Reach3 docker environment

This is the development version meaning each container must be built from scratch, and each container has everything required to alter the code and compile it (inside a container).

Production images are made out of these development containers, see export

Notes

In order to build the environment it is required to have access to private Reach3 repo, therefore an access token must be provided by setting TOKEN env variable.

Reach3 instance must have following containers running:

1. reach
2. rr
3. freeswitch
4. timescale

If you want to have UI, then also one (or both, if you're curious) of:

1. reach-ui
2. reach-ui-jh

In order to run automated test following containers must be running (in addition to above):

1. busytone
2. freeswitch-agents

Optional containers (but highly recommended):

1. kamailio
2. nginx-ingress
3. grafana

Continous integration:

1. ezuce-ci

Architecture

All images are derived from base-os container, reach3 and busytone in turn are derived from erlang container.

Installation

You need to have Docker version at least 1.9.0 (as this setup relies on docker network heavily). Also you probably want to add your local user to docker group with usermod -aG docker $(whoami)

export TOKEN=... (see above)
./build.sh
./run.sh

# You know what are you doing, right?
./hosts.sh >> /etc/hosts

Use of environment variables

The intent is to specify docker build flags e.g. to pass --no-cache to forcefully rebuild.

BUILD_FLAGS -- flags to pass to `docker build` command.

Check also README.md in each folder to understand the particular container.

Production images

It is possible to create so-called production images out of development images. See export/README.md