opsforge ANViL

The opsforge anvil (a renaming of the opsforge stack 2.0) is an all-in-one development solution, that aims to deliver a development platform that is:

• using browser accessible web-apps (no terminal emulator or IDE needed / even on Windows)
• built on docker-native and container-based services
• structured and configured by docker-compose
• relying on OSS-only components
• swarm-compatible for scale-out needs (e.g larger dev environments)
• has in-built logging and audit (for collaboration and record keeping)
• highly customisable (modular)
• easly reproducable and replacable
• version controlled
• platform and infrastructure agnostic
• can run on localhost and AWS the same way

Project names and images	Project Health	Comments
opsforge/anvil	Build health: Code health:	Ubuntu container with tools and shell
docker.elastic.co/logstash/logstash:6.0.1 docker.elastic.co/elasticsearch/elasticsearch:6.0.1 docker.elastic.co/kibana/kibana:6.0.1	N/A	Logging and collector engine
opsforge/splunk-light	N/A	Logging and collector engine

Components:

Name	Purpose	Accessibility
(plugin) Cloud9 IDE	Development interface	Browser window
anvil	ZSH (shell) through Butterfly, pre-built CLIs and git control	Browser window or Docker shell
(plugin) Apache Guacamole	Remote desktop, VNC, TTY, etc access (optional)	Browser window
(plugin) portainer	Docker engine access and container access	Browser window or API
(plugin) ELK	Audit and shell logging (optional)	Browser window or API
(plugin) Splunk Light (C)	Audit and shell logging (optional)	Browser window or API

Deployment

Pre-deployment steps

It's recommended to create a new folder in the destination directory first. The name of the folder can be anything the OS supports.

If using ELK

1. cd into the folder you created
2. mkdir -p ./es_data
3. chmod -R 777 ./es_data

This will resolve any issues with ownership when the ES container is starting up.

If using Splunk Light

1. cd into the folder you created
2. mkdir -p ./splunk-data && touch ./splunk-data/splunk-launch.conf

This will resolve launch issues with the Splunk Light container.

Deploying the stack

The recommended way to deploy anvil is through docker-compose. Make sure your machine has the latest version of docker engine and docker-compose installed. You can check from your command line using docker -v and docker-compose -v

Either clone this project repository to your machine using git or just download the docker-compose.yaml file from the project root to a folder on your machine, cd into that folder and type docker-compose -p anvil up -d

Removal

To delete the stack, simply cd into your project folder and type docker-compose -p anvil down. Make sure you remove content from any persisted data mounts manually.

Usage

hostname will be either localhost or if you deployed to another machine, the DNS record / IP for that.

Service	Address
Splunk Light	http://localhost:8000
anvil (core)	http://localhost:8001
Cloud9	http://localhost:8002
Apache Guacamole	http://localhost:8003/guacamole/
portainer	http://localhost:9000
Kibana	http://localhost:5601
Logstash (for sending logs and metrics in)	http://localhost:5000

Customization

You are free to comment out / add in parts of the docker-compose to disable or remove services on startup. You can apply your changes with a simple docker-compose -p anvil up -d --remove-orphaned on an active deployment.

You can also make use of portainer on port 9000. After logging in and connecting in to the local docker engine, you can stop / start / create / remove containers on-demand through the web UI or its API.

WIP (to-be-done)

• Guacamole doesn't play nice with private DNS servers (public records seem to be fine)
• Desktop copy/paste doesn't seem to work in Guacamole and drive sharing is flaky
• anvil container doesn't include a puppet cli yet
• Logstash doesn't support multiline TCP intake since its previous version update. A line sent to the listener remains a single line.
• Create a read-the-docs page for the project
• YouTube demo video