describe table issue, hive_system table, set command and version fuction questions

[Todd Wilson <to...@coffingdw.com>]

Hello Everyone:

 

I'm using Hive .9.  MapR ODBC driver against Cloudera.  Relatively new to
Hadoop, been looking at it about a week to add it to our query tool.  

 

CREATE DATABASE test;


CREATE TABLE test.table1
      (col1  int);

describe table1;

I'm getting the message "Table table1 does not exist .  I'm trying to get
the columns for this table which is in another database other than default.
I understand version .10 has SHOW COLUMNS sysntax to get columns in any
database, but is there a way to do this on .9. 

 

describe test.table1;
I don't think this is valid syntax so I was expecting an error message.
Give "Table test does not exist".

 

-----

What is this hive_system database and hive_system table?  Is this a Cloudera
table?  I can seem to find any information on it anywhere.

A SELECT from this table will give configuration information.

 

-----

SET command appears to run like this (not documented in HQL as far as I can
tell).

 

set hive.enforce.bucketing = false;

 

.but the configuration setting doesn't change.  I thought you could override
some of these configuration variable at the session level and I'd like to
learn how to do this.

 

When using set -v through the CLI it doesn't seem to return this
hive.enforce.bucketing configuration setting.

 

-----

Finally, a command like SELECT Version() exposed through HQL to show the
current version of hive would be very helpful.  I can get this by querying
this hive_system table and doing some string functions in .NET to get it,
but if there is something like select version that would be great.

 

Thank you for listening.

 

Best Regards,

 

Todd