I never tried Docker (looking around now), but the approach is extremely dirty for a few reasons.
1. Process supervision should be handled in Docker (or something that is designed to do supervision), not in the `while true` loop. Idea on hanging on *.log files is terrible. Maybe there is a reason to stop the container when application stops?
2. There must be a better way to handle docker logs of multiple programs. If not, run both instances in different containers.
3. The thing serves configuration files to arbitrary clients that ask for them. It doesn't even log multiple downloads of the same key (though serves the private keys over SSL, which makes it hilarious on the purpose of SSL here).
Point 3 is openvpn specific and can be acceptable for example (though I am still lost why bother with SSL). However, points 1 and 2 show how to seriously misuse Docker. Either Docker or the setup is flawed (I suspect the latter).
Please do not take this article as an example how to do things in Docker. There must be better way, in more or less every step.
1. Process supervision is kept very simple, because there are many other tutorials and demos out there using Supervisor and other systems. I'd love to know what is inherently wrong with using `while true` and hanging on logs. Personally, replacing Supervisor (or monit, or god, or $yourfavoriteprocessmonitor) with 3 lines of shell script is fine with me, especially as the focus is to show that Docker can now run OpenVPN.
2. Handle multiple log streams: indeed! I considered running two separate containers, and the only reason why I didn't do it is to simplify the setup process. I also explored the other option (using two separate containers for TCP and UDP connections), and it works almost the same; it looks like:
docker run openvpn genkeys
docker run -volumes-from $CID openvpn tcp
docker run -volumes-from $CID openvpn udp
docker run -volumes-from $CID openvpn serveconfig
Hopefully, the way the project is laid out makes it fairly trivial for someone to change it to run each openvpn process in its own container.
3. Here again, it's a (not-so-good) trade-off. In many cases, you need something simple and straightforward to move the configuration file to the device, e.g. when the device is a phone or tablet. Downloading a combined file over HTTPS is the simplest solution I found (even if in the Android case, it is slightly crippled by the Download Manager bug). It uses SSL to prevent eavesdropping. It protects against a system which would passively record e.g. WiFi traffic. It doesn't protect against a system which would actively detect such a transfer, and immediately initiate the same transfer, before you have the opportunity to stop the `serveconfig` container. I considered adding some HTTP basic auth, but that would have required a "real" HTTP server in `serveconfig` (or to hack further with socat, but I'm already quite ashamed of the current hack, TBH).
I agree that the setup can be greatly improved (what can't?), but I disagree with "there must be better way in every step". This is certainly not the setup that you want if you have a large number of clients, or if you are in an extremely compromised network (then again, it's only a concern during key exchange), but I don't see why it can't be acceptable for 99% of the personal users out there.
One tiny nitpick: you might want something like: dhcp-option DNS 8.8.8.8 to go with your redirect-gateway def1 .
This is because all of your DNS traffic will be redirected over your VPN as well. If you happened to have been assigned a local only DNS by your home router or cable/dsl provider, DNS will be broken when your VPN connects. Use a globally accessible one like 8.8.8.8 and the dhcp-option to tell openVPN to switch your DNS on connect.
I don't fully get it, but am trying to follow along. Basically, s small box you can install 1-x s-Linux software apps on, and deploy it on another machine, or inside a VM on that machine.
It's not a real box, or hardware, but a small chunk of software that essential is a pre-made .iso with whatever single, or multi-packackaged goody you desire?
I don't get, with spinning up a VPS being a few clicks, what is the advantage? Can't you make images of your VPS on amazon, or any of the other cloud providers, and save a snapshot of your config? What is the difference?
These are totally posable? Where do they get their real hardware resources from, such as RAM and drive? If I set up a LAMP server in a Docker container, allocate 50GB of memory to Apache, and drop that docker into a micro instance on Amazon, what is going to happen?
There are numerous reasons for why containerization (even containerization on-top of a vm) is a wonderful thing.
The first thing is that containers allow you to securely isolate software and its configuration without impacting that software's usage of resources the same way a vm does.
Isolation of configuration is extremely valuable here, you now have a collection of containers that aren't polluting your host system with its dependencies and configuration, now containers can inter-operate with each other over known protocols and upgrade/downgrade their configurations or software without any impact to the surrounding host system.
Security is a big one too, containerization allows you to completely isolate the permissions of the given application - if something is compromised only that one thing and its container are compromised. Fix the issue, throw away the old container, build a new one, deploy it, voila!
The second reason containers are awesome is the ability to build, test, and "deploy" your software. Your build process (if you've actually worked on it!) with Docker is now to have it part of your automated testing cycle, build cycle, and packaging cycle where when you, say, tag a release - an automated build gets kicked off that builds the whole application in a Docker image with the explicitly listed dependencies (as they may be updated!!!), if that succeeds without error then go on to automated/QA testing, if that passes then packaging as a docker image ready to be installed. You drop it on the server and you're ready to go!
[EDIT] I use a similar set-up to the author but I'm on FreeBSD; in FreeBSD we have ZFS and ezjails which provides most of the same functionality (but slightly more "cool" because of ZFS) as Docker but as a set of shell scripts, you can have certain jails start up with the host system.
I run OpenVPN, TOR, a web server with OwnCloud on it, and a few other experiments on the machine and it's a joy to be able to throw away a jail and build a new one, or create new jails whenever I want to experiment but never worry about it polluting the host system and also not worry about performance impact on my cheap hardware.
This looks fun, but if this is the only thing you are going to be using on your extremely small Joyent VM, why waste time and complexity putting your OpenVPN setup inside Docker? Seems unnecessary.
That's not the only thing I'm doing on this VM; as I pointed out, I'm consolidating lots of little hacks and projects that are running all over the place in different environments. I put each thing in its own container. Benefits:
- I can shutdown one of those "experiments" without affecting others.
- If I want to migrate one of those things elsewhere (because it turns out to need more resources), it's already "Dockerized" (i.e. trivial to redeploy elsewhere).
- When I remove one of those things, I don't have to worry about leftover packages, dependencies, files, etc. lying around.
Otherwise it would indeed be a poor use of my time :-)
I now don't have to know how to set up an OpenVPN service. I can just use jpetazzo's. There's a benefit in Docker as a distribution platform (many companies are doing this right now) that is pretty compelling.
> Joyent Ubuntu image comes with an “optimized kernel”. It might be optimized, but it doesn’t have AUFS support, so you want to install an official Ubuntu kernel instead
IPSec is one of the few instances where I have encountered consistent ( irregular ) kernel panics and when a panic ensues from a containerized app.. it is of course the 'host' kernel that is panicking. So all your containers are hosed.
I only run VPNs through virtualized kernel instances now; if they fail, the hypervisor restarts them. Nothing else affected.
Just trying to understand Docker and hence the question. How is it different from the HPUX or Solaris Package managers? Maybe this question itself shows my age :)
1. Process supervision should be handled in Docker (or something that is designed to do supervision), not in the `while true` loop. Idea on hanging on *.log files is terrible. Maybe there is a reason to stop the container when application stops?
2. There must be a better way to handle docker logs of multiple programs. If not, run both instances in different containers.
3. The thing serves configuration files to arbitrary clients that ask for them. It doesn't even log multiple downloads of the same key (though serves the private keys over SSL, which makes it hilarious on the purpose of SSL here).
Point 3 is openvpn specific and can be acceptable for example (though I am still lost why bother with SSL). However, points 1 and 2 show how to seriously misuse Docker. Either Docker or the setup is flawed (I suspect the latter).
Please do not take this article as an example how to do things in Docker. There must be better way, in more or less every step.