rojimboo: I'm not sure you realise that wear and tear will occur regardless on the live iso. Live iso doesn't make the drive read only, there will still be a bunch written and read. It's just at shutdown, all those written things are lost for the next boot. Non-mechanical drives fare much better anyways with wear and tear as there are no moving parts, so I'm not sure where you get your estimates from for a few months lifetimes for a USB key or by extension, SSDs...
And all I'm saying there should be a way to make a persistent live usb read-only where relevant, or just setup the security in such a way that a normal user can't install anything (simply not give them sudo priviledges) or change anything major. So it's definitely worth considering if you're adamant about having cushy firefox custom settings for users in a live iso that apparently can't be done in CLI...
From what I understand, the live mode just writes into memory and doesn't save it anywhere on any drive, but
pretends to do so - like if you make changes to a file or even delete a file, it just pretends that it doesn't exist, but obviously it didn't delete it, because it's still on the drive. So it's basically like using toram, except that it doesn't read everything into RAM during booting, and may keep reading more whenever it needs it. Isn't that right? So f it's not changing anything then how, and for that matter,
why would it write anything to the drive?
And I think I need the users to be able to use sudo, because they need to be able to boot with toram occasionally, and then be able to copy everything from the USB drive onto a new one to make a clone of it in case they need backups (I'll provide specific instructions on how to do it, but it will need to be possible).
rojimboo: It took me 30seconds to google two solutions for you based on reading the error messages.
GUI:
System settings >> Software and Updates >> Other Software and removing each duplicate entry from the list.
CLI:
In your /etc/apt/sources.list, you have duplicate components. Edit and remove duplicate lines.
Reboot and rerun sudo apt update and sudo apt upgrade and see what happens this time.
Thanks, I'll try that when I get the chance. It's just that this stuff has taken priority so everything else has a back seat until it's finished.
rojimboo: I think the first error message is about 32-bit architecture and the repo being 64-bit. I would hazard a guess you are running in 32-bit architecture Ubuntu for some reason. Check with 'uname -m'. Alternatively edit the /etc/apt/sources.list and replace and add 'deb [arch=amd64]' instead of 'deb' in the offending line.
I know that my Linux and processor are both 64-bit, and uname -m says "x86_64". I'll check that file though.
rojimboo: Do these errors stop you from upgrading all your packages? Or are they just warnings and sudo apt upgrade completes ok? Because your cubic install has a major systemd dependency issue, that should be fixed by upgrading the offending packages with their own dependencies...You need to sort out your linux install first, at a basic level.
They
seem to upgrade fine, because afterwards it doesn't come back again requesting the same upgrades.
rojimboo: Keep trying it with Cubic though, I think that's your most user-friendly option.
I wouldn't call it user-friendly at all. Live Kit is user-friendly, because all I have to do is use my OS as normal, get it into the exact state that I want it to be, and then run a script to make the ISO. Can Cubic do all that? No. You even admitted that it's going to be very difficult if not virtually impossible to configure some of the settings and stuff, not to mention that I would need to know all commands corresponding to every single thing that I want to do, if there even is a command for it. I swear, I can't imagine why any moron would create a utility that works in any way
other than the way Live Kit and Ubuntu Imager work. They're extremely intuitive, and there's just no possible advantage to doing it this other, indirect way with a text-only interface. I mean really, what's the advantage of doing things the hard way when the same thing (and more) can be accomplished the easy way? But of course it's just my luck that the only program(s) that can work the easy and good way aren't compatible with UEFI! Is the whole world filled with masochists and imbeciles that want everything to be impossible, just to make it so that every avenue I attempt to follow forces me into dead ends that lead nowhere?!?!?!
In any case, I'll look into it anyway IF I can actually get it installed and working. If not then I don't know how I'll get UEFI to work. I may just have to leave the stupid thing to be only compatible with CSM. But regarding that, are there computers that aren't compatible with Legacy/BIOS/CSM? If so then how common are they, and if not then is there any large chance at all that it will happen in the future? I mean I wouldn't
think that anyone would want to make a motherboard which is incompatible with it, because wouldn't that cause a HUGE amount of stuff to be unable to boot? Frankly, I don't see the point of why UEFI was ever invented. The old way worked fine and was MUCH simpler, so it seems like UEFI is just an enormous overcomplication with little or no advantage. Am I right?
rojimboo: But honestly, if updating and upgrading normal packages is such a hurdle (on the most beginner-friendly distro haha), I suddenly have misgivings about you managing to do even the slightly trickier stuff...You might just have to go back to basics, bite the bullet, and find linux docs to understand the 101 stuff, for Ubuntu for example. I don't know how much experience you have with computers and PCs, or what kind of programming you do, or how tech savvy you are, but you might want to seriously consider taking some time to learn the basics first before deep diving head first into these complex things.
First of all, I'm a very good and very experienced programmer, but I
don't have much experience with Linux, simply because I'm new to it. I've spent time here and there, to an extent, just learning various things about it, but I've mainly been focusing on this specific project, because it's EXTREMELY important to me that I get it to work! It MUST work! And frankly, getting it compatible with Linux is just the final step in a very long list of very horrible steps that have all been incredibly arduous, and the whole thing has taken over a year of constant headaches, when it really should have taken maybe a month or two all together! But now I'm about 99% of the way there, so I hope you can see why I'm feeling a bit impatient and I just want the damned thing to FINALLY work!
rojimboo: I applaud your efforts at extracting information from others on a vidya gaming forum that's not even Linux friendly and quite off-topic to boot (no pun intended) but really at some point you need to ask yourself - Is this the best way of doing things? and - Can I spend my time more effectively? Because it seems like your rig and linux install is quite a mess and that you're not really learning a lot.
I don't see a more time-effective way of trying to get this thing to work, unless there's an easier solution. Do you have one? I'm learning a lot in some ways, but then only a fraction of the things that I learn, or the ideas that I try, actually work (maybe 5% of them), but that's only because all of this stuff seems to be defective and not as advertised, and not doing what I would expect it to do (or what any reasonable person would have designed it to do, or would expect it to do based on the description of what all this stuff is and what it's
supposed to do, but there are too many incompatibilities that aren't even obvious until one attempts everything from every angle only to
finally realize that it can't be done!).
rojimboo: First thing I would probably do, is to make a list of hard requirements for this project. Preferably from the point of view of the technical feasiblity of doing it as a live iso in either persistent or non-persistent mode, and figure out if all the requirements can be met with an existing tool that you are able to understand and use. And then go from there.
What do you think I've been doing? I've listed the things that I need, so I obviously have figured out what needs to be done, but just now
how it needs to be done. The reason why I'm haphazardly trying every method imaginable is because in my experience, only a very small percentage of methods - usually only one of them, if even that - will actually work, but there's no way to know which one until I try
everything (or at least everything that seems like it can feasibly have any chance of accomplishing all of the things that I want to do).