Managing Docker With Portainer

A handy tool if you're new to Docker and/or CLI

The whole world seems to be switching to Docker. Instead of having to master the intricacies of Linux, with the implicit hazard that sooner or later you’ll paint yourself into a corner and brick a major subsystem, Docker containers are low inertia. You can fire stuff up and if it doesn’t do what you want it’s trivial to down the current config, erase the containers, and start fresh.

Being an old person, we’re talking the internet didn’t HAVE a web interface for the first ten years I used it, I prefer command line apps for many things. But when I started using Docker I couldn’t easily visualize what I was doing in the same way I can with Linux. Then I discovered Portainer, which formerly gave out free Business Edition licenses for five or fewer machines. These days the gratis option is three machines with Community Edition, and that works just fine for learning.

My desktop is fairly potent - it might be twelve years old, but a twelve core Xeon with 128GB feels fast and roomy. My Docker setup right now is what you see above - a ten core virtual machine with 40GB of ram. I formerly had it as part of the host OS, but it fights with Tailscale for control of the network.

My workstation has an eight year old GPU - a 6GB Nvidia 1060GTX. This morning I waged an epic battle with ChatGPT, Docker, and Google. I am surprised but pleased that I actually got this thing to run.

But getting Portainer to see that there’s a GPU available?

*shrug*

I want the GPU available in Docker so I can accelerate Dify, but this sort of thing has always been a godawful nuisance that consumes hours of my time and leaves me at a dead end. I’m going to step away from this, maybe for a day, maybe for a quarter, then try again.

This evening I’ll fire up Brain and I’m sure GPU accelerated Docker will just work. That may become the Dify platform of choice around here.

Conclusion:

Ignore my grousing about GPU stuff, if you’re debating learning Docker, Portainer will make that much easier.