apache/pulsar · warning · ReplicationException.UnavailableException

Interrupted while contacting metadata store

Error message

Interrupted while contacting metadata store

What it means

Thrown by markLedgerReplicated when the thread waiting on the blocking metadata-store future is interrupted; the InterruptedException cause carries the interruption. The method re-asserts the interrupt flag (Thread.currentThread().interrupt()) before wrapping, so interruption is not swallowed. This is ReplicationException.UnavailableException, i.e. the operation did not complete because the caller stopped waiting.

Source

Thrown at pulsar-metadata/src/main/java/org/apache/pulsar/metadata/bookkeeper/PulsarLedgerUnderreplicationManager.java:485

                    }
                }
            }
        } catch (ExecutionException ee) {
            if (ee.getCause() instanceof MetadataStoreException.NotFoundException) {
                // this is ok
            } else if (ee.getCause() instanceof MetadataStoreException.BadVersionException) {
                // if this is the case, some has marked the ledger
                // for rereplication again. Leave the underreplicated
                // znode in place, so the ledger is checked.
            } else {
                log.error().exception(ee).log("Error deleting underreplicated ledger node");
                throw new ReplicationException.UnavailableException("Error contacting metadata store", ee);
            }
        } catch (TimeoutException ex) {
            throw new ReplicationException.UnavailableException("Error contacting metadata store", ex);
        } catch (InterruptedException ie) {
            Thread.currentThread().interrupt();
            throw new ReplicationException.UnavailableException("Interrupted while contacting metadata store", ie);
        } finally {
            releaseUnderreplicatedLedger(ledgerId);
        }
    }

    /**
     * Get a list of all the underreplicated ledgers which have been
     * marked for rereplication, filtered by the predicate on the replicas list.
     *
     * <p>Replicas list of an underreplicated ledger is the list of the bookies which are part of
     * the ensemble of this ledger and are currently unavailable/down.
     *
     * @param predicate filter to use while listing under replicated ledgers. 'null' if filtering is not required.
     * @return an iterator which returns underreplicated ledgers.
     */
    @Override
    public Iterator<UnderreplicatedLedger> listLedgersToRereplicate(final Predicate<List<String>> predicate) {
        final Queue<String> queue = new LinkedList<>();

View on GitHub (pinned to 820761864e)

Solutions

1. Check whether the broker/auditor was shutting down at the time; if so this is expected and needs no fix.
2. Avoid interrupting the thread running markLedgerReplicated; let in-flight rereplication finish before calling shutdownNow().
3. Re-check the interrupt flag in your caller (it is preserved) and abort gracefully rather than retrying immediately.
4. If interruptions come from your own executor, use awaitTermination with a generous timeout after shutdown() before shutdownNow().

Example fix

// before: swallowing the exception and losing the interrupt flag
catch (ReplicationException.UnavailableException e) { /* ignore */ }

// after: detect interruption and stop the loop
} catch (ReplicationException.UnavailableException e) {
    if (Thread.currentThread().isInterrupted()) {
        log.info("Interrupted during markLedgerReplicated, stopping rereplication loop");
        return;
    }
    throw e;
}

// Java: skip the call when the thread is already interrupted
if (Thread.currentThread().isInterrupted()) {
    return; // don't start blocking metadata-store work while interrupted
}

try {
    manager.markLedgerReplicated(ledgerId);
} catch (ReplicationException.UnavailableException e) {
    if (Thread.currentThread().isInterrupted()) {
        Thread.currentThread().interrupt(); // preserve flag, stop work
        return;
    }
    throw e; // genuine store unavailability
}

When it happens

Trigger: Calling markLedgerReplicated(ledgerId) while the calling thread (typically the auditor's scheduled executor or a shutdown hook) is interrupted, e.g. during broker shutdown, executor.shutdownNow(), or a task cancelled while blocked in .get(BLOCKING_CALL_TIMEOUT) on store.delete/store.get.

Common situations: Broker/auditor shutdown racing with ledger replication completion; BookKeeper client closing its callback executors; tests cancelling futures; thread pools being torn down while rereplication work is in flight.

Related errors

• Interrupted while reading ledgers at path ${path}
• IOException
• Interrupted while connecting metadata store
• Interrupted initializing OAuth2 IdP TLS factory
• Interrupted at fetching schema info for <SchemaUtils.getStri

AI-assisted analysis of apache/pulsar@820761864e (2026-09-06). Data as JSON: /api/errors/93ee5aba79d5f16e. Report an issue: GitHub.