rs-healthcheck-client

HTTP health check client for monitoring service dependencies with configurable thresholds

Installation

[dependencies]
philiprehberger-healthcheck-client = "0.3.0"

Usage

use philiprehberger_healthcheck_client::HealthChecker;

#[tokio::main]
async fn main() {
    let mut checker = HealthChecker::new();
    checker
        .add_http("api", "http://localhost:8080/health")
        .add_tcp("database", "127.0.0.1", 5432)
        .add_tcp("redis", "127.0.0.1", 6379);

    let report = checker.check_all().await;
    println!("{}", report.summary());
}

Custom Timeouts and Expected Status

use philiprehberger_healthcheck_client::HealthChecker;

let mut checker = HealthChecker::new();
checker
    .add_http_with("api", "http://localhost:8080/health", 200, 3000)
    .add_tcp_with("db", "127.0.0.1", 5432, 2000);

HTTP Method (HEAD vs GET)

use philiprehberger_healthcheck_client::{HealthChecker, HttpMethod};

let mut checker = HealthChecker::new();
checker.add_http_with_method("api", "http://localhost:8080/health", HttpMethod::Head);

Status Code Range

use philiprehberger_healthcheck_client::HealthChecker;

let mut checker = HealthChecker::new();
// Accept any 2xx response
checker.add_http_status_range("api", "http://localhost:8080/health", 200, 299);

Latency Aggregates

let report = checker.check_all().await;
println!("p50: {:?}ms", report.latency_p50());
println!("p95: {:?}ms", report.latency_p95());

Custom Checks

use philiprehberger_healthcheck_client::{HealthChecker, CheckResult, HealthStatus};

let mut checker = HealthChecker::new();
checker.add_custom("disk-space", || {
    CheckResult {
        name: String::new(),
        status: HealthStatus::Healthy,
        latency_ms: 0,
        message: Some("80% free".to_string()),
        timestamp: 0,
    }
});

Check a Single Service

if let Some(result) = checker.check_one("api").await {
    println!("{}: {}", result.name, result.status);
}

JSON Output for /health Endpoints

let report = checker.check_all().await;
let json = report.to_json();
// Serve `json` as the response body for your /health endpoint

API

Type / Method	Description
HealthStatus	Enum: Healthy, Degraded, Unhealthy
CheckResult	Result of a single check (name, status, latency, message, timestamp)
Check	Check definition enum: Http, Tcp, Custom
HttpMethod	Enum: Get, Head
StatusMatch	Enum: Exact(code), Range(min, max)
HealthChecker::new()	Create a new checker with no checks
.add_http(name, url)	Add HTTP GET check (expects 200, 5s timeout)
.add_http_with(name, url, status, timeout_ms)	Add HTTP GET check with custom settings
.add_http_with_method(name, url, method)	Add HTTP check with the given HttpMethod
.add_http_status_range(name, url, min, max)	Add HTTP check accepting any code in [min, max]
.add_tcp(name, host, port)	Add TCP check (5s timeout)
.add_tcp_with(name, host, port, timeout_ms)	Add TCP check with custom timeout
.add_custom(name, fn)	Add a custom check function
.failure_threshold(n)	Set consecutive failure threshold
.check_all()	Run all checks in parallel, return HealthReport
.check_one(name)	Run a single check by name
HealthReport.overall	Overall status derived from all checks
HealthReport.is_healthy()	Returns true if overall status is Healthy
HealthReport.is_degraded()	Returns true if overall status is Degraded
HealthReport.failed_checks()	Returns only Unhealthy check results
HealthReport.unhealthy_checks()	Returns non-healthy check results
HealthReport.healthy_checks()	Returns only Healthy check results
HealthReport.latency_p50()	Median latency across checks (ms), or None if empty
HealthReport.latency_p95()	95th percentile latency (ms), or None if empty
HealthReport.to_json()	Serialize report to JSON
HealthReport.summary()	Human-readable summary string

Development

cargo test
cargo clippy -- -D warnings

Support

If you find this project useful:

⭐ Star the repo

🐛 Report issues

💡 Suggest features

❤️ Sponsor development

🌐 All Open Source Projects

💻 GitHub Profile

🔗 LinkedIn Profile

License

MIT