hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sergio Pena <>
Subject Fwd: How is LOCATION used in CREATE INDEX ?
Date Fri, 27 May 2016 17:03:03 GMT
Hi All,

Anyone knows what is the use of the LOCATION keyword when creating an index?

There is an issue with CREATE INDEX when we specify the LOCATION to a table
location, and then rebuild the index. It writes unwanted data to it.

Also, if we use '/user/hive/warehouse' in the LOCATION, and then we DROP
INDEX, then the location is deleted.

I haven't found if the LOCATION is to store indexes data, or it is a table
location. When looking at the AST, a TOK_TABLELOCATION is added. This seems
like a table location, right?

hive> select * from encrypted;
1       sergio

hive> create index enc_index on table encrypted(id) as
'org.apache.hadoop.hive.ql.index.compact.CompactIndexHandler' with deferred
rebuild location '/user/hive/warehouse/encrypted';

hive> alter index enc_index on encrypted rebuild;

hive> select * from encrypted;
1       hdfs://localhost:9000/user/hive/warehouse/encrypted/000000_0

- Sergio

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message