xintongsong commented on a change in pull request #9105:
[FLINK-13241][Yarn/Mesos] Fix Yarn/MesosResourceManager setting managed memory
size into wrong configuration instance.
URL: https://github.com/apache/flink/pull/9105#discussion_r305173128
##########
File path:
flink-runtime/src/main/java/org/apache/flink/runtime/resourcemanager/ResourceManager.java
##########
@@ -185,6 +199,10 @@ public ResourceManager(
this.jmResourceIdRegistrations = new HashMap<>(4);
this.taskExecutors = new HashMap<>(8);
this.taskExecutorGatewayFutures = new HashMap<>(8);
+
+ this.defaultTaskManagerMemoryMB =
ConfigurationUtils.getTaskManagerHeapMemory(flinkConfig).getMebiBytes();
+ this.numberOfTaskSlots =
flinkConfig.getInteger(TaskManagerOptions.NUM_TASK_SLOTS);
+ this.slotsPerWorker =
updateTaskManagerConfigAndCreateWorkerSlotProfiles(this.flinkConfig,
defaultTaskManagerMemoryMB, numberOfTaskSlots);
Review comment:
Yes, the background of this calculating `slotsPerWorker` on RM is that, we
need to know the resource profiles of slots for `PendingTaskManagerSlot` before
TM is started and registered. For standalone, we don't have any pending task
manager slot because RM cannot actively start any TM.
----------------------------------------------------------------
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]
With regards,
Apache Git Services
