In this post I will report my experiences and the challenges I had to fight regarding my first steps to CoreOS
CoreOS is an operating system built to run docker containers with clustering and high availability in mind.
I will write a tutorial on setting up CoreOS clusters manually any time soon. For now I just assume you followed this Guide to have a 3 machine CoreOS Cluster on EC2.
Let’s start with the basics
While you just by some clicks should have a running CoreOS cluster now, let’s make sure you connect to it the right way:
You should check, if your ssh-agent has the right key
This should output your key. If it doesn’t, just add it by typing:
and check again.
If everything is okay, then ssh into one of your instances via ssh
Note: To interact with your CoreOS node, you have to use the `core` user.
Now you should be logged in. Yay!
Listing your machines
To list your machines in your cluster type:
Your output should look like this:
1 2 3 4 5
In the first column you’ll see the abbreviated form of the hash representing your machine
the second one is the IP of that machine
and the third one contains metadata (if present)
You’ll probably be a little impatient, so we’ll start now by running a docker container:
1 2 3 4 5 6 7 8 9 10
Yay a command prompt of the newly created ubuntu container :-)
But that’S definetly not the way you want to run docker containers on CoreOS. Enter systemd and fleet.