TeraFlow Traffic Engineering Service

This service is mean as an example of a Teraflow Service made in Erlang.

The Traffic Engineering service is tested on Ubuntu 20.04. Follow the instructions below to build, test, and run this service on your local environment.

Build

First the TeraFlow protocol buffer code must have been generated:

$ ../../proto/generate_code_erlang.sh

Then the TE service can be built:

$ rebar3 compile

Execute Unit Tests

$ rebar3 eunit

Run Service Console

First you need to crete a configuration file if not already done, and customize it if required:

$ cp config/dev.config.template config/dev.config

Then you can start the service in console mode:

$ rebar3 shell

Docker

Build Image

The docker image must be built from the root of the Teraflow project:

$ docker build -t te:dev -f src/te/Dockerfile .

Run a shell from inside the container

$ docker run -ti --rm --entrypoint sh te:dev

Run Docker Container

$ docker run -d --name te --init te:dev

Open a Console to a Docker Container's Service

$ docker exec -it te /tfte/bin/tfte remote_console

Show Logs

$ docker logs te

Kubernetes

Open a Console

$ kubectl --namespace tfs exec -ti $(kubectl --namespace tfs get pods --selector=app=teservice -o name) -c server -- /tfte/bin/tfte remote_console

Show Logs

$ kubectl --namespace tfs logs $(kubectl --namespace tfs get pods --selector=app=teservice -o name) -c server

Teraflow

To build and deploy the TE service as part of Teraflow, the following line must be added or uncomented in your my_deploy.sh:

export TFS_COMPONENTS="${TFS_COMPONENTS} te"