Thank you @vesper1978 . I ended up doing the above.
And thank you @phillxnet.
In all honesty - pardon me for saying so: I will end up proceeding without Rockstor. At least for the foreseeable future.
In order to make that constructive, let me try to explain my reasons.
I am commenting as both a software and a devops engineer. By that I mean: I understand my share about the process(es) involved in software development.
- No UEFI support. Recent Linux kernels support EFISTUB, they can be booted directly from a suitable boot loader.
My BIOS doesn’t even support booting BIOS devices as primary devices. This is a problem: I have needed to be able to fallback. I have needed my multiboot, rather than playing around with live USBs in case anything happens.
- I set Rockstor up. I started it. I imported my BTRFS Raid 1. This went well.
Then I tried to rename my pool. Here I am.
As you have pointed out, the command to rename a volume is reasonably simple. There is no way I am going to go around squinting to guess which pool I am on. Nope. No.
BTRFs commands are proving to be straightforward.
The UI lacks flexibility. It has absolutely positioned elements, HTML tables without dynamic options. It’s pretty rigid. Exposing the stylesheet for change would make it themable.
Componentizing the interface with a view library (like React) would make it easier to scale.
I was about to propose contributing at least a couple of PRs toward that.
But as it stands, a fresh Linux is a clearer starting point. I am just better off starting my btrfs-and-nginx-powered solution.
I hate the way I sound when saying that, but. It is true. I have the need so this is what I’ll do.
(now, if Rockstor comes a little further, perhaps with the rebase to OpenSuse? I really hope I’ll have to look back. I was rooting for something like that).