fsk119 commented on a change in pull request #15197:
URL: https://github.com/apache/flink/pull/15197#discussion_r595776842
##########
File path: flink-table/flink-sql-client/src/test/resources/sql/set.q
##########
@@ -44,6 +44,14 @@ CREATE TABLE hive_table (
# list the configured configuration
set;
+execution.attached=true
+execution.savepoint.ignore-unclaimed-state=false
+execution.shutdown-on-attached-exit=false
+execution.target=remote
+jobmanager.rpc.address=localhost
Review comment:
After digging into the code, the
```
execution.savepoint.ignore-unclaimed-state=false
execution.shutdown-on-attached-exit=false
execution.target=remote
```
is set by the `DefaultCli`.
In `DefaultContext`, it collects avaliable CLI and determine which CLI
should be used. In the test, it just uses `DefaultCli`. Then it will convert
the CLI to the configuration with specified options in deployment entry. In
this way, we put the options list above into the configuration.
For more details, you can refer to the
`DefaultContext#createExecutionConfig`.
The `jobmanager.rpc.address=localhost` is added by MiniCluster. In the test
we create a `MiniClusterResource`. In the
`MiniClusterResource#startMiniCluster`, it create a `MiniCluster` with
configuration. The used configuration that uses the 'localhost' as
`jobmanager.rpc.address`.
Therefore, I think all the options in the list are generated by default,
which is consistent on different machines. For the first part, it is
influenced by YAML Deployment Entry. For the second part, it used the
`localhost` as the default value.
----------------------------------------------------------------
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]
