smtwilio commented on code in PR #19811:
URL: https://github.com/apache/hudi/pull/19811#discussion_r3964375148
##########
hudi-utilities/src/main/java/org/apache/hudi/utilities/streamer/HoodieMultiTableStreamer.java:
##########
@@ -461,28 +484,204 @@ private static String resetTarget(Config configuration,
String database, String
/**
* Creates actual HoodieDeltaStreamer objects for every table/topic and does
incremental sync.
+ *
+ * <p>In continuous mode each table's sync blocks until it is shut down, so
the tables are synced concurrently.
+ * Otherwise the tables are synced sequentially, one after another.
*/
public void sync() {
+ try {
+ if (continuousMode) {
+ syncContinuously();
+ } else {
+ syncSequentially();
+ }
+ } finally {
+ log.info("Ingestion was successful for topics: {}", successTables);
+ if (!failedTables.isEmpty()) {
+ log.error("Ingestion failed for topics: {}", failedTables);
+ }
+ }
+ }
+
+ private void syncSequentially() {
for (TableExecutionContext context : tableExecutionContexts) {
+ String table = Helpers.getTableWithDatabase(context);
HoodieStreamer streamer = null;
try {
streamer = new HoodieStreamer(context.getConfig(), jssc,
Option.ofNullable(context.getProperties()));
streamer.sync();
- successTables.add(Helpers.getTableWithDatabase(context));
- streamer.shutdownGracefully();
+ successTables.add(table);
} catch (Exception e) {
- log.error("error while running MultiTableDeltaStreamer for table: {}",
context.getTableName(), e);
- failedTables.add(Helpers.getTableWithDatabase(context));
+ log.error("error while running MultiTableDeltaStreamer for table: {}",
table, e);
+ failedTables.add(table);
} finally {
if (streamer != null) {
- streamer.shutdownGracefully();
+ shutdownQuietly(streamer, table);
}
}
}
+ }
+
+ /**
+ * Syncs all tables concurrently, one thread per table. Used for continuous
mode where each table's sync blocks
+ * indefinitely.
+ *
+ * <p>When {@code --fail-fast-on-continuous} is enabled, the first table
failure fails the whole job. The sibling
+ * streamers are shut down and a {@link HoodieException} is thrown so the
caller can exit with a non-zero status.
+ * Otherwise, every table is synced independently and a single failure does
not affect the others.
+ */
+ private void syncContinuously() {
+ if (tableExecutionContexts.isEmpty()) {
+ return;
+ }
+ // Streamer instances are registered from worker threads, so a thread-safe
list is required.
+ final List<HoodieStreamer> streamerInstances = new
CopyOnWriteArrayList<>();
+ // Set once fail fast trips, so tasks that register their streamer
afterwards stop before starting the sync.
+ final AtomicBoolean shutdownRequested = new AtomicBoolean(false);
+ final ExecutorService executor =
Executors.newFixedThreadPool(tableExecutionContexts.size(),
+ new CustomizedThreadFactory("multi-table-streamer", true));
+ boolean terminated = false;
+ try {
+ final List<CompletableFuture<Void>> tableFutures =
tableExecutionContexts.stream()
+ .map(context -> CompletableFuture.runAsync(
+ () -> runTableSync(context, streamerInstances,
shutdownRequested), executor))
+ .collect(Collectors.toList());
+
+ if (failFastOnContinuousMode) {
+ log.info("Fail fast enabled in continuous mode. The whole job fails on
any single table failure");
+ awaitFailFast(tableFutures);
+ } else {
+ CompletableFuture.allOf(tableFutures.toArray(new
CompletableFuture[0])).join();
+ }
+ } finally {
+ // On an abnormal exit the siblings are still ingesting, since
FutureUtils.allOf only cancels their futures.
+ // Stopping them here rather than in a catch covers every such exit,
including an Error, which the workers do
+ // not catch; it is a no-op on the success path because each table has
already shut its ingestion service down.
+ shutdownRequested.set(true);
+ shutdownStreamers(streamerInstances);
+ // Wait for every worker thread to finish (including its finally
cleanup) before returning, so sync() does not
+ // return while a table is still writing and main() then stops the
shared Spark context under it.
+ terminated = shutdownExecutor(executor);
+ }
+ // If the workers never terminated, ingestion may still be running. Fail
loudly instead of returning as if the
+ // cleanup succeeded, so the caller does not silently proceed to Spark
teardown with live writers.
+ if (!terminated) {
+ throw new HoodieException("Timed out shutting down table ingestion
workers in continuous mode");
+ }
+ }
+
+ /**
+ * Syncs one table on the calling worker thread. Rethrows only under fail
fast; otherwise the failure is recorded
+ * in {@link #failedTables} and the sibling tables carry on.
+ */
+ private void runTableSync(TableExecutionContext context,
List<HoodieStreamer> streamerInstances, AtomicBoolean shutdownRequested) {
+ String table = Helpers.getTableWithDatabase(context);
+ // The tables now log concurrently into one driver log, so name the worker
after the table it is syncing.
+ Thread.currentThread().setName("multi-table-streamer-" + table);
+ HoodieStreamer streamer = null;
+ try {
+ streamer = new HoodieStreamer(context.getConfig(), jssc,
Option.ofNullable(context.getProperties()));
+ streamerInstances.add(streamer);
+ // Register before checking the flag so a concurrent shutdownStreamers()
always sees this streamer.
+ if (shutdownRequested.get()) {
+ return;
+ }
+ streamer.sync();
+ // A streamer registered just before fail fast tripped can reach here
without ever ingesting.
+ // shutdown() call will be a no-op because its ingestion service hadn't
started yet.
+ // Don't count that as a success.
+ if (!shutdownRequested.get()) {
+ successTables.add(table);
+ }
+ } catch (Exception e) {
+ log.error("error while running MultiTableDeltaStreamer for table: {}",
table, e);
+ failedTables.add(table);
+ if (failFastOnContinuousMode) {
+ // Name the table so the thrown exception identifies the culprit, not
the siblings torn down after it.
+ throw new HoodieException("Table sync failed in continuous mode for
table: " + table, e);
+ }
+ } finally {
+ if (streamer != null) {
+ shutdownQuietly(streamer, table);
+ }
+ }
+ }
+
+ /**
+ * Waits until either every table sync finishes successfully or the first
one fails. On the first failure, the
+ * remaining streamers are shut down and a {@link HoodieException} is
thrown. {@link FutureUtils#allOf} only trips
+ * on an <em>exceptional</em> completion, so a table that terminates
normally (e.g. via a
+ * {@link PostWriteTerminationStrategy}) does not abort its siblings.
+ */
+ private static void awaitFailFast(List<CompletableFuture<Void>>
tableFutures) {
+ try {
+ FutureUtils.allOf(tableFutures).join();
+ } catch (CompletionException e) {
+ Throwable cause = unwrapCompletionException(e);
+ // An Error is rethrown as is rather than boxed, so the JVM-level
failure reaches the caller unchanged.
+ if (cause instanceof Error) {
+ throw (Error) cause;
+ }
+ log.error("error while running MultiTableDeltaStreamer, shutting down
remaining tables as fail fast is enabled", cause);
+ throw new HoodieException("Fail fast is enabled and a table sync failed
in continuous mode.", cause);
+ }
+ }
+
+ /**
+ * Releases a streamer's resources, logging rather than propagating a
failure to do so. Closing can throw, and this
+ * runs in a {@code finally} on the failure path where escaping would mask
the table failure and, in
+ * {@link #syncSequentially()}, abort the tables not synced yet.
+ */
+ private static void shutdownQuietly(HoodieStreamer streamer, String table) {
+ try {
+ streamer.shutdownGracefully();
+ } catch (Exception e) {
+ log.warn("error while shutting down the streamer for table: {}", table,
e);
+ }
+ }
+
+ // A worker failure reaches the waiter wrapped in CompletionException, and
FutureUtils.allOf re-wraps it, so the
+ // real cause can sit under more than one layer.
+ private static Throwable unwrapCompletionException(CompletionException e) {
+ Throwable cause = e;
+ while (cause instanceof CompletionException && cause.getCause() != null) {
+ cause = cause.getCause();
+ }
+ return cause;
+ }
+
+ /**
+ * Two-phase shutdown of the per-table executor: wait for the running syncs
to finish, then force-cancel any that
+ * ignore interruption. Bounded by {@link
Constants#SHUTDOWN_TIMEOUT_SECONDS} so a stuck table cannot hang the job.
+ *
+ * @return true if all workers terminated, false if any were still running
when the timeout elapsed.
+ */
+ private static boolean shutdownExecutor(ExecutorService executor) {
+ executor.shutdown();
+ try {
+ if (executor.awaitTermination(Constants.SHUTDOWN_TIMEOUT_SECONDS,
TimeUnit.SECONDS)) {
+ return true;
+ }
+ executor.shutdownNow();
+ if (executor.awaitTermination(Constants.SHUTDOWN_TIMEOUT_SECONDS,
TimeUnit.SECONDS)) {
+ return true;
+ }
+ log.error("executor service did not terminate after shutdown");
+ return false;
+ } catch (InterruptedException e) {
+ executor.shutdownNow();
+ Thread.currentThread().interrupt();
+ return false;
+ }
+ }
- log.info("Ingestion was successful for topics: {}", successTables);
- if (!failedTables.isEmpty()) {
- log.info("Ingestion failed for topics: {}", failedTables);
+ private static void shutdownStreamers(List<HoodieStreamer>
streamerInstances) {
Review Comment:
Right on both counts, I left the wording behind. Renamed to
`interruptAllIngestion` and updated the message as well.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
