[FLINK-37110][cdc-runtime] Fixed the problem of using the same executor to process schema change requests and flush event requests, resulting in blocking timeout.

Author: linjc13

flink-cdc-runtime/src/main/java/org/apache/flink/cdc/runtime/operators/schema/common/SchemaRegistry.java

@@ -52,6 +52,7 @@
 import java.util.concurrent.CompletableFuture;
 import java.util.concurrent.ConcurrentHashMap;
 import java.util.concurrent.ExecutorService;
+import java.util.concurrent.Executors;
 import java.util.concurrent.TimeoutException;
 import java.util.function.BooleanSupplier;
 
@@ -96,6 +97,9 @@ public abstract class SchemaRegistry implements OperatorCoordinator, Coordinatio
     protected transient SchemaManager schemaManager;
     protected transient TableIdRouter router;
 
+    /** Executor service to execute handle event from operator. */
+    private final ExecutorService runInEventFromOperatorExecutor;
+
     protected SchemaRegistry(
             OperatorCoordinator.Context context,
             String operatorName,
@@ -111,6 +115,7 @@ protected SchemaRegistry(
         this.routingRules = routingRules;
         this.rpcTimeout = rpcTimeout;
         this.behavior = schemaChangeBehavior;
+        this.runInEventFromOperatorExecutor = Executors.newSingleThreadExecutor();
     }
 
     // ---------------
@@ -130,6 +135,7 @@ public void start() throws Exception {
     public void close() throws Exception {
         LOG.info("Closing SchemaRegistry - {}.", operatorName);
         coordinatorExecutor.shutdown();
+        runInEventFromOperatorExecutor.shutdown();
     }
 
     // ------------------------------
@@ -235,6 +241,7 @@ public final CompletableFuture<CoordinationResponse> handleCoordinationRequest(
             CoordinationRequest request) {
         CompletableFuture<CoordinationResponse> future = new CompletableFuture<>();
         runInEventLoop(
+                coordinatorExecutor,
                 () -> {
                     if (request instanceof GetEvolvedSchemaRequest) {
                         handleGetEvolvedSchemaRequest((GetEvolvedSchemaRequest) request, future);
@@ -253,6 +260,7 @@ public final CompletableFuture<CoordinationResponse> handleCoordinationRequest(
     public final void handleEventFromOperator(
             int subTaskId, int attemptNumber, OperatorEvent event) {
         runInEventLoop(
+                runInEventFromOperatorExecutor,
                 () -> {
                     if (event instanceof FlushSuccessEvent) {
                         handleFlushSuccessEvent((FlushSuccessEvent) event);
@@ -297,7 +305,11 @@ public final void executionAttemptReady(
     public final void checkpointCoordinator(
             long checkpointId, CompletableFuture<byte[]> completableFuture) throws Exception {
         LOG.info("Going to start checkpoint No.{}", checkpointId);
-        runInEventLoop(() -> snapshot(completableFuture), "Taking checkpoint - %d", checkpointId);
+        runInEventLoop(
+                coordinatorExecutor,
+                () -> snapshot(completableFuture),
+                "Taking checkpoint - %d",
+                checkpointId);
     }
 
     @Override
@@ -325,6 +337,7 @@ public final void resetToCheckpoint(long checkpointId, @Nullable byte[] checkpoi
      * directly, make sure you're running heavy logics inside, or the entire job might hang!
      */
     protected void runInEventLoop(
+            final ExecutorService coordinatorExecutor,
             final ThrowingRunnable<Throwable> action,
             final String actionName,
             final Object... actionNameFormatParameters) {