Hi,  I have a kerberized cluster, where each application is working fine (hdfs, yarn, hive). Accidentally via kadmin.local I re-exported the keytab for the spark user. xst -k ....  And now the Clouder Manager of course complains because Spark cannot start, the keytab stored in CM database is different and does not match with the Kerberos DB.   So how to:  - replace the keytab in Cloudera Manager database with the new one which I just created? or..  - force Cloudera Manager to re-create keytab for the Spark application?   Thanks T.   ... View more

Hi I have a dataframe (loaded CSV) where the inferredSchema filled the column names from the file. I am trying to get rid of white spaces from column names - because otherwise the DF cannot be saved as parquet file - and did not find any usefull method for renaming.   The method withColumnRenamed("Company ID","Company_ID") works, but I need to repeat it for every column in the dataframe. I tried to to use toDF method, such as:   val dfnew = df.toDF( df.columns.map( a => a.replace(" ","_") ) ); but it failed.,   Any ideas?   ... View more

Hi guys,  has anybody experience how to setup an access from SQL Developer to Impala via JDBC/ODBC? I have read a discussion forum (outdated) where it states that Oracle does not support Impala, better said Oracle has a whitelist of connectors,  Impala is not there.   If SQL Developer cannot be used, do you have any suggestion what tool should use a Oracle developer who needs to access Imapal (and dont want to use Hue, neither impala-shell)?   Thanks T.   ... View more