无法使用 Spark.Net UDF 和 HDInsight 群集
Cannot use Spark.Net UDFs and HDInsight cluster
我已经尝试 运行 prod env 中包含来自 https://github.com/dotnet/spark/blob/master/examples/Microsoft.Spark.CSharp.Examples/Sql/Batch/Basic.cs 的代码的简单应用程序
应用程序 运行 运行良好并向标准输出发出输出，直到此代码在遇到第一个 UDF 时崩溃。
感谢您就此分享任何见解。
环境。
代码打包使用
dotnet publish -c Release -f netcoreapp3.1 -r ubuntu.16.04-x64
HDInsight 群集 HDI 4.0、Spark 2.4
-- 使用 https://docs.microsoft.com/en-us/dotnet/spark/tutorials/hdinsight-deployment
中的指南设置服务器
spark-submit --master yarn --conf spark.yarn.appMasterEnv.DOTNET_ASSEMBLY_SEARCH_PATHS="./app/publish.zip" --archives wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/publish.zip#mySparkApp --class org.apache.spark.deploy.dotnet.DotnetRunner wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/microsoft-spark-2.4.x-0.12.1.jar wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/publish.zip mySparkApp
（以及绝望中的各种变体，--deploy-mode cluster，各种路径等等，什么都不管用）
标准输出：
...
+---+-----+
|age| name|
+---+-----+
| 22|Ricky|
| 36| Jeff|
| 62|Geddy|
+---+-----+
[2020-10-28T09:15:10.1478641Z] [wn0-hdinsi] [Error] [JvmBridge] JVM method execution failed: Nonstatic method 'showString' failed for class '41' when called with 3 arguments ([Index=1, Type=Int32, Value=20], [Index=2, Type=Int32, Value=20], [Index=3, Type=Boolean, Value=False], )
[2020-10-28T09:15:10.1480587Z] [wn0-hdinsi] [Error] [JvmBridge] org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 16.0 failed 4 times, most recent failure: Lost task 0.3 in stage 16.0 (TID 210, wn0-hdinsi.xwccrqijnmqujdjghwrza0nzbb.fx.internal.cloudapp.net, executor 2): org.apache.spark.api.python.PythonException: System.NullReferenceException: Object reference not set to an instance of an object.
at Microsoft.Spark.Utils.UdfSerDe.<>c.b__10_0(TypeData td) in //src/csharp/Microsoft.Spark/Utils/UdfSerDe.cs:line 262
at System.Collections.Concurrent.ConcurrentDictionary2.GetOrAdd(TKey key, Func2 valueFactory)
at Microsoft.Spark.Utils.UdfSerDe.DeserializeType(TypeData typeData) in //src/csharp/Microsoft.Spark/Utils/UdfSerDe.cs:line 258
at Microsoft.Spark.Utils.UdfSerDe.Deserialize(UdfData udfData) in //src/csharp/Microsoft.Spark/Utils/UdfSerDe.cs:line 160
at Microsoft.Spark.Utils.CommandSerDe.DeserializeUdfs[T](UdfWrapperData data, Int32& nodeIndex, Int32& udfIndex) in //src/csharp/Microsoft.Spark/Utils/CommandSerDe.cs:line 333
at Microsoft.Spark.Utils.CommandSerDe.Deserialize[T](Stream stream, SerializedMode& serializerMode, SerializedMode& deserializerMode, String& runMode) in /_/src/csharp/Microsoft.Spark/Utils/CommandSerDe.cs:line 306
at Microsoft.Spark.Worker.Processor.CommandProcessor.ReadSqlCommands(PythonEvalType evalType, Stream stream) in D:\a\s\src\csharp\Microsoft.Spark.Worker\Processor\CommandProcessor.cs:line 188
at Microsoft.Spark.Worker.Processor.CommandProcessor.ReadSqlCommands(PythonEvalType evalType, Stream stream, Version version) in D:\a\s\src\csharp\Microsoft.Spark.Worker\Processor\CommandProcessor.cs:line 98
at Microsoft.Spark.Worker.Processor.CommandProcessor.Process(Stream stream) in D:\a\s\src\csharp\Microsoft.Spark.Worker\Processor\CommandProcessor.cs:line 43
at Microsoft.Spark.Worker.Processor.PayloadProcessor.Process(Stream stream) in D:\a\s\src\csharp\Microsoft.Spark.Worker\Processor\PayloadProcessor.cs:line 82
at Microsoft.Spark.Worker.TaskRunner.ProcessStream(Stream inputStream, Stream outputStream, Version version, Boolean& readComplete) in D:\a\s\src\csharp\Microsoft.Spark.Worker\TaskRunner.cs:line 143
at org.apache.spark.api.python.BasePythonRunner$ReaderIterator.handlePythonException(PythonRunner.scala:456)
at org.apache.spark.sql.execution.python.PythonUDFRunner$$anon.read(PythonUDFRunner.scala:81)
at org.apache.spark.sql.execution.python.PythonUDFRunner$$anon.read(PythonUDFRunner.scala:64)
at org.apache.spark.api.python.BasePythonRunner$ReaderIterator.hasNext(PythonRunner.scala:410)
at org.apache.spark.InterruptibleIterator.hasNext(InterruptibleIterator.scala:37)
at scala.collection.Iterator$$anon.hasNext(Iterator.scala:440)
at scala.collection.Iterator$$anon.hasNext(Iterator.scala:409)
at scala.collection.Iterator$$anon.hasNext(Iterator.scala:409)
at org.apache.spark.sql.catalyst.expressions.GeneratedClass$GeneratedIteratorForCodegenStage2.processNext(Unknown Source)
at org.apache.spark.sql.execution.BufferedRowIterator.hasNext(BufferedRowIterator.java:43)
at org.apache.spark.sql.execution.WholeStageCodegenExec$$anonfun$$anon.hasNext(WholeStageCodegenExec.scala:636)
at org.apache.spark.sql.execution.SparkPlan$$anonfun.apply(SparkPlan.scala:255)
at org.apache.spark.sql.execution.SparkPlan$$anonfun.apply(SparkPlan.scala:247)
at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$$anonfun$apply.apply(RDD.scala:836)
at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$$anonfun$apply.apply(RDD.scala:836)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:324)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:288)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:324)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:288)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90)
at org.apache.spark.scheduler.Task.run(Task.scala:123)
at org.apache.spark.executor.Executor$TaskRunner$$anonfun.apply(Executor.scala:408)
at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:414)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
------切--------
我的问题确实是路径问题。对于遇到同样问题的任何其他人，我通过将带有 UDF 的 dll（可以是与通用 spark 应用程序相同的 dll）必须列在“--files”中来使它工作。所以本质上你需要一个带有程序集的 zip 文件，然后直接链接到 dll。可能有更聪明的方法，但这对我有用（当 运行ning 在集群模式下）：
spark-submit --deploy-mode cluster --master yarn --files wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/mySparkApp.dll --class org.apache.spark.deploy.dotnet.DotnetRunner wasbs:/ /xxx@yyy.blob.core.windows.net/SparkJobs/microsoft-spark-2.4.x-0.12.1.jar wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/publish.zip mySparkApp
错误是因为找不到你代码的dll
两件事，首先是纱线模式。在 DOTNET_ASSEMBLY_SEARCH_PATHS 的开头导致用户主目录被添加到路径之前，所以它不是 currentdirectory/app/publish.zip 所以如果它不同那么它会在错误的地方查找。
其次确保 publish.zip 不包含文件夹，并且带有 udf 的 dll 位于 zip 的顶层。
我不会将 zip 放在应用程序文件夹中，而是使用当前文件夹，不用担心 DOTNET_ASSEMBLY_SEARCH_PATHS
要进行演练，请务必遵循：
https://docs.microsoft.com/en-us/dotnet/spark/tutorials/hdinsight-deployment
我已经尝试 运行 prod env 中包含来自 https://github.com/dotnet/spark/blob/master/examples/Microsoft.Spark.CSharp.Examples/Sql/Batch/Basic.cs 的代码的简单应用程序 应用程序 运行 运行良好并向标准输出发出输出，直到此代码在遇到第一个 UDF 时崩溃。 感谢您就此分享任何见解。
环境。 代码打包使用
dotnet publish -c Release -f netcoreapp3.1 -r ubuntu.16.04-x64
HDInsight 群集 HDI 4.0、Spark 2.4 -- 使用 https://docs.microsoft.com/en-us/dotnet/spark/tutorials/hdinsight-deployment
中的指南设置服务器spark-submit --master yarn --conf spark.yarn.appMasterEnv.DOTNET_ASSEMBLY_SEARCH_PATHS="./app/publish.zip" --archives wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/publish.zip#mySparkApp --class org.apache.spark.deploy.dotnet.DotnetRunner wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/microsoft-spark-2.4.x-0.12.1.jar wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/publish.zip mySparkApp
（以及绝望中的各种变体，--deploy-mode cluster，各种路径等等，什么都不管用）
标准输出： ...
+---+-----+
|age| name|
+---+-----+
| 22|Ricky|
| 36| Jeff|
| 62|Geddy|
+---+-----+
[2020-10-28T09:15:10.1478641Z] [wn0-hdinsi] [Error] [JvmBridge] JVM method execution failed: Nonstatic method 'showString' failed for class '41' when called with 3 arguments ([Index=1, Type=Int32, Value=20], [Index=2, Type=Int32, Value=20], [Index=3, Type=Boolean, Value=False], )
[2020-10-28T09:15:10.1480587Z] [wn0-hdinsi] [Error] [JvmBridge] org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 16.0 failed 4 times, most recent failure: Lost task 0.3 in stage 16.0 (TID 210, wn0-hdinsi.xwccrqijnmqujdjghwrza0nzbb.fx.internal.cloudapp.net, executor 2): org.apache.spark.api.python.PythonException: System.NullReferenceException: Object reference not set to an instance of an object.
at Microsoft.Spark.Utils.UdfSerDe.<>c.b__10_0(TypeData td) in //src/csharp/Microsoft.Spark/Utils/UdfSerDe.cs:line 262
at System.Collections.Concurrent.ConcurrentDictionary2.GetOrAdd(TKey key, Func2 valueFactory)
at Microsoft.Spark.Utils.UdfSerDe.DeserializeType(TypeData typeData) in //src/csharp/Microsoft.Spark/Utils/UdfSerDe.cs:line 258
at Microsoft.Spark.Utils.UdfSerDe.Deserialize(UdfData udfData) in //src/csharp/Microsoft.Spark/Utils/UdfSerDe.cs:line 160
at Microsoft.Spark.Utils.CommandSerDe.DeserializeUdfs[T](UdfWrapperData data, Int32& nodeIndex, Int32& udfIndex) in //src/csharp/Microsoft.Spark/Utils/CommandSerDe.cs:line 333
at Microsoft.Spark.Utils.CommandSerDe.Deserialize[T](Stream stream, SerializedMode& serializerMode, SerializedMode& deserializerMode, String& runMode) in /_/src/csharp/Microsoft.Spark/Utils/CommandSerDe.cs:line 306
at Microsoft.Spark.Worker.Processor.CommandProcessor.ReadSqlCommands(PythonEvalType evalType, Stream stream) in D:\a\s\src\csharp\Microsoft.Spark.Worker\Processor\CommandProcessor.cs:line 188
at Microsoft.Spark.Worker.Processor.CommandProcessor.ReadSqlCommands(PythonEvalType evalType, Stream stream, Version version) in D:\a\s\src\csharp\Microsoft.Spark.Worker\Processor\CommandProcessor.cs:line 98
at Microsoft.Spark.Worker.Processor.CommandProcessor.Process(Stream stream) in D:\a\s\src\csharp\Microsoft.Spark.Worker\Processor\CommandProcessor.cs:line 43
at Microsoft.Spark.Worker.Processor.PayloadProcessor.Process(Stream stream) in D:\a\s\src\csharp\Microsoft.Spark.Worker\Processor\PayloadProcessor.cs:line 82
at Microsoft.Spark.Worker.TaskRunner.ProcessStream(Stream inputStream, Stream outputStream, Version version, Boolean& readComplete) in D:\a\s\src\csharp\Microsoft.Spark.Worker\TaskRunner.cs:line 143
at org.apache.spark.api.python.BasePythonRunner$ReaderIterator.handlePythonException(PythonRunner.scala:456)
at org.apache.spark.sql.execution.python.PythonUDFRunner$$anon.read(PythonUDFRunner.scala:81)
at org.apache.spark.sql.execution.python.PythonUDFRunner$$anon.read(PythonUDFRunner.scala:64)
at org.apache.spark.api.python.BasePythonRunner$ReaderIterator.hasNext(PythonRunner.scala:410)
at org.apache.spark.InterruptibleIterator.hasNext(InterruptibleIterator.scala:37)
at scala.collection.Iterator$$anon.hasNext(Iterator.scala:440)
at scala.collection.Iterator$$anon.hasNext(Iterator.scala:409)
at scala.collection.Iterator$$anon.hasNext(Iterator.scala:409)
at org.apache.spark.sql.catalyst.expressions.GeneratedClass$GeneratedIteratorForCodegenStage2.processNext(Unknown Source)
at org.apache.spark.sql.execution.BufferedRowIterator.hasNext(BufferedRowIterator.java:43)
at org.apache.spark.sql.execution.WholeStageCodegenExec$$anonfun$$anon.hasNext(WholeStageCodegenExec.scala:636)
at org.apache.spark.sql.execution.SparkPlan$$anonfun.apply(SparkPlan.scala:255)
at org.apache.spark.sql.execution.SparkPlan$$anonfun.apply(SparkPlan.scala:247)
at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$$anonfun$apply.apply(RDD.scala:836)
at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$$anonfun$apply.apply(RDD.scala:836)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:324)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:288)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:324)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:288)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90)
at org.apache.spark.scheduler.Task.run(Task.scala:123)
at org.apache.spark.executor.Executor$TaskRunner$$anonfun.apply(Executor.scala:408)
at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:414)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
------切-------- 我的问题确实是路径问题。对于遇到同样问题的任何其他人，我通过将带有 UDF 的 dll（可以是与通用 spark 应用程序相同的 dll）必须列在“--files”中来使它工作。所以本质上你需要一个带有程序集的 zip 文件，然后直接链接到 dll。可能有更聪明的方法，但这对我有用（当 运行ning 在集群模式下）： spark-submit --deploy-mode cluster --master yarn --files wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/mySparkApp.dll --class org.apache.spark.deploy.dotnet.DotnetRunner wasbs:/ /xxx@yyy.blob.core.windows.net/SparkJobs/microsoft-spark-2.4.x-0.12.1.jar wasbs://xxx@yyy.blob.core.windows.net/SparkJobs/publish.zip mySparkApp
错误是因为找不到你代码的dll
两件事，首先是纱线模式。在 DOTNET_ASSEMBLY_SEARCH_PATHS 的开头导致用户主目录被添加到路径之前，所以它不是 currentdirectory/app/publish.zip 所以如果它不同那么它会在错误的地方查找。
其次确保 publish.zip 不包含文件夹，并且带有 udf 的 dll 位于 zip 的顶层。
我不会将 zip 放在应用程序文件夹中，而是使用当前文件夹，不用担心 DOTNET_ASSEMBLY_SEARCH_PATHS
要进行演练，请务必遵循：
https://docs.microsoft.com/en-us/dotnet/spark/tutorials/hdinsight-deployment