What causes "Collect worker resource snapshot failed: {}" in apache/seatunnel?

Check the exception message to identify which worker failed to report.

Collect worker resource snapshot failed

Error message

Collect worker resource snapshot failed: {}

What it means

PendingDiagnosticsCollector.collectWorkerResourceSnapshot builds a per-worker resource diagnostic projection from the master's ResourceManager. On exception, it logs 'Collect worker resource snapshot failed', marks the snapshot unavailable, and returns it so the collector can still emit a partial cluster snapshot.

Solutions

Check the exception message to identify which worker failed to report.
Confirm the worker process is running and reachable; restart it if dead.
Re-run diagnostics after the cluster stabilizes to capture the full worker snapshot.
If only some workers appear in snapshots, check master-worker RPC timeouts and network stability.

Example fix

// before: diagnose with missing worker data
// after: restore worker then re-collect
bin/seatunnel.sh --shutdown-worker <address> ; bin/seatunnel.sh --start-worker  // then rerun pending diagnostics

Defensive patterns

Strategy: try-catch

Validate before calling

// Filter null/incomplete worker profiles before building the snapshot
List<WorkerProfile> profiles = workers.stream().filter(Objects::nonNull).collect(Collectors.toList());
if (profiles.isEmpty()) { log.warn("no worker profiles available; snapshot will be marked unavailable"); }

Try / catch

try {
    snapshot = collector.collectWorkerResourceSnapshot(resourceManager);
    if (!snapshot.isAvailable()) {
        // re-run later or fall back to last good snapshot
    }
} catch (Exception e) {
    log.warn("diagnostic collection degraded: {}", e.getMessage());
}

Prevention

Monitor worker liveness so resource queries don't hit dead nodes.
Avoid running diagnostics during worker scale events.
Treat snapshot.available=false as a signal to re-collect later.
Increase master-worker RPC timeout if profiles are large or networks are slow.

When it happens

Trigger: Called from collectClusterSnapshot; fails when iterating worker profiles throws — a worker disconnected mid-iteration, its WorkerProfile is null/incomplete, or an RPC to fetch worker resource info fails.

Common situations: Flapping workers during scale-up/down, network partitions isolating a worker from the master, or running diagnostics on an already-degraded cluster where some workers cannot answer resource queries.

Related errors

AI-assisted analysis of apache/seatunnel@cf67b549a7 (2026-09-10). Data as JSON: /api/errors/463f875be648757a. Report an issue: GitHub.

Appendix: source

Thrown at seatunnel-engine/seatunnel-engine-server/src/main/java/org/apache/seatunnel/engine/server/diagnostic/PendingDiagnosticsCollector.java:296

        snapshot.setCollectedAt(System.currentTimeMillis());
        if (resourceManager == null) {
            return snapshot;
        }
        try {
            Map<Address, WorkerProfile> registerWorker = resourceManager.getRegisterWorker();
            if (registerWorker == null) {
                return snapshot;
            }
            List<WorkerResourceDiagnostic> workers =
                    registerWorker.entrySet().stream()
                            .map(entry -> convertWorker(entry.getKey(), entry.getValue()))
                            .filter(worker -> worker != null)
                            .sorted(Comparator.comparing(WorkerResourceDiagnostic::getAddress))
                            .collect(Collectors.toList());
            snapshot.setAvailable(true);
            snapshot.setWorkers(workers);
        } catch (Exception e) {
            log.warn("Collect worker resource snapshot failed: {}", ExceptionUtils.getMessage(e));
        }
        return snapshot;
    }

    private static WorkerResourceDiagnostic convertWorker(
            Address registeredAddress, WorkerProfile workerProfile) {
        if (workerProfile == null) {
            return null;
        }
        WorkerResourceDiagnostic diagnostic = new WorkerResourceDiagnostic();
        Address address = workerProfile.getAddress();
        if (address == null) {
            address = registeredAddress;
        }
        diagnostic.setAddress(address == null ? "UNKNOWN" : address.toString());
        if (workerProfile.getAttributes() != null) {
            diagnostic.setTags(new HashMap<>(workerProfile.getAttributes()));
        } else {

View on GitHub (pinned to cf67b549a7)