zhuzhurk commented on a change in pull request #10232: [FLINK-14314][runtime]
Allocate shared slot resources respecting the resources of all vertices in the
group
URL: https://github.com/apache/flink/pull/10232#discussion_r348507615
##########
File path:
flink-table/flink-table-planner-blink/src/test/java/org/apache/flink/table/planner/runtime/utils/BatchAbstractTestBase.java
##########
@@ -47,7 +47,7 @@
private static Configuration getConfiguration() {
Configuration config = new Configuration();
- config.setString(TaskManagerOptions.LEGACY_MANAGED_MEMORY_SIZE,
"100m");
+ config.setString(TaskManagerOptions.LEGACY_MANAGED_MEMORY_SIZE,
"200m");
Review comment:
This change is needed because the some table batch tests have a pipelined
region(corresponding to a slot sharing group) with managed memory sum to be
more than 100m, so that these cases will pend at slot allocation.
Previously they passed because it allocates shared slot with resources of
the first arrived task slot request. This, however, make memory OOM possible to
happen.
----------------------------------------------------------------
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
