Tracker Checker: handle Health Check timeouts

[josecelano]

Parent issue: #677

You can run a Tracker Checker with:

TORRUST_CHECKER_CONFIG='{
    "udp_trackers": [],
    "http_trackers": [],
    "health_checks": ["http://127.0.0.1:3030/sleep"]
}' cargo run --bin tracker_checker

If the server does not reply in 5 seconds you will see an error like this:

$ TORRUST_CHECKER_CONFIG='{
    "udp_trackers": [],
    "http_trackers": [],
    "health_checks": ["http://127.0.0.1:3030/sleep"]
}' cargo run --bin tracker_checker
    Finished dev [optimized + debuginfo] target(s) in 0.08s
     Running `target/debug/tracker_checker`
Running checks for trackers ...
UDP trackers ...
HTTP trackers ...
Health checks ...
✗ - Health API at http://127.0.0.1:3030/sleep is failing: reqwest::Error { kind: Request, url: Url { scheme: "http", cannot_be_a_base: false, username: "", password: None, host: Some(Ipv4(127.0.0.1)), port: Some(3030), path: "/sleep", query: None, fragment: None }, source: TimedOut }

You can use this example app to simulate the server.

This is the function for the check:

async fn run_health_check(url: Url, console: &Console) -> Result<(), CheckError> {
    let client = HttpClient::builder().timeout(Duration::from_secs(5)).build().unwrap();

    let colored_url = url.to_string().yellow();

    match client.get(url.clone()).send().await {
        Ok(response) => {
            if response.status().is_success() {
                console.println(&format!("{} - Health API at {} is OK", "✓".green(), colored_url));
                Ok(())
            } else {
                console.eprintln(&format!(
                    "{} - Health API at {} is failing: {:?}",
                    "✗".red(),
                    colored_url,
                    response
                ));
                Err(CheckError::HealthCheckError { url })
            }
        }
        Err(err) => {
            console.eprintln(&format!(
                "{} - Health API at {} is failing: {:?}",
                "✗".red(),
                colored_url,
                err
            ));
            Err(CheckError::HealthCheckError { url })
        }
    }
}

It contains this line:

let client = HttpClient::builder().timeout(Duration::from_secs(5)).build().unwrap();

We have to catch the error instead of unwrapping. Otherwise, the Tracker Checker is halted. The Tracker Checker should finish the whole report even if one service is not available.

[josecelano]

This was implemented by @da2ce7:

Example:

TORRUST_CHECKER_CONFIG='{
    "udp_trackers": [],
    "http_trackers": [],
    "health_checks": ["http://127.0.0.1:3030/sleep", "https://google.es"]
}' cargo run --bin tracker_checker

Output:

    Finished `dev` profile [optimized + debuginfo] target(s) in 0.09s
     Running `target/debug/tracker_checker`
2024-09-11T08:11:26.887637Z  INFO torrust_tracker::console::clients::checker::service: Running checks for trackers ...

[
  {
    "Health": {
      "Err": {
        "url": "http://127.0.0.1:3030/sleep",
        "result": {
          "Err": "Heath check failed to get a response: reqwest::Error { kind: Request, url: \"http://127.0.0.1:3030/sleep\", source: TimedOut }"
        }
      }
    }
  },
  {
    "Health": {
      "Ok": {
        "url": "https://google.es/",
        "result": {
          "Ok": "200 OK"
        }
      }
    }
  }
]