apache/iceberg · error · LockException

Interrupted while creating lock on table %s.%s

Error message

Interrupted while creating lock on table %s.%s

What it means

MetastoreLock uses Tasks.foreach with retries to run the create-lock loop; if the thread is interrupted while (re)issuing the lock request (checkLock/wait) call, the code restores the interrupt flag and throws this LockException wrapping the InterruptedException. Like error 2123, it means lock creation was aborted by interruption, not by metastore state.

Source

Thrown at hive-metastore/src/main/java/org/apache/iceberg/hive/MetastoreLock.java:344

                  Thread.currentThread().interrupt();
                  interrupted.set(true);
                  LOG.warn(
                      "Interrupted while trying to find lock for table {}.{}",
                      databaseName,
                      tableName,
                      e);
                  throw new LockException(
                      e,
                      "Interrupted while trying to find lock for table %s.%s",
                      databaseName,
                      tableName);
                }
              } catch (InterruptedException e) {
                Thread.currentThread().interrupt();
                interrupted.set(true);
                LOG.warn(
                    "Interrupted while creating lock on table {}.{}", databaseName, tableName, e);
                throw new LockException(
                    e, "Interrupted while creating lock on table %s.%s", databaseName, tableName);
              }
            },
            LockException.class);

    // This should be initialized always, or exception should be thrown.
    LOG.debug("Lock {} created for table {}.{}", lockInfo, databaseName, tableName);
    return lockInfo;
  }

  /**
   * Search for the locks using HMSClient.showLocks identified by the agentInfo. If the lock is
   * there, then a {@link LockInfo} object is returned. If the lock is not found <code>null</code>
   * is returned.
   *
   * @return The {@link LockInfo} for the found lock, or <code>null</code> if nothing found
   */
  private LockInfo findLock() throws LockException, InterruptedException {

View on GitHub (pinned to 86d9c8fc54)

Solutions

1. Eliminate the interruption source (don't shutdownNow while commits are in flight; drain first).
2. Reduce lock contention so the create-lock wait is short and less exposed to cancellation.
3. Retry the commit after the interrupt clears; note the thread's interrupt flag is re-set by this code.
4. Wrap commit calls in a retry policy that tolerates LockException caused by interruption during shutdown only if the commit is safe to re-run (Iceberg commits are).

Example fix

// before
future.cancel(true); // interrupts commit thread mid-lock
// after
future.cancel(false); // or wait for commit completion before cancelling

if (Thread.currentThread().isInterrupted()) {
  throw new CancellationException("Skipping commit; thread interrupted");
}

try {
  table.commit(apply);
} catch (LockException e) {
  if (e.getCause() instanceof InterruptedException) {
    Thread.currentThread().interrupt();
    throw new CancellationException("Commit interrupted during lock creation");
  }
  throw e;
}

When it happens

Trigger: Calling commit on a Hive-catalog table when the thread is interrupted inside the metastore checkLock heartbeat/wait RPC within the Tasks retry loop — executor shutdown, query cancellation, or Future.cancel(true).

Common situations: Long lock waits (contended table) being cancelled mid-wait; Spark/Flink job cancellation interrupting the committing task; JVM shutdown hooks interrupting background commit threads.

Related errors

• Interrupted while trying to find lock for table %s.%s
• Interrupted during commit
• Hive lock heartbeat thread not active
• Interrupted finding locks to unlock {}.{}
• Interrupted in SQL query

AI-assisted analysis of apache/iceberg@86d9c8fc54 (2026-09-12). Data as JSON: /api/errors/715657d16194722c. Report an issue: GitHub.