Single CoreOS Droplet on DigitalOcean
Introduction
I just finished converting my blog to run on Docker and sit on top of CoreOS. Since it’s my blog and I don’t actually want to spend money on a cluster of VMs, I use just a single server (single instance/droplet) of CoreOS. Below is a list of steps on how to do the same on DigitalOcean. I took instructions from DigitalOcean’s guide on setting up a CoreOS cluster for my case.
Prerequisites
Docker
I assume you have whatever application you want to run in a Docker image. Docker makes it super easy to replicate an environment. I have the same thing running on my laptop as on DigitalOcean except for some SSL files.
One downside of Docker though is the image size. When I first started with Docker, my image size was 1.4G. I’ve done some work to half the size for now while utilizing the image cache as much as I can.
SSH Keys
This section in DigitalOcean’s guide is applicable. Below is what it says about SSH keys:
Every CoreOS server that you create will need to have at least one SSH public key installed during its creation process. The key(s) will be installed to the core
user’s authorized keys file, and you will need the corresponding private key(s) to log in to your CoreOS server.
If you do not already have any SSH keys associated with your DigitalOcean account, do so now by following steps 1-3 of this tutorial: How To Use SSH Keys with DigitalOcean Droplets. Then you will want to add your private key to your SSH agent on your client machine by running the following command:
ssh-add
Write a Cloud-Config File
For our purpose of a single server CoreOS, the cloud-config file is really just applicable for configuring CoreOS’s reboot strategy after installing an update. There’s a lot more to cloud-config that’s documented here, but we don’t need that right now.
Minimal Cloud-Config
My cloud-config file is very simple and only configures the reboot-strategy:
#cloud-config
coreos:
update:
reboot-strategy: reboot
Create Droplet on DigitalOcean
If you have the prerequisite and the minimal cloud-config file, then you’re ready to create a new droplet on DigitalOcean.
You would go through the typical steps:
- Give the droplet a name
- Select a droplet size
- Select region
- Then when it comes to the “Available Settings” section, check the “Enable User Data” checkbox and put the content of cloud-config in there.
- Select CoreOS image in the next section
- Add your SSH key
- Click the “Create Droplet” button
SSH to CoreOS
Your DigitalOcean Control Panel should show the IP of the droplet. You can SSH into CoreOS by typing:
ssh core@[ip]
Setup systemd Service to Run Docker
The last major step is to setup a systemd service to automatically run a docker container to serve your application.
Login to Docker Hub
If you are not using a private image on Docker Hub, then you can skip this step.
Since downloading Docker images can take a while, it’s a good idea to grab the image prior to doing anything else. For example, I grab my private image from Docker Hub this way:
docker login
docker pull dannysu/mystuff
You’ll be prompted for username, password and email for the docker login
command.
Writing the systemd Service File
A minimal systemd service file that you can use as an example is shown below. You’ll save it as mystuff.service
. It basically defines what to run when you type systemctl start mystuff
or systemctl stop mystuff
. Save this file to /etc/systemd/system/mystuff.service
on your CoreOS instance.
[Unit]
Description=My Cool Stuff
After=docker.service
[Service]
User=core
TimeoutStartSec=0
KillMode=none
EnvironmentFile=/etc/environment
ExecStart=/home/core/mystuff.service.sh ${COREOS_PUBLIC_IPV4}
ExecStop=/usr/bin/docker stop mystuff
[Install]
WantedBy=default.target
In my example, ExecStart runs a shell script named mystuff.service.sh. I use it to decide what docker commands to run. You can use it as an example:
#!/bin/bash
if [ -n "$(docker ps -l -q)" ]; then
/usr/bin/docker start -i mystuff
else
/usr/bin/docker login -u dannysu
/usr/bin/docker pull dannysu/mystuff
/usr/bin/docker run --name mystuff -p $1:80:80 dannysu/mystuff nginx
fi
Finally, you can start the service and make it run upon reboot as well:
systemctl start mystuff
systemctl enable mystuff