[ 
https://issues.apache.org/jira/browse/RANGER-988?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15320148#comment-15320148
 ] 

Madhan Neethiraj commented on RANGER-988:
-----------------------------------------

Please ensure the HDFS service in Ranger (like cl1_hadoop) has following 
service-configurations set correctly:

{code}
fs.default.name=hdfs://<cluster-name>
dfs.nameservices=<cluster-name>
dfs.ha.namenodes.<cluster-name>=nn1,nn2
dfs.namenode.rpc-address.<cluster-name>.nn1=<namenode1-host:8020>
dfs.namenode.rpc-address.<cluster-name>.nn2=<namenode2-host:8020>
dfs.client.failover.proxy.provider.<cluster-name>.nn2=org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider
 
{code}

Replace the text within '<' and '>' with appropriate values.

> In namenode HA enabled HDFS cluster, Ranger connection test fails if the name 
> of the name service is used in Namenode URL.
> --------------------------------------------------------------------------------------------------------------------------
>
>                 Key: RANGER-988
>                 URL: https://issues.apache.org/jira/browse/RANGER-988
>             Project: Ranger
>          Issue Type: Bug
>          Components: Ranger
>    Affects Versions: 0.5.0
>            Reporter: Kalexin Baoerjiin
>
> When connecting to a namenode high availability enabled HDFS cluster (with 
> kerberos enabled) when the name of namenode is used in Namenode URL, test 
> connection fails.  
> {code}
> 2016-05-19 13:23:53,131 [timed-executor-pool-0] DEBUG 
> org.apache.ranger.services.hdfs.RangerServiceHdfs (RangerServiceHdfs.java:52) 
> - <== RangerServiceHdfs.validateConfig Service: (hdp_240_hadoop )
> 2016-05-19 13:23:53,132 [timed-executor-pool-0] DEBUG 
> apache.ranger.services.hdfs.client.HdfsResourceMgr (HdfsResourceMgr.java:42) 
> - <== HdfsResourceMgr.testConnection ServiceName: 
> hdp_240_hadoopConfigs{[email protected], password=platfora, 
> fs.default.name=hdfs://nameservice1:8020, hadoop.security.authorization=true, 
> hadoop.security.authentication=kerberos, 
> hadoop.security.auth_to_local=RULE:[1:$1@$0]([email protected])s/.*/ambari-qa/RULE:[1:$1@$0]([email protected])s/.*/ambari-qa/RULE:[1:$1@$0]([email protected])s/.*/hbase/RULE:[1:$1@$0]([email protected])s/.*/hbase/RULE:[1:$1@$0]([email protected])s/.*/hdfs/RULE:[1:$1@$0]([email protected])s/.*/hdfs/RULE:[1:$1@$0]([email protected])s/.*/spark/RULE:[1:$1@$0]([email protected])s/.*/spark/RULE:[1:$1@$0](.*@PLATFORA.COM)s/@.*//RULE:[2:$1@$0]([email protected])s/.*/ams/RULE:[2:$1@$0]([email protected])s/.*/ams/RULE:[2:$1@$0]([email protected])s/.*/hdfs/RULE:[2:$1@$0]([email protected])s/.*/hbase/RULE:[2:$1@$0]([email protected])s/.*/hive/RULE:[2:$1@$0]([email protected])s/.*/mapred/RULE:[2:$1@$0]([email protected])s/.*/hdfs/RULE:[2:$1@$0]([email protected])s/.*/hdfs/RULE:[2:$1@$0]([email protected])s/.*/yarn/RULE:[2:$1@$0]([email protected])s/.*/hdfs/RULE:[2:$1@$0]([email protected])s/.*/yarn/RULE:[2:$1@$0]([email protected])s/.*/yarn/DEFAULT,
>  dfs.datanode.kerberos.principal=dn/[email protected], 
> dfs.namenode.kerberos.principal=nn/[email protected], 
> dfs.secondary.namenode.kerberos.principal=nn/[email protected],
>  hadoop.rpc.protection=authentication, commonNameForCertificate=}
> 2016-05-19 13:23:53,132 [timed-executor-pool-0] INFO  
> org.apache.ranger.plugin.client.BaseClient (BaseClient.java:100) - Init 
> Login: using username/password
> 2016-05-19 13:23:53,208 [timed-executor-pool-0] DEBUG 
> apache.ranger.services.hdfs.client.HdfsClient (HdfsClient.java:152) - <== 
> HdfsClient listFilesInternal Error : java.lang.IllegalArgumentException: 
> java.net.UnknownHostException: nameservice1
> {code}



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

Reply via email to