Hi,
I am using kylin 1.6 and trying to define a topN measure on string column
(in SUM | ORDER BY field), which is giving me Number format exception. I
understand that kylin is expecting a numeric column to do sum operation on
it. But, I just wanted to ask that is there any way to define topN on
st
谢谢你的关注和说明。
我hive 版本是 2.1 而且是自己根据源码编译的，是否是版本问题？
错误日志如下:
Container: container_e05_1490775930310_0016_01_01 on SND-0081_54270
=
LogType:stderr
Log Upload Time:Wed Mar 29 03:43:15 -0500 2017
LogLength:760
Log Contents:
SLF4J:
Thanks ShaoFeng Shi!!
I will review your changes and apply it to:
https://github.com/albertoRamon/Kylin/tree/master/KylinWithSQuirreL
https://github.com/albertoRamon/Kylin/tree/master/KylinWithMain
And will send you a commit with these new tech notes
2017-03-29 10:16 GMT+01:00 ShaoFeng Shi :
Kylin supports Hadoop 2.6 and 2.7, as stated in
https://kylin.apache.org/docs20/install/hadoop_env.html
We never try Hadoop 1.x, because that is quite old. I'm not sure whether it
can run with that.
If you do need, at least a re-compile and package is needed (after changing
hadoop version in pom.x
Right, putting 125 dimensions on fact table to one cube is crazy; Usually
when dimension > 12, you need do some optimization to control the
combination numbers (by default combination number = 2^N, N is the number
of dimensions).
2017-03-29 15:04 GMT+08:00 Fan Xie :
>
> Hi suresh, currently kyli
Hi Alberto,
Sorry for the late response; I merged the patch today, and made minor
update on it for some wording. Now these two documents are on Kylin
document page. Let me know if have any question. Thanks for your
contribution!
2017-03-27 7:25 GMT+08:00 Alberto Ramón :
> Uppss "*ezmlm-reject:
To see actual jobs of YARN: localhost:8088
To see historical jobs (has been finished): localhost:19888
(check this ports with your Hadoop distribution and actual configuration)
2017-03-29 8:26 GMT+01:00 Bart :
> To see what YARN is saying I need to kill my process. It shows me the
> problem
> wi
To see what YARN is saying I need to kill my process. It shows me the problem
with memory, but I'm not sure whether it is True or just because killing the
process.
I've created a database with one table of fact and one dimension (with date
obły). I pull two "facts" inside table of fact and two dat
mr job failed， get more detail in your mr job history
中国移动广东有限公司 网管中心 梁猛
13802880...@139.com
发件人： kylin_liu
发送时间： 2017-03-28 17:53
收件人： dev
主题： KYLIN 测试 sample 的时候碰到这个坑，请教
执行到这步的时候出现异常，请问这个是怎么回事呢？是MR 问题，还是HBASE 或者 KYLIN 问题
2017-03-28 17:36:58 GMT+8
#17 Step Name: Convert Cuboid Data to HFile
<
Hi suresh, currently kylin could not handle fact table with 125 dimensions yet.
Normal we suggest users to build cubes with no more than 16 dimensions. You can
refer to this blog :
http://kylin.apache.org/docs20/howto/howto_optimize_cubes.html to learn how to
reduce the dimensions number of yo
10 matches
Mail list logo
