oasis
oasis copied to clipboard
oasislinux
Bootable USB image?
It would be great to try to boot Oasis from USB to test it out on a real system. How difficult would it be to set something like that up?
If this doesn't exist already, it would be also handy to have some sort of script or process that sets up an image ready for flashing to USB (kind of like what buildroot can do).
Yeah, I agree that it would be nice to have bootable USB images.
I think the main issue would be building a kernel that supports a bunch of different hardware. Right now, you are expected to configure your kernel by building in the drivers you need. So, I'd probably need to look into some initramfs solution.
YESS! I almost opened an issue to ask this! I would like a USB disk so I can directly install into my machine;
@michaelforney why don't you take a look and see how void-linux and alpine do? if you need testing, I can help you with my laptop;
I think the main issue would be building a kernel that supports a bunch of different hardware. Right now, you are expected to configure your kernel by building in the drivers you need.
@michaelforney A version of Oasis that we could boot via Qemu or VirtualBox would be extremely useful, and then you wouldn't need to support real hardware (yet)!
There is a way to use dracut with minimal dependencies (and certainly without systemd). While at first Dracut might look like an overkill, it really isn't. Dracut is cross platform. I am using it on Alpine and is the default on Void Linux .
If there is some positive feedback that dracut could be a desirable direction, I can help. Might need to create a dracut module to use hotplugd instead of udev.
Its 2024 and I'm absoluetly here for this.
I've tried Oasis in QEMU and it was great. Now I want to install it to disk. However, while compared to my normie friends I'm an elite hacker, compared to people who are "comfortable building their own kernel" I'm the normie. I'd actually love to attempt building Oasis and a linux kernal, but with 2 small kids and limited time thats probably not going to happen. I dont want to cave in and "just install Alpine" (like most people tell me to). Dont get me wrong, Apline is great, but Oasis is greaterer..er.
The only other thing that might help, other than a bootable image, would be a more detailed walk-through / installation guide. Better yet, a video tutorial of the process form start to finish. I know that's asking alot, but watch out! If I find myself determined enough to try and build Oasis from source then I'll probably also be createing a long issue thread asking for help along the way with many many dumb questions.
Its 2024 and I'm absoluetly here for this.
I've tried Oasis in QEMU and it was great. Now I want to install it to disk. However, while compared to my normie friends I'm an elite hacker, compared to people who are "comfortable building their own kernel" I'm the normie. I'd actually love to attempt building Oasis and a linux kernal, but with 2 small kids and limited time thats probably not going to happen. I dont want to cave in and "just install Alpine" (like most people tell me to). Dont get me wrong, Apline is great, but Oasis is greaterer..er.
The only other thing that might help, other than a bootable image, would be a more detailed walk-through / installation guide. Better yet, a video tutorial of the process form start to finish. I know that's asking alot, but watch out! If I find myself determined enough to try and build Oasis from source then I'll probably also be createing a long issue thread asking for help along the way with many many dumb questions.
@pfr-dev I would love that too and share the same thoughts. An ISO image would be so much easier for everyone. Building it on your own sounds like a very complicated task. I would love an ISO or a detailed tutorial/video about. An ISO image will open SO MANY DOORS for many many people...
Thanks everyone.
Its 2024 and I'm absoluetly here for this.
I've tried Oasis in QEMU and it was great. Now I want to install it to disk. However, while compared to my normie friends I'm an elite hacker, compared to people who are "comfortable building their own kernel" I'm the normie. I'd actually love to attempt building Oasis and a linux kernal, but with 2 small kids and limited time thats probably not going to happen. I dont want to cave in and "just install Alpine" (like most people tell me to). Dont get me wrong, Apline is great, but Oasis is greaterer..er.
The only other thing that might help, other than a bootable image, would be a more detailed walk-through / installation guide. Better yet, a video tutorial of the process form start to finish. I know that's asking alot, but watch out! If I find myself determined enough to try and build Oasis from source then I'll probably also be createing a long issue thread asking for help along the way with many many dumb questions.
Here's my perspective:
- It doesn't really matter to me who uses oasis and who doesn't. I made this system for myself and published it in case it's useful for others.
- I don't really see much value in an ISO image. You'd still need to do the same install process once you booted it up, so this would only make it easier to test oasis on a real system instead of qemu. I expect that it's not too difficult to make such an image if you want to contribute. But you'll have to do the research on initramfs, generic kernel config, how to make a bootable USB image, etc.
- You mentioned your time is limited. Mine too. It's hard for me not to read this as "I don't have a lot of time, so please spend a bunch of your time for me". I have a hundred other things to do, and making a detailed install guide/video is a lot of work.
Overall, I'd say oasis is for developers who want to fully understand their system and have the ability to modify or replace components. If this isn't you, then I'd ask yourself what you're hoping to gain from oasis. What specifically don't you like about Alpine that you think will be better with oasis?
An ISO image will open SO MANY DOORS for many many people...
I'm worried that those who can't open this first door will find 10 more doors immediately behind that one.
Building it on your own sounds like a very complicated task.
This makes it sound like you haven't actually tried.
Here's my perspective:
It doesn't really matter to me who uses oasis and who doesn't. I made this system for myself and published it in case it's useful for others.
Valid point.
You mentioned your time is limited. Mine too. It's hard for me not to read this as "I don't have a lot of time, so please spend a bunch of your time for me". I have a hundred other things to do, and making a detailed install guide/video is a lot of work.
That is not what I was insinuating at all and I do apologise if that's how it has been received. I was simply wanting to show my appreciation for the project and support the idea of a Live ISO because, as I made reference to in my comment, I'm not yet comfortable with building kernels or linux systems from bare bones. This isn't to say that I can't do it, I'd certainly like to try. I only wanted to support the idea in case it was something you might consider, which would ultimately make it easier for the less experienced users to dip their toes in to such a minimal and bespoke system. I fully respect your decision and accept your reasoning behind making it. As for a detailed guide/video, I suppose this was just me expressing my desire to feel more confident in giving it a go. A wish more-so than a request, meant more with a sentiment of envy and eagerness than of contempt or dissatisfaction. Again, I do apologise if it came across that way.
Overall, I'd say oasis is for developers who want to fully understand their system and have the ability to modify or replace components. If this isn't you, then I'd ask yourself what you're hoping to gain from oasis. What specifically don't you like about Alpine that you think will be better with oasis?
I'm certainly not a developer, I'm merely a hobbiest/OS nerd who likes to tinker with less mainstream systems to learn how they work, how they are put together and how to modify them to suit specific needs. So I guess I'm in the right place and that is why I am interested in Oasis.
Recently I did install Alpine after trying my hardest to love Tribblix (which is still my favourite OS, it just wasn't compatible with what I needed on my old tiny notebook). I love Alpine, it's great. I just like the design principles behind Oasis a little more. The use of sbase instead of busybox, cproc insead of gcc or clang, BearSSL instead of LibreSSL. The goal of prioitising smaller and cleaner code base is something that really resonates with me, so does the decision not to include a package manager. I use NetBSD as my daily driver so I'm already quite familiar with pkgsrc. Pretty much everything in the README ticks all of my boxes.
I know I could tweak Alpine and use many of the same tools, in fact I have done so already. I'm also trying to build velox on Alpine which you may have noticed from the issue I've created over on that git repo (I'll update that soon). I will at some point attempt to install Oasis on this particular machine because then I wont feel the need to change much.
Anyway, I love your work and please don't feel like you owe me, or anyone else, anything. I just believe in the motto "if you don't ask, you don't get" so it's always worth asking.
Cheers
EDIT: Grammar
What specifically don't you like about Alpine that you think will be better with oasis?
In my case I have a feeling it will be less bloated and use even less resources from the computer than Alpine
Hey folks,
It's worth mentioning that it's certainly possible to compile some of the 'software packages' from the Oasis userland in other distros likes Alpine. The vis editor is even available on ubuntu and it features structural regular expressions. Another example is the rc shell, amazing software.
