GitHub user liuyq2step created a discussion: StreamPark UI Requires Manual 
Savepoint Path Input Even Though state.savepoints.dir is Configured

I have configured both state.savepoints.dir and 
execution.checkpointing.savepoint-dir in my Flink configuration, and the 
JobManager logs confirm these configurations are loaded successfully.  ps : 
Checkpoints are working normally.

However, when manually triggering a savepoint through the StreamPark Web UI, I 
still must manually fill in a savepoint path. The UI shows the path field as 
"optional," but leaving it empty results in an error. If I provide an HDFS path 
(e.g., hdfs://uat-cluster/bg/flink-savepoints), the savepoint is created 
successfully with the _metadata file generated.

This suggests that while the underlying Flink configuration is correct, the 
StreamPark UI might not be correctly utilizing the configured default savepoint 
directory when the path field is left empty, or there might be a bug in how the 
optional field is handled.

StreamPark Version: 2.1.6
Flink Version: 1.20.2
Execution Mode: yarn-application

GitHub link: https://github.com/apache/streampark/discussions/4307

----
This is an automatically sent email for [email protected].
To unsubscribe, please send an email to: [email protected]

Reply via email to