Error + Send + Sync>> {
|
||||
let containers = container::get_available_containers(&self.docker).await;
|
||||
|
||||
Ok(containers
|
||||
.into_iter()
|
||||
.map(|container| DockerContainerDto {
|
||||
id: container.id,
|
||||
image: container.image,
|
||||
name: container.name,
|
||||
})
|
||||
.collect())
|
||||
}
|
||||
|
||||
/// Gets container metrics for all containers
|
||||
pub async fn get_container_metrics(
|
||||
&self,
|
||||
) -> Result<Vec<DockerContainerMetricDto>, Box<dyn Error + Send + Sync>> {
|
||||
let containers = container::get_available_containers(&self.docker).await;
|
||||
let mut metrics = Vec::new();
|
||||
|
||||
for container in containers {
|
||||
// Get network stats (you'll need to implement this in container.rs)
|
||||
let network_stats = container::get_network_stats(&self.docker, &container.id).await?;
|
||||
// Get CPU stats (you'll need to implement this in container.rs)
|
||||
let cpu_stats = container::get_cpu_stats(&self.docker, &container.id).await?;
|
||||
|
||||
// Get current status by inspecting the container
|
||||
let status = match self
|
||||
.docker
|
||||
.inspect_container(&container.id, None::<InspectContainerOptions>)
|
||||
.await
|
||||
{
|
||||
Ok(container_info) => {
|
||||
// Extract status from container state and convert to string
|
||||
container_info
|
||||
.state
|
||||
.and_then(|state| state.status)
|
||||
.map(|status_enum| {
|
||||
match status_enum {
|
||||
bollard::models::ContainerStateStatusEnum::CREATED => "created",
|
||||
bollard::models::ContainerStateStatusEnum::RUNNING => "running",
|
||||
bollard::models::ContainerStateStatusEnum::PAUSED => "paused",
|
||||
bollard::models::ContainerStateStatusEnum::RESTARTING => {
|
||||
"restarting"
|
||||
}
|
||||
bollard::models::ContainerStateStatusEnum::REMOVING => "removing",
|
||||
bollard::models::ContainerStateStatusEnum::EXITED => "exited",
|
||||
bollard::models::ContainerStateStatusEnum::DEAD => "dead",
|
||||
bollard::secret::ContainerStateStatusEnum::EMPTY => todo!(),
|
||||
}
|
||||
.to_string()
|
||||
})
|
||||
.unwrap_or_else(|| "unknown".to_string())
|
||||
}
|
||||
Err(_) => "unknown".to_string(),
|
||||
};
|
||||
|
||||
metrics.push(DockerContainerMetricDto {
|
||||
id: container.id,
|
||||
status: status,
|
||||
network: network_stats,
|
||||
cpu: cpu_stats,
|
||||
});
|
||||
}
|
||||
|
||||
Ok(metrics)
|
||||
}
|
||||
|
||||
/// Gets the number of running containers
|
||||
pub async fn get_container_count(&self) -> Result<usize, Box<dyn Error + Send + Sync>> {
|
||||
let containers = container::get_available_containers(&self.docker).await;
|
||||
Ok(containers.len())
|
||||
}
|
||||
|
||||
/// Restarts a specific container by ID
|
||||
pub async fn restart_container(
|
||||
&self,
|
||||
container_id: &str,
|
||||
) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
container::restart_container(&self.docker, container_id).await
|
||||
}
|
||||
|
||||
/// Gets total network statistics across all containers
|
||||
pub async fn get_total_network_stats(
|
||||
&self,
|
||||
) -> Result<(u64, u64), Box<dyn Error + Send + Sync>> {
|
||||
let metrics = self.get_container_metrics().await?;
|
||||
|
||||
let net_in_total: u64 = metrics.iter().map(|m| m.network.rx_bytes).sum();
|
||||
let net_out_total: u64 = metrics.iter().map(|m| m.network.tx_bytes).sum();
|
||||
|
||||
Ok((net_in_total, net_out_total))
|
||||
}
|
||||
}
|
||||
|
||||
// Keep these as utility functions if needed, but they should use DockerManager internally
|
||||
impl DockerContainerDto {
|
||||
/// Returns the container ID
|
||||
pub fn id(&self) -> &str {
|
||||
&self.id
|
||||
}
|
||||
|
||||
/// Returns the image name
|
||||
pub fn image(&self) -> &str {
|
||||
&self.image
|
||||
}
|
||||
|
||||
/// Returns the container name
|
||||
pub fn name(&self) -> &str {
|
||||
&self.name
|
||||
}
|
||||
}
|
59
WatcherAgent/src/docker/serverclientcomm.rs
Normal file
59
WatcherAgent/src/docker/serverclientcomm.rs
Normal file
@@ -0,0 +1,59 @@
|
||||
//! Server-client communication utilities for WatcherAgent
|
||||
//!
|
||||
//! Handles server commands, Docker image updates, and container management using the Bollard library.
|
||||
//!
|
||||
use crate::models::ServerMessage;
|
||||
|
||||
use super::container::{restart_container, update_docker_image};
|
||||
use bollard::query_parameters::{CreateImageOptions, RestartContainerOptions};
|
||||
use bollard::Docker;
|
||||
use std::error::Error;
|
||||
|
||||
/// Handles a message from the backend server and dispatches the appropriate action.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `docker` - Reference to a Bollard Docker client.
|
||||
/// * `msg` - The server message to handle.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(), Box<dyn Error + Send + Sync>>` - Ok if handled successfully, error otherwise.
|
||||
pub async fn handle_server_message(
|
||||
docker: &Docker,
|
||||
msg: ServerMessage,
|
||||
) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
let msg = msg.clone();
|
||||
println!("Handling server message: {:?}", msg);
|
||||
|
||||
// Handle different message types
|
||||
match msg.message_type.as_str() {
|
||||
"update_image" => {
|
||||
if let Some(image_name) = msg.data.get("image").and_then(|v| v.as_str()) {
|
||||
println!("Received update command for image: {}", image_name);
|
||||
// Call your update_docker_image function here
|
||||
update_docker_image(docker, image_name).await?;
|
||||
Ok(())
|
||||
} else {
|
||||
Err("Missing image name in update message".into())
|
||||
}
|
||||
}
|
||||
"restart_container" => {
|
||||
if let Some(image_name) = msg.data.get("image").and_then(|v| v.as_str()) {
|
||||
println!("Received restart command for image: {}", image_name);
|
||||
// Call your update_docker_image function here
|
||||
update_docker_image(docker, image_name).await?;
|
||||
Ok(())
|
||||
} else {
|
||||
Err("Missing image name in update message".into())
|
||||
}
|
||||
}
|
||||
"stop_agent" => {
|
||||
println!("Received stop agent command");
|
||||
// Implement graceful shutdown
|
||||
std::process::exit(0);
|
||||
}
|
||||
_ => {
|
||||
eprintln!("Unknown message type: {}", msg.message_type);
|
||||
Err(format!("Unknown message type: {}", msg.message_type).into())
|
||||
}
|
||||
}
|
||||
}
|
206
WatcherAgent/src/docker/stats.rs
Normal file
206
WatcherAgent/src/docker/stats.rs
Normal file
@@ -0,0 +1,206 @@
|
||||
use bollard::query_parameters::{ListContainersOptions, StatsOptions};
|
||||
use bollard::Docker;
|
||||
use futures_util::stream::TryStreamExt;
|
||||
use serde::{Deserialize, Serialize};
|
||||
use std::error::Error;
|
||||
|
||||
#[derive(Debug, Serialize, Deserialize, Clone)]
|
||||
pub struct ContainerCpuInfo {
|
||||
pub container_id: String,
|
||||
pub cpu_usage_percent: f64,
|
||||
pub system_cpu_usage: u64,
|
||||
pub container_cpu_usage: u64,
|
||||
pub online_cpus: u32,
|
||||
}
|
||||
|
||||
#[derive(Debug, Serialize, Deserialize, Clone)]
|
||||
pub struct ContainerNetworkInfo {
|
||||
pub container_id: String,
|
||||
pub rx_bytes: u64,
|
||||
pub tx_bytes: u64,
|
||||
pub rx_packets: u64,
|
||||
pub tx_packets: u64,
|
||||
pub rx_errors: u64,
|
||||
pub tx_errors: u64,
|
||||
}
|
||||
|
||||
/// Get container statistics for all containers using an existing Docker client
|
||||
pub async fn get_container_stats(
|
||||
docker: &Docker,
|
||||
) -> Result<(Vec<ContainerCpuInfo>, Vec<ContainerNetworkInfo>), Box<dyn Error + Send + Sync>> {
|
||||
let containers = docker
|
||||
.list_containers(Some(ListContainersOptions {
|
||||
all: true,
|
||||
..Default::default()
|
||||
}))
|
||||
.await?;
|
||||
|
||||
let mut cpu_infos = Vec::new();
|
||||
let mut net_infos = Vec::new();
|
||||
|
||||
for container in containers {
|
||||
let id = container.id.unwrap_or_default();
|
||||
|
||||
// Skip if no ID
|
||||
if id.is_empty() {
|
||||
continue;
|
||||
}
|
||||
|
||||
let mut stats_stream = docker.stats(
|
||||
&id,
|
||||
Some(StatsOptions {
|
||||
stream: false,
|
||||
one_shot: true,
|
||||
}),
|
||||
);
|
||||
|
||||
if let Some(stats) = stats_stream.try_next().await? {
|
||||
// CPU Info
|
||||
if let (Some(cpu_stats), Some(precpu_stats)) = (&stats.cpu_stats, &stats.precpu_stats) {
|
||||
if let (Some(cpu_usage), Some(pre_cpu_usage)) =
|
||||
(&cpu_stats.cpu_usage, &precpu_stats.cpu_usage)
|
||||
{
|
||||
let cpu_delta = cpu_usage
|
||||
.total_usage
|
||||
.unwrap_or(0)
|
||||
.saturating_sub(pre_cpu_usage.total_usage.unwrap_or(0));
|
||||
|
||||
let system_delta = cpu_stats
|
||||
.system_cpu_usage
|
||||
.unwrap_or(0)
|
||||
.saturating_sub(precpu_stats.system_cpu_usage.unwrap_or(0));
|
||||
|
||||
let online_cpus = cpu_stats.online_cpus.unwrap_or(1);
|
||||
|
||||
let cpu_percent = if system_delta > 0 && online_cpus > 0 {
|
||||
(cpu_delta as f64 / system_delta as f64) * online_cpus as f64 * 100.0
|
||||
} else {
|
||||
0.0
|
||||
};
|
||||
|
||||
cpu_infos.push(ContainerCpuInfo {
|
||||
container_id: id.clone(),
|
||||
cpu_usage_percent: cpu_percent,
|
||||
system_cpu_usage: cpu_stats.system_cpu_usage.unwrap_or(0),
|
||||
container_cpu_usage: cpu_usage.total_usage.unwrap_or(0),
|
||||
online_cpus,
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
// Network Info
|
||||
if let Some(networks) = stats.networks {
|
||||
for (_name, net) in networks {
|
||||
net_infos.push(ContainerNetworkInfo {
|
||||
container_id: id.clone(),
|
||||
rx_bytes: net.rx_bytes.unwrap(),
|
||||
tx_bytes: net.tx_bytes.unwrap(),
|
||||
rx_packets: net.rx_packets.unwrap(),
|
||||
tx_packets: net.tx_packets.unwrap(),
|
||||
rx_errors: net.rx_errors.unwrap(),
|
||||
tx_errors: net.tx_errors.unwrap(),
|
||||
});
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Ok((cpu_infos, net_infos))
|
||||
}
|
||||
|
||||
/// Get container statistics for a specific container
|
||||
pub async fn get_single_container_stats(
|
||||
docker: &Docker,
|
||||
container_id: &str,
|
||||
) -> Result<(Option<ContainerCpuInfo>, Option<ContainerNetworkInfo>), Box<dyn Error + Send + Sync>>
|
||||
{
|
||||
let mut stats_stream = docker.stats(
|
||||
container_id,
|
||||
Some(StatsOptions {
|
||||
stream: false,
|
||||
one_shot: true,
|
||||
}),
|
||||
);
|
||||
|
||||
if let Some(stats) = stats_stream.try_next().await? {
|
||||
let mut cpu_info = None;
|
||||
let mut net_info = None;
|
||||
|
||||
// CPU Info
|
||||
if let (Some(cpu_stats), Some(precpu_stats)) = (&stats.cpu_stats, &stats.precpu_stats) {
|
||||
if let (Some(cpu_usage), Some(pre_cpu_usage)) =
|
||||
(&cpu_stats.cpu_usage, &precpu_stats.cpu_usage)
|
||||
{
|
||||
let cpu_delta = cpu_usage
|
||||
.total_usage
|
||||
.unwrap_or(0)
|
||||
.saturating_sub(pre_cpu_usage.total_usage.unwrap_or(0));
|
||||
|
||||
let system_delta = cpu_stats
|
||||
.system_cpu_usage
|
||||
.unwrap_or(0)
|
||||
.saturating_sub(precpu_stats.system_cpu_usage.unwrap_or(0));
|
||||
|
||||
let online_cpus = cpu_stats.online_cpus.unwrap_or(1);
|
||||
|
||||
let cpu_percent = if system_delta > 0 && online_cpus > 0 {
|
||||
(cpu_delta as f64 / system_delta as f64) * online_cpus as f64 * 100.0
|
||||
} else {
|
||||
0.0
|
||||
};
|
||||
|
||||
cpu_info = Some(ContainerCpuInfo {
|
||||
container_id: container_id.to_string(),
|
||||
cpu_usage_percent: cpu_percent,
|
||||
system_cpu_usage: cpu_stats.system_cpu_usage.unwrap_or(0),
|
||||
container_cpu_usage: cpu_usage.total_usage.unwrap_or(0),
|
||||
online_cpus,
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
// Network Info
|
||||
if let Some(networks) = stats.networks {
|
||||
// Take the first network interface (usually eth0)
|
||||
if let Some((_name, net)) = networks.into_iter().next() {
|
||||
net_info = Some(ContainerNetworkInfo {
|
||||
container_id: container_id.to_string(),
|
||||
rx_bytes: net.rx_bytes.unwrap(),
|
||||
tx_bytes: net.tx_bytes.unwrap(),
|
||||
rx_packets: net.rx_packets.unwrap(),
|
||||
tx_packets: net.tx_packets.unwrap(),
|
||||
rx_errors: net.rx_errors.unwrap(),
|
||||
tx_errors: net.tx_errors.unwrap(),
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
Ok((cpu_info, net_info))
|
||||
} else {
|
||||
Ok((None, None))
|
||||
}
|
||||
}
|
||||
|
||||
/// Get total network statistics across all containers
|
||||
pub async fn get_total_network_stats(
|
||||
docker: &Docker,
|
||||
) -> Result<(u64, u64), Box<dyn Error + Send + Sync>> {
|
||||
let (_, net_infos) = get_container_stats(docker).await?;
|
||||
|
||||
let total_rx: u64 = net_infos.iter().map(|net| net.rx_bytes).sum();
|
||||
let total_tx: u64 = net_infos.iter().map(|net| net.tx_bytes).sum();
|
||||
|
||||
Ok((total_rx, total_tx))
|
||||
}
|
||||
|
||||
/// Get average CPU usage across all containers
|
||||
pub async fn get_average_cpu_usage(docker: &Docker) -> Result<f64, Box<dyn Error + Send + Sync>> {
|
||||
let (cpu_infos, _) = get_container_stats(docker).await?;
|
||||
|
||||
if cpu_infos.is_empty() {
|
||||
return Ok(0.0);
|
||||
}
|
||||
|
||||
let total_cpu: f64 = cpu_infos.iter().map(|cpu| cpu.cpu_usage_percent).sum();
|
||||
Ok(total_cpu / cpu_infos.len() as f64)
|
||||
}
|
@@ -2,6 +2,29 @@ use anyhow::Result;
|
||||
use std::error::Error;
|
||||
use sysinfo::System;
|
||||
|
||||
/// # CPU Hardware Module
|
||||
///
|
||||
/// This module provides CPU information collection for WatcherAgent, including load, temperature, and system uptime.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **CPU Detection:** Identifies CPU model and core count.
|
||||
/// - **Metric Collection:** Queries CPU load, temperature, and uptime.
|
||||
/// - **Error Handling:** Graceful fallback if metrics are unavailable.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `current_load`: CPU usage as a percentage (**0.0–100.0**)
|
||||
/// - `current_temp`: CPU temperature in **degrees Celsius (°C)**
|
||||
/// - `uptime`: System uptime in **seconds (s)**
|
||||
///
|
||||
/// CPU statistics for the host system.
|
||||
///
|
||||
/// # Fields
|
||||
/// - `name`: CPU model name (string)
|
||||
/// - `cores`: Number of physical CPU cores (integer)
|
||||
/// - `current_load`: CPU usage as a percentage (**0.0–100.0**)
|
||||
/// - `current_temp`: CPU temperature in **degrees Celsius (°C)**
|
||||
/// - `uptime`: System uptime in **seconds (s)**
|
||||
/// - `host_name`: Hostname of the system (string)
|
||||
#[derive(Debug)]
|
||||
pub struct CpuInfo {
|
||||
pub name: Option<String>,
|
||||
@@ -12,6 +35,10 @@ pub struct CpuInfo {
|
||||
pub host_name: Option<String>,
|
||||
}
|
||||
|
||||
/// Collects CPU information (model, cores, load, temperature, uptime).
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<CpuInfo, Box<dyn Error + Send + Sync>>` - CPU statistics or error if unavailable.
|
||||
pub async fn get_cpu_info() -> Result<CpuInfo, Box<dyn Error + Send + Sync>> {
|
||||
let mut sys = System::new_all();
|
||||
|
||||
@@ -33,12 +60,23 @@ pub async fn get_cpu_info() -> Result<CpuInfo, Box<dyn Error + Send + Sync>> {
|
||||
})
|
||||
}
|
||||
|
||||
/// Queries system for current CPU load (percentage).
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `sys` - Mutable reference to sysinfo::System
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<f64, Box<dyn Error + Send + Sync>>` - CPU load as percentage.
|
||||
pub async fn get_cpu_load(sys: &mut System) -> Result<f64, Box<dyn Error + Send + Sync>> {
|
||||
sys.refresh_cpu_all();
|
||||
tokio::task::yield_now().await; // Allow other tasks to run
|
||||
Ok(sys.global_cpu_usage() as f64)
|
||||
}
|
||||
|
||||
/// Attempts to read CPU temperature from system sensors (Linux only).
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<f64, Box<dyn Error + Send + Sync>>` - CPU temperature in degrees Celsius (°C).
|
||||
pub async fn get_cpu_temp() -> Result<f64, Box<dyn Error + Send + Sync>> {
|
||||
println!("Attempting to get CPU temperature...");
|
||||
|
||||
|
@@ -7,6 +7,27 @@ use sysinfo::{Component, Components, Disk, Disks};
|
||||
use serde::Serialize;
|
||||
|
||||
|
||||
/// # Disk Hardware Module
|
||||
///
|
||||
/// This module provides disk information collection for WatcherAgent, including total and per-disk statistics and temperature data.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **Disk Enumeration:** Lists all physical disks and their properties.
|
||||
/// - **Usage Calculation:** Computes total and per-disk usage, available space, and usage percentage.
|
||||
/// - **Temperature Monitoring:** Associates disk components with temperature sensors if available.
|
||||
///
|
||||
/// ## Units
|
||||
/// - All sizes are in **bytes** unless otherwise noted.
|
||||
/// - Temperatures are in **degrees Celsius (°C)**.
|
||||
///
|
||||
/// Summary of disk statistics for the system.
|
||||
///
|
||||
/// # Fields
|
||||
/// - `total_size`: Total disk size in bytes (all disks > 100MB)
|
||||
/// - `total_used`: Total used disk space in bytes
|
||||
/// - `total_available`: Total available disk space in bytes
|
||||
/// - `total_usage`: Usage percentage (0.0–100.0)
|
||||
/// - `detailed_info`: Vector of [`DiskInfoDetailed`] for each disk
|
||||
#[derive(Serialize, Debug)]
|
||||
pub struct DiskInfo {
|
||||
pub total_size: Option<f64>,
|
||||
@@ -16,6 +37,12 @@ pub struct DiskInfo {
|
||||
pub detailed_info: Vec<DiskInfoDetailed>,
|
||||
}
|
||||
|
||||
/// Collects disk information for all detected disks, including usage and temperature.
|
||||
///
|
||||
/// This function enumerates all disks, calculates usage statistics, and attempts to associate temperature sensors with disk components.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<DiskInfo, Box<dyn std::error::Error + Send + Sync>>` - Disk statistics and details, or error if collection fails.
|
||||
pub async fn get_disk_info() -> Result<DiskInfo, Box<dyn std::error::Error + Send + Sync>> {
|
||||
let disks = Disks::new_with_refreshed_list();
|
||||
let mut detailed_info = Vec::new();
|
||||
@@ -37,28 +64,12 @@ pub async fn get_disk_info() -> Result<DiskInfo, Box<dyn std::error::Error + Sen
|
||||
component_disk_label: String::new(),
|
||||
component_disk_temperature: 0.0,
|
||||
});
|
||||
|
||||
println!(
|
||||
"Disk_Name: {:?}:\n---- Disk_Kind: {},\n---- Total: {},\n---- Available: {},\n---- Used: {}, \n---- Mount_Point: {:?}",
|
||||
disk.name(),
|
||||
disk.kind(),
|
||||
disk.total_space(),
|
||||
disk.available_space(),
|
||||
disk_used,
|
||||
disk.mount_point()
|
||||
);
|
||||
}
|
||||
|
||||
// Get component temperatures
|
||||
let components = Components::new_with_refreshed_list();
|
||||
for component in &components {
|
||||
if let Some(temperature) = component.temperature() {
|
||||
println!(
|
||||
"Component_Label: {}, Temperature: {}°C",
|
||||
component.label(),
|
||||
temperature
|
||||
);
|
||||
|
||||
// Update detailed info with temperature data if it matches a disk component
|
||||
for disk_info in &mut detailed_info {
|
||||
if component.label().contains(&disk_info.disk_name) {
|
||||
|
@@ -2,6 +2,29 @@ use anyhow::Result;
|
||||
use nvml_wrapper::Nvml;
|
||||
use std::error::Error;
|
||||
|
||||
/// # GPU Hardware Module
|
||||
///
|
||||
/// This module provides GPU information collection for WatcherAgent, including load, temperature, and VRAM statistics.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **GPU Detection:** Identifies GPU model and capabilities.
|
||||
/// - **Metric Collection:** Queries GPU load, temperature, and VRAM usage using NVML (NVIDIA only).
|
||||
/// - **Error Handling:** Graceful fallback if GPU or NVML is unavailable.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `current_load`: GPU usage as a percentage (**0.0–100.0**)
|
||||
/// - `current_temp`: GPU temperature in **degrees Celsius (°C)**
|
||||
/// - `vram_total`: Total VRAM in **bytes**
|
||||
/// - `vram_used`: Used VRAM in **bytes**
|
||||
///
|
||||
/// GPU statistics for the host system.
|
||||
///
|
||||
/// # Fields
|
||||
/// - `name`: GPU model name (string)
|
||||
/// - `current_load`: GPU usage as a percentage (**0.0–100.0**)
|
||||
/// - `current_temp`: GPU temperature in **degrees Celsius (°C)**
|
||||
/// - `vram_total`: Total VRAM in **bytes**
|
||||
/// - `vram_used`: Used VRAM in **bytes**
|
||||
#[derive(Debug)]
|
||||
pub struct GpuInfo {
|
||||
pub name: Option<String>,
|
||||
@@ -11,6 +34,12 @@ pub struct GpuInfo {
|
||||
pub vram_used: Option<f64>,
|
||||
}
|
||||
|
||||
/// Collects GPU information (load, temperature, VRAM) using NVML.
|
||||
///
|
||||
/// This function attempts to query the first NVIDIA GPU using NVML. If unavailable, it returns a fallback with only the detected GPU name.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<GpuInfo, Box<dyn Error + Send + Sync>>` - GPU statistics or fallback if unavailable.
|
||||
pub async fn get_gpu_info() -> Result<GpuInfo, Box<dyn Error + Send + Sync>> {
|
||||
match get_gpu_metrics() {
|
||||
Ok((gpu_temp, gpu_load, vram_used, vram_total)) => {
|
||||
@@ -37,6 +66,10 @@ pub async fn get_gpu_info() -> Result<GpuInfo, Box<dyn Error + Send + Sync>> {
|
||||
}
|
||||
}
|
||||
|
||||
/// Queries NVML for GPU metrics: temperature, load, VRAM used/total.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(f64, f64, f64, f64), Box<dyn Error + Send + Sync>>` - Tuple of (temperature °C, load %, VRAM used bytes, VRAM total bytes).
|
||||
pub fn get_gpu_metrics() -> Result<(f64, f64, f64, f64), Box<dyn Error + Send + Sync>> {
|
||||
let nvml = Nvml::init();
|
||||
if let Ok(nvml) = nvml {
|
||||
|
@@ -3,25 +3,56 @@ use std::error::Error;
|
||||
use anyhow::Result;
|
||||
use sysinfo::System;
|
||||
|
||||
/// # Memory Hardware Module
|
||||
///
|
||||
/// This module provides memory information collection for WatcherAgent, including total, used, and free RAM.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **Memory Detection:** Queries system for total, used, and free RAM.
|
||||
/// - **Usage Calculation:** Computes memory usage percentage.
|
||||
/// - **Error Handling:** Graceful fallback if metrics are unavailable.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `total`, `used`, `free`: RAM in **bytes**
|
||||
///
|
||||
/// Memory statistics for the host system.
|
||||
///
|
||||
/// # Fields
|
||||
/// - `total`: Total RAM in **bytes**
|
||||
/// - `used`: Used RAM in **bytes**
|
||||
/// - `free`: Free RAM in **bytes**
|
||||
#[derive(Debug)]
|
||||
pub struct MemoryInfo {
|
||||
pub total: Option<f64>,
|
||||
pub total_size: Option<f64>,
|
||||
pub used: Option<f64>,
|
||||
pub free: Option<f64>,
|
||||
pub current_load: Option<f64>,
|
||||
}
|
||||
|
||||
pub async fn get_memory_info() -> Result<MemoryInfo> {
|
||||
/// Collects memory information (total, used, free RAM).
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<MemoryInfo>` - Memory statistics or error if unavailable.
|
||||
pub async fn get_memory_info() -> Result<MemoryInfo, Box<dyn Error + Send + Sync>> {
|
||||
let mut sys = System::new();
|
||||
sys.refresh_memory();
|
||||
|
||||
Ok(MemoryInfo {
|
||||
total: Some(sys.total_memory() as f64),
|
||||
total_size: Some(sys.total_memory() as f64),
|
||||
used: Some(sys.used_memory() as f64),
|
||||
free: Some(sys.free_memory() as f64),
|
||||
current_load: Some(get_memory_usage(&mut sys).unwrap() as f64)
|
||||
})
|
||||
}
|
||||
|
||||
pub fn _get_memory_usage(sys: &mut System) -> Result<f64, Box<dyn Error + Send + Sync>> {
|
||||
/// Computes memory usage percentage from sysinfo::System.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `sys` - Mutable reference to sysinfo::System
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<f64, Box<dyn Error + Send + Sync>>` - Memory usage as percentage.
|
||||
pub fn get_memory_usage(sys: &mut System) -> Result<f64, Box<dyn Error + Send + Sync>> {
|
||||
sys.refresh_memory();
|
||||
Ok((sys.used_memory() as f64 / sys.total_memory() as f64) * 100.0)
|
||||
}
|
||||
|
@@ -14,6 +14,23 @@ pub use memory::get_memory_info;
|
||||
pub use network::get_network_info;
|
||||
pub use network::NetworkMonitor;
|
||||
|
||||
/// # Hardware Module
|
||||
///
|
||||
/// This module aggregates all hardware subsystems for WatcherAgent, providing unified collection and access to CPU, GPU, memory, disk, and network statistics.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **Subsystem Aggregation:** Combines all hardware modules into a single struct for easy access.
|
||||
/// - **Unified Collection:** Provides a single async method to collect all hardware metrics at once.
|
||||
///
|
||||
/// Aggregated hardware statistics for the host system.
|
||||
///
|
||||
/// # Fields
|
||||
/// - `cpu`: CPU statistics (see [`CpuInfo`])
|
||||
/// - `gpu`: GPU statistics (see [`GpuInfo`])
|
||||
/// - `memory`: Memory statistics (see [`MemoryInfo`])
|
||||
/// - `disk`: Disk statistics (see [`DiskInfo`])
|
||||
/// - `network`: Network statistics (see [`NetworkInfo`])
|
||||
/// - `network_monitor`: Rolling monitor for network bandwidth
|
||||
#[derive(Debug)]
|
||||
pub struct HardwareInfo {
|
||||
pub cpu: cpu::CpuInfo,
|
||||
@@ -25,6 +42,10 @@ pub struct HardwareInfo {
|
||||
}
|
||||
|
||||
impl HardwareInfo {
|
||||
/// Collects all hardware statistics asynchronously.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<HardwareInfo, Box<dyn Error + Send + Sync>>` - Aggregated hardware statistics or error if any subsystem fails.
|
||||
pub async fn collect() -> Result<Self, Box<dyn Error + Send + Sync>> {
|
||||
let mut network_monitor = network::NetworkMonitor::new();
|
||||
Ok(Self {
|
||||
|
@@ -2,6 +2,24 @@ use std::error::Error;
|
||||
use std::result::Result;
|
||||
use std::time::Instant;
|
||||
|
||||
/// # Network Hardware Module
|
||||
///
|
||||
/// This module provides network information collection for WatcherAgent, including interface enumeration and bandwidth statistics.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **Interface Detection:** Lists all network interfaces.
|
||||
/// - **Bandwidth Monitoring:** Tracks receive/transmit rates using a rolling monitor.
|
||||
/// - **Error Handling:** Graceful fallback if metrics are unavailable.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `rx_rate`, `tx_rate`: Network bandwidth in **bytes per second (B/s)**
|
||||
///
|
||||
/// Network statistics for the host system.
|
||||
///
|
||||
/// # Fields
|
||||
/// - `interfaces`: List of network interface names (strings)
|
||||
/// - `rx_rate`: Receive bandwidth in **bytes per second (B/s)**
|
||||
/// - `tx_rate`: Transmit bandwidth in **bytes per second (B/s)**
|
||||
#[derive(Debug)]
|
||||
pub struct NetworkInfo {
|
||||
pub interfaces: Option<Vec<String>>,
|
||||
@@ -9,6 +27,13 @@ pub struct NetworkInfo {
|
||||
pub tx_rate: Option<f64>,
|
||||
}
|
||||
|
||||
|
||||
/// Rolling monitor for network bandwidth statistics.
|
||||
///
|
||||
/// # Fields
|
||||
/// - `prev_rx`: Previous received bytes
|
||||
/// - `prev_tx`: Previous transmitted bytes
|
||||
/// - `last_update`: Timestamp of last update
|
||||
#[derive(Debug)]
|
||||
pub struct NetworkMonitor {
|
||||
prev_rx: u64,
|
||||
@@ -23,6 +48,7 @@ impl Default for NetworkMonitor {
|
||||
}
|
||||
|
||||
impl NetworkMonitor {
|
||||
/// Creates a new `NetworkMonitor` for bandwidth tracking.
|
||||
pub fn new() -> Self {
|
||||
Self {
|
||||
prev_rx: 0,
|
||||
@@ -31,6 +57,10 @@ impl NetworkMonitor {
|
||||
}
|
||||
}
|
||||
|
||||
/// Updates the network usage statistics and returns current rx/tx rates.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(f64, f64), Box<dyn Error>>` - Tuple of (rx_rate, tx_rate) in bytes per second.
|
||||
pub fn update_usage(&mut self) -> Result<(f64, f64), Box<dyn Error>> {
|
||||
let (current_rx, current_tx) = get_network_bytes()?;
|
||||
let elapsed = self.last_update.elapsed().as_secs_f64();
|
||||
@@ -55,6 +85,13 @@ impl NetworkMonitor {
|
||||
}
|
||||
}
|
||||
|
||||
/// Collects network information (interfaces, rx/tx rates) using a monitor.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `monitor` - Mutable reference to a `NetworkMonitor`
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<NetworkInfo, Box<dyn Error>>` - Network statistics or error if unavailable.
|
||||
pub async fn get_network_info(monitor: &mut NetworkMonitor) -> Result<NetworkInfo, Box<dyn Error>> {
|
||||
let (rx_rate, tx_rate) = monitor.update_usage()?;
|
||||
Ok(NetworkInfo {
|
||||
|
@@ -1,22 +1,54 @@
|
||||
/// WatcherAgent - A Rust-based system monitoring agent
|
||||
/// This agent collects hardware metrics and sends them to a backend server.
|
||||
/// It supports CPU, GPU, RAM, disk, and network metrics.
|
||||
/// # WatcherAgent
|
||||
///
|
||||
/// **WatcherAgent** is a cross-platform system monitoring agent written in Rust.
|
||||
///
|
||||
/// ## Overview
|
||||
/// This agent collects real-time hardware metrics (CPU, GPU, RAM, disk, network) and communicates with a backend server for registration, reporting, and remote control. It is designed for deployment in environments where automated monitoring and remote management of system resources is required.
|
||||
///
|
||||
/// ## Features
|
||||
/// - **Hardware Metrics:** Collects CPU, GPU, RAM, disk, and network statistics using platform-specific APIs.
|
||||
/// - **Docker Integration:** Detects and manages its own Docker container, supports image updates and container restarts.
|
||||
/// - **Server Communication:** Registers with a backend server, sends periodic heartbeats, and reports metrics securely.
|
||||
/// - **Remote Commands:** Listens for and executes commands from the backend (e.g., update image, restart container, stop agent).
|
||||
///
|
||||
/// ## Modules
|
||||
/// - [`api`]: Handles HTTP communication with the backend server (registration, heartbeat, metrics, commands).
|
||||
/// - [`hardware`]: Collects hardware metrics from the host system (CPU, GPU, RAM, disk, network).
|
||||
/// - [`metrics`]: Orchestrates metric collection and reporting.
|
||||
/// - [`models`]: Defines data structures for server communication and metrics.
|
||||
/// - [`docker`]: Integrates with Docker for container management and agent lifecycle.
|
||||
///
|
||||
/// ## Usage
|
||||
/// Run the agent with the backend server URL as an argument:
|
||||
/// ```sh
|
||||
/// watcheragent <server-url>
|
||||
/// ```
|
||||
///
|
||||
/// The agent will register itself, start collecting metrics, and listen for remote commands.
|
||||
pub mod api;
|
||||
pub mod docker;
|
||||
pub mod hardware;
|
||||
pub mod metrics;
|
||||
pub mod models;
|
||||
pub mod serverclientcomm;
|
||||
|
||||
use bollard::Docker;
|
||||
use std::env;
|
||||
use std::error::Error;
|
||||
use std::marker::Send;
|
||||
use std::marker::Sync;
|
||||
use std::result::Result;
|
||||
use tokio::task::JoinHandle;
|
||||
use bollard::Docker;
|
||||
|
||||
use crate::serverclientcomm::{get_current_image};
|
||||
|
||||
/// Awaits a spawned asynchronous task and flattens its nested `Result` type.
|
||||
///
|
||||
/// This utility is used to handle the result of a `tokio::spawn`ed task that itself returns a `Result`,
|
||||
/// propagating any errors from both the task and its execution.
|
||||
///
|
||||
/// # Type Parameters
|
||||
/// * `T` - The type returned by the task on success.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `handle` - The `JoinHandle` of the spawned task.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<T, Box<dyn Error + Send + Sync>>` - The result of the task, or an error if the task failed or panicked.
|
||||
async fn flatten<T>(
|
||||
handle: JoinHandle<Result<T, Box<dyn Error + Send + Sync>>>,
|
||||
) -> Result<T, Box<dyn Error + Send + Sync>> {
|
||||
@@ -27,28 +59,28 @@ async fn flatten<T>(
|
||||
}
|
||||
}
|
||||
|
||||
/// Main entry point for the WatcherAgent application.
|
||||
///
|
||||
/// This function performs the following steps:
|
||||
/// 1. Initializes the Docker client for container management.
|
||||
/// 2. Detects the current running image version.
|
||||
/// 3. Parses command-line arguments to obtain the backend server URL.
|
||||
/// 4. Registers the agent with the backend server and retrieves its server ID and IP address.
|
||||
/// 5. Spawns background tasks for:
|
||||
/// - Listening for remote commands from the server
|
||||
/// - Sending periodic heartbeat signals
|
||||
/// - Collecting and reporting hardware metrics
|
||||
/// 6. Waits for all background tasks to complete and logs their results.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `server-url` - The URL of the backend server to register and report metrics to (passed as a command-line argument).
|
||||
///
|
||||
/// # Errors
|
||||
/// Returns an error if registration or any background task fails, or if required arguments are missing.
|
||||
#[tokio::main]
|
||||
async fn main() -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
// Initialize Docker client
|
||||
let docker = Docker::connect_with_local_defaults()
|
||||
.map_err(|e| format!("Failed to connect to Docker: {}", e))?;
|
||||
|
||||
// Get current image version
|
||||
let client_version = match get_current_image(&docker).await {
|
||||
Ok(Some(version)) => version,
|
||||
Ok(None) => {
|
||||
eprintln!("Warning: No image version found");
|
||||
"unknown".to_string()
|
||||
}
|
||||
Err(e) => {
|
||||
eprintln!("Warning: Could not get current image version: {}", e);
|
||||
"unknown".to_string()
|
||||
}
|
||||
};
|
||||
println!("Client Version: {}", client_version);
|
||||
|
||||
// Parse command-line arguments
|
||||
let args: Vec<String> = env::args().collect();
|
||||
// args[0] is the binary name, args[1] is the first actual argument
|
||||
if args.len() < 2 {
|
||||
eprintln!("Usage: {} <server-url>", args[0]);
|
||||
return Err("Missing server URL argument".into());
|
||||
@@ -56,22 +88,41 @@ async fn main() -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
let server_url = &args[1];
|
||||
println!("Server URL: {:?}", server_url);
|
||||
|
||||
// Registration
|
||||
// Registration with backend server
|
||||
let (server_id, ip) = match api::register_with_server(&server_url).await {
|
||||
Ok((id, ip)) => (id, ip),
|
||||
Ok((id, ip)) => {
|
||||
println!("Registered with server. ID: {}, IP: {}", id, ip);
|
||||
(id, ip)
|
||||
},
|
||||
Err(e) => {
|
||||
eprintln!("Fehler bei der Registrierung am Server: {e}");
|
||||
return Err(e);
|
||||
}
|
||||
};
|
||||
|
||||
// Initialize Docker (optional - agent can run without Docker)
|
||||
let docker_manager = docker::DockerManager::new_optional();
|
||||
|
||||
// Get current image version
|
||||
let client_version = if let Some(ref docker_manager) = docker_manager {
|
||||
docker_manager.get_client_version().await
|
||||
} else {
|
||||
"unknown".to_string()
|
||||
};
|
||||
println!("Client Version: {}", client_version);
|
||||
|
||||
// Start background tasks
|
||||
// Start server listening for commands
|
||||
let listening_handle = tokio::spawn({
|
||||
let docker = docker.clone();
|
||||
let server_url = server_url.to_string();
|
||||
async move { api::listening_to_server(&docker, &server_url).await }
|
||||
});
|
||||
// Start server listening for commands (only if Docker is available)
|
||||
let listening_handle = if let Some(docker_manager) = docker_manager {
|
||||
tokio::spawn({
|
||||
let docker = docker_manager.docker.clone();
|
||||
let server_url = server_url.to_string();
|
||||
async move { api::listening_to_server(&docker, &server_url).await }
|
||||
})
|
||||
} else {
|
||||
println!("Docker not available, skipping server command listener.");
|
||||
tokio::spawn(async { Ok(()) }) // Dummy task
|
||||
};
|
||||
|
||||
// Start heartbeat in background
|
||||
let heartbeat_handle = tokio::spawn({
|
||||
|
@@ -1,11 +1,31 @@
|
||||
/// # Metrics Module
|
||||
///
|
||||
/// This module orchestrates the collection and reporting of hardware and network metrics for WatcherAgent.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **Metric Collection:** Gathers real-time statistics from all hardware subsystems (CPU, GPU, RAM, disk, network).
|
||||
/// - **Reporting:** Periodically sends metrics to the backend server using the API module.
|
||||
/// - **Error Handling:** Robust to hardware failures and network errors, with retry logic and logging.
|
||||
///
|
||||
/// ## Usage
|
||||
/// The [`Collector`] struct is instantiated in the main loop and runs as a background task, continuously collecting and reporting metrics.
|
||||
use std::error::Error;
|
||||
use std::time::Duration;
|
||||
|
||||
use crate::api;
|
||||
//use crate::docker::DockerInfo;
|
||||
use crate::hardware::network::NetworkMonitor;
|
||||
use crate::hardware::HardwareInfo;
|
||||
use crate::models::MetricDto;
|
||||
|
||||
/// Main orchestrator for hardware and network metric collection and reporting.
|
||||
///
|
||||
/// The `Collector` struct manages the state required to collect metrics and send them to the backend server. It maintains a network monitor for bandwidth tracking, the agent's server ID, and its IP address.
|
||||
///
|
||||
/// # Fields
|
||||
/// - `network_monitor`: Tracks network usage rates (rx/tx).
|
||||
/// - `server_id`: Unique server ID assigned by the backend.
|
||||
/// - `ip_address`: IP address of the agent.
|
||||
pub struct Collector {
|
||||
network_monitor: NetworkMonitor,
|
||||
server_id: i32,
|
||||
@@ -13,6 +33,14 @@ pub struct Collector {
|
||||
}
|
||||
|
||||
impl Collector {
|
||||
/// Creates a new `Collector` instance for metric collection and reporting.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `server_id` - The server ID assigned by the backend.
|
||||
/// * `ip_address` - The IP address of the agent.
|
||||
///
|
||||
/// # Returns
|
||||
/// A new `Collector` ready to collect and report metrics.
|
||||
pub fn new(server_id: i32, ip_address: String) -> Self {
|
||||
Self {
|
||||
network_monitor: NetworkMonitor::new(),
|
||||
@@ -21,6 +49,15 @@ impl Collector {
|
||||
}
|
||||
}
|
||||
|
||||
/// Runs the main metrics collection loop, periodically sending metrics to the backend server.
|
||||
///
|
||||
/// This function continuously collects hardware and network metrics, sends them to the backend, and handles errors gracefully. It uses a configurable interval and retries on failures.
|
||||
///
|
||||
/// # Arguments
|
||||
/// * `base_url` - The base URL of the backend server.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<(), Box<dyn Error + Send + Sync>>` - Ok if metrics are sent successfully.
|
||||
pub async fn run(&mut self, base_url: &str) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
loop {
|
||||
println!(
|
||||
@@ -40,6 +77,12 @@ impl Collector {
|
||||
}
|
||||
}
|
||||
|
||||
/// Collects hardware and network metrics from all subsystems.
|
||||
///
|
||||
/// This function queries the hardware module for CPU, GPU, RAM, disk, and network statistics, and packages them into a [`MetricDto`] for reporting.
|
||||
///
|
||||
/// # Returns
|
||||
/// * `Result<MetricDto, Box<dyn Error + Send + Sync>>` - The collected metrics or an error if hardware info is unavailable.
|
||||
pub async fn collect(&mut self) -> Result<MetricDto, Box<dyn Error + Send + Sync>> {
|
||||
let hardware = match HardwareInfo::collect().await {
|
||||
Ok(hw) => hw,
|
||||
@@ -59,11 +102,11 @@ impl Collector {
|
||||
gpu_load: hardware.gpu.current_load.unwrap_or_default(),
|
||||
gpu_temp: hardware.gpu.current_temp.unwrap_or_default(),
|
||||
gpu_vram_size: hardware.gpu.vram_total.unwrap_or_default(),
|
||||
gpu_vram_usage: hardware.gpu.vram_used.unwrap_or_default(),
|
||||
ram_load: hardware.memory.used.unwrap_or_default(),
|
||||
ram_size: hardware.memory.total.unwrap_or_default(),
|
||||
gpu_vram_load: hardware.gpu.current_load.unwrap_or_default(),
|
||||
ram_load: hardware.memory.current_load.unwrap_or_default(),
|
||||
ram_size: hardware.memory.total_size.unwrap_or_default(),
|
||||
disk_size: hardware.disk.total_size.unwrap_or_default(),
|
||||
disk_usage: hardware.disk.total_used.unwrap_or_default(),
|
||||
disk_usage: hardware.disk.total_usage.unwrap_or_default(),
|
||||
disk_temp: 0.0, // not supported
|
||||
net_rx: hardware.network.rx_rate.unwrap_or_default(),
|
||||
net_tx: hardware.network.tx_rate.unwrap_or_default(),
|
||||
|
@@ -1,10 +1,31 @@
|
||||
/// # Models Module
|
||||
///
|
||||
/// This module defines all data structures (DTOs) used for communication between WatcherAgent and the backend server, as well as hardware metrics and Docker container info.
|
||||
///
|
||||
/// ## Responsibilities
|
||||
/// - **DTOs:** Define payloads for registration, metrics, heartbeat, and server commands.
|
||||
/// - **Units:** All struct fields are documented with their units for clarity and API compatibility.
|
||||
/// - **Docker Info:** Structures for representing Docker container state and statistics.
|
||||
///
|
||||
/// ## Usage
|
||||
/// These types are serialized/deserialized for HTTP communication and used throughout the agent for data exchange.
|
||||
use crate::docker::stats;
|
||||
|
||||
use serde::{Deserialize, Serialize};
|
||||
|
||||
// Data structures matching the C# DTOs
|
||||
/// Registration data sent to the backend server.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `id`: Unique server identifier (integer)
|
||||
/// - `ip_address`: IPv4 or IPv6 address (string)
|
||||
/// - `cpu_type`: CPU model name (string)
|
||||
/// - `cpu_cores`: Number of physical CPU cores (integer)
|
||||
/// - `gpu_type`: GPU model name (string)
|
||||
/// - `ram_size`: Total RAM size in **megabytes (MB)**
|
||||
#[derive(Serialize, Debug)]
|
||||
pub struct RegistrationDto {
|
||||
#[serde(rename = "id")]
|
||||
pub id: i32,
|
||||
pub server_id: i32,
|
||||
#[serde(rename = "ipAddress")]
|
||||
pub ip_address: String,
|
||||
#[serde(rename = "cpuType")]
|
||||
@@ -17,6 +38,24 @@ pub struct RegistrationDto {
|
||||
pub ram_size: f64,
|
||||
}
|
||||
|
||||
/// Hardware and network metrics data sent to the backend server.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `server_id`: Unique server identifier (integer)
|
||||
/// - `ip_address`: IPv4 or IPv6 address (string)
|
||||
/// - `cpu_load`: CPU usage as a percentage (**0.0–100.0**)
|
||||
/// - `cpu_temp`: CPU temperature in **degrees Celsius (°C)**
|
||||
/// - `gpu_load`: GPU usage as a percentage (**0.0–100.0**)
|
||||
/// - `gpu_temp`: GPU temperature in **degrees Celsius (°C)**
|
||||
/// - `gpu_vram_size`: Total GPU VRAM in **bytes**
|
||||
/// - `gpu_vram_load`: GPU Usage of VRAM as a percentage (**0.0–100.0**)
|
||||
/// - `ram_load`: RAM usage as a percentage (**0.0–100.0**)
|
||||
/// - `ram_size`: Total RAM in **bytes**
|
||||
/// - `disk_size`: Total disk size in **bytes**
|
||||
/// - `disk_usage`: Used disk space in **bytes**
|
||||
/// - `disk_temp`: Disk temperature in **degrees Celsius (°C)** (if available)
|
||||
/// - `net_rx`: Network receive rate in **bytes per second (B/s)**
|
||||
/// - `net_tx`: Network transmit rate in **bytes per second (B/s)**
|
||||
#[derive(Serialize, Debug)]
|
||||
pub struct MetricDto {
|
||||
#[serde(rename = "serverId")]
|
||||
@@ -33,8 +72,8 @@ pub struct MetricDto {
|
||||
pub gpu_temp: f64,
|
||||
#[serde(rename = "gpu_Vram_Size")]
|
||||
pub gpu_vram_size: f64,
|
||||
#[serde(rename = "gpu_Vram_Usage")]
|
||||
pub gpu_vram_usage: f64,
|
||||
#[serde(rename = "gpu_Vram_Load")]
|
||||
pub gpu_vram_load: f64,
|
||||
#[serde(rename = "ram_Load")]
|
||||
pub ram_load: f64,
|
||||
#[serde(rename = "ram_Size")]
|
||||
@@ -51,6 +90,13 @@ pub struct MetricDto {
|
||||
pub net_tx: f64,
|
||||
}
|
||||
|
||||
/// Detailed disk information for each detected disk.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `disk_total_space`: Total disk space in **bytes**
|
||||
/// - `disk_available_space`: Available disk space in **bytes**
|
||||
/// - `disk_used_space`: Used disk space in **bytes**
|
||||
/// - `component_disk_temperature`: Disk temperature in **degrees Celsius (°C)**
|
||||
#[derive(Serialize, Debug)]
|
||||
pub struct DiskInfoDetailed {
|
||||
pub disk_name: String,
|
||||
@@ -63,6 +109,11 @@ pub struct DiskInfoDetailed {
|
||||
pub component_disk_temperature: f32,
|
||||
}
|
||||
|
||||
/// Response containing server ID and IP address.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `id`: Unique server identifier (integer)
|
||||
/// - `ip_address`: IPv4 or IPv6 address (string)
|
||||
#[derive(Deserialize)]
|
||||
pub struct IdResponse {
|
||||
pub id: i32,
|
||||
@@ -70,12 +121,24 @@ pub struct IdResponse {
|
||||
pub ip_address: String,
|
||||
}
|
||||
|
||||
/// Heartbeat message data sent to the backend server.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `ip_address`: IPv4 or IPv6 address (string)
|
||||
#[derive(Serialize)]
|
||||
pub struct HeartbeatDto {
|
||||
#[serde(rename = "IpAddress")]
|
||||
pub ip_address: String,
|
||||
}
|
||||
|
||||
/// Hardware summary data for diagnostics and registration.
|
||||
///
|
||||
/// ## Units
|
||||
/// - `cpu_type`: CPU model name (string)
|
||||
/// - `cpu_cores`: Number of physical CPU cores (integer)
|
||||
/// - `gpu_type`: GPU model name (string)
|
||||
/// - `ram_size`: Total RAM size in **megabytes (MB)**
|
||||
/// - `ip_address`: IPv4 or IPv6 address (string)
|
||||
#[derive(Serialize, Debug)]
|
||||
pub struct HardwareDto {
|
||||
pub cpu_type: String,
|
||||
@@ -85,6 +148,12 @@ pub struct HardwareDto {
|
||||
pub ip_address: String,
|
||||
}
|
||||
|
||||
/// Command message received from the backend server.
|
||||
///
|
||||
/// ## Fields
|
||||
/// - `message_type`: Type of command (e.g., "update_image", "restart_container", "stop_agent")
|
||||
/// - `data`: Command payload (arbitrary JSON)
|
||||
/// - `message_id`: Unique identifier for acknowledgment
|
||||
#[derive(Debug, Deserialize, Clone)]
|
||||
pub struct ServerMessage {
|
||||
// Define your message structure here
|
||||
@@ -93,9 +162,46 @@ pub struct ServerMessage {
|
||||
pub message_id: String, // Add an ID for acknowledgment
|
||||
}
|
||||
|
||||
/// Acknowledgment payload sent to the backend server for command messages.
|
||||
///
|
||||
/// ## Fields
|
||||
/// - `message_id`: Unique identifier of the acknowledged message
|
||||
/// - `status`: Status string ("success", "error", etc.)
|
||||
/// - `details`: Additional details or error messages
|
||||
#[derive(Debug, Serialize, Clone)]
|
||||
pub struct Acknowledgment {
|
||||
pub message_id: String,
|
||||
pub status: String, // "success" or "error"
|
||||
pub details: String,
|
||||
}
|
||||
}
|
||||
|
||||
/// Docker container information for agent and managed containers.
|
||||
///
|
||||
/// ## Fields
|
||||
/// - `ID`: Container ID (first 12 hex digits, integer)
|
||||
/// - `image`: Docker image name (string)
|
||||
/// - `Name`: Container name (string)
|
||||
/// - `Status`: Container status ("running", "stopped", etc.)
|
||||
/// - `_net_in`: Network receive rate in **bytes per second (B/s)**
|
||||
/// - `_net_out`: Network transmit rate in **bytes per second (B/s)**
|
||||
/// - `_cpu_load`: CPU usage as a percentage (**0.0–100.0**)
|
||||
#[derive(Debug, Serialize, Clone)]
|
||||
pub struct DockerContainerRegistrationDto {
|
||||
pub server_id: u32,
|
||||
pub containers: Vec<DockerContainerDto>,
|
||||
}
|
||||
|
||||
#[derive(Debug, Serialize, Clone)]
|
||||
pub struct DockerContainerDto {
|
||||
pub id: String,
|
||||
pub image: String,
|
||||
pub name: String,
|
||||
}
|
||||
|
||||
#[derive(Debug, Serialize, Clone)]
|
||||
pub struct DockerContainerMetricDto {
|
||||
pub id: String,
|
||||
pub status: String, // "running";"stopped";others
|
||||
pub network: stats::ContainerNetworkInfo,
|
||||
pub cpu: stats::ContainerCpuInfo,
|
||||
}
|
||||
|
@@ -1,227 +0,0 @@
|
||||
use crate::models::{ServerMessage};
|
||||
|
||||
use std::error::Error;
|
||||
|
||||
use bollard::Docker;
|
||||
use bollard::query_parameters::{CreateImageOptions, RestartContainerOptions, InspectContainerOptions};
|
||||
use futures_util::StreamExt;
|
||||
|
||||
pub async fn handle_server_message(docker: &Docker, msg: ServerMessage) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
let msg = msg.clone();
|
||||
println!("Handling server message: {:?}", msg);
|
||||
|
||||
// Handle different message types
|
||||
match msg.message_type.as_str() {
|
||||
"update_image" => {
|
||||
if let Some(image_name) = msg.data.get("image").and_then(|v| v.as_str()) {
|
||||
println!("Received update command for image: {}", image_name);
|
||||
// Call your update_docker_image function here
|
||||
update_docker_image(docker, image_name).await?;
|
||||
Ok(())
|
||||
} else {
|
||||
Err("Missing image name in update message".into())
|
||||
}
|
||||
}
|
||||
"restart_container" => {
|
||||
println!("Received restart container command");
|
||||
// Call your restart_container function here
|
||||
restart_container(docker).await?;
|
||||
Ok(())
|
||||
}
|
||||
"stop_agent" => {
|
||||
println!("Received stop agent command");
|
||||
// Implement graceful shutdown
|
||||
std::process::exit(0);
|
||||
}
|
||||
_ => {
|
||||
eprintln!("Unknown message type: {}", msg.message_type);
|
||||
Err(format!("Unknown message type: {}", msg.message_type).into())
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
pub async fn update_docker_image(docker: &Docker, image: &str) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
println!("Updating to {}", image);
|
||||
|
||||
// 1. Pull new image
|
||||
let mut stream = docker.create_image(
|
||||
Some(CreateImageOptions {
|
||||
from_image: Some(image.to_string()),
|
||||
..Default::default()
|
||||
}),
|
||||
None,
|
||||
None,
|
||||
);
|
||||
|
||||
// Use the stream with proper trait bounds
|
||||
while let Some(result) = StreamExt::next(&mut stream).await {
|
||||
match result {
|
||||
Ok(progress) => {
|
||||
if let Some(status) = progress.status {
|
||||
println!("Pull status: {}", status);
|
||||
}
|
||||
}
|
||||
Err(e) => {
|
||||
eprintln!("Error pulling image: {}", e);
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// 2. Restart the current container
|
||||
let _ = restart_container(docker).await;
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
pub async fn get_current_image(docker: &Docker) -> Result<Option<String>, Box<dyn Error + Send + Sync>> {
|
||||
// Get the current container ID from /proc/self/cgroup
|
||||
let container_id = match std::fs::read_to_string("/proc/self/cgroup") {
|
||||
Ok(content) => {
|
||||
content
|
||||
.lines()
|
||||
.find_map(|line| {
|
||||
// Look for any line that might contain container information
|
||||
if line.contains("docker") || line.contains("crio") || line.contains("containerd") || line.contains("kubepods") {
|
||||
// Extract the container ID from the line
|
||||
extract_container_id(line)
|
||||
} else {
|
||||
None
|
||||
}
|
||||
})
|
||||
}
|
||||
Err(e) => {
|
||||
eprintln!("Error reading cgroup file: {}", e);
|
||||
return Ok(None);
|
||||
}
|
||||
};
|
||||
|
||||
// Debug: Print what we found
|
||||
println!("Container ID search result: {:?}", container_id);
|
||||
|
||||
let container_id = match container_id {
|
||||
Some(id) if !id.is_empty() => {
|
||||
println!("Found container ID: '{}'", id);
|
||||
id
|
||||
}
|
||||
_ => {
|
||||
// Debug the cgroup file content
|
||||
if let Ok(content) = std::fs::read_to_string("/proc/self/cgroup") {
|
||||
eprintln!("Cgroup file content for debugging:");
|
||||
for (i, line) in content.lines().enumerate() {
|
||||
eprintln!("Line {}: {}", i, line);
|
||||
}
|
||||
}
|
||||
eprintln!("Could not find valid container ID in cgroup");
|
||||
return Ok(None);
|
||||
}
|
||||
};
|
||||
|
||||
// Clean up the container ID - remove any non-hex characters
|
||||
let clean_container_id: String = container_id
|
||||
.chars()
|
||||
.filter(|c| c.is_ascii_hexdigit())
|
||||
.collect();
|
||||
|
||||
// Validate the container ID (should be 64 characters for full ID, but short ID might work too)
|
||||
if clean_container_id.is_empty() {
|
||||
eprintln!("Container ID contains no hex characters: '{}'", container_id);
|
||||
return Ok(None);
|
||||
}
|
||||
|
||||
println!("Using container ID: '{}' (cleaned: '{}')", container_id, clean_container_id);
|
||||
|
||||
// Try with both the original and cleaned ID
|
||||
let ids_to_try = vec![&clean_container_id, &container_id];
|
||||
|
||||
for id in ids_to_try {
|
||||
println!("Attempting to inspect container with ID: '{}'", id);
|
||||
|
||||
match docker.inspect_container(id, None::<InspectContainerOptions>).await {
|
||||
Ok(container_info) => {
|
||||
if let Some(config) = container_info.config {
|
||||
if let Some(image) = config.image {
|
||||
println!("Successfully found image: {}", image);
|
||||
return Ok(Some(image));
|
||||
}
|
||||
}
|
||||
eprintln!("Container inspected but no image found in config");
|
||||
return Ok(Some("unknown".to_string()));
|
||||
}
|
||||
Err(e) => {
|
||||
eprintln!("Error inspecting container with ID '{}': {}", id, e);
|
||||
// Continue to try the next ID
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
eprintln!("All attempts to inspect container failed");
|
||||
Ok(None)
|
||||
}
|
||||
|
||||
fn extract_container_id(line: &str) -> Option<String> {
|
||||
let parts: Vec<&str> = line.split('/').collect();
|
||||
|
||||
if let Some(last_part) = parts.last() {
|
||||
let last_part = last_part.trim();
|
||||
|
||||
println!("Processing cgroup line part: '{}'", last_part);
|
||||
|
||||
// Common patterns for container IDs in cgroups:
|
||||
|
||||
// Pattern 1: Full container ID (64-character hex)
|
||||
if last_part.len() == 64 && last_part.chars().all(|c| c.is_ascii_hexdigit()) {
|
||||
println!("Found full container ID: {}", last_part);
|
||||
return Some(last_part.to_string());
|
||||
}
|
||||
|
||||
// Pattern 2: docker-<short_id>.scope
|
||||
if last_part.starts_with("docker-") && last_part.ends_with(".scope") {
|
||||
let id = last_part
|
||||
.trim_start_matches("docker-")
|
||||
.trim_end_matches(".scope")
|
||||
.to_string();
|
||||
println!("Found docker scope ID: {}", id);
|
||||
return Some(id);
|
||||
}
|
||||
|
||||
// Pattern 3: <short_id> (12-character hex or similar)
|
||||
if last_part.chars().all(|c| c.is_ascii_hexdigit()) && last_part.len() >= 12 {
|
||||
println!("Found hex ID: {}", last_part);
|
||||
return Some(last_part.to_string());
|
||||
}
|
||||
|
||||
// Pattern 4: Try to extract ID from any string that looks like it contains a hex ID
|
||||
if let Some(caps) = regex::Regex::new(r"[a-fA-F0-9]{12,64}")
|
||||
.ok()
|
||||
.and_then(|re| re.find(last_part))
|
||||
{
|
||||
let id = caps.as_str().to_string();
|
||||
println!("Extracted ID via regex: {}", id);
|
||||
return Some(id);
|
||||
}
|
||||
|
||||
// Pattern 5: If nothing else matches, try the last part as-is
|
||||
if !last_part.is_empty() {
|
||||
println!("Using last part as ID: {}", last_part);
|
||||
return Some(last_part.to_string());
|
||||
}
|
||||
}
|
||||
|
||||
None
|
||||
}
|
||||
|
||||
pub async fn restart_container(docker: &Docker) -> Result<(), Box<dyn Error + Send + Sync>> {
|
||||
if let Ok(container_id) = std::env::var("HOSTNAME") {
|
||||
println!("Restarting container {}", container_id);
|
||||
if let Err(e) = docker.restart_container(&container_id, Some(RestartContainerOptions { signal: None, t: Some(0) }))
|
||||
.await
|
||||
{
|
||||
eprintln!("Failed to restart container: {}", e);
|
||||
}
|
||||
} else {
|
||||
eprintln!("No container ID found (HOSTNAME not set?)");
|
||||
}
|
||||
|
||||
Ok(())
|
||||
}
|
Reference in New Issue
Block a user