LuciferYang commented on PR #38079: URL: https://github.com/apache/spark/pull/38079#issuecomment-1265055601
https://github.com/LuciferYang/spark/actions/runs/3171106629/jobs/5164233303 <img width="1085" alt="image" src="https://user-images.githubusercontent.com/1475305/193519835-cf39e088-f706-4d19-b041-d431099c8b1b.png";> It seems that my guess is correct. It is effective to manually filter some unnecessary jars from classpath, such as `selenium`, `junit`, `mockito` or others. So I have the following two ideas: 1. Add a new configuration item(`spark.testing.classpath.blacklist`) to specify a keyword list and use it in the current scenario to dynamically specify and filter some unnecessary jars. 2. Try to add unnecessary artifacts to the `excludeDependencies` and ensure the yarn module will not use it. https://github.com/apache/spark/blob/087082254170e46d4162a47fa9691cc298e08afe/project/SparkBuild.scala#L917-L922 @HyukjinKwon @Yikun Do you have any other suggestions？ In addition, I found that the prefix of a large number of jar packages under the classpath is'/home/runner/.cache/coursier/v1/https/maven-central.storage-download.googleapis.com/maven2/, if we can change a short identification , it may also solve this issue. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
![https://user-images.githubusercontent.com/1475305/193519835-cf39e088-f706-4d19-b041-d431099c8b1b.png"](https://user-images.githubusercontent.com/1475305/193519835-cf39e088-f706-4d19-b041-d431099c8b1b.png"){kind=link}