// 实时指标收集器
// 收集 Amber 运行时的各种性能指标，支持 Prometheus 导出
use std::collections::BTreeMap;
use std::sync::{Arc, Mutex, atomic::Ordering};
use anyhow::{Result, Context};
use serde::{Deserialize, Serialize};
use tracing::{info, warn, error, debug};
use uuid::Uuid;
use std::sync::atomic::AtomicU64;
use std::time::SystemTime;
/// Tenant ID type
pub type TenantId = String;
/// Cluster ID type
pub type ClusterId = String;
/// 指标收集器
#[derive(Debug)]
pub struct MetricsCollector {
/// 总请求数
pub requests_total: AtomicU64,
/// 总延迟时间（毫秒）
pub total_latency_ms: AtomicU64,
/// 当前活跃连接数
pub active_connections: AtomicU64,
/// 内存使用量（字节）
pub memory_usage_bytes: AtomicU64,
/// CPU 使用率（百分比）
pub cpu_usage_percent: AtomicU64,
/// 最后更新时间
pub last_update: SystemTime,
}
impl MetricsCollector {
/// 创建新的指标收集器
pub fn new() -> Self {
Self {
requests_total: AtomicU64::new(0),
total_latency_ms: AtomicU64::new(0),
active_connections: AtomicU64::new(0),
memory_usage_bytes: AtomicU64::new(0),
cpu_usage_percent: AtomicU64::new(0),
last_update: SystemTime::now(),
}
}
/// 记录请求指标
///
/// # Arguments
///
/// * `latency` - 请求延迟时间
/// * `status` - 请求状态
pub fn record_request(&mut self, latency: Duration, status: RequestStatus) {
self.requests_total.fetch_add(1, Ordering::SeqCst);
self.total_latency_ms
.fetch_add(latency.as_millis() as u64, Ordering::SeqCst);
self.last_update = SystemTime::now();
}
/// 记录内存使用情况
///
/// # Arguments
///
/// * `bytes` - 内存使用量（字节）
pub fn record_memory_usage(&mut self, bytes: u64) {
self.memory_usage_bytes
.fetch_add(bytes, Ordering::SeqCst);
self.last_update = SystemTime::now();
}
/// 更新活跃连接数
///
/// # Arguments
///
/// * `count` - 活跃连接数
pub fn update_active_connections(&mut self, count: u64) {
self.active_connections.store(count, Ordering::SeqCst);
self.last_update = SystemTime::now();
}
/// 更新 CPU 使用率
///
/// # Arguments
///
/// * `percent` - CPU 使用率（0-100）
pub fn update_cpu_usage(&mut self, percent: f64) {
let percent: _ = percent.round() as u64;
self.cpu_usage_percent.store(percent, Ordering::SeqCst);
self.last_update = SystemTime::now();
}
/// 获取平均请求延迟（毫秒）
pub fn get_average_latency_ms(&self) -> f64 {
let total_requests: _ = self.requests_total.load(Ordering::SeqCst);
if total_requests == 0 {
return 0.0;
}
let total_latency: _ = self.total_latency_ms.load(Ordering::SeqCst);
total_latency as f64 / total_requests as f64
}
/// 获取当前指标快照
pub fn get_metrics_snapshot(&self) -> MetricsSnapshot {
MetricsSnapshot {
requests_total: self.requests_total.load(Ordering::SeqCst),
total_latency_ms: self.total_latency_ms.load(Ordering::SeqCst),
active_connections: self.active_connections.load(Ordering::SeqCst),
memory_usage_bytes: self.memory_usage_bytes.load(Ordering::SeqCst),
cpu_usage_percent: self.cpu_usage_percent.load(Ordering::SeqCst),
last_update: self.last_update,
}
}
/// 导出 Prometheus 格式的指标
///
/// # Returns
///
/// 返回 Prometheus 格式的指标字符串
pub fn export_prometheus(&self) -> Result<String> {
let snapshot: _ = self.get_metrics_snapshot();
let average_latency: _ = self.get_average_latency_ms();
let output: _ = format!(
"# HELP amberjs_requests_total Total number of requests processed\n\
# TYPE amberjs_requests_total counter\n\
amberjs_requests_total {}\n\
# HELP amberjs_request_duration_ms_total Total request duration in milliseconds\n\
# TYPE amberjs_request_duration_ms_total counter\n\
amberjs_request_duration_ms_total {}\n\
# HELP amberjs_request_duration_ms Average request duration in milliseconds\n\
# TYPE amberjs_request_duration_ms gauge\n\
amberjs_request_duration_ms {}\n\
# HELP amberjs_active_connections Number of active connections\n\
# TYPE amberjs_active_connections gauge\n\
amberjs_active_connections {}\n\
# HELP amberjs_memory_usage_bytes Memory usage in bytes\n\
# TYPE amberjs_memory_usage_bytes gauge\n\
amberjs_memory_usage_bytes {}\n\
# HELP amberjs_cpu_usage_percent CPU usage percentage\n\
# TYPE amberjs_cpu_usage_percent gauge\n\
amberjs_cpu_usage_percent {}\n",
snapshot.requests_total,
snapshot.total_latency_ms,
average_latency,
snapshot.active_connections,
snapshot.memory_usage_bytes,
snapshot.cpu_usage_percent
);
Ok(output)
}
}
impl Default for MetricsCollector {
fn default() -> Self {
Self::new()
}
}
/// 请求状态
#[derive(Debug, Clone, Serialize, Deserialize)]
pub enum RequestStatus {
Success,
Error(String),
}
/// 指标快照
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct MetricsSnapshot {
pub requests_total: u64,
pub total_latency_ms: u64,
pub active_connections: u64,
pub memory_usage_bytes: u64,
pub cpu_usage_percent: u64,
pub last_update: SystemTime,
}
#[cfg(test)]
mod tests {
#[test]
fn test_metrics_collector_creation() {
let collector: _ = MetricsCollector::new();
assert_eq!(collector.requests_total.load(Ordering::SeqCst), 0);
assert_eq!(collector.total_latency_ms.load(Ordering::SeqCst), 0);
assert_eq!(collector.active_connections.load(Ordering::SeqCst), 0);
}
#[test]
fn test_record_request() {
let mut collector = MetricsCollector::new();
collector.record_request(
Duration::from_millis(150),
RequestStatus::Success,
);
assert_eq!(collector.requests_total.load(Ordering::SeqCst), 1);
assert_eq!(collector.total_latency_ms.load(Ordering::SeqCst), 150);
}
#[test]
fn test_get_average_latency() {
let mut collector = MetricsCollector::new();
collector.record_request(
Duration::from_millis(100),
RequestStatus::Success,
);
collector.record_request(
Duration::from_millis(200),
RequestStatus::Success,
);
collector.record_request(
Duration::from_millis(150),
RequestStatus::Success,
);
let average: _ = collector.get_average_latency_ms();
assert_eq!(average, 150.0); // (100 + 200 + 150) / 3
}
#[test]
fn test_export_prometheus() {
let mut collector = MetricsCollector::new();
collector.record_request(
Duration::from_millis(150),
RequestStatus::Success,
);
collector.update_active_connections(5);
collector.record_memory_usage(1024 * 1024); // 1MB
collector.update_cpu_usage(45.5);
let output: _ = collector.export_prometheus().unwrap();
// 验证 Prometheus 格式
assert!(output.contains("amberjs_requests_total 1"));
assert!(output.contains("amberjs_request_duration_ms_total 150"));
assert!(output.contains("amberjs_active_connections 5"));
assert!(output.contains("amberjs_memory_usage_bytes 1048576"));
assert!(output.contains("amberjs_cpu_usage_percent 46")); // 45.5 舍入为 46
// 验证 HELP 和 TYPE 注释
assert!(output.contains("# HELP"));
assert!(output.contains("# TYPE"));
}
#[test]
fn test_memory_usage() {
let mut collector = MetricsCollector::new();
collector.record_memory_usage(2048);
assert_eq!(collector.memory_usage_bytes.load(Ordering::SeqCst), 2048);
collector.record_memory_usage(1024);
assert_eq!(collector.memory_usage_bytes.load(Ordering::SeqCst), 3072);
}
#[test]
fn test_active_connections() {
let mut collector = MetricsCollector::new();
collector.update_active_connections(10);
assert_eq!(collector.active_connections.load(Ordering::SeqCst), 10);
collector.update_active_connections(5);
assert_eq!(collector.active_connections.load(Ordering::SeqCst), 5);
}
}
/// Enterprise Metrics Collector
/// 企业级指标收集器，支持多租户和集群级监控
/// 集群指标
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct ClusterMetrics {
pub cluster_id: ClusterId,
pub timestamp: SystemTime,
pub total_nodes: u32,
pub ready_nodes: u32,
pub cpu_usage_percent: f64,
pub memory_usage_percent: f64,
pub network_io_mbps: f64,
pub disk_io_ops: u64,
pub active_pods: u32,
pub failed_pods: u32,
pub pending_pods: u32,
}
/// 租户指标
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct TenantMetrics {
pub tenant_id: TenantId,
pub timestamp: SystemTime,
pub cpu_usage_percent: f64,
pub memory_usage_percent: f64,
pub storage_usage_gb: f64,
pub network_throughput_mbps: f64,
pub active_sessions: u32,
pub request_rate: f64,
pub error_rate: f64,
pub response_time_ms: f64,
}
/// 告警配置
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct AlertConfig {
pub name: String,
pub description: String,
pub severity: AlertSeverity,
pub condition: AlertCondition,
pub threshold: f64,
pub duration_seconds: u64,
pub enabled: bool,
}
/// 告警严重程度
#[derive(Debug, Clone, Serialize, Deserialize)]
pub enum AlertSeverity {
Critical,
Warning,
Info,
}
/// 告警条件
#[derive(Debug, Clone, Serialize, Deserialize)]
pub enum AlertCondition {
CpuUsageAbove,
MemoryUsageAbove,
DiskUsageAbove,
ErrorRateAbove,
ResponseTimeAbove,
PodFailureRateAbove,
}
/// 告警事件
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct Alert {
pub id: Uuid,
pub config_name: String,
pub severity: AlertSeverity,
pub message: String,
pub timestamp: SystemTime,
pub resolved: bool,
pub cluster_id: Option<ClusterId>,
pub tenant_id: Option<TenantId>,
}
/// 告警动作
#[derive(Debug, Clone, Serialize, Deserialize)]
pub enum AlertAction {
Notify,
Scale,
Restart,
Suspend,
Log,
}
/// 企业级指标收集器
#[derive(Debug)]
pub struct EnterpriseMetricsCollector {
prometheus_registry: Arc<PrometheusRegistry>,
cluster_collectors: std::collections::BTreeMap<ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector, ClusterId, MetricsCollector>>,
tenant_collectors: std::collections::BTreeMap<TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector, TenantId, MetricsCollector>>,
alert_configs: std::collections::BTreeMap<String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig, String, AlertConfig>>,
alerts: std::collections::BTreeMap<Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert, Uuid, Alert>>,
}
/// Prometheus 注册表
#[derive(Debug)]
pub struct PrometheusRegistry {
metrics: Arc<std::sync::Mutex<std::collections::BTreeMap<String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String>>>,
}
impl PrometheusRegistry {
pub fn new() -> Self {
Self {
metrics: Arc::new(Mutex::new(std::sync::Mutex::new(std::collections::BTreeMap::new()))
}
}
pub fn register_metric(&self, name: String, value: String) {
let mut metrics = self.metrics.lock().unwrap();
metrics.insert(name, value);
}
pub fn get_metrics(&self) -> String {
let metrics: _ = self.metrics.lock().unwrap();
metrics
.iter()
.map(|(k, v)| format!("{} {}", k, v))
.collect::<Vec<_>()
.join("\n")
}
}
impl EnterpriseMetricsCollector {
/// 创建新的企业级指标收集器
pub fn new() -> Self {
Self {
prometheus_registry: Arc::new(Mutex::new(PrometheusRegistry::new()))
cluster_collectors: std::collections::BTreeMap::new(),
tenant_collectors: std::collections::BTreeMap::new(),
alert_configs: std::collections::BTreeMap::new(),
alerts: std::collections::BTreeMap::new(),
}
}
/// 收集集群指标
pub async fn collect_cluster_metrics(&self, cluster_id: &ClusterId) -> Result<ClusterMetrics> {
let collector: _ = self
.cluster_collectors
.get(cluster_id)
.context("Cluster collector not found")?;
let snapshot: _ = collector.get_metrics_snapshot();
// 模拟集群指标收集（实际实现中会从 Kubernetes API 获取）
let metrics: _ = ClusterMetrics {
cluster_id: cluster_id.clone(),
timestamp: SystemTime::now(),
total_nodes: 10,
ready_nodes: 9,
cpu_usage_percent: snapshot.cpu_usage_percent as f64,
memory_usage_percent: (snapshot.memory_usage_bytes as f64 / (1024.0 * 1024.0 * 1024.0)) * 100.0,
network_io_mbps: 100.0,
disk_io_ops: 1000,
active_pods: snapshot.active_connections as u32,
failed_pods: 0,
pending_pods: 1,
};
Ok(metrics)
}
/// 收集租户指标
pub async fn collect_tenant_metrics(&self, tenant_id: &TenantId) -> Result<TenantMetrics> {
let collector: _ = self
.tenant_collectors
.get(tenant_id)
.context("Tenant collector not found")?;
let snapshot: _ = collector.get_metrics_snapshot();
// 模拟租户指标收集（实际实现中会从租户级监控获取）
let metrics: _ = TenantMetrics {
tenant_id: tenant_id.clone(),
timestamp: SystemTime::now(),
cpu_usage_percent: snapshot.cpu_usage_percent as f64,
memory_usage_percent: (snapshot.memory_usage_bytes as f64 / (1024.0 * 1024.0 * 1024.0)) * 100.0,
storage_usage_gb: 10.0,
network_throughput_mbps: 50.0,
active_sessions: snapshot.active_connections as u32,
request_rate: snapshot.requests_total as f64 / 60.0,
error_rate: 0.01,
response_time_ms: collector.get_average_latency_ms(),
};
Ok(metrics)
}
/// 设置告警规则
pub async fn setup_alerts(&mut self, configs: Vec<AlertConfig>) -> Result<()> {
for config in configs {
self.alert_configs.insert(config.name.clone(), config);
info!("Added alert rule: {}", config.name);
}
Ok(())
}
/// 处理告警事件
pub async fn handle_alert(&mut self, alert: Alert) -> Result<AlertAction> {
let action: _ = match alert.severity {
AlertSeverity::Critical => AlertAction::Scale,
AlertSeverity::Warning => AlertAction::Notify,
AlertSeverity::Info => AlertAction::Log,
};
self.alerts.insert(alert.id, alert);
info!("Handled alert event: {}", action_to_string(&action));
Ok(action)
}
/// 检查告警条件
pub async fn check_alerts(&mut self) -> Result<Vec<Alert> {
let mut triggered_alerts = Vec::new();
for (name, config) in &self.alert_configs {
if !config.enabled {
continue;
}
let should_alert: _ = match &config.condition {
AlertCondition::CpuUsageAbove => self.check_cpu_threshold(config.threshold),
AlertCondition::MemoryUsageAbove => self.check_memory_threshold(config.threshold),
AlertCondition::ErrorRateAbove => self.check_error_rate_threshold(config.threshold),
_ => false,
};
if should_alert {
let alert: _ = Alert {
id: Uuid::new_v4(),
config_name: name.clone(),
severity: config.severity.clone(),
message: format!(
"Alert condition met: {} above threshold {}",
name, config.threshold
),
timestamp: SystemTime::now(),
resolved: false,
cluster_id: None,
tenant_id: None,
};
triggered_alerts.push(alert);
}
}
Ok(triggered_alerts)
}
/// 检查 CPU 阈值
fn check_cpu_threshold(&self, threshold: f64) -> bool {
for collector in self.cluster_collectors.values() {
let cpu_usage: _ = collector.cpu_usage_percent.load(Ordering::SeqCst);
if cpu_usage as f64 > threshold {
return true;
}
}
false
}
/// 检查内存阈值
fn check_memory_threshold(&self, threshold: f64) -> bool {
for collector in self.cluster_collectors.values() {
let memory_bytes: _ = collector.memory_usage_bytes.load(Ordering::SeqCst);
let memory_gb: _ = memory_bytes as f64 / (1024.0 * 1024.0 * 1024.0);
if memory_gb > threshold {
return true;
}
}
false
}
/// 检查错误率阈值
fn check_error_rate_threshold(&self, threshold: f64) -> bool {
for collector in self.cluster_collectors.values() {
let requests: _ = collector.requests_total.load(Ordering::SeqCst);
// 简化错误率计算
let error_rate: _ = if requests > 0 { 0.05 } else { 0.0 };
if error_rate > threshold {
return true;
}
}
false
}
/// 添加集群收集器
pub fn add_cluster_collector(&mut self, cluster_id: ClusterId, collector: MetricsCollector) {
self.cluster_collectors.insert(cluster_id, collector);
}
/// 添加租户收集器
pub fn add_tenant_collector(&mut self, tenant_id: TenantId, collector: MetricsCollector) {
self.tenant_collectors.insert(tenant_id, collector);
}
/// 导出所有指标到 Prometheus
pub fn export_all_metrics(&self) -> Result<String> {
let mut output = String::new();
// 导出集群指标
for (cluster_id, collector) in &self.cluster_collectors {
let cluster_metrics: _ = format!(
"# Cluster: {}\n{}",
cluster_id,
collector.export_prometheus()?
);
output.push_str(&cluster_metrics);
}
// 导出租户指标
for (tenant_id, collector) in &self.tenant_collectors {
let tenant_metrics: _ = format!(
"# Tenant: {}\n{}",
tenant_id,
collector.export_prometheus()?
);
output.push_str(&tenant_metrics);
}
Ok(output)
}
}
/// 告警动作转字符串
fn action_to_string(action: &AlertAction) -> &'static str {
match action {
AlertAction::Notify => "Notify",
AlertAction::Scale => "Scale",
AlertAction::Restart => "Restart",
AlertAction::Suspend => "Suspend",
AlertAction::Log => "Log",
}
}
#[cfg(test)]
mod enterprise_tests {
use std::collections::{HashMap, BTreeMap};
use std::time::Duration;
use std::sync::atomic::Ordering;
#[tokio::test]
async fn test_enterprise_metrics_collector() {
let mut collector = EnterpriseMetricsCollector::new();
// 添加测试集群
let cluster_collector: _ = MetricsCollector::new();
collector.add_cluster_collector("test-cluster".to_string(), cluster_collector);
// 添加测试租户
let tenant_collector: _ = MetricsCollector::new();
collector.add_tenant_collector("test-tenant".to_string(), tenant_collector);
// 测试集群指标收集
let cluster_metrics: _ = collector
.collect_cluster_metrics(&"test-cluster".to_string())
.await
.unwrap();
assert_eq!(cluster_metrics.cluster_id, "test-cluster");
assert!(cluster_metrics.total_nodes > 0);
// 测试租户指标收集
let tenant_metrics: _ = collector
.collect_tenant_metrics(&"test-tenant".to_string())
.await
.unwrap();
assert_eq!(tenant_metrics.tenant_id, "test-tenant");
assert!(tenant_metrics.cpu_usage_percent >= 0.0);
}
#[tokio::test]
async fn test_alert_setup() {
let mut collector = EnterpriseMetricsCollector::new();
let alert_configs: _ = vec![AlertConfig {
name: "high_cpu".to_string(),
description: "CPU usage above 80%".to_string(),
severity: AlertSeverity::Warning,
condition: AlertCondition::CpuUsageAbove,
threshold: 80.0,
duration_seconds: 60,
enabled: true,
}];
collector.setup_alerts(alert_configs).await.unwrap();
assert_eq!(collector.alert_configs.len(), 1);
assert!(collector.alert_configs.contains_key("high_cpu"));
}
#[test]
fn test_prometheus_registry() {
let registry: _ = PrometheusRegistry::new();
registry.register_metric("test_metric".to_string(), "100".to_string());
let metrics: _ = registry.get_metrics();
assert!(metrics.contains("test_metric 100"));
}
}