Help Center/ Data Lake Insight/ FAQs/ Flink Jobs/ Flink Jar Jobs/ Why Does a Flink Jar Package Conflict Result in Job Submission Failure?
Updated on 2024-11-15 GMT+08:00

Why Does a Flink Jar Package Conflict Result in Job Submission Failure?

Symptom

The dependency of your Flink job conflicts with a built-in dependency of the DLI Flink platform. As a result, the job submission fails.

Solution

Check whether there are conflicting JAR files.

DLI Flink provides a series of pre-installed dependency packages in the DLI service to support various data processing and analysis tasks.

If the JAR file you upload contains a package that already exists in the DLI Flink runtime platform, it will prompt a Flink Jar package conflict, resulting in job submission failure.

Delete the repetitive package by referring to the dependency package information provided in the Data Lake Insight User Guide and then upload the package.

For details about DLI built-in dependencies, see Built-in Dependencies.