The fastest and most consistent way to launch DivvyCloud is to use Docker Compose. This quick start guide provides instructions on how to use Docker Compose to launch DivvyCloud and all of its service dependencies as containers in less than 15 minutes.
This quickstart method is intended for trial environments, not production environments. If you wish to set up a production environment, please refer to our enterprise set-up documentation.
Instructions for installing on macOS and windows can be found in the Docker Installation Guide.
NOTE: docker-compose in automatically included with their application.
In the remaining documentation you will see docker-compose invoked with its full path.
this can be simplified on macOS by only using the command name.
# from
sudo -E /usr/local/bin/docker-compose up
# to
docker-compose upDivvyCloud supports two primary Linux distributions. They are:
- Ubuntu 14.04+
- CentOS 6+ (see note at end of instructions)
DivvyCloud requires an instance with at least:
- 4 cores
- 6 Gb of memory
- 20 Gb root volume
If using AWS, we recommend using a m4.xlarge instance, which has 4 cores and 16 Gb, and attaching a volume with 30 Gb.
First, you need to install Docker 2.2 and Docker-Compose to use them for your installation. (For more information about Docker-Compose, please see https://docs.docker.com/compose/install/)
After logging into your DivvyCloud instance, run the following commands to install Docker and Docker-Compose:
sudo curl -sSL https://get.docker.com/ | sh
sudo curl -L "https://github.com/docker/compose/releases/download/1.14.0/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
sudo chmod +x /usr/local/bin/docker-composeAfter installing Docker and Docker-Compose, you will need to download the DivvyCloud repository from GitHub:
git clone https://github.com/DivvyCloud/QuickStart.git divvycloudWhen registering your installation, DivvyCloud checks the values of certain registration variables, specifically your company information, your name, and your email. For example, Jane Doe at Acme Corporation would use the following export variables:
export COMPANY_NAME="Acme Corporation"
export CONTACT_NAME="Jane Doe"
export CONTACT_EMAIL="jane.doe@acmecorp.com"Now you are ready to start DivvyCloud. You can run DivvyCloud in the foreground using the first command or in the background by using the second command. Both commands assume you are in the 'divvycloud' directory.
To run DivvyCloud in the foreground and see logging information in your terminal, use:
sudo -E /usr/local/bin/docker-compose upor to run DivvyCloud in the background, use:
sudo -E /usr/local/bin/docker-compose up -dAfter DivvyCloud has completed its launch, you can configure your DivvyCloud installation by connecting to your instance using a browser via its public IP address, e.g., http://[ip_address_of_your_DivvyCloud_installation]/
NOTE: If you are running docker-compose locally, you can connect via http://localhost:8001.
The first page you will see is an administrator account creation page. On the page, you will enter your name and email address and create a userid and password. Your userid can be your email address, but it doesn't have to be. Your password must be at least 12 characters in length.
After you have created your administrator account, you will be asked to log into DivvyCloud with your administrator account credentials. When you do, you can add your cloud accounts to start harvesting data and explore bot templates to select bots to customize and/or activate (see http://docs.divvycloud.com under "Cloud Support" and "DivvyCloud => Templates Listing" respectively.)
To stop DivvyCloud, use the following from the 'divvycloud' directory:
sudo -E /usr/local/bin/docker-compose down -dTo upgrade DivvyCloud, use the following from the 'divvycloud' directory:
sudo -E /usr/local/bin/docker-compose pull
sudo -E /usr/local/bin/docker-compose down
sudo -E /usr/local/bin/docker-compose up -dDivvyCloud can be started with plugins to customize your installation in several ways.
In the context of docker-compose, we can mount a directory called plugins into the interface server container when it starts. Reloading plugin changes requires restarting the whole suite as described above.
Mounting volumes can be done by passing arguments when starting a container. However,
since we're using the compose file it's best to add a single line to the docker-compose.yml
file.
interfaceserver:
image: divvycloud/botfactory
environment:
VIRTUAL_ENV: /
VIRTUAL_HOST: "*"
entrypoint:
- /entrypoint.sh
links:
- redis:redis
- mysql:mysql
ports:
- 8001:8001/tcp
command:
- ./uwsgi.sh
volumes: <-- this is new!
- ./plugins:/plugins <-- this is new!
Add a directory called plugins in the cloned repo directory.
An example plugin which will render a new page in the application can be found on GitHub called
SkeletonPlugin. Clone this plugin into the
plugins directory you just made.
cd plugins
git clone https://github.com/DivvyCloud/SkeletonPlugin.git helloworld
Restart the suite.
Logs will be placed in the directory where docker-compose up is invoked from
and are named by application process and date.
# Example
./logs/DivvyInterfaceServer-2016-11-02-0.log
If you would like to use Docker as a non-root user, you will need to add your user, e.g., jane_doe, to the "docker" group with something like:
sudo usermod -aG docker jane_doeThen, you will need to log out and back in for the change to take effect.
Of note, adding a user to the "docker" group will grant that user the ability to run containers, which can be used to obtain root privileges on the docker host. See https://docs.docker.com/engine/security/security/#docker-daemon-attack-surface for more information.
SE Linux prevents Docker from writing MySQL data to the host system. The workaround is to run this command from the 'divvycloud' directory:
chcon -Rt svirt_sandbox_file_t data