如何修复Dataflow无法序列化我的DoFn？

Question

当我运行我的Dataflow管道时,我得到以下异常抱怨我的DoFn无法序列化.我该如何解决？

这是堆栈跟踪:

Caused by: java.lang.IllegalArgumentException: unable to serialize contrail.dataflow.AvroMRTransforms$AvroReducerDoFn@bba0fc2
    at com.google.cloud.dataflow.sdk.util.SerializableUtils.serializeToByteArray(SerializableUtils.java:51)
    at com.google.cloud.dataflow.sdk.util.SerializableUtils.ensureSerializable(SerializableUtils.java:81)
    at com.google.cloud.dataflow.sdk.runners.DirectPipelineRunner$Evaluator.ensureSerializable(DirectPipelineRunner.java:784)
    at com.google.cloud.dataflow.sdk.transforms.ParDo.evaluateHelper(ParDo.java:1025)
    at com.google.cloud.dataflow.sdk.transforms.ParDo.evaluateSingleHelper(ParDo.java:963)
    at com.google.cloud.dataflow.sdk.transforms.ParDo.access$000(ParDo.java:441)
    at com.google.cloud.dataflow.sdk.transforms.ParDo$1.evaluate(ParDo.java:951)
    at com.google.cloud.dataflow.sdk.transforms.ParDo$1.evaluate(ParDo.java:946)
    at com.google.cloud.dataflow.sdk.runners.DirectPipelineRunner$Evaluator.visitTransform(DirectPipelineRunner.java:611)
    at com.google.cloud.dataflow.sdk.runners.TransformTreeNode.visit(TransformTreeNode.java:200)
    at com.google.cloud.dataflow.sdk.runners.TransformTreeNode.visit(TransformTreeNode.java:196)
    at com.google.cloud.dataflow.sdk.runners.TransformHierarchy.visit(TransformHierarchy.java:109)
    at com.google.cloud.dataflow.sdk.Pipeline.traverseTopologically(Pipeline.java:204)
    at com.google.cloud.dataflow.sdk.runners.DirectPipelineRunner$Evaluator.run(DirectPipelineRunner.java:584)
    at com.google.cloud.dataflow.sdk.runners.DirectPipelineRunner.run(DirectPipelineRunner.java:328)
    at com.google.cloud.dataflow.sdk.runners.DirectPipelineRunner.run(DirectPipelineRunner.java:70)
    at com.google.cloud.dataflow.sdk.Pipeline.run(Pipeline.java:145)
    at contrail.stages.DataflowStage.stageMain(DataflowStage.java:51)
    at contrail.stages.NonMRStage.execute(NonMRStage.java:130)
    at contrail.stages.NonMRStage.run(NonMRStage.java:157)
    at org.apache.hadoop.util.ToolRunner.run(ToolRunner.java:65)
    at contrail.stages.ValidateGraphDataflow.main(ValidateGraphDataflow.java:139)
    ... 6 more
Caused by: java.io.NotSerializableException: org.apache.hadoop.mapred.JobConf
    at java.io.ObjectOutputStream.writeObject0(ObjectOutputStream.java:1183)
    at java.io.ObjectOutputStream.defaultWriteFields(ObjectOutputStream.java:1547)
    at java.io.ObjectOutputStream.writeSerialData(ObjectOutputStream.java:1508)
    at java.io.ObjectOutputStream.writeOrdinaryObject(ObjectOutputStream.java:1431)
    at java.io.ObjectOutputStream.writeObject0(ObjectOutputStream.java:1177)
    at java.io.ObjectOutputStream.writeObject(ObjectOutputStream.java:347)
    at com.google.cloud.dataflow.sdk.util.SerializableUtils.serializeToByteArray(SerializableUtils.java:47)
    ... 27 more

Answer 1

加上杰里米所说的......

Serializable问题的另一个常见原因是在非静态上下文中使用匿名DoFn.匿名内部类具有指向封闭类的隐式指针,这将导致它也被序列化.

Answer 2

如果滚动堆栈跟踪,其中一个原因会清楚地标识不可序列化的数据.

Caused by: java.io.NotSerializableException: org.apache.hadoop.mapred.JobConf

问题是我的DoFn正在构造函数中获取JobConf实例并将其存储在实例变量中.我假设JobConf是可序列化的,但事实证明它不是.

要解决这个问题,我做了以下几点

• 我将JobConf成员变量标记为瞬态,以便它不会被序列化.
• 我创建了一个byte []类型的单独变量来存储JobConf的序列化版本
• 在我的构造函数中,我将JobConf序列化为byte []并将其存储在实例变量中.
• 我重写了startBundle并从byte []反序列化了JobConf

这是我的DoFn 的要点.