zhangjun0x01 commented on a change in pull request #2229:
URL: https://github.com/apache/iceberg/pull/2229#discussion_r575039399
##########
File path: flink/src/test/java/org/apache/iceberg/flink/FlinkTestBase.java
##########
@@ -78,7 +79,9 @@ protected TableEnvironment getTableEnv() {
.build();
TableEnvironment env = TableEnvironment.create(settings);
-
env.getConfig().getConfiguration().set(FlinkTableOptions.TABLE_EXEC_ICEBERG_INFER_SOURCE_PARALLELISM,
false);
+ env.getConfig().getConfiguration()
+
.set(FlinkTableOptions.TABLE_EXEC_ICEBERG_INFER_SOURCE_PARALLELISM, false)
+ .set(CoreOptions.DEFAULT_PARALLELISM, 1);
Review comment:
In my test, all test methods in
`org.apache.iceberg.flink.TestFlinkTableSource` that return more than one
record of filter push down will be affected. We need to sort the returned
results to do assertion.
In addition, in
`TestRewriteDataFilesAction#testRewriteLargeTableHasResiduals` ,
([link](https://github.com/apache/iceberg/blob/master/flink/src/test/java/org/apache/iceberg/flink/actions/TestRewriteDataFilesAction.java#L251)),I
have two `INSERT INTO`, but it generated 8 data files. I haven't look up the
reason yet.
I haven't checked whether there are similar problems in other methods. I
think set the parallelism to 1, it is consistent with the original state , it
can avoid unknown and unstable results.
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
