Hi Megh, Thanks for the logs. I found a discrepancy in HDP logs,as per "Query counters on HDP Cluster.txt" ,the Mapper spawned is 64 but from the application_1627033595344_12159_hdp.log,I observe it is 127. VERTICES MODE STATUS TOTAL COMPLETED RUNNING PENDING FAILED KILLED ---------------------------------------------------------------------------------------------- Map 1 .......... container SUCCEEDED 64 64 0 0 0 0 Reducer 2 ...... container SUCCEEDED 1 1 0 0 0 0 ---------------------------------------------------------------------------------------------- grep "Event:TASK_STARTED]: vertexName=Map 1" syslog_dag_1627033595344_12159_1 | wc -l 127 grep "Event:TASK_FINISHED]: vertexName=Map 1" syslog_dag_1627033595344_12159_1 | wc -l 127 Question: have you provided correct application log for HDP ? As I see a discrepancy. Assuming you have provided correct set of logs for HDO,I compared between CDP and HDP,I see that input records for each maper is more in CDP then HDP that is the reason it took more tie in CDP. I have collected samle taskID and INPUT_RECORDS_PROCESSED for each maper. application_1627885418606_0016_cdp.log ====== each Mapper took close to 1 min to finish taskId=task_1627885418606_0016_1_00_000059 INPUT_RECORDS_PROCESSED=2012469 taskId=task_1627885418606_0016_1_00_000061 INPUT_RECORDS_PROCESSED=2010369 taskId=task_1627885418606_0016_1_00_000026 INPUT_RECORDS_PROCESSED=2021105 application_1627033595344_12159_hdp.log ========== taskId=task_1627033595344_12159_1_00_000062 INPUT_RECORDS_PROCESSED=1010250 taskId=task_1627033595344_12159_1_00_000126 INPUT_RECORDS_PROCESSED=376062 Thanks, Asish ... View more

I would need complete application logs,HS2 logs and beeline trace to analyse. The snippet provided does not give much info. ... View more

Are you checking from beeline/any other JDBC client ?   You are seeing this because your window is not properly formatted. ... View more

Hi Sai,   Please follow https://community.cloudera.com/t5/Support-Questions/Does-hive-support-Photo-or-images-datatypes/td-p/221473   Copy the data to hdfs instead of loadpath and check,   Thanks, Asish ... View more

Please provide the below information:    1. What is the version? HDP/CDP/CDH ?    2. provide " scriptpro"    3. What si the error recieved ? ... View more

Can you provide the below details:    1. hiveserver2 logs  2, beeline trace 3. application logs 4. Version of HDP/CDP ... View more

Are you not able to connect to hiveserver2 from beelne ?   You are not able to view tables in the databases? Please provide a screenshot and also the beeline trace and url. ... View more

hi megh   Can you  share application logs and beeline trace .   Thaanks, Asish ... View more

Please provide driver DEBUG   To enable logging in the driver please modify the odbc.ini file and under the [Driver] section set LogLevel to 6 and LogPath to point to a writeable directory where you would like to driver to create the log files in. For example: [Driver] LogLevel=6 LogPath=/tmp   ... View more

By default hive displays in UTC. If you want to use specific timestamp,you can run below command   SELECT from_utc_timestamp(cast(from_unixtime(cast(1623943533 AS bigint)) as TIMESTAMP),"Asia/Kolkata") ; ... View more

I believe this issue is the result of upgrade.   Did you upgrade from 3.0.1 to 3.1.0 ?   In which version you were using Embedded mode ? Can you recreate the metadata if this is not production and check ? ... View more

I see the two large tables are scanned and it is done at Map 1 phase  and scanned around 9,569,213,817 records.   predicate:(concat(trim(col1), trim(col2)) is not null and CAST( date_partition_col AS DATE) BETWEEN DATE'2021-02-04' AND DATE'2021-05-10') ==> scanning entire large table   Can you try below one: create table mydb.mytbl as select col1,col2,...col18 where here date_partition_col between date_sub('2021-05-10',95) and '2021-05-10 from from mydb.tbl1 a inner join mydb.tbl2 b on concat(trim(col1),trim(col2))=colb;   ... View more

@Raul23 Is  datagrip a Java or non Java application? If it is a Java application. you need to use JDBC driver and use ODBC driver,if it is a non-java application. Download ODBC driver from https://www.cloudera.com/downloads/connectors/hive/odbc/2-6-9.html Driver JDBC drivers from https://www.cloudera.com/downloads/connectors/hive/jdbc/2-6-13.html   Please select the latest driver. Please follow https://docs.cloudera.com/HDPDocuments/other/connectors/hive-jdbc/2.6.7/Simba-Hive-JDBC-Install-and-Configuration-Guide.pdf to configure ... View more

can you collect STATS and re-run the query?? ... View more

Can you please try the below command:    select count(*) from   data2;    It should return one single row . ... View more