spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Umesh Kacha (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (SPARK-6995) Spark driver program throws exception cant get Master Kerberose principal for use as renewer
Date Sun, 19 Apr 2015 07:19:58 GMT

     [ https://issues.apache.org/jira/browse/SPARK-6995?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Umesh Kacha updated SPARK-6995:
-------------------------------
    Description: 
I am having thousands of files residing in HDFS which I would like to read/process/store back
to HDFS using Spark. I have the following Java Spark driver program:

SparkConf sparkConf = new SparkConf().setAppName("MyApp").setMaster("spark://myhost1.com");
JavaSparkContext sc = new JavaSparkContext(sparkConf);
JavaRDD<MySchemaPOJO> schemaRDD = sc.textFile("hdfs://path/to/file").map(
      new Function<String,MySchemaPOJO>() {
         public MySchemaPOJO call(String line) throws Exception {
         String[] fields = line.split("\\t");
         MySchemaPOJO schemaPOJO = new MySchemaPOJO();
         //set required fields on POJO
         return schemaPOJO;
         }
});
System.out.println("POJO list are " + schemaRDD.collect());
schemaRDD.saveAsTextFile("hdfs://path/to/save/file");
sc.stop();

I have started Master/Worker on myhost1.com and I am running this driver program on IntelliJ
on another host myhost2.com. When I start driver program everything works fine until schemaRDD.collect()
and it throws exception as

Exception in thread 'main' java.io.IOException : Cant get Master 
         Kerberose principal to use as renewer.
I did kinit on both host where I am running driver and master/worker. My driver program can
connect to Master also only when actual actions gets called on RDD it throws above exception.
Please guide I am new to Spark. Thanks in advance.

  was:
I am having thousands of files residing in HDFS which I would like to read/process/store back
to HDFS using Spark. I have the following Java Spark driver program:

SparkConf sparkConf = new SparkConf().setAppName("MyApp").setMaster("spark://myhost1.com");
JavaSparkContext sc = new JavaSparkContext(sparkConf);
JavaRDD<MySchemaPOJO> schemaRDD = sc.textFile("hdfs://path/to/file").map(
      new Function<String,MySchemaPOJO>() {
         public MySchemaPOJO call(String line) throws Exception {
         String[] fields = line.split("\\t");
         MySchemaPOJO schemaPOJO = new MySchemaPOJO();
         //set required fields on POJO
         return schemaPOJO;
         }
});
System.out.println("POJO list are " + schemaRDD.collect());
schemaRDD.saveAsTextFile("hdfs://path/to/save/file");
sc.stop();

I have started Master/Worker on myhost1.com and I am running this driver program on IntelliJ
on another host myhost2.com. When I start driver program everything works fine until schemaRDD.collect()
and it throws exception as

Exception in thread 'main' java.io.IOException : Cant get Master 
         Kerberose principal to use as renewer.
I did kinit on both host where I am running driver and master/worker. My driver program can
connect to Master also only when actual actions gets called on RDD it throws above exception.
Please guide I am new to Spark.


> Spark driver program throws exception cant get Master Kerberose principal for use as
renewer
> --------------------------------------------------------------------------------------------
>
>                 Key: SPARK-6995
>                 URL: https://issues.apache.org/jira/browse/SPARK-6995
>             Project: Spark
>          Issue Type: Question
>          Components: Java API
>    Affects Versions: 1.1.0
>         Environment: Linux,IntelliJ
>            Reporter: Umesh Kacha
>
> I am having thousands of files residing in HDFS which I would like to read/process/store
back to HDFS using Spark. I have the following Java Spark driver program:
> SparkConf sparkConf = new SparkConf().setAppName("MyApp").setMaster("spark://myhost1.com");
> JavaSparkContext sc = new JavaSparkContext(sparkConf);
> JavaRDD<MySchemaPOJO> schemaRDD = sc.textFile("hdfs://path/to/file").map(
>       new Function<String,MySchemaPOJO>() {
>          public MySchemaPOJO call(String line) throws Exception {
>          String[] fields = line.split("\\t");
>          MySchemaPOJO schemaPOJO = new MySchemaPOJO();
>          //set required fields on POJO
>          return schemaPOJO;
>          }
> });
> System.out.println("POJO list are " + schemaRDD.collect());
> schemaRDD.saveAsTextFile("hdfs://path/to/save/file");
> sc.stop();
> I have started Master/Worker on myhost1.com and I am running this driver program on IntelliJ
on another host myhost2.com. When I start driver program everything works fine until schemaRDD.collect()
and it throws exception as
> Exception in thread 'main' java.io.IOException : Cant get Master 
>          Kerberose principal to use as renewer.
> I did kinit on both host where I am running driver and master/worker. My driver program
can connect to Master also only when actual actions gets called on RDD it throws above exception.
Please guide I am new to Spark. Thanks in advance.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org


Mime
View raw message