The uri provided in your log is wrong, check
http://10.40.35.54:9103/tasklog?attemptid=attempt_201209171421_0029_m_00_0&start=-8193
Shouldn't this misleading uri issue be a bug?
On Tue, Sep 18, 2012 at 11:14 PM, Aniket Daoo wrote:
> Hi,
>
>
>
> I have a ROW FORMAT SERDE table created using th
A presentation on Hive vs Pig by a committer on both projects is here:
http://hortonworks.com/blog/hadoop-features-large-at-stanford-xldb/
Russell Jurney
twitter.com/rjurney
russell.jur...@gmail.com
datasyndrome.com
On Sep 18, 2012, at 3:01 PM, Aniket Mokashi wrote:
(Probably not what you are l
I have created HIVE-3465 for this issue.
On Mon, Sep 17, 2012 at 1:17 AM, MiaoMiao wrote:
> Yes, same problem here, sorry I missed your point and tested only once.
>
> On Fri, Sep 14, 2012 at 10:13 PM, Kaufman Ng wrote:
>> Did you run the insert into query twice? And did it append?
>>
>> From w
Here's the fix: just fix the hive-log4j.properties in your hive install.
replace this line:
log4j.appender.EventCounter=org.apache.hadoop.metrics.jvm.EventCounter
with this:
log4j.appender.EventCounter=org.apache.hadoop.log.metrics.EventCounter
On Mon, Sep 17, 2012 at 10:34 AM, Robin Verlangen
(Probably not what you are looking for) Check -
http://www.larsgeorge.com/2009/10/hive-vs-pig.html
~Aniket
On Fri, Sep 14, 2012 at 2:28 PM, Russell Jurney wrote:
> A detailed post comparing Pig/Hive performance from last week:
> http://hortonworks.com/blog/pig-performance-and-optimization-analys
I agree that we can use $.C.D & C.D directly but if the key is not present
and I have several such keys then it is better to not create so many
columns.
On Tue, Sep 18, 2012 at 5:47 AM, Connell, Chuck wrote:
> OK, but keep in mind that you could use
> get_json_object(whole-json-field, $.C.D). If
Hi,
I have a ROW FORMAT SERDE table created using the following DDL.
CREATE external TABLE multivalset_u6
(
col1 string,
col2 string,
col3 string,
col4 string,
col5 string
)
ROW FORMAT SERDE 'org.apache.hadoop.hive.contrib.serde2.RegexSerDe'
WITH SERDEPROPERTIES
(
"input.regex" = "(.*)\\t(.*)\\t~
OK, but keep in mind that you could use get_json_object(whole-json-field,
$.C.D). If using a serde, then you can say just C.D directly.
Chuck Connell
Nuance R&D Data Team
Burlington, MA
From: Mark Golden [mailto:mark.golden1...@gmail.com]
Sent: Monday, September 17, 2012 7:31 PM
To: Connell, Ch
Is there any solution for my issue?
Otherwise I will quit my hive exploration and move on with other product
for my requirement.
I thought of giving it a try instead of Pig as I was not comfortable in
writing complex data mining logic using Pig.
If there is no other way I can get Hive on my machin
Yes, it and all (hive-lib and hadoop jars) should be seen in class path.Please
set.
And if antlr 3.0.1 does not work, you will have to checkout source to be able
to build it with new antlr, otherwise compile and runtime would clash.
However I agree today there was some issue with hive trunk buil
10 matches
Mail list logo
