Github user KurtYoung commented on a diff in the pull request:
https://github.com/apache/flink/pull/2571#discussion_r82134943
--- Diff:
flink-runtime/src/main/java/org/apache/flink/runtime/resourcemanager/ResourceManager.java
---
@@ -245,22 +255,26 @@ public TaskExecutorGateway call() throws Exception {
resourceID,
taskExecutorAddress, leaderSessionID, resourceManagerLeaderId);
throw new Exception("Invalid leader
session id");
}
- return
getRpcService().connect(taskExecutorAddress, TaskExecutorGateway.class).get(5,
TimeUnit.SECONDS);
+ return
getRpcService().connect(taskExecutorAddress, TaskExecutorGateway.class)
+ .get(timeout.toMilliseconds(),
timeout.getUnit());
}
}).handleAsync(new BiFunction<TaskExecutorGateway, Throwable,
RegistrationResponse>() {
@Override
public RegistrationResponse apply(TaskExecutorGateway
taskExecutorGateway, Throwable throwable) {
if (throwable != null) {
return new
RegistrationResponse.Decline(throwable.getMessage());
} else {
- WorkerType oldWorker =
taskExecutorGateways.remove(resourceID);
- if (oldWorker != null) {
+ WorkerRegistration oldRegistration =
taskExecutors.remove(resourceID);
+ if (oldRegistration != null) {
// TODO :: suggest old
taskExecutor to stop itself
slotManager.notifyTaskManagerFailure(resourceID);
--- End diff --
notifyTaskManagerFailure is called again in
slotManager.registerTaskExecutor, maybe we should only keep one of these
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
