Naresh P R created CARBONDATA-925:
-------------------------------------
Summary: CarbonEnv is static & shared among all the Sessions.
Cached relation in 1 session is not getting refreshed when another session is
adding/dropping column
Key: CARBONDATA-925
URL: https://issues.apache.org/jira/browse/CARBONDATA-925
Project: CarbonData
Issue Type: Bug
Environment: Spark 2.1
Reporter: Naresh P R
WIth multiple beeline connects in single thriftserver, CarbonEnv object is
static. When user performs alter operation in 1 beeline, latest updated schema
information is not reflecting 2nd beeline.
Assume following concurrent operations
1) Beeline => select query on table1 (CarbonDatasourceRelation is cached)
2) spark-sql => add new column on table1 (this will refresh carbonMetastore
available in carbonenv)
3) Beeline => Select query on table1 (This operation fails because it does not
know whether the carbonMetastore is refreshed, hence cached
CarbonDatasourceRelation will be returned which will not have newly added
columns in schema fields)
Corrected this problem by changing CarbonEnv Object to class which will be
instantiated for each Session
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
