Apache Tez
  1. Apache Tez
  2. TEZ-591

Provide mode specific diagnostic information to the Tez client

    Details

    • Type: Wish Wish
    • Status: Closed
    • Priority: Minor Minor
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 0.2.0
    • Component/s: None
    • Labels:
      None

      Description

      While developing Pig on Tez, I found it's hard to debug DAG failures due to lack of diagnostic information. Currently, the MR Pig client reports the backend error message when there is a job failure. For example, if I have a UDF that throws a runtime exception, I will see the following stack trace in the front-end log file-

      Pig Stack Trace
      ---------------
      ERROR 1066: Unable to open iterator for alias b. Backend error : FAIL IT NOW!
      ...
      Caused by: java.lang.RuntimeException: FAIL IT NOW!
          at Kill.exec(Kill.java:9)
          at Kill.exec(Kill.java:6)
          at org.apache.pig.backend.hadoop.executionengine.physicalLayer.expressionOperators.POUserFunc.getNext(POUserFunc.java:334)
          at org.apache.pig.backend.hadoop.executionengine.physicalLayer.expressionOperators.POUserFunc.getNext(POUserFunc.java:383)
          at org.apache.pig.backend.hadoop.executionengine.physicalLayer.PhysicalOperator.getNext(PhysicalOperator.java:346)
          at org.apache.pig.backend.hadoop.executionengine.physicalLayer.relationalOperators.POForEach.processPlan(POForEach.java:372)
          at org.apache.pig.backend.hadoop.executionengine.physicalLayer.relationalOperators.POForEach.getNext(POForEach.java:297)
          at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.runPipeline(PigGenericMapBase.java:283)
          at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.map(PigGenericMapBase.java:278)
          at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.map(PigGenericMapBase.java:64)
          at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:144)
          at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:771)
          at org.apache.hadoop.mapred.MapTask.run(MapTask.java:375)
          at org.apache.hadoop.mapred.Child$4.run(Child.java:255)
          at java.security.AccessController.doPrivileged(Native Method)
          at javax.security.auth.Subject.doAs(Subject.java:396)
          at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1132)
      

      Basically, I'd like to do something similar in Tez Pig.

      If there are multiple failed vertices and tasks, it may be not possible to propagate all the backend exceptions to the frontend. But would it be possible to propagate some of first ones at least? Perhaps one per failed vertex? Given that DAGStatus.getDiagnostics() returns a list of Strings, it seems feasible.

      1. TEZ-591.1.patch
        12 kB
        Hitesh Shah

        Issue Links

          Activity

          Hitesh Shah made changes -
          Status Resolved [ 5 ] Closed [ 6 ]
          Hitesh Shah made changes -
          Status In Progress [ 3 ] Resolved [ 5 ]
          Fix Version/s 0.2.0 [ 12324349 ]
          Resolution Fixed [ 1 ]
          Hitesh Shah made changes -
          Status Open [ 1 ] In Progress [ 3 ]
          Hitesh Shah made changes -
          Attachment TEZ-591.1.patch [ 12612071 ]
          Hitesh Shah made changes -
          Assignee Hitesh Shah [ hitesh ]
          Cheolsoo Park made changes -
          Link This issue is required by PIG-3446 [ PIG-3446 ]
          Cheolsoo Park made changes -
          Field Original Value New Value
          Description While deploying Pig on Tez, I found it's hard to debug DAG failures due to lack of diagnostic information. Currently, the MR Pig client reports the backend error message when there is a job failure. For example, if I have a UDF that throws a runtime exception, I will see the following stack trace in the front-end log file-
          {code}
          Pig Stack Trace
          ---------------
          ERROR 1066: Unable to open iterator for alias b. Backend error : FAIL IT NOW!
          ...
          Caused by: java.lang.RuntimeException: FAIL IT NOW!
              at Kill.exec(Kill.java:9)
              at Kill.exec(Kill.java:6)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.expressionOperators.POUserFunc.getNext(POUserFunc.java:334)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.expressionOperators.POUserFunc.getNext(POUserFunc.java:383)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.PhysicalOperator.getNext(PhysicalOperator.java:346)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.relationalOperators.POForEach.processPlan(POForEach.java:372)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.relationalOperators.POForEach.getNext(POForEach.java:297)
              at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.runPipeline(PigGenericMapBase.java:283)
              at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.map(PigGenericMapBase.java:278)
              at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.map(PigGenericMapBase.java:64)
              at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:144)
              at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:771)
              at org.apache.hadoop.mapred.MapTask.run(MapTask.java:375)
              at org.apache.hadoop.mapred.Child$4.run(Child.java:255)
              at java.security.AccessController.doPrivileged(Native Method)
              at javax.security.auth.Subject.doAs(Subject.java:396)
              at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1132)
          {code}
          Basically, I'd like to do something similar in Tez Pig.

          If there are multiple failed vertices and tasks, it may be not possible to propagate all the backend exceptions to the frontend. But would it be possible to propagate some of first ones at least? Perhaps one per failed vertex? Given that DAGStatus.getDiagnostics() returns a list of Strings, it seems feasible.



          While developing Pig on Tez, I found it's hard to debug DAG failures due to lack of diagnostic information. Currently, the MR Pig client reports the backend error message when there is a job failure. For example, if I have a UDF that throws a runtime exception, I will see the following stack trace in the front-end log file-
          {code}
          Pig Stack Trace
          ---------------
          ERROR 1066: Unable to open iterator for alias b. Backend error : FAIL IT NOW!
          ...
          Caused by: java.lang.RuntimeException: FAIL IT NOW!
              at Kill.exec(Kill.java:9)
              at Kill.exec(Kill.java:6)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.expressionOperators.POUserFunc.getNext(POUserFunc.java:334)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.expressionOperators.POUserFunc.getNext(POUserFunc.java:383)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.PhysicalOperator.getNext(PhysicalOperator.java:346)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.relationalOperators.POForEach.processPlan(POForEach.java:372)
              at org.apache.pig.backend.hadoop.executionengine.physicalLayer.relationalOperators.POForEach.getNext(POForEach.java:297)
              at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.runPipeline(PigGenericMapBase.java:283)
              at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.map(PigGenericMapBase.java:278)
              at org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase.map(PigGenericMapBase.java:64)
              at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:144)
              at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:771)
              at org.apache.hadoop.mapred.MapTask.run(MapTask.java:375)
              at org.apache.hadoop.mapred.Child$4.run(Child.java:255)
              at java.security.AccessController.doPrivileged(Native Method)
              at javax.security.auth.Subject.doAs(Subject.java:396)
              at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1132)
          {code}
          Basically, I'd like to do something similar in Tez Pig.

          If there are multiple failed vertices and tasks, it may be not possible to propagate all the backend exceptions to the frontend. But would it be possible to propagate some of first ones at least? Perhaps one per failed vertex? Given that DAGStatus.getDiagnostics() returns a list of Strings, it seems feasible.



          Cheolsoo Park created issue -

            People

            • Assignee:
              Hitesh Shah
              Reporter:
              Cheolsoo Park
            • Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development