faas/guide/deployment_digitalocean.md
Marko Mudrinić ca15008adc guide,do: switch to ubuntu-17-10-x64
Signed-off-by: Marko Mudrinić <mudrinic.mare@gmail.com>
2018-03-23 21:26:26 +00:00

162 lines
7.2 KiB
Markdown

# Deployment guide for DigitalOcean
In this guide we will be using the `docker-machine` tool to provision a number of Docker Swarm nodes then we'll connect them together and deploy OpenFaaS. Before you get started - sign up to [Digital Ocean here to get free credits](https://m.do.co/c/8d4e75e9886f). Once you've signed up come back to the tutorial.
### A foreword on security
These instructions are for a development environment. If you plan to expose OpenFaaS on the public Internet you need to enable basic authentication with a proxy such as Kong or Traefik at a minimum. TLS is also highly recomended and freely available with LetsEncrypt.org. [Kong guide](https://github.com/openfaas/faas/blob/master/guide/kong_integration.md) [Traefik guide](https://github.com/openfaas/faas/blob/master/guide/traefik_integration.md).
## Create DigitalOcean API Access Token
Follow the [DigitalOcean instructions here](https://www.digitalocean.com/community/tutorials/how-to-use-the-digitalocean-api-v2#how-to-generate-a-personal-access-token) to create a Personal Access Token with **Read** and **Write** permissions, give the token a descriptive name for example `openfaas-getting-started`.
Now set an environment variable with the new token value.
```
$ export DOTOKEN=738cb0cd2jfhu84c33hu...
```
> If you want to make this permanent, you can insert the value into your `~/.bash_profile` file.
## Install Docker Machine
Type in `docker-machine` to see if you already have the tool installed this is normally bundled with Docker for Mac/Windows. If not then you can download [Docker Machine here](https://docs.docker.com/machine/install-machine/).
## Create Docker Nodes
Use Docker Machine to create Docker hosts or nodes. On Digital Ocean your hosts or VMs (Virtual Machines) are called *Droplets* and will run a full version of Linux. Note: you'll be able to connect to any of your droplets with `ssh` later on.
The example below creates 3 droplets in the NYC3 zone, if you want to deploy only one Droplet change `"1 2 3"` to `"1"`.
This process will take a few minutes as Droplets are created and Docker installed.
```
for i in 1 2 3; do
docker-machine create \
--driver digitalocean \
--digitalocean-image ubuntu-17-10-x64 \
--digitalocean-tags openfaas-getting-started \
--digitalocean-region=nyc3 \
--digitalocean-access-token $DOTOKEN \
node-$i;
done
```
List the newly created Docker nodes.
```
$ docker-machine ls
NAME ACTIVE DRIVER STATE URL SWARM DOCKER ERRORS
node-1 - digitalocean Running tcp://104.131.69.233:2376 v17.07.0-ce
node-2 - digitalocean Running tcp://104.131.115.146:2376 v17.07.0-ce
node-3 - digitalocean Running tcp://159.203.168.121:2376 v17.07.0-ce
```
Refer to the [documentation](https://docs.docker.com/machine/drivers/digital-ocean/) for more detailed information on the DigitalOcean options for docker-machine.
# Create your Docker Swarm
A Docker Swarm can contain as little as a single master node and begins by running the `docker swarm init` command. It's important if you have more than one node that you specify an `--advertise-addr` value.
Intialize Docker Swarm on `node-1`.
```
$ docker-machine ssh node-1 -- docker swarm init --advertise-addr $(docker-machine ip node-1)
```
> If you opted to deploy a single node, then skip to the next section.
When deploying more than a single Docker host take a note of the command to add a worker to the Swarm. This output contains your *join token*.
> If you lose it you can get a new one any time with the command: `docker swarm join-token worker` or `manager`.
```
Swarm initialized: current node (je5vne1f974fea60ca75q2cac) is now a manager.
To add a worker to this swarm, run the following command:
docker swarm join --token SWMTKN-1-239v0epdnhuol2ldguttncoaleovy29hnwyglde0kba1owc9ng-9488z5we2invwcn69f5flq7uu 104.131.69.233:2377
To add a manager to this swarm, run 'docker swarm join-token manager' and follow the instructions.
```
Add `node-2` to the Swarm, using the `docker swarm join..` command returned when initializing the master.
```
$ docker-machine ssh node-2 -- docker swarm join --token SWMTKN-1-239v0epdnhuol2ldguttncoaleovy29hnwyglde0kba1owc9ng-9488z5we2invwcn69f5flq7uu 104.131.69.233:2377
```
Repeat for `node-3`.
```
$ docker-machine ssh node-3 -- docker swarm join --token SWMTKN-1-239v0epdnhuol2ldguttncoaleovy29hnwyglde0kba1owc9ng-9488z5we2invwcn69f5flq7uu 104.131.69.233:2377
```
## Configure Docker CLI to use remote Swarm
Run this command each time you open a new shell, this tells Docker where your remote Swarm is.
```
eval $(docker-machine env node-1)
```
## Deploy the OpenFaaS Stack
This command clones the OpenFaaS Github repository then checkouts out a stable release before deploying a Docker stack. Docker Swarm will automatically distribute your functions and OpenFaaS services across the cluster.
```
$ git clone https://github.com/alexellis/faas && \
cd faas && \
git checkout 0.7.1 && \
./deploy_stack.sh
```
## Test the UI
Within a few seconds (or minutes if on a poor WiFi connection) the API gateway and sample functions will be deployed to the Swarm cluster running on DigitalOcean.
Access the Gateway UI via the IP address returned by `docker-machine ip node-1` (you can also access via `node-2` and `node-3`):
```
$ echo http://$(docker-machine ip node-1):8080
```
Prometheus metrics can be viewed on port 9090 on a master. Fetch the IP like this:
```
$ echo http://$(docker-machine ip node-1):9090
```
## Deleting OpenFaaS Droplets
You can use `docker-machine` to delete any created Droplets if are finished with your OpenFaaS deployment.
```
docker-machine rm node-1 node-2 node-3
```
## Advanced
### Create a Load Balancer
Digital Ocean provide their own *Load Balancers* which mean you only need to share or map one IP address to your DNS records or internal applications. They can also apply health-checks which ensure traffic is only routed to healthy nodes.
From the DigitalOcean console Networking page, open the Load Balancers tab and click *Create Load Balancer*.
Give the balancer a name and select the Droplets which will be balanced using the `openfaas-getting-started` tag and `NYC3` region (these were values passed to docker-machine when creating the nodes).
![create_lb](https://user-images.githubusercontent.com/83862/30240233-274c4dc0-9564-11e7-8881-54bce652392f.jpg)
Update the forwarding rules to point at the Gateway on `8080` and Prometheus dashboard on `9090`:
![forwarding_rules](https://user-images.githubusercontent.com/83862/30240106-0eb71242-9562-11e7-846e-093627026a7c.jpg)
Expand the Advanced section and update the health check to use port `8080`.
![health_checks](https://user-images.githubusercontent.com/83862/30240104-0e98e3d0-9562-11e7-89b6-c266384e35d8.jpg)
Click `Create Load Balancer` and after a few minutes your balancer will be available.
![balancer_ready](https://user-images.githubusercontent.com/83862/30240232-2747becc-9564-11e7-867a-c3ac220f2ae3.png)
You can now access the OpenFaaS Gateway, Prometheus dashboard and all functions via the load balanced IP address. For example from the balancer above:
- Gateway: http://45.55.124.29:8080
- Prometheus: http://45.55.124.29:9090