@Simon Jespersen You are having two forward slashes in your curl call after https://localhost:9091, just use one forward slash and run again the your curl call curl -k -i -H 'Content-Type: application/json'-XPUT -d '{"id":"cdb54c9a-0158-1000-5566-c45ca9692f85","state":"RUNNING"}' https://localhost:9091/nifi-api/flow/process-groups/a9d5c45f-015b-1000-0000-00006d9844d3 If you are still facing some issues then follow the below steps to start/stop processor groups in kerberos HDF2.1.1. if our HDF is Kerberised then we need to pass our access token with CURL api call. Steps to Start/Stop Processor Group:- 1.First do kinit in your NiFi node bash$ kinit 2.Check the validity of the kerberos ticket and make sure your your ticket is valid bash$ klist 3. Now create access token bash$ token=`curl -k -X POST --negotiate -u : https://localhost:9091/nifi-api/access/kerberos` https://localhost:9091/nifi/ 4.Use the created token in your curl call to start processor group bash$ curl -k --header "Authorization: Bearer $token" -i -H 'Content-Type: application/json' -XPUT -d'{"id":"cdb54c9a-0158-1000-5566-c45ca9692f85","state":"RUNNING"}' https://localhost:9091/nifi-api/flow/process-groups/cdb54c9a-0158-1000-5566-c45ca9692f85 5.Use the created token in your curl call to stop processor group bash$ curl -k --header "Authorization: Bearer $token" -i -H 'Content-Type: application/json' -XPUT -d'{"id":"cdb54c9a-0158-1000-5566-c45ca9692f85","state":"STOPPED"}' https://localhost:9091/nifi-api/flow/process-groups/cdb54c9a-0158-1000-5566-c45ca9692f85 ... View more

@Gayathri Devi Could you try with below query as we are reading 2018-02-27T02:00 value and converting as timestamp value. Query:- hive> select from_unixtime(unix_timestamp('2018-02-27T02:00',"yyyy-MM-dd'T'hh:mm"),'yyyy-MM-dd hh:mm:ss'); +----------------------+--+ | _c0 | +----------------------+--+ | 2018-02-27 02:00:00 | +----------------------+--+ (or) By using regexp_replace function we can replace T in your timestamp value hive> select regexp_replace('2018-02-27T02:00','T',' '); +-------------------+--+ | _c0 | +-------------------+--+ | 2018-02-27 02:00 | +-------------------+--+ And use concat function to add missing :00 value to make above value as hive timestamp. hive> select concat(regexp_replace('2018-02-27T02:00','T',' '),":00"); +----------------------+--+ | _c0 | +----------------------+--+ | 2018-02-27 02:00:00 | +----------------------+--+ ... View more

@Mark As you are trying to insert values having semicolon(;) in it, hive thinks semi colon will be end of statement even if you escape semicolon with back slash(\). To insert values with semicolon use unicode for semicolon \u003B in your insert values statement and back slash to escape /,space,). Insert statement:- hive> insert into semicolon values ('Mozilla\/5\.0\ \(iPhone\u003B\ CPU\ iPhone\ OS\ 5_0\)'); hive> select * from semicolon; +------------------------------------------+--+ | a | +------------------------------------------+--+ | Mozilla/5.0 (iPhone; CPU iPhone OS 5_0) | +------------------------------------------+--+ (or) keep your data file into HDFS directory and create semicolon table with string datatype, pointing to that HDFS directory. Semicolon Table reading from HDFS directory:- hive> select * from semicolon; +------------------------------------------+--+ | a | +------------------------------------------+--+ | Mozilla/5.0 (iPhone; CPU iPhone OS 5_0) | +------------------------------------------+--+ Results will be same from both ways. ... View more