rust-lang/rust · error · anyhow::Error

Job failed

Error message

Job failed

What it means

Thrown by run_workflow_locally after the spawned docker run.sh process (configured with the job's image and environment) exits with a non-zero status. The message carries no detail because the underlying failure already printed its own output to the inherited stdout/stderr of the docker container.

Source

Thrown at src/ci/citool/src/main.rs:135

        let value = match value {
            Value::Bool(value) => value.to_string(),
            Value::Number(value) => value.to_string(),
            Value::String(value) => value.clone(),
            _ => panic!("Unexpected type for environment variable {key} Only bool/number/string is supported.")
        };
        (key.clone(), value)
    }));

    init_submodule_if_needed("src/llvm-project/")?;

    let mut cmd = Command::new(Path::new(DOCKER_DIRECTORY).join("run.sh"));
    cmd.arg(job.image());
    cmd.envs(custom_env);

    eprintln!("Executing {cmd:?}");

    let result = cmd.spawn()?.wait()?;
    if !result.success() { Err(anyhow::anyhow!("Job failed")) } else { Ok(()) }
}

fn upload_ci_metrics(cpu_usage_csv: &Path) -> anyhow::Result<()> {
    let usage = load_cpu_usage(cpu_usage_csv).context("Cannot load CPU usage from input CSV")?;
    eprintln!("CPU usage\n{usage:?}");

    let avg = if !usage.is_empty() { usage.iter().sum::<f64>() / usage.len() as f64 } else { 0.0 };
    eprintln!("CPU usage average: {avg}");

    upload_datadog_metric("avg-cpu-usage", avg).context("Cannot upload Datadog metric")?;

    Ok(())
}

fn postprocess_metrics(
    metrics_path: PathBuf,
    parent: Option<String>,
    job_name: Option<String>,

View on GitHub (pinned to 7088e4b63a)

Solutions

1. Scroll up through the docker container output to find the real failing command or test output.
2. Re-run with the same job name after fixing the identified build/test failure.
3. If the container errored on environment setup, verify the Docker image is current (./x.py setup, or rebuild the image).
4. Check host resources (df -h, free -h); dist jobs need substantial disk and RAM.

Example fix

null

match run_workflow_locally(db, job_type, name) {
    Ok(()) => {}
    Err(e) if e.to_string() == "Job failed" => {
        // The real failure was printed by the container; surface exit guidance.
        eprintln!("Local job exited non-zero; review container output above.");
        std::process::exit(1);
    }
    Err(e) => return Err(e),
}

When it happens

Trigger: The job's build/test/CI script inside the Docker container failed: a compilation error, a test failure, a missing dependency, an out-of-memory or disk-space exhaustion, or a network error fetching artifacts.

Common situations: Reproducing a failing CI job locally where the same test failure occurs; Docker image out of date relative to the repo; insufficient host resources (disk/RAM) for a dist build; a flaky test.

Related errors

• Job {name} not found. The following jobs are available: {}
• Only Linux jobs can be executed locally
• Cannot get jobs of workflow run {workflow_run_id}: {status}
• duplicate job name `{job_name}` in section `{section}`
• PR job `{}` differs from corresponding Auto job `{}` in conf

AI-assisted analysis of rust-lang/rust@7088e4b63a (2026-08-10). Data as JSON: /api/errors/62b6afa142e88f5e. Report an issue: GitHub.