apache/cassandra · warning · RuntimeException

Failed to submit background compaction task

Error message

Failed to submit background compaction task

What it means

When the compaction executor rejects a submitted background compaction task (executor saturated or shutting down), rejectTask calls task.rejected() and then throws this RuntimeException. The exception (including any thrown by rejected()) is recorded via addAsyncError rather than propagating to the caller.

Source

Thrown at src/java/org/apache/cassandra/db/compaction/CompactionManager.java:483

                task.execute(active);
            }
            catch (Throwable t)
            {
                addAsyncError(t);
            }
            finally
            {
                if (toComplete.decrementAndGet() == 0)
                    complete(true, asyncErrors.get());
            }
        }

        private void rejectTask(AbstractCompactionTask task, AtomicInteger toComplete)
        {
            try
            {
                task.rejected();
                throw new RuntimeException("Failed to submit background compaction task");
            }
            catch (Throwable t) // make sure we catch exceptions thrown by task.rejected() as well
            {
                addAsyncError(t);
            }
            finally
            {
                if (toComplete.decrementAndGet() == 0)
                    complete(false, asyncErrors.get());
            }
        }

        private void complete(boolean submitNew, Throwable error)
        {
            compactingCF.remove(cfs);
            if (error == null)
            {
                toSignalWhenDone.setSuccess(null);

View on GitHub (pinned to 88fd0f6a0e)

Solutions

1. Check whether the node is shutting down or autocompaction is disabled (nodetool statusautocompaction); re-enable and let compaction resume.
2. Inspect async error logs for the underlying rejection reason and address executor saturation (increase compaction throughput headroom).
3. Retry compaction after the node stabilizes; rejected tasks are normally resubmitted later.

// before triggering compaction programmatically
if (storageService.isStarting() || !compactionManager.isEnabled()) skipSubmission();

try { scheduleCompaction(); } catch (RuntimeException e) { if (e.getMessage().equals("Failed to submit background compaction task")) { retryAfterStabilization(); } }

When it happens

Trigger: Background compaction submission to the executor's queue is rejected — typically because the node is shutting down, the executor was paused/disabled, or the task queue is full.

Common situations: Node shutdown or decommission racing with compaction submission, disabling autocompaction while tasks are in flight, severe executor backlog on overloaded nodes.

Related errors

• ${executor.name} not terminated
• Unable to terminate fast tasks within 1 minute.
• UnsupportedOperationException
• Cannot resize core pool size of SEPExecutor
• Maximum number of workers must not be negative

AI-assisted analysis of apache/cassandra@88fd0f6a0e (2026-09-10). Data as JSON: /api/errors/3b8318ebe21cae8d. Report an issue: GitHub.