hadoop-yarn-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Vasu Mariyala (JIRA)" <j...@apache.org>
Subject [jira] [Created] (YARN-1361) Not able to run the map reduce jobs on multiple mini map reduce clusters
Date Tue, 29 Oct 2013 18:10:25 GMT
Vasu Mariyala created YARN-1361:
-----------------------------------

             Summary: Not able to run the map reduce jobs on multiple mini map reduce clusters
                 Key: YARN-1361
                 URL: https://issues.apache.org/jira/browse/YARN-1361
             Project: Hadoop YARN
          Issue Type: Task
          Components: resourcemanager
            Reporter: Vasu Mariyala
            Priority: Critical


As part of testing backup validation on the primary and dr clusters, we do the following

a) Start two mini clusters with each mini cluster containing hdfs, map reduce and hbase mini
clusters.
b) Run the backup map reduce job on both the primary and secondary test mini clusters
c) These jobs interfere with each other and the second job fails due to the following

The application id is created by using the time stamp (static variable) and a job counter
(instance specific always starts with zero). Since a unit test is ran on a single jvm, these
jobs get the same application id and the second job fails with tokens already exist error.



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Mime
View raw message