Compare commits
11 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 5be67fed64 | |||
| cac836601b | |||
| bd22ce265b | |||
| bbc8b7b1cb | |||
| 5dd8cadef3 | |||
| fefe30ec51 | |||
| fb40cce748 | |||
| eaa057b284 | |||
| f23a1b5cec | |||
| 3f98f68b51 | |||
| 3d38a7a984 |
6
Cargo.lock
generated
6
Cargo.lock
generated
@@ -279,7 +279,7 @@ checksum = "a1d728cc89cf3aee9ff92b05e62b19ee65a02b5702cff7d5a377e32c6ae29d8d"
|
||||
|
||||
[[package]]
|
||||
name = "cm-dashboard"
|
||||
version = "0.1.172"
|
||||
version = "0.1.182"
|
||||
dependencies = [
|
||||
"anyhow",
|
||||
"chrono",
|
||||
@@ -301,7 +301,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "cm-dashboard-agent"
|
||||
version = "0.1.172"
|
||||
version = "0.1.182"
|
||||
dependencies = [
|
||||
"anyhow",
|
||||
"async-trait",
|
||||
@@ -324,7 +324,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "cm-dashboard-shared"
|
||||
version = "0.1.172"
|
||||
version = "0.1.182"
|
||||
dependencies = [
|
||||
"chrono",
|
||||
"serde",
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
[package]
|
||||
name = "cm-dashboard-agent"
|
||||
version = "0.1.172"
|
||||
version = "0.1.183"
|
||||
edition = "2021"
|
||||
|
||||
[dependencies]
|
||||
|
||||
@@ -386,7 +386,7 @@ impl DiskCollector {
|
||||
/// Get SMART data for drives
|
||||
async fn get_smart_data_for_drives(&self, physical_drives: &[PhysicalDrive], mergerfs_pools: &[MergerfsPool]) -> HashMap<String, SmartData> {
|
||||
let mut smart_data = HashMap::new();
|
||||
|
||||
|
||||
// Collect all drive names
|
||||
let mut all_drives = std::collections::HashSet::new();
|
||||
for drive in physical_drives {
|
||||
@@ -413,23 +413,30 @@ impl DiskCollector {
|
||||
|
||||
/// Get SMART data for a single drive
|
||||
async fn get_smart_data(&self, drive_name: &str) -> Result<SmartData, CollectorError> {
|
||||
let output = Command::new("sudo")
|
||||
.args(&["smartctl", "-a", &format!("/dev/{}", drive_name)])
|
||||
.output()
|
||||
use tracing::info;
|
||||
|
||||
// Use direct smartctl (no sudo) - service has CAP_SYS_RAWIO capability
|
||||
// For NVMe drives, specify device type explicitly
|
||||
let mut cmd = Command::new("smartctl");
|
||||
if drive_name.starts_with("nvme") {
|
||||
cmd.args(&["-d", "nvme", "-a", &format!("/dev/{}", drive_name)]);
|
||||
} else {
|
||||
cmd.args(&["-a", &format!("/dev/{}", drive_name)]);
|
||||
}
|
||||
|
||||
let output = cmd.output()
|
||||
.map_err(|e| CollectorError::SystemRead {
|
||||
path: format!("SMART data for {}", drive_name),
|
||||
error: e.to_string(),
|
||||
})?;
|
||||
|
||||
let output_str = String::from_utf8_lossy(&output.stdout);
|
||||
let error_str = String::from_utf8_lossy(&output.stderr);
|
||||
|
||||
// Debug logging for SMART command results
|
||||
debug!("SMART output for {}: status={}, stdout_len={}, stderr={}",
|
||||
drive_name, output.status, output_str.len(), error_str);
|
||||
|
||||
|
||||
if !output.status.success() {
|
||||
debug!("SMART command failed for {}: {}", drive_name, error_str);
|
||||
if drive_name.starts_with("nvme") {
|
||||
info!("NVMe SMART failed for {}: status={}, stderr={}",
|
||||
drive_name, output.status, String::from_utf8_lossy(&output.stderr));
|
||||
}
|
||||
// Return unknown data rather than failing completely
|
||||
return Ok(SmartData {
|
||||
health: "UNKNOWN".to_string(),
|
||||
@@ -514,12 +521,19 @@ impl DiskCollector {
|
||||
}
|
||||
}
|
||||
|
||||
Ok(SmartData {
|
||||
health,
|
||||
serial_number,
|
||||
let smart_data = SmartData {
|
||||
health: health.clone(),
|
||||
serial_number: serial_number.clone(),
|
||||
temperature_celsius: temperature,
|
||||
wear_percent,
|
||||
})
|
||||
};
|
||||
|
||||
if drive_name.starts_with("nvme") {
|
||||
info!("NVMe SMART parsed for {}: serial={:?}, temp={:?}, health={}, wear={:?}",
|
||||
drive_name, serial_number, temperature, health, wear_percent);
|
||||
}
|
||||
|
||||
Ok(smart_data)
|
||||
}
|
||||
|
||||
/// Populate drives data into AgentData
|
||||
|
||||
@@ -117,8 +117,37 @@ impl SystemdCollector {
|
||||
}
|
||||
}
|
||||
|
||||
// Docker containers are now collected as top-level services below
|
||||
// Keeping nginx as sub-services for now
|
||||
if service_name.contains("docker") && active_status == "active" {
|
||||
let docker_containers = self.get_docker_containers();
|
||||
for (container_name, container_status) in docker_containers {
|
||||
// For now, docker containers have no additional metrics
|
||||
// Future: could add memory_mb, cpu_percent, restart_count, etc.
|
||||
let metrics = Vec::new();
|
||||
|
||||
sub_services.push(SubServiceData {
|
||||
name: container_name.clone(),
|
||||
service_status: self.calculate_service_status(&container_name, &container_status),
|
||||
metrics,
|
||||
});
|
||||
}
|
||||
|
||||
// Add Docker images
|
||||
let docker_images = self.get_docker_images();
|
||||
for (image_name, image_status, image_size) in docker_images {
|
||||
let mut metrics = Vec::new();
|
||||
metrics.push(SubServiceMetric {
|
||||
label: "size".to_string(),
|
||||
value: 0.0, // Size as string in name instead
|
||||
unit: None,
|
||||
});
|
||||
|
||||
sub_services.push(SubServiceData {
|
||||
name: format!("{} ({})", image_name, image_size),
|
||||
service_status: self.calculate_service_status(&image_name, &image_status),
|
||||
metrics,
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
// Create complete service data
|
||||
let service_data = ServiceData {
|
||||
@@ -140,21 +169,6 @@ impl SystemdCollector {
|
||||
}
|
||||
}
|
||||
|
||||
// Collect Docker containers as top-level services
|
||||
let docker_containers = self.get_docker_containers();
|
||||
for (container_name, container_status) in docker_containers {
|
||||
let service_data = ServiceData {
|
||||
name: container_name.clone(),
|
||||
memory_mb: 0.0, // TODO: Could add container memory via docker stats
|
||||
disk_gb: 0.0, // TODO: Could add container disk usage
|
||||
user_stopped: false,
|
||||
service_status: self.calculate_service_status(&container_name, &container_status),
|
||||
sub_services: Vec::new(),
|
||||
};
|
||||
agent_data.services.push(service_data.clone());
|
||||
complete_service_data.push(service_data);
|
||||
}
|
||||
|
||||
// Update cached state
|
||||
{
|
||||
let mut state = self.state.write().unwrap();
|
||||
@@ -759,10 +773,10 @@ impl SystemdCollector {
|
||||
fn get_docker_containers(&self) -> Vec<(String, String)> {
|
||||
let mut containers = Vec::new();
|
||||
|
||||
// Check if docker is available (use sudo for permissions)
|
||||
// Check if docker is available (cm-agent user is in docker group)
|
||||
// Use -a to show ALL containers (running and stopped)
|
||||
let output = Command::new("sudo")
|
||||
.args(&["docker", "ps", "-a", "--format", "{{.Names}},{{.Status}}"])
|
||||
let output = Command::new("docker")
|
||||
.args(&["ps", "-a", "--format", "{{.Names}},{{.Status}}"])
|
||||
.output();
|
||||
|
||||
let output = match output {
|
||||
@@ -799,6 +813,55 @@ impl SystemdCollector {
|
||||
|
||||
containers
|
||||
}
|
||||
|
||||
/// Get docker images as sub-services
|
||||
fn get_docker_images(&self) -> Vec<(String, String, String)> {
|
||||
let mut images = Vec::new();
|
||||
// Check if docker is available (cm-agent user is in docker group)
|
||||
let output = Command::new("docker")
|
||||
.args(&["images", "--format", "{{.Repository}}:{{.Tag}},{{.Size}}"])
|
||||
.output();
|
||||
|
||||
let output = match output {
|
||||
Ok(out) if out.status.success() => out,
|
||||
Ok(_) => {
|
||||
return images;
|
||||
}
|
||||
Err(_) => {
|
||||
return images;
|
||||
}
|
||||
};
|
||||
|
||||
let output_str = match String::from_utf8(output.stdout) {
|
||||
Ok(s) => s,
|
||||
Err(_) => return images,
|
||||
};
|
||||
|
||||
for line in output_str.lines() {
|
||||
if line.trim().is_empty() {
|
||||
continue;
|
||||
}
|
||||
|
||||
let parts: Vec<&str> = line.split(',').collect();
|
||||
if parts.len() >= 2 {
|
||||
let image_name = parts[0].trim();
|
||||
let size = parts[1].trim();
|
||||
|
||||
// Skip <none>:<none> images (dangling images)
|
||||
if image_name.contains("<none>") {
|
||||
continue;
|
||||
}
|
||||
|
||||
images.push((
|
||||
format!("image_{}", image_name),
|
||||
"active".to_string(), // Images are always "active" (present)
|
||||
size.to_string()
|
||||
));
|
||||
}
|
||||
}
|
||||
|
||||
images
|
||||
}
|
||||
}
|
||||
|
||||
#[async_trait]
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
[package]
|
||||
name = "cm-dashboard"
|
||||
version = "0.1.172"
|
||||
version = "0.1.183"
|
||||
edition = "2021"
|
||||
|
||||
[dependencies]
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
[package]
|
||||
name = "cm-dashboard-shared"
|
||||
version = "0.1.172"
|
||||
version = "0.1.183"
|
||||
edition = "2021"
|
||||
|
||||
[dependencies]
|
||||
|
||||
Reference in New Issue
Block a user