Move 17 library crates into crates/, CLI binary into bin/fxt, consolidate 10 test crates into testing/, split config crate from deployment config files. Root directory reduced from 38+ to ~17 directories. All Cargo.toml paths and build.rs proto refs updated. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
247 lines
7.5 KiB
Rust
247 lines
7.5 KiB
Rust
//! Resilience and Recovery Metrics Collection
|
|
//!
|
|
//! Tracks recovery times, error rates, and system behavior under stress.
|
|
|
|
use hdrhistogram::Histogram;
|
|
use serde::{Deserialize, Serialize};
|
|
use std::sync::Arc;
|
|
use std::time::{Duration, Instant};
|
|
use tokio::sync::RwLock;
|
|
|
|
/// Recovery metrics for measuring system resilience
|
|
#[derive(Debug, Clone, Serialize, Deserialize)]
|
|
pub struct RecoveryMetrics {
|
|
/// Time taken to detect the failure
|
|
pub detection_time: Duration,
|
|
/// Time taken to recover from the failure
|
|
pub recovery_time: Duration,
|
|
/// Total downtime experienced
|
|
pub total_downtime: Duration,
|
|
/// Number of retry attempts before recovery
|
|
pub retry_attempts: u32,
|
|
/// Whether circuit breaker activated
|
|
pub circuit_breaker_activated: bool,
|
|
/// Whether graceful degradation occurred
|
|
pub graceful_degradation: bool,
|
|
/// Data consistency maintained during failure
|
|
pub data_consistency_maintained: bool,
|
|
}
|
|
|
|
impl RecoveryMetrics {
|
|
/// Create new recovery metrics
|
|
pub fn new() -> Self {
|
|
Self {
|
|
detection_time: Duration::ZERO,
|
|
recovery_time: Duration::ZERO,
|
|
total_downtime: Duration::ZERO,
|
|
retry_attempts: 0,
|
|
circuit_breaker_activated: false,
|
|
graceful_degradation: false,
|
|
data_consistency_maintained: true,
|
|
}
|
|
}
|
|
|
|
/// Calculate uptime percentage
|
|
pub fn uptime_percentage(&self) -> f64 {
|
|
let total_time = self.total_downtime + Duration::from_secs(1); // Avoid div by zero
|
|
let uptime = total_time.saturating_sub(self.total_downtime);
|
|
(uptime.as_secs_f64() / total_time.as_secs_f64()) * 100.0
|
|
}
|
|
|
|
/// Check if meets 99.9% uptime SLA
|
|
pub fn meets_sla(&self, target_uptime: f64) -> bool {
|
|
self.uptime_percentage() >= target_uptime
|
|
}
|
|
}
|
|
|
|
impl Default for RecoveryMetrics {
|
|
fn default() -> Self {
|
|
Self::new()
|
|
}
|
|
}
|
|
|
|
/// Comprehensive resilience metrics
|
|
pub struct ResilienceMetrics {
|
|
/// Recovery time histogram (microseconds)
|
|
recovery_times: Arc<RwLock<Histogram<u64>>>,
|
|
/// Detection time histogram (microseconds)
|
|
detection_times: Arc<RwLock<Histogram<u64>>>,
|
|
/// Error count by category
|
|
error_counts: Arc<RwLock<std::collections::HashMap<String, u64>>>,
|
|
/// Circuit breaker activations
|
|
circuit_breaker_count: Arc<RwLock<u64>>,
|
|
/// Total test scenarios run
|
|
total_scenarios: Arc<RwLock<u64>>,
|
|
/// Successful recoveries
|
|
successful_recoveries: Arc<RwLock<u64>>,
|
|
}
|
|
|
|
impl ResilienceMetrics {
|
|
/// Create new resilience metrics collector
|
|
pub fn new() -> Self {
|
|
Self {
|
|
recovery_times: Arc::new(RwLock::new(
|
|
Histogram::<u64>::new(5).expect("Failed to create recovery histogram"),
|
|
)),
|
|
detection_times: Arc::new(RwLock::new(
|
|
Histogram::<u64>::new(5).expect("Failed to create detection histogram"),
|
|
)),
|
|
error_counts: Arc::new(RwLock::new(std::collections::HashMap::new())),
|
|
circuit_breaker_count: Arc::new(RwLock::new(0)),
|
|
total_scenarios: Arc::new(RwLock::new(0)),
|
|
successful_recoveries: Arc::new(RwLock::new(0)),
|
|
}
|
|
}
|
|
|
|
/// Record recovery metrics from a test scenario
|
|
pub async fn record_recovery(&self, metrics: &RecoveryMetrics) {
|
|
// Record recovery time
|
|
if self
|
|
.recovery_times
|
|
.write()
|
|
.await
|
|
.record(u64::try_from(metrics.recovery_time.as_micros()).unwrap_or(u64::MAX))
|
|
.is_ok()
|
|
{
|
|
// Recorded successfully
|
|
}
|
|
|
|
// Record detection time
|
|
if self
|
|
.detection_times
|
|
.write()
|
|
.await
|
|
.record(u64::try_from(metrics.detection_time.as_micros()).unwrap_or(u64::MAX))
|
|
.is_ok()
|
|
{
|
|
// Recorded successfully
|
|
}
|
|
|
|
// Increment circuit breaker count if activated
|
|
if metrics.circuit_breaker_activated {
|
|
*self.circuit_breaker_count.write().await += 1;
|
|
}
|
|
|
|
// Increment successful recoveries if recovered
|
|
if metrics.recovery_time > Duration::ZERO {
|
|
*self.successful_recoveries.write().await += 1;
|
|
}
|
|
|
|
*self.total_scenarios.write().await += 1;
|
|
}
|
|
|
|
/// Record an error occurrence
|
|
pub async fn record_error(&self, category: &str) {
|
|
let mut counts = self.error_counts.write().await;
|
|
*counts.entry(category.to_string()).or_insert(0) += 1;
|
|
}
|
|
|
|
/// Get mean recovery time
|
|
pub async fn mean_recovery_time(&self) -> Duration {
|
|
let hist = self.recovery_times.read().await;
|
|
// hist.mean() returns f64, convert safely
|
|
let mean_micros = hist.mean();
|
|
let mean_u64 = mean_micros as u64;
|
|
Duration::from_micros(mean_u64)
|
|
}
|
|
|
|
/// Get p99 recovery time
|
|
pub async fn p99_recovery_time(&self) -> Duration {
|
|
let hist = self.recovery_times.read().await;
|
|
Duration::from_micros(hist.value_at_quantile(0.99))
|
|
}
|
|
|
|
/// Get success rate
|
|
pub async fn success_rate(&self) -> f64 {
|
|
let total = *self.total_scenarios.read().await as f64;
|
|
if total == 0.0 {
|
|
return 0.0;
|
|
}
|
|
let successful = *self.successful_recoveries.read().await as f64;
|
|
(successful / total) * 100.0
|
|
}
|
|
|
|
/// Get circuit breaker activation rate
|
|
pub async fn circuit_breaker_rate(&self) -> f64 {
|
|
let total = *self.total_scenarios.read().await as f64;
|
|
if total == 0.0 {
|
|
return 0.0;
|
|
}
|
|
let activations = *self.circuit_breaker_count.read().await as f64;
|
|
(activations / total) * 100.0
|
|
}
|
|
|
|
/// Generate summary report
|
|
pub async fn generate_report(&self) -> String {
|
|
let mean_recovery = self.mean_recovery_time().await;
|
|
let p99_recovery = self.p99_recovery_time().await;
|
|
let success_rate = self.success_rate().await;
|
|
let cb_rate = self.circuit_breaker_rate().await;
|
|
let total = *self.total_scenarios.read().await;
|
|
|
|
format!(
|
|
r#"
|
|
Resilience Metrics Summary
|
|
==========================
|
|
Total Scenarios: {}
|
|
Success Rate: {:.2}%
|
|
Circuit Breaker Activation Rate: {:.2}%
|
|
|
|
Recovery Times:
|
|
Mean: {:?}
|
|
P99: {:?}
|
|
|
|
Error Counts:
|
|
"#,
|
|
total, success_rate, cb_rate, mean_recovery, p99_recovery,
|
|
)
|
|
}
|
|
}
|
|
|
|
impl Default for ResilienceMetrics {
|
|
fn default() -> Self {
|
|
Self::new()
|
|
}
|
|
}
|
|
|
|
/// Recovery timer for measuring recovery phases
|
|
pub struct RecoveryTimer {
|
|
start: Instant,
|
|
detection_time: Option<Duration>,
|
|
recovery_time: Option<Duration>,
|
|
}
|
|
|
|
impl RecoveryTimer {
|
|
/// Start a new recovery timer
|
|
pub fn start() -> Self {
|
|
Self {
|
|
start: Instant::now(),
|
|
detection_time: None,
|
|
recovery_time: None,
|
|
}
|
|
}
|
|
|
|
/// Mark failure detection
|
|
pub fn mark_detection(&mut self) {
|
|
self.detection_time = Some(self.start.elapsed());
|
|
}
|
|
|
|
/// Mark recovery completion
|
|
pub fn mark_recovery(&mut self) {
|
|
self.recovery_time = Some(self.start.elapsed());
|
|
}
|
|
|
|
/// Build recovery metrics
|
|
pub fn build_metrics(self) -> RecoveryMetrics {
|
|
RecoveryMetrics {
|
|
detection_time: self.detection_time.unwrap_or(Duration::ZERO),
|
|
recovery_time: self.recovery_time.unwrap_or(Duration::ZERO),
|
|
total_downtime: self.recovery_time.unwrap_or(Duration::ZERO),
|
|
retry_attempts: 0,
|
|
circuit_breaker_activated: false,
|
|
graceful_degradation: false,
|
|
data_consistency_maintained: true,
|
|
}
|
|
}
|
|
}
|