Fwd: Kylin: sample data in docker

[hongbin ma <su...@gmail.com>]

forward to mail list  for further discussion.

---------- Forwarded message ----------
From: Daniel Carroza <dc...@stratio.com>
Date: Wed, Feb 11, 2015 at 5:38 PM
Subject: Kylin: sample data in docker
To: subuliu@gmail.com


Hi Hongbin,

I'm testing Kylin with the docker containers you provided and sample data
is not loaded. I'm following the tutorial in
https://github.com/sequenceiq/docker-kylin.

Another question: Is there any problem if I deploy the container like this?

kylin-deploy-cluster 1


kind regards


Daniel Carroza


<http://www.stratio.com/>
Vía de las Dos Castillas, 33, Ática 4, 3ª Planta.
28224 Pozuelo de Alarcón. Madrid.
Tel: +34 91 828 64 73 // *@stratiobd <https://twitter.com/StratioBD>*

Re: Kylin: sample data in docker

["Shi, Shaofeng" <sh...@ebay.com>]

Hi Daniel,

Thanks for evaluating Kylin.

Let me answer your second question at first, which is relatively simple;
"kylin-deploy-cluster 1² is good to use (and I also prefer this). If you
give a bigger number like 3, it will start a hadoop cluster with 3 nodes;
while the Kylin is only started in the first container (³amb0²); So, it is
good to go with 1 node if the data set is small.

Okay now let¹s talk about your first question. I think there was a problem
in the initial docker image, which caused the sample data was not
populated automatically. Now we have a new version v0.6.4 (hasn¹t been
updated in docker hub); You can build the new image easily with the new
docker file:

1. Git clone https://github.com/sequenceiq/docker-kylin.git, and then
checkout the 0.6.4 branch
2. Build the docker image in your local with docker build command, for
example: docker build -t sequenceiq/kylin:0.6.4 .
3. After the image be built, you can see it with: docker images
4. Manually edit the "ambari-functions² file in this git project, in line
4 change the image name to the name of the one you just build.
5. Source the "ambari-functions² file you just updated.
6. kylin-deploy-cluster 1
7. When Kylin is started, logon with ADMIN/KYLIN, you should see the
sample cubes like ³test_kylin_cube_without_slr_empty" in the web UI. Then
select a cube and click ³Build², this will trigger the hadoop jobs.
8. When the cube build is finished, you would see its status changed to
³active²; Repeat until all cubes be built. Finally you can write some SQLs
and query them against Kylin.

I have contact sequenceiq, ideally the new image will be published to
docker hub soon. Later we will have an image built with Kylin binary
package, that would accelerate the docker deploy. If you have any
comments, just drop email to dev@kylin.incubator.apache.org
<de...@kylin.incubator.apache.org>.

Shaofeng, Shi


On 2/11/15, 6:43 PM, "hongbin ma" <su...@gmail.com> wrote:

>forward to mail list  for further discussion.
>
>---------- Forwarded message ----------
>From: Daniel Carroza <dc...@stratio.com>
>Date: Wed, Feb 11, 2015 at 5:38 PM
>Subject: Kylin: sample data in docker
>To: subuliu@gmail.com
>
>
>Hi Hongbin,
>
>I'm testing Kylin with the docker containers you provided and sample data
>is not loaded. I'm following the tutorial in
>https://github.com/sequenceiq/docker-kylin.
>
>Another question: Is there any problem if I deploy the container like
>this?
>
>kylin-deploy-cluster 1
>
>
>kind regards
>
>
>Daniel Carroza
>
>
><http://www.stratio.com/>
>Vía de las Dos Castillas, 33, Ática 4, 3ª Planta.
>28224 Pozuelo de Alarcón. Madrid.
>Tel: +34 91 828 64 73 // *@stratiobd <https://twitter.com/StratioBD>*