predictionio-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sebastian Fix <sebastian....@smartgate.ch>
Subject Re: Exception: Job aborted due to stage failure: Task 0 in stage 3.0 failed 1 times,
Date Sun, 09 Jul 2017 09:31:24 GMT
Hi Kenneth & Vaghawan

Thanks for your answer. You will find the error attached.

@Vaghawan

At the Documentation the Event Data Requirements are set to:

  *   Users' view events
  *   Users' buy events
  *   Items' with categories properties
  *   Constraint unavailableItems set events

The Documentation set the following parameter. Are all of these parameter required?

I uploaded with a .json import:

  *   Product-Items
  *   User
  *   User Buy Events.

[cid:f2700cf6-ceb1-4133-ace4-df99b21972e3@eurprd07.prod.outlook.com]

[cid:70f40559-adf4-4123-9de8-a7492ada1674@eurprd07.prod.outlook.com]
[cid:46c9e810-e251-4bd6-80b3-88282f0b210a@eurprd07.prod.outlook.com]

[cid:f3166fe2-5934-4179-999b-129fbb50a665@eurprd07.prod.outlook.com]

In the pio.log file i also found this exception:

2017-07-08 10:50:15,153 ERROR org.apache.spark.scheduler.TaskSetManager [task-result-getter-1]
- Task 0 in stage 3.0 failed 1 times; aborting job
2017-07-09 11:21:10,476 WARN  org.apache.predictionio.workflow.WorkflowUtils$ [main] - Environment
variable MYSQL_JDBC_DRIVER is pointing to a nonexistent file /users/sebastian/Development/Recommendation/PredictionIO-0.11.0-incubating/lib/mysql-connector-ja$
2017-07-09 11:21:10,487 INFO  org.apache.predictionio.tools.Runner$ [main] - Submission command:
/users/sebastian/Development/Recommendation/PredictionIO-0.11.0-incubating/vendors/spark-1.6.3-bin-hadoop2.6/bin/spark-submit
--class org.apache.predictionio.wo$
2017-07-09 11:21:12,665 INFO  org.apache.predictionio.controller.Engine [main] - Extracting
datasource params...
2017-07-09 11:21:12,754 INFO  org.apache.predictionio.workflow.WorkflowUtils$ [main] - No
'name' is found. Default empty String will be used.
2017-07-09 11:21:12,805 INFO  org.apache.predictionio.controller.Engine [main] - Datasource
params: (,org.example.recommendation.DataSourceParams@2c715e84)
2017-07-09 11:21:12,805 INFO  org.apache.predictionio.controller.Engine [main] - Extracting
preparator params...
2017-07-09 11:21:12,806 INFO  org.apache.predictionio.controller.Engine [main] - Preparator
params: (,Empty)
2017-07-09 11:21:12,819 INFO  org.apache.predictionio.controller.Engine [main] - Extracting
serving params...
2017-07-09 11:21:12,819 INFO  org.apache.predictionio.controller.Engine [main] - Serving params:
(,Empty)
2017-07-09 11:21:14,148 INFO  Remoting [sparkDriverActorSystem-akka.actor.default-dispatcher-2]
- Starting remoting
2017-07-09 11:21:14,301 INFO  Remoting [sparkDriverActorSystem-akka.actor.default-dispatcher-2]
- Remoting started; listening on addresses :[akka.tcp://sparkDriverActorSystem@192.168.1.35:59998]
2017-07-09 11:21:15,256 INFO  org.apache.predictionio.controller.Engine$ [main] - EngineWorkflow.train
2017-07-09 11:21:15,257 INFO  org.apache.predictionio.controller.Engine$ [main] - DataSource:
org.example.recommendation.DataSource@1b13467c
2017-07-09 11:21:15,257 INFO  org.apache.predictionio.controller.Engine$ [main] - Preparator:
org.example.recommendation.Preparator@1fedf0a4
2017-07-09 11:21:15,257 INFO  org.apache.predictionio.controller.Engine$ [main] - AlgorithmList:
List(org.example.recommendation.Algorithm@1d901f20)
2017-07-09 11:21:15,258 INFO  org.apache.predictionio.controller.Engine$ [main] - Data sanity
check is on.
2017-07-09 11:21:15,857 INFO  org.apache.predictionio.controller.Engine$ [main] - org.example.recommendation.TrainingData
supports data sanity check. Performing check.
2017-07-09 11:21:16,693 ERROR org.apache.spark.executor.Executor [Executor task launch worker-2]
- Exception in task 0.0 in stage 3.0 (TID 9)
java.lang.NullPointerException



I’ll have a look later this day at the link you send me. I’ll also try to check the database
what data are in there.


Thanks for your help
Sebastian




Am 09.07.2017 um 05:01 schrieb Kenneth Chan <kenneth@apache.org<mailto:kenneth@apache.org>>:

Hi Sebastian,

would you post the full stack trace?
would you go to Spark master and look into more detailed Spark log of the failure? it may
give more hint where in engine code trigger the  null pointer exception

You could also follow here to debug and see if the data is imported as expected, or add some
debug message in your engine to get more info
https://predictionio.incubator.apache.org/customize/troubleshooting/



On Sat, Jul 8, 2017 at 5:12 AM, Vaghawan Ojha <vaghawan781@gmail.com<mailto:vaghawan781@gmail.com>>
wrote:
Hi,

As much as I can understand from the error message, the Null Exception happens when you have
wrong data or no data at all. Please make sure everything is fine in your data.

Sometimes, the event store can store the data you provide, but the template you might be using
may need other kind of data, so please make sure you're following the right doc and providing
the right kind of data.

Thanks


On Sat, Jul 8, 2017 at 2:39 PM, Sebastian Fix <sebastian.fix@smartgate.ch<mailto:sebastian.fix@smartgate.ch>>
wrote:
Hey Keneth

Thank for the answer.

I’managed to import the following data:


After running „pio build“ i get the following error during the „pio train“ process:

[ERROR] [Executor] Exception in task 0.0 in stage 3.0 (TID 9)
[WARN] [TaskSetManager] Lost task 0.0 in stage 3.0 (TID 9, localhost): java.lang.NullPointerException

[ERROR] [TaskSetManager] Task 0 in stage 3.0 failed 1 times; aborting job
Exception in thread "main" org.apache.spark.SparkException: Job aborted due to stage failure:
Task 0 in stage 3.0 failed 1 times, most recent failure: Lost task 0.0 in stage 3.0 (TID 9,
localhost): java.lang.NullPointerException

Driver stacktrace:
at org.apache.spark.scheduler.DAGScheduler.org<http://org.apache.spark.scheduler.dagscheduler.org/>$apache$spark$scheduler$DAGScheduler$$failJobAndIndependentStages(DAGScheduler.scala:1431)
Caused by: java.lang.NullPointerException


Best regards
Sebastian


Am 08.07.2017 um 09:52 schrieb Kenneth Chan <kenneth@apache.org<mailto:kenneth@apache.org>>:

re: " i’m always getting an exception with the initial load from User & Product"
what's the error you got?

On Fri, Jul 7, 2017 at 12:46 PM, Sebastian Fix <sebastian.fix@smartgate.ch<mailto:sebastian.fix@smartgate.ch>>
wrote:
Hello together,

I finally got the System running and would like to import some initial data now.

I’m using the E-Commerce-Template: http://predictionio.incubator.apache.org/templates/ecommercerecommendation/quickstart/

I've the following data’s i would like to import:

  *   User data

  *   Products data

  *   Transaction data

When i try to use the pio import --appid 1 --input import/data-user.json command i’m always
getting an exception with the initial load from User & Product. My .json file is following
the structure mentioned on the tutorial.


  *   For User Data: {"event":"$set","entityType":"user","entityId":"user001","eventTime":““}
  *   For Products  {"event":"$set","entityType“:"item","entityId“:"product001","eventTime":““}

  *   For Transactions: {"entityId“:"user001","entityType":"user","event":"buy","targetEntityType":"item","targetEntityId“:"product001“}

Only the Import for Transactions is working I found this template here: https://gist.github.com/vaghawan/0a5fb8ddb85e03631dd500d7c8f0677d#file-data-sample-json

So my quesiton, how can I import Data’s like


  *   User Data
  *   User Categories
  *   Product Data
  *   Product Categories

with a .json file? What is the file structure?

Thanks for any help,
Sebastian







Mime
View raw message