Skip to content

rust.unbounded_retry

Stability High

Detects retry loops without maximum attempt limits.

Why It Matters

Unbounded retries:

Infinite loops — Never give up on permanent failures
Amplify load — Retry storms overwhelm upstream
Waste resources — CPU/network spent on hopeless retries

Example

// ❌ Before (unbounded retry)
async fn fetch_data() -> Data {
    loop {
        match client.get("/data").send().await {
            Ok(data) => return data,
            Err(_) => tokio::time::sleep(Duration::from_secs(1)).await,
        }
    }
}

// ✅ After (with max retries and backoff)
use backoff::ExponentialBackoff;

async fn fetch_data() -> Result<Data, Error> {
    let backoff = ExponentialBackoff {
        max_elapsed_time: Some(Duration::from_secs(60)),
        ..Default::default()
    };

    backoff::future::retry(backoff, || async {
        client.get("/data").send().await
            .map_err(backoff::Error::transient)
    }).await
}

What Unfault Detects

Loop-based retries without counter
Missing max_retries configuration
No exponential backoff strategy

Auto-Fix

Unfault can add retry limits using backoff crate.