apache/pulsar · error · IllegalArgumentException

Per instance CPU requested, %s, for function is greater than

Error message

Per instance CPU requested, %s, for function is greater than the maximum required, %s

What it means

validateMaxResourcesRequired enforces a worker-configured maximum CPU per function instance. If functionDetails.resources.cpu exceeds functionInstanceMaxResources.cpu, Pulsar throws IllegalArgumentException with both values to prevent oversized functions from being admitted to Kubernetes.

Source

Thrown at pulsar-functions/runtime/src/main/java/org/apache/pulsar/functions/runtime/kubernetes/KubernetesRuntimeFactory.java:481

                            String.format("Per instance RAM requested is not specified. "
                                    + "Must specify RAM requested for function to be at least %s", minRam));
                } else if (functionDetails.getResources().getRam() < minRam) {
                    throw new IllegalArgumentException(
                            String.format("Per instance RAM requested, %s, "
                                            + "for function is less than the minimum required, %s",
                                    functionDetails.getResources().getRam(), minRam));
                }
            }
        }
    }

    void validateMaxResourcesRequired(FunctionDetails functionDetails) {
        if (functionInstanceMaxResources != null) {
            Double maxCpu = functionInstanceMaxResources.getCpu();
            Long maxRam = functionInstanceMaxResources.getRam();

            if (maxCpu != null && functionDetails.getResources().getCpu() > maxCpu) {
                throw new IllegalArgumentException(
                        String.format(
                                "Per instance CPU requested, %s, for function is greater than the maximum required, %s",
                                functionDetails.getResources().getCpu(), maxCpu));
            }

            if (maxRam != null && functionDetails.getResources().getRam() > maxRam) {
                throw new IllegalArgumentException(
                        String.format(
                                "Per instance RAM requested, %s, for function is greater than the maximum required, %s",
                                functionDetails.getResources().getRam(), maxRam));
            }
        }
    }

    void validateResourcesGranularityAndProportion(FunctionDetails functionDetails) {
        final long baseMillis = 1000;
        long multiples = 0L;
        if (functionInstanceResourceGranularities != null) {

View on GitHub (pinned to 820761864e)

Solutions

Reduce the function's --cpu to at most the configured maximum.
Split the workload into more function instances with lower per-instance CPU instead of raising per-instance CPU.
Ask the cluster admin to raise functionInstanceMaxResources.cpu if the cap is too low.
Check whether a units mistake (millicores vs cores) inflated the requested value.

Example fix

// before
--cpu 8
// after
--cpu 2  (assuming max cpu is 2) -- or increase --instances to scale out

Defensive patterns

Strategy: validation

Validate before calling

Double maxCpu = workerCfg.getFunctionInstanceMaxResources().getCpu();
if (maxCpu != null && details.getResources().getCpu() > maxCpu) {
    throw new IllegalArgumentException("cpu " + details.getResources().getCpu() + " > allowed max " + maxCpu);
}

Try / catch

try { admin.functions().updateFunction(config); } catch (IllegalArgumentException e) { if (e.getMessage().contains("greater than the maximum")) { /* lower cpu / scale out instances */ } throw e; }

Prevention

Cap CPU in deployment templates to the cluster maximum
Prefer horizontal scale (more instances) over raising per-instance CPU
Watch for core-vs-millicore unit mistakes

When it happens

Trigger: Submitting/updating a function whose --cpu is greater than the cluster's configured functionInstanceMaxResources.cpu (doAdmissionChecks -> validateMaxResourcesRequired, line 481).

Common situations: Scaling CPU up without checking cluster caps; copying configs from a cluster without maximums; accidental unit mistakes (cores vs millicores); cluster admin tightening max resources after deployment.

Related errors

AI-assisted analysis of apache/pulsar@820761864e (2026-09-06). Data as JSON: /api/errors/66a281f77f9908d7. Report an issue: GitHub.