[ https://issues.apache.org/jira/browse/CASSANDRA-3851?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13200442#comment-13200442 ]
Samarth Gahire commented on CASSANDRA-3851: ------------------------------------------- Got It Actually in cassandra-trunk we are handling it as {code} File outputdir = new File(getOutputLocation() + File.separator + keyspace + File.separator + ConfigHelper.getOutputColumnFamily(conf)); //dir must be named by ks/cf for the loader {code} That is the reason it is creating the keyspace name properly. So Its a bug in cassandra-1.1. > Wrong Keyspace name is generated while streaming the sstables using > BulkOutputFormat. > ------------------------------------------------------------------------------------- > > Key: CASSANDRA-3851 > URL: https://issues.apache.org/jira/browse/CASSANDRA-3851 > Project: Cassandra > Issue Type: Bug > Components: Hadoop, Tools > Affects Versions: 1.1 > Reporter: Samarth Gahire > Assignee: Brandon Williams > Priority: Minor > Labels: bulkloader, hadoop, sstableloader > Fix For: 1.1 > > Original Estimate: 48h > Remaining Estimate: 48h > > I have merge the committed changes of > [CASSANDRA-3828|https://issues.apache.org/jira/browse/CASSANDRA-3828] into my > cassadra-trunk. Also the changes for the OutputLocation. > But when I tried to load the sstables with hadoop job it results into the > following exception: > {code} > 12/02/04 11:19:12 INFO mapred.JobClient: map 6% reduce 0% > 12/02/04 11:19:14 INFO mapred.JobClient: Task Id : > attempt_201202041114_0001_m_000001_1, Status : FAILED > java.lang.RuntimeException: Could not retrieve endpoint ranges: > at > org.apache.cassandra.hadoop.BulkRecordWriter$ExternalClient.init(BulkRecordWriter.java:252) > at > org.apache.cassandra.io.sstable.SSTableLoader.stream(SSTableLoader.java:117) > at > org.apache.cassandra.io.sstable.SSTableLoader.stream(SSTableLoader.java:112) > at > org.apache.cassandra.hadoop.BulkRecordWriter.close(BulkRecordWriter.java:182) > at > org.apache.cassandra.hadoop.BulkRecordWriter.close(BulkRecordWriter.java:167) > at > org.apache.hadoop.mapred.MapTask$NewDirectOutputCollector.close(MapTask.java:650) > at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:765) > at org.apache.hadoop.mapred.MapTask.run(MapTask.java:369) > at org.apache.hadoop.mapred.Child$4.run(Child.java:259) > at java.security.AccessController.doPrivileged(Native Method) > at javax.security.auth.Subject.doAs(Subject.java:396) > at > org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1059) > at org.apache.hadoop.mapred.Child.main(Child.java:253) > Caused by: InvalidRequestException (*why:There is no ring for the keyspace: > tmp*) > at > org.apache.cassandra.thrift.Cassandra$describe_ring_result.read(Cassandra.java:24053) > at > org.apache.thrift.TServiceClient.receiveBase(TServiceClient.java:78) > at > org.apache.cassandra.thrift.Cassandra$Client.recv_describe_ring(Cassandra.java:1065) > at > org.apache.cassandra.thrift.Cassandra$Client.describe_ring(Cassandra.java:1052) > at > org.apache.cassandra.hadoop.BulkRecordWriter$ExternalClient.init(BulkRecordWriter.java:225) > ... 12 more > {code} > After looking into the code I figured out that as we are setting the > OUTPUTLOCATION with system property "java.io.tmpdir" the output directory is > getting created as: /tmp/Keyspace_Name > So in SSTableLoader while generating the kespace name like > {code} > this.keyspace = directory.getParentFile().getName(); > {code} > It is setting the keyspace name as "tmp" and results into the above exception. > I have changed the code as: > {code}this.keyspace = directory.getName();{code} > and it works perfect. > But I am wondering how it was working fine previously? Am I doing anything > wrong ? or is it a bug? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira