Ngone51 opened a new pull request #28257: [SPARK-31485][CORE] Avoid application hang if only partial barrier tasks launched URL: https://github.com/apache/spark/pull/28257 <!-- Thanks for sending a pull request! Here are some tips for you: 1. If this is your first time, please read our contributor guidelines: https://spark.apache.org/contributing.html 2. Ensure you have added or run the appropriate tests for your PR: https://spark.apache.org/developer-tools.html 3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][SPARK-XXXX] Your PR title ...'. 4. Be sure to keep the PR description updated to reflect all changes. 5. Please write your PR title to summarize what this PR proposes. 6. If possible, provide a concise example to reproduce the issue for a faster review. 7. If you want to add a new configuration, please read the guideline first for naming configurations in 'core/src/main/scala/org/apache/spark/internal/config/ConfigEntry.scala'. --> ### What changes were proposed in this pull request? <!-- Please clarify what changes you are proposing. The purpose of this section is to outline the changes and how this PR fixes the issue. If possible, please consider writing useful notes for better and faster reviews in your PR. See the examples below. 1. If you refactor some codes with changing classes, showing the class hierarchy will help reviewers. 2. If you fix some SQL features, you can provide some references of other DBMSes. 3. If there is design documentation, please add the link. 4. If there is a discussion in the mailing list, please add the link. --> Use `dagScheduler.taskSetFailed` to abort a barrier stage instead of throwing exception within `resourceOffers`. ### Why are the changes needed? <!-- Please clarify why the changes are needed. For instance, 1. If you propose a new API, clarify the use case for a new API. 2. If you fix a bug, you can clarify why it is a bug. --> Any non fatal exception thrown within Spark RPC framework can be swallowed: https://github.com/apache/spark/blob/100fc58da54e026cda87832a10e2d06eaeccdf87/core/src/main/scala/org/apache/spark/rpc/netty/Inbox.scala#L202-L211 The method `TaskSchedulerImpl.resourceOffers` is also within the scope of Spark RPC framework. Thus, throw exception inside `resourceOffers` won't fail the application. As a result, if a barrier stage fail the require check at `require(addressesWithDescs.size == taskSet.numTasks, ...)`, the barrier stage will fail the check again and again util all tasks from `TaskSetManager` dequeued. But since the barrier stage isn't really executed, the application will hang. The issue can be reproduced by the following test: ```scala initLocalClusterSparkContext(2) val rdd0 = sc.parallelize(Seq(0, 1, 2, 3), 2) val dep = new OneToOneDependency[Int](rdd0) val rdd = new MyRDD(sc, 2, List(dep), Seq(Seq("executor_h_0"),Seq("executor_h_0"))) rdd.barrier().mapPartitions { iter => BarrierTaskContext.get().barrier() iter }.collect() ``` ### Does this PR introduce any user-facing change? <!-- If yes, please clarify the previous behavior and the change this PR proposes - provide the console output, description and/or an example to show the behavior difference if possible. If no, write 'No'. --> Yes, application hang previously but fail-fast after this fix. ### How was this patch tested? <!-- If tests were added, say they were added here. Please make sure to add some test cases that check the changes thoroughly including negative and positive cases if possible. If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future. If tests were not added, please describe why they were not added and/or why it was difficult to add. --> Added a regression test.
---------------------------------------------------------------- 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] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
