rkhachatryan commented on a change in pull request #14847:
URL: https://github.com/apache/flink/pull/14847#discussion_r574494137



##########
File path: 
flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/SchedulerBase.java
##########
@@ -908,38 +909,56 @@ public void reportCheckpointMetrics(
         // will be restarted by the CheckpointCoordinatorDeActivator.
         checkpointCoordinator.stopCheckpointScheduler();
 
+        final CompletableFuture<Collection<ExecutionState>> 
executionGraphTerminationFuture =
+                FutureUtils.combineAll(
+                        StreamSupport.stream(
+                                        
executionGraph.getAllExecutionVertices().spliterator(),
+                                        false)
+                                
.map(ExecutionVertex::getCurrentExecutionAttempt)
+                                .map(Execution::getTerminalStateFuture)
+                                .collect(Collectors.toList()));
+
         final CompletableFuture<String> savepointFuture =
                 checkpointCoordinator
                         .triggerSynchronousSavepoint(advanceToEndOfEventTime, 
targetDirectory)
                         .thenApply(CompletedCheckpoint::getExternalPointer);
 
-        final CompletableFuture<JobStatus> terminationFuture =
-                executionGraph
-                        .getTerminationFuture()
-                        .handle(
-                                (jobstatus, throwable) -> {
-                                    if (throwable != null) {
-                                        log.info(
-                                                "Failed during stopping job {} 
with a savepoint. Reason: {}",
-                                                jobGraph.getJobID(),
-                                                throwable.getMessage());
-                                        throw new 
CompletionException(throwable);
-                                    } else if (jobstatus != 
JobStatus.FINISHED) {
-                                        log.info(
-                                                "Failed during stopping job {} 
with a savepoint. Reason: Reached state {} instead of FINISHED.",
-                                                jobGraph.getJobID(),
-                                                jobstatus);
-                                        throw new CompletionException(
-                                                new FlinkException(
-                                                        "Reached state "
-                                                                + jobstatus
-                                                                + " instead of 
FINISHED."));
-                                    }
-                                    return jobstatus;
-                                });
-
         return savepointFuture
-                .thenCompose((path) -> terminationFuture.thenApply((jobStatus 
-> path)))
+                .thenCompose(
+                        path ->
+                                executionGraphTerminationFuture
+                                        .handle(
+                                                (executionStates, throwable) 
-> {
+                                                    Set<ExecutionState> 
nonFinishedStates =
+                                                            
extractNonFinishedStates(
+                                                                    
executionStates);
+                                                    if (throwable != null) {
+                                                        log.info(
+                                                                "Failed during 
stopping job {} with a savepoint. Reason: {}",
+                                                                
jobGraph.getJobID(),
+                                                                
throwable.getMessage());
+                                                        throw new 
CompletionException(throwable);
+                                                    } else if 
(!nonFinishedStates.isEmpty()) {
+                                                        log.info(
+                                                                "Failed while 
stopping job {} after successfully creating a savepoint. A global failover is 
going to be triggered. Reason: One or more states ended up in the following 
termination states instead of FINISHED: {}",
+                                                                
jobGraph.getJobID(),
+                                                                
nonFinishedStates);
+                                                        FlinkException
+                                                                
inconsistentFinalStateException =
+                                                                        new 
FlinkException(
+                                                                               
 String.format(
+                                                                               
         "Inconsistent execution state after stopping with savepoint. A global 
fail-over was triggered to recover the job %s.",
+                                                                               
         jobGraph
+                                                                               
                 .getJobID()));
+                                                        
executionGraph.failGlobal(
+                                                                
inconsistentFinalStateException);

Review comment:
       I'm a bit concerned about thread-safety here: the statement seems 
non-thread-safe but it can be executed by any thread that terminates some 
`Execution`.
   
   Previously, this block didn't access any state.
   
   Should we use `handleAsync` with `mainThreadExecutor` here?

##########
File path: 
flink-runtime/src/main/java/org/apache/flink/runtime/scheduler/SchedulerBase.java
##########
@@ -908,38 +909,56 @@ public void reportCheckpointMetrics(
         // will be restarted by the CheckpointCoordinatorDeActivator.
         checkpointCoordinator.stopCheckpointScheduler();
 
+        final CompletableFuture<Collection<ExecutionState>> 
executionGraphTerminationFuture =
+                FutureUtils.combineAll(
+                        StreamSupport.stream(
+                                        
executionGraph.getAllExecutionVertices().spliterator(),

Review comment:
       Just a question, we could theoretically iterate over regions instead of 
all vertices, here, right?

##########
File path: 
flink-tests/src/test/java/org/apache/flink/test/checkpointing/SavepointITCase.java
##########
@@ -600,6 +616,171 @@ public void testSubmitWithUnknownSavepointPath() throws 
Exception {
         }
     }
 
+    @Test
+    public void testStopWithSavepointFailingInSnapshotCreation() throws 
Exception {
+        testStopWithFailingSourceInOnePipeline(
+                new SnapshotFailingInfiniteTestSource(),
+                folder.newFolder(),
+                // two restarts expected:
+                // 1. task failure restart
+                // 2. job failover triggered by the CheckpointFailureManager
+                2,
+                assertInSnapshotCreationFailure());
+    }
+
+    @Test
+    public void testStopWithSavepointFailingAfterSnapshotCreation() throws 
Exception {

Review comment:
       This test fails with a timeout if I run it in IDE without other tests 
(while others don't; and also if I run it `SavepointITCase` it succeeds).




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
[email protected]


Reply via email to