Can you share your hive version?
On Sat, Jun 8, 2013 at 12:03 AM, Stephen Sprague wrote:
> I would venture to say if you haven't got a reply nobody particularly has
> anything useful to add.
>
> you have a meta data error there. the error message shows you the table
> name. You're going to hav
I just took a look at the error stack and hive schema definition.
>From error, it looks like you are hitting the char length limit on postgres
schema table SERDE_PARAMS.
if your values for the column on postgres are more than 4000 chars then I
would recommend you alter your postgres meta data tab
Hi,
"if your values for the column on postgres are more than 4000 chars then I
would recommend you alter your postgres meta data table to have a bigger limit."
How to do the above?
Actually, while looking at the logs, I found out that
/usr/lib/hive/scripts/metastore/upgrade/postgres location,
Hi,
I have a table stored as SEQUENCEFILE in hive-0.10,* facts520_normal_seq*
Now, i wish to create another table stored as a SEQUENCEFILE itself, but
compressed using the Gzip codec.
So, i set the compression codec and type as BLOCK and then executed the
following query:
*SET hive.exec.compres
You should consult the PostgreSQL documentation:
http://www.postgresql.org/docs/9.1/static/sql-altertable.html
http://www.postgresql.org/docs/9.1/static/datatype-character.html
Essentially, connect to your PostgreSQL instance as a super-user and issue:
ALTER TABLE serde_params ALTER COLUMN para
Hi,
I'm using hive 0.10.0 over hadoop 1.0.4.
I have created a couple of test tables and found that various join queries
that refer to virtual columns fail. For example the query:
SELECT * FROM a JOIN b ON b.rownumber = a.number;
works but the following three queries all fail.
SELECT *,a.BLOCK
Documentation is here
https://cwiki.apache.org/confluence/display/Hive/CompressedStorage. Performance
overhead is trivial for larger amounts of data but may be magnified as data
size gets smaller. Typically where you gain is data transfers between nodes and
disk reads/writes. Again, the larger
Hi,
Using hive 0.10.0 over hadoop 1.0.4
I have a (non-partitioned) table with loads of columns.
I would like to create a partitioned table with the same set of columns.
So the approach that I have been taking is to use "CREATE TABLE copy LIKE
original;"
then I can use ALTER TABLE to change the l
If a table is not partitioned and then you want to partition the table on
the data already written but data is not in partition format, that is not
doable.
Best approach would be, create a new table definition with the partition
columns you want.
turn on the dynamic partitioning system before you
Hi,
Can you please point to documentation on Dynamic partitioning.
I dont fully understand meaning of values for these parameters.
Regards
Richa
On Mon, Jun 10, 2013 at 7:08 PM, Nitin Pawar wrote:
> If a table is not partitioned and then you want to partition the table on
> the data already
You need to create the partitioned table and then copy the rows into it.
create table foo_staging (int x, int y);
create table foo(int x) partitioned by (int y) clustered by (x) into 16
buckets;
set hive.exec.dynamic.partition=true;
set hive.exec.dynamic.partition.mode=nonstrict;
set hive.enforc
Hi,
On running the following query I am getting multiple records with same
value of F1
SELECT F1, COUNT(*)
FROM
(
SELECT F1, F2, COUNT(*)
FROM TABLE1
GROUP BY F1, F2
) a
GROUP BY F1;
As per what I understand there are multiple number of records based on
number of reducers.
Replicating the test
You might be hitting into
https://issues.apache.org/jira/browse/HIVE-4033in which case its
recommended that you upgrade to 0.11 where in this bug is
fixed.
On Mon, Jun 10, 2013 at 1:57 AM, Peter Marron <
peter.mar...@trilliumsoftware.com> wrote:
> Hi,
>
> ** **
>
> I’m using hive 0.10.0 ove
--Original Message--
From: Chris Kudelka
To: user@hive.apache.org
ReplyTo: user@hive.apache.org
Subject: unsubscribe
Sent: 10 Jun 2013 4:53 PM
Sent from my BlackBerry® wireless handheld from Glo Mobile.
On Mon, Jun 10, 2013 at 12:48 AM, Sachin Sudarshana wrote:
what's the header of the first sequence file look like?
*$ dfs -cat
/user/hive/warehouse/facts_520.db/test3facts520_gzip_seq/00_0 | head
*
Sachin, it works
SET hive.exec.compress.output=true;
SET mapred.output.compression.codec=org.apache.hadoop.io.compress.GzipCodec;
SET mapred.output.compression.type=BLOCK;
create table data1_seq STORED AS SEQUENCEFILE as select * from date1;
hadoop fs -cat /user/hive/warehouse/data1_seq/00_0
Sachin, it works
SET hive.exec.compress.output=true;
SET mapred.output.compression.codec=org.apache.hadoop.io.compress.GzipCodec;
SET mapred.output.compression.type=BLOCK;
create table data1_seq STORED AS SEQUENCEFILE as select * from date1;
hadoop fs -cat /user/hive/warehouse/data1_seq/00_0
Unsubscribe
Sent from my BlackBerry® wireless handheld from Glo Mobile.
-Original Message-
From: Alexander Pivovarov
Date: Mon, 10 Jun 2013 11:37:30
To:
Reply-To: user@hive.apache.org
Subject: Re: Sequence file compression in Hive
Sachin, it works
SET hive.exec.compress.output=true;
S
I got it. Makes total sense!
Thank you for your explanation.
On Fri, 2013-05-31 at 20:04 +, Sanjay Subramanian wrote:
> Hi
> Hive reads and writes to HDFSŠand by definition HDFS is write once and
> immutable after that.
> So like an RDBMS there is no concept of an update rows.
> However if u
Hello list,
I have a file stored in my HDFS which contains some urls. File
looks like this :
abc.in
xyz.net
http://tariq.com
http://tariq.in/sompath
And i'm trying to get the hostnames from these urls using *parse_url*. It
works fine except for the urls which do not contain any scheme. S
It is not a valid URL if it does not have a scheme and can not be parsed.
SELECT if (column like 'http%', column, concat( 'http://', column) ) as
column might do what you need.
On Mon, Jun 10, 2013 at 5:59 PM, Mohammad Tariq wrote:
> Hello list,
>
> I have a file stored in my HDFS whi
Hello Edward,
Thank you so much for the quick response. I'll try it out. But I
would like to know, is it something Hive specific?Links do work without a
scheme, like *hive.apache.org*.
Thank again.
Warm Regards,
Tariq
cloudfront.blogspot.com
On Tue, Jun 11, 2013 at 3:40 AM, Edward Capri
23 matches
Mail list logo
