tillrohrmann commented on a change in pull request #13592:
URL: https://github.com/apache/flink/pull/13592#discussion_r505257420
##########
File path:
flink-yarn/src/main/java/org/apache/flink/yarn/YarnResourceManagerDriver.java
##########
@@ -228,21 +221,27 @@ public void deregisterApplication(ApplicationStatus
finalStatus, @Nullable Strin
@Override
public CompletableFuture<YarnWorkerNode>
requestResource(TaskExecutorProcessSpec taskExecutorProcessSpec) {
- final Optional<Resource> containerResourceOptional =
getContainerResource(taskExecutorProcessSpec);
+ checkInitialized();
+
final CompletableFuture<YarnWorkerNode> requestResourceFuture =
new CompletableFuture<>();
- if (containerResourceOptional.isPresent()) {
-
resourceManagerClient.addContainerRequest(getContainerRequest(containerResourceOptional.get()));
+ final
Optional<TaskExecutorProcessSpecContainerResourcePriorityAdapter.PriorityAndResource>
priorityAndResourceOpt =
+
taskExecutorProcessSpecContainerResourcePriorityAdapter.getPriorityAndResource(taskExecutorProcessSpec);
+
+ if (!priorityAndResourceOpt.isPresent()) {
+ requestResourceFuture.completeExceptionally(
+ new
ResourceManagerException(String.format("Could not compute the container
Resource from the given TaskExecutorProcessSpec %s.",
taskExecutorProcessSpec)));
Review comment:
True the logs exist. However, it might be not that easy to correlate
these logging statements with the logging of the `ResourceManagerException` at
a different place. If there is currently only one reason why the operation
fails, maybe we can state it in the exception 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.
For queries about this service, please contact Infrastructure at:
[email protected]
