data-analytics-steps

0-a) Assuming that the current working directory is /home/mahmood/cloudsuite_v3/
0-b) Extract data-analytics.zip
unzip data-analytics.zip -d .

1) docker images
if the output is empty proceed to step3.

2) docker rmi -f IMAGE_IDS

3) docker load -i /home/mahmood/cloudsuite_v3/ubuntu.tar

4) docker inspect ubuntu:14.04
be sure that the CMD section is not empty (or is not NULL)

5) vim data-analytics/dataset/Dockerfile (The first line is very important)
FROM ubuntu:10.04
RUN apt-get update && apt-get install wget -y
# create data directory
RUN mkdir /data
RUN cd /data \
  && wget -nv http://parsa.epfl.ch/cloudsuite/software/enwiki-20100904-pages-articles1.xml.bz2
VOLUME ["/data"]


6) docker build -t data-analytics:dataset ./data-analytics/dataset
the last lines of a successful build looks like
Step 5 : VOLUME /data
 ---> Running in c1101481f462
 ---> 1db2326dc82b
Removing intermediate container c1101481f462
Successfully built 1db2326dc82b


7) docker create --name data data-analytics:dataset

8) put serf and jdk in the same directory as data-analytics/base
serf_0.5.0_linux_amd64.zip
jdk-7u79-b15-linux-x64.tar.gz

9) Modify base/Dockerfile (-- means delete line and ++ means add line)
-- RUN curl -Lso serf.zip https://dl.bintray.com/mitchellh/serf/0.5.0_linux_amd64.zip \
--    && unzip serf.zip -d /bin \
--    && rm serf.zip
++ COPY serf_0.5.0_linux_amd64.zip serf_0.5.0_linux_amd64.zip

++ RUN unzip serf_0.5.0_linux_amd64.zip -d /bin \

++    && rm serf_0.5.0_linux_amd64.zip

-- RUN cd $SRC_DIR && curl -LO "$JDK_URL/$JDK_VER/$JDK_VER2-linux-x64.tar.gz" -H 'Cookie: oraclelicense=accept-securebackup-cookie' \
--  && tar xzf $JDK_VER2-linux-x64.tar.gz && mv jdk1* $JAVA_HOME && rm -f $JDK_VER2-linux-x64.tar.gz \
++ COPY jdk-7u79-linux-x64.tar.gz jdk-7u79-linux-x64.tar.gz
++ RUN tar xzf jdk-7u79-linux-x64.tar.gz && mv jdk1* $JAVA_HOME && rm -f jdk-7u79-linux-x64.tar.gz \



10) docker build -t data-analytics:base ./data-analytics/base

11) docker run -d -t --dns 127.0.0.1 -P --name master -h master.cloudsuite.com --volumes-from data data-analytics:master

12) docker exec -it master bash
You should now see the master's prompt like this
mahmood@orca:cloudsuite_v3$ docker exec -it master bash
root@master:/# ls
bin  boot  data  dev  etc  home  lib  lib64  media  mnt  opt  proc  root  run  r
root@master:/# ls /opt
new_analytic  new_analytic.tar.gz
root@master:/# ./run.sh
Type the number of slave nodes, followed by [ENTER]:
......

13) Enter 0 as the number of slaves (the system is in single machine mode) and then it will continue to download the huge wikipedia file