A short guide for proxying to a different nomad cluster

when developing the UI

ui/DEVELOPMENT_MODE.md

@@ -0,0 +1,42 @@
+## Running the Web UI in development mode against a production Nomad cluster
+
+:warning: **Running the Web UI in development mode is only necessary when debugging
+issues. Unless you are debugging an issue, please only use the Web UI contained
+in the Nomad binary.** :warning:
+
+The production Web UI concatenates and minifies JavaScript and CSS. This can make errors
+cryptic or useless. In development mode, files are as expected and stack traces are useful.
+
+Debugging Web UI issues with the Web UI in development mode is done in three steps:
+
+  1. Cloning the Nomad Repo
+  2. Setting up your environment (or using Vagrant)
+  3. Serving the Web UI locally while proxying to the production Nomad cluster
+
+### Cloning the Nomad Repo
+
+The Web UI is part of the same repo as Nomad itself. Clone the repo
+[using Github](https://help.github.com/articles/cloning-a-repository/).
+
+### Setting up your environment
+
+The [Web UI README](README.md) includes a list of software prerequisites and instructions
+for running the UI locally or with the Vagrant VM.
+
+### Serving the Web UI locally while proxying to the production Nomad cluster
+
+Serving the Web UI is done with a single command in the `/ui` directory.
+
+  - **Local:** `ember serve`
+  - **Vagrant:** `ember serve --watch polling --port 4201`
+
+However, this will use the [Mirage fixtures](http://www.ember-cli-mirage.com/) as a backend.
+To use your own Nomad cluster as a backend, use the proxy option.
+
+  - **Local:** `ember serve --proxy https://demo.nomadproject.io`
+  - **Vagrant:** `ember serve --watch polling --port 4201 --proxy https://demo.nomadproject.io`
+
+The Web UI will now be accessible from your host machine.
+
+  - **Local:** [http://localhost:4200](http://localhost:4200)
+  - **Vagrant:** [http://localhost:4201](http://localhost:4201)