[ https://issues.apache.org/jira/browse/CARBONDATA-1460?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Sangeeta Gulia closed CARBONDATA-1460. -------------------------------------- Resolution: Invalid > Drop column in alter table working incorrectly when connected to same thrift > using different beeline sessions > ------------------------------------------------------------------------------------------------------------- > > Key: CARBONDATA-1460 > URL: https://issues.apache.org/jira/browse/CARBONDATA-1460 > Project: CarbonData > Issue Type: Bug > Components: data-query > Affects Versions: 1.2.0 > Environment: Spark-2.1 > Reporter: Sangeeta Gulia > Assignee: anubhav tarar > Priority: Minor > > I am trying to do concurrent testing on the same table. For that, I have > started my thrift server with the given command: > sudo /home/hduser/spark-2.1.0-bin-hadoop2.7/bin/spark-submit --master > spark://host-name:7077 --class > org.apache.carbondata.spark.thriftserver.CarbonThriftServer > carbondata_2.11-1.2.0-SNAPSHOT-shade-hadoop2.2.0.jar <carbon-store-path> > and two different nodes are connecting to same thrift using two beeline > sessions. > While beeline1 session executes a query : alter table uniqdata drop > columns(cust_id); > Then after execution of the query, beeline1 session gets error: > Error: org.apache.spark.sql.AnalysisException: cannot resolve '`cust_id`' > given input columns: [bigint_column1, double_column1, dob, doj, > active_emui_version, decimal_column2, bigint_column2, integer_column1, > cust_name, double_column2, decimal_column1]; line 1 pos 7; > While beeline2 session is able to get data for cust_id from that table using > the below command: > select cust_id from uniqdata; > But, when both the beeline session try to run "describe table uniqdata" they > see the same result which does not include cust_id as a column in the table. -- This message was sent by Atlassian JIRA (v6.4.14#64029)