apache/seatunnel · error · IllegalArgumentException

Already an MDCCallable

Error message

Already an MDCCallable

What it means

MDCTracer.tracing(context, Callable) rejects a delegate already wrapped as MDCCallable via IllegalArgumentException. It prevents nested MDC wrapping of Callables, which would snapshot/restore the wrong MDC state on completion.

Source

Thrown at seatunnel-api/src/main/java/org/apache/seatunnel/api/tracing/MDCTracer.java:81

    public static MDCRunnable tracing(MDCContext context, Runnable delegate) {
        if (delegate instanceof MDCRunnable) {
            throw new IllegalArgumentException("Already an MDCRunnable");
        }
        return new MDCRunnable(context, delegate);
    }

    public static <V> MDCCallable<V> tracing(Callable<V> delegate) {
        return tracing(MDCContext.current(), delegate);
    }

    public static <V> MDCCallable<V> tracing(Long jobId, Callable<V> delegate) {
        return tracing(MDCContext.of(jobId), delegate);
    }

    public static <V> MDCCallable<V> tracing(MDCContext context, Callable<V> delegate) {
        if (delegate instanceof MDCCallable) {
            throw new IllegalArgumentException("Already an MDCCallable");
        }
        return new MDCCallable<>(context, delegate);
    }

    public static MDCExecutor tracing(Executor delegate) {
        return tracing(MDCContext.current(), delegate);
    }

    public static MDCExecutor tracing(Long jobId, Executor delegate) {
        return tracing(MDCContext.of(jobId), delegate);
    }

    public static MDCExecutor tracing(MDCContext context, Executor delegate) {
        if (delegate instanceof MDCExecutor) {
            throw new IllegalArgumentException("Already an MDCExecutor");
        }
        return new MDCExecutor(context, delegate);
    }

View on GitHub (pinned to cf67b549a7)

Solutions

1. Trace the Callable only once, at the outermost layer
2. Guard with `if (delegate instanceof MDCCallable) use it else tracing(...)`
3. Strip the earlier wrap and pass the unwrapped Callable

Example fix

// before
MDCCallable<V> c = MDCTracer.tracing(jobId, MDCTracer.tracing(jobId, call));
// after
MDCCallable<V> c = MDCTracer.tracing(jobId, call);

MDCCallable<V> safeWrap(Long jobId, Callable<V> c) {
    return c instanceof MDCCallable ? (MDCCallable<V>) c : MDCTracer.tracing(jobId, c);
}

boolean isWrapped(Callable<?> c) {
    return c instanceof MDCCallable;
}

try {
    return MDCTracer.tracing(jobId, callable);
} catch (IllegalArgumentException e) {
    return (MDCCallable<V>) callable;
}

When it happens

Trigger: Calling tracing(jobId, callable) or tracing(context, callable) where callable is already an MDCCallable from a previous tracing() call.

Common situations: Double-wrapping tasks submitted through both a helper utility and the executor wrapper; caching already-traced Callables and re-tracing them.

Understand the failure class

Background: "Must be a positive integer", "Invalid value", "Unsupported": the invalid-argument-value error family, when a library rejects the value you pass — this error's family across 35 libraries.

Related errors

• Already an MDCRunnable
• Already an MDCExecutor
• Already an MDCConsumer
• Already an MDCFunction
• Already an MDCPredicate

AI-assisted analysis of apache/seatunnel@cf67b549a7 (2026-09-10). Data as JSON: /api/errors/d7b8b0927127fb89. Report an issue: GitHub.