Prerequisites

Tested on Ubuntu 16.04. Need ubuntu 16.04 update 4 or above

Install git

sudo apt-get install -y git

Install Docker CE https://docs.docker.com/install/linux/docker-ce/ubuntu/#set-up-the-repository

Create Docker Swarm

sudo docker swarm init --advertise-addr <ip>

is one of the ip on this Linux For example, for Virtualbox it can be host-only or bridged ip or docker0 ip

To list node joined:

sudo docker node ls

Get the App

Get HPCC-Docker-Stack to your host Linux:

git clone https://github.com/hpcc-systems/hpcc-docker

Run HPCC Cluster

Start a HPCC Cluster

cd to HPCC-Docker-Stack and run

sudo docker stack deploy -c docker-stack.yml hpcc

"hpcc" is the app name. You can give other name.

Wait for all containers start...

sudo docker service ls

All service with prefix "hpcc_" should be started by checking "REPLICAS" column.

To configure the cluster to go bin/:

./cluster_config.sh

If it deploys successfully it should print all HPCC node's running status To access ECLWatch: http://localhost:8010

There are several help scripts under bin/

cluster_config.sh

Configure HPCC Cluster, setup ansible hosts file and stop/start HPCC cluster

cluster_query.sh

Query entries of Docker/HPCC cluster For example list all components defined in environmentx.ml:

./cluster_query -q comp -c .*

Get node ips of HPCC cluster:

./cluster_query -q ip -g HPCC

Get docker container id from ip

./cluster_query -q id -p <ip>

Get docker container id for admin node

./cluster_query -q id -g admin

You can use to access the container:

sudo docker exec -it <id> /bin/bash

Just type "exit" to return the host

cluster_run.sh

This script is for start/stop/restart HPCC as well as query status. It also can start configmgr.

To start/stop/restart or show status HPCC cluster:

./cluster_run.sh [start|stop|restart|status]

To performan above action for particlar node give contianer id or ip:

./cluster_run.sh -p <ip> [start|stop|restart|status]

./cluster_run.sh -id <id> [start|stop|restart|status]

To performace an action for a component:

./cluster_run.sh -p <ip> -c <comp name> [start|stop|restart|status]

For example

./cluster_run.sh -p <ip> -c mydali status

To start configmgr:

./cluster_run.sh -c configmgr start

Type Ctrl C to stop it

Replace "hpcc" with your app name if is different.

cluster_env.sh

This will : stop HPCC cluster transfer /etc/HPCCSystems/source/environment.xml on admin node to all HPCC cluster nodes start HPCC cluster

Stop a HPCC Cluster

cd to HPCC-Docker-Stack and run

sudo docker stack rm hpcc