isunjin commented on a change in pull request #6684: [FLINK-10205] Batch
Job: InputSplit Fault tolerant for DataSource…
URL: https://github.com/apache/flink/pull/6684#discussion_r221389763
##########
File path:
flink-runtime/src/test/java/org/apache/flink/runtime/jobmaster/JobMasterTest.java
##########
@@ -704,6 +716,66 @@ public void
testResourceManagerConnectionAfterRegainingLeadership() throws Excep
}
}
+ private JobGraph createDataSourceJobGraph() {
+ final TextInputFormat inputFormat = new TextInputFormat(new
Path("."));
+ final InputFormatVertex producer = new
InputFormatVertex("Producer");
+ new TaskConfig(producer.getConfiguration()).setStubWrapper(new
UserCodeObjectWrapper<InputFormat<?, ?>>(inputFormat));
+ producer.setInvokableClass(DataSourceTask.class);
+
+ final JobVertex consumer = new JobVertex("Consumer");
+ consumer.setInvokableClass(NoOpInvokable.class);
+ consumer.connectNewDataSetAsInput(producer,
DistributionPattern.POINTWISE, ResultPartitionType.BLOCKING);
+
+ final JobGraph jobGraph = new JobGraph(producer, consumer);
+ jobGraph.setAllowQueuedScheduling(true);
+
+ return jobGraph;
+ }
+
+ /**
+ * Tests the {@link JobMaster#requestNextInputSplit(JobVertexID,
ExecutionAttemptID)}
+ * validate that it will get same result for a different retry
+ */
+ @Test
+ public void testRequestNextInputSplitWithDataSourceFailover() throws
Exception {
+
+ final JobGraph dataSourceJobGraph = createDataSourceJobGraph();
+ testJobMasterAPIWithMockExecution(dataSourceJobGraph, (tdd,
jobMaster) ->{
+ try{
+ final JobMasterGateway gateway =
jobMaster.getSelfGateway(JobMasterGateway.class);
+
+ final TaskInformation taskInformation =
tdd.getSerializedTaskInformation()
+
.deserializeValue(getClass().getClassLoader());
+ JobVertexID vertexID =
taskInformation.getJobVertexId();
+
+ //get the previous split
+ SerializedInputSplit split1 =
gateway.requestNextInputSplit(vertexID, tdd.getExecutionAttemptId()).get();
+
+ //start a new version of this execution
+ ExecutionGraph executionGraph =
jobMaster.getExecutionGraph();
+ Execution execution =
executionGraph.getRegisteredExecutions().get(tdd.getExecutionAttemptId());
+ ExecutionVertex executionVertex =
execution.getVertex();
+
+ long version = execution.getGlobalModVersion();
+ gateway.updateTaskExecutionState(new
TaskExecutionState(dataSourceJobGraph.getJobID(), tdd.getExecutionAttemptId(),
ExecutionState.FINISHED)).get();
+ Execution newExecution =
executionVertex.resetForNewExecution(System.currentTimeMillis(), version);
+
+ //get the new split
+ SerializedInputSplit split2 =
gateway.requestNextInputSplit(vertexID, newExecution.getAttemptId()).get();
+
+
Assert.assertArrayEquals(split1.getInputSplitData(),
split2.getInputSplitData());
+
+ //get the new split3
Review comment:
add more assert
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
