Hive connectivity issue

Comments

3 comments

  • Konsta Danyliuk

    Hello,

    Please review our documentation for Connecting and Importing from Hive.

    It seems you have to pay attention at the following part:

    "If high availability (HA) for the Hive metastore is enabled on cluster, users must provide both thrift URLs separated by commas while creating the Hive connection."

    In case of further questions, please also attach screenshot of your Hive connection. 

    0
    Comment actions Permalink
  • Sudipkumar H Ghanvat

    No issue with connection . we are able to make connection successful. 

    after connection while importing data or creating data link using hive connection we can see Database name and tables belongs to it.

    we are getting error while after selecting table name and clicking on next retrieve fileds/data.

    0
    Comment actions Permalink
  • Joel Stewart

    It appears that nameservice1 is an HA alias for your Hadoop Cluster, likely for HDFS. If that's the case, can you please ensure that you've provided the relevant HA parameters to the Hive Connection Custom Properties in Datameer? 

    Here are some example properties that would be for consideration: 

    dfs.nameservices=nameservice1
    dfs.ha.namenodes.nameservice1=nn0,nn1
    dfs.namenode.rpc-address.nameservice1.nn0=ip-<host-name>:<port>
    dfs.namenode.rpc-address.nameservice1.nn1=ip-<host-name>:<port>
    dfs.client.failover.proxy.provider.nameservice1=org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider

    You may have different node aliases from nn0 and nn1 so you may need to adjust for these. 

    0
    Comment actions Permalink

Please sign in to leave a comment.