@Saravana V These are all INFO messages, makes me believe that since additional datanodes have been added, rebalance is taking place. See if you can find any exception in the HDFS logs. ... View more

@Aditya Sirna Thanks for your reply. there is no such directory inside /usr/hdp/. Does it refer the old version in cache. hdp]$ ls -lrt total 12 drwxr-xr-x 3 hadoop hadoop 4096 Sep 22 19:56 share drwxr-xr-x 14 root root 4096 Sep 28 21:15 2.6.4.25-1 drwxr-xr-x 2 hadoop hadoop 4096 Oct 3 18:12 current Thanks. ... View more

@Geoffrey Shelton Okot I see the port mentioned in config is correct and how to check the port in java code. ... View more

@Saravana V In the below I used the hive user and password, Access to HiveServer2 in kerberized cluster using hive CLI has been deprecated. Example One $ beeline $ ! connect jdbc:hive2://{hive_host}:10000/test Here you MUST be the test database owner as you will be prompted for username / password see below prompts Hs2 in kerberized cluster Note the hive principal and REALM $ beeline Beeline version 1.2.1000.2.5.3.0-37 by Apache Hive beeline> ! connect jdbc:hive2://france.paris.fr:10000/;principal=hive/france.paris.fr@PARIS.FR Connecting to jdbc:hive2://france.paris.fr:10000/;principal=hive/france.paris.fr@PARIS.FR Will be prompted for username/password Enter username for jdbc:hive2://france.paris.fr:10000/;principal=hive/france.paris.fr@PARIS.FR:{username} Enter password for jdbc:hive2://france.paris.fr:10000/;principal=hive/france.paris.fr@PARIS.FR:{password} Connected to: Apache Hive (version 1.2.1000.2.5.3.0-37) Driver: Hive JDBC (version 1.2.1000.2.5.3.0-37) Transaction isolation: TRANSACTION_REPEATABLE_READ 0: jdbc:hive2://france.paris.fr:10000/> show databases; +----------------+ | database_name | +----------------+ | default | | agricole | | achats | +----------------+--+ 3 rows selected (2.863 seconds) 0: jdbc:hive2://london.nakuru.com:10000> use agricole; ... View more

@Geoffrey Shelton Okot What you have mentioned is for any specific path. Do you have any procedure through which we can change the block size of existing cluster which includes all old files ? Please post it here if any you are aware!! ... View more

HIVE needs to store meta data such as Table structures, OOZIE needs to store Job details etc. you can go into the db schemas of each product and see what they store. Its a persistent mechanism for storing meta data, please look at the internal architecture of each component, it will be clear. ... View more

@Jay Kumar SenSharma Thanks for heads up. Ambari agents are running as Non root (hadoop) user the directory was created when we tried to srart the spark2 service and the owner was assigned as spark user. Now I've changed the owner as hadoop user and the service came up. thanks a lot 🙂 ... View more

@Jay Kumar SenSharma Thanks for your reply. Yeah I could see the option to restart the required services after modifying the configs. Thanks much 🙂 ... View more