Bootstrap container with grafanalib

Author: Carlos León

.gitignore

@@ -127,3 +127,7 @@ docker-swarm.box
 *.tfvars
 join.sh
 output-virtualbox-iso
+
+# Ignore grafanalib output
+/graphs/__pycache__/
+/graphs/*.json

graphs/Dockerfile

@@ -0,0 +1,20 @@
+FROM python:3.4-alpine
+
+MAINTAINER Container Solutions [email protected]
+
+# Install basic dependencies
+RUN apk add -U curl git parallel
+
+# Silence parallel citation notice
+RUN mkdir /root/.parallel && \
+    touch /root/.parallel/will-cite
+
+# Install pip
+RUN curl https://bootstrap.pypa.io/get-pip.py | python
+
+# Install grafanalib
+RUN pip install git+https://github.com/weaveworks/grafanalib@82556ddfbbd6134837d280a7999d35c45cc3c87e
+
+WORKDIR /opt/code
+
+CMD ["/bin/sh", "-c"]

graphs/README.md

@@ -0,0 +1,30 @@
+# Grafana Dashboards
+We have included a set of dashboards for Grafana in this demo application.
+Most of the source of these dashboards look exactly the same. For the sake of
+keeping our code DRY, we have decided to generate our dashboards using [grafanalib](https://github.com/weaveworks/grafanalib).
+
+
+# Requirements
+If you don't have Python installed on your computer but do have Docker, then Docker and Make will
+make your life easier today:
+
+| what   | version   |
+| ------ | --------- |
+| make   | `>= 4.1`  |
+| docker | `>= 17`   |
+
+# Getting Started
+
+## The Base Image
+All the tooling required to generate the dashboards is inside a container. Build it with like this:
+
+```
+docker build -t weaveworks/grafanalib .
+```
+
+## Generating Dashboards
+Make sure that you run the following commands from within the `graphs/` directory:
+
+```
+docker run --rm -it -v ${PWD}:/opt/code weaveworks/grafanalib /bin/sh -c 'ls /opt/code/*.dashboard.py | parallel generate-dashboard -o {.}.json {}'
+```