Hello, I’ve been here a while and I already have a server running for almost a year! now the bosses want me to install another server for an expansion the next town over. I confess I know next to nothing about docker, or kubernetes but its the hot new thing in servers, and from what I can see orthanc is docker friendly. so I guess my question as in the topic is:
- Docker vs “Normal” installation
- Pros
- Cons
The Metal I have available is a dell T40 with a 40 gigs of ram, 4 tb mirrored raid for image storage and a 1 tb drive for os. the first server is identical and is running fine with ubuntu binaries, and that’s what I’m inclined to use because its what I’m comfortable, but I’m also curious of the docker install benefits.
any comments are welcome.
If you haven’t read this yet, you might want to read that over first: https://book.orthanc-server.com/users/docker.html
The Osimis BitBucket Repo has sample set-ups, including for Docker here: https://bitbucket.org/osimis/orthanc-setup-samples/src/master/
If you just search Docker Hub for Orthanc, you’ll find a number of publicly available images: https://hub.docker.com/search?q=orthanc&type=image
There is a little bit of a learning curve to use Docker, but not much really and there is excellent documentation: https://docs.docker.com/
Very easy to setup on UBUNTU.
Hello,
In my opinion, the very big benefit of Docker is the easiness for deployment. Indeed, everything is wrapped up and you just have to set some environment variables depending on the specific need of the customer. With Docker, you won’t hear anymore “it works on my machine, it should works on your one”.
I’ll be honest, I have always deployed Orthanc through Docker, so my opinion is probably biased, but I don’t regret the learning of Docker.
Here are some other benefits you could think about:
- if you consider a professional use of Docker, you probably use a true DB in place of SQLite (Postgres in my case), the use of Docker also ease the deployment of that DB…
- if you plan to upgrade your setup in the future, Kubernetes could be interesting and so containers is the way to go…
- if you need to execute some scripts from time to time (for clean up, data processing,…) Docker will allow you to reproduce the exact same setup on another machine (or on the same machine) in order to try/validate your scripts
HTH,