palordrolap

Tentatively yes.

I did once manage to mount an external USB NTFS drive to a VirtualBox-hosted copy of Windows 7 and was actually able to defrag it. I assume I also ran a quick disk check before that, but it was a long time ago now.

Before I did it, I backed up everything important off the drive to another location just in case. I’d recommend you do the same.

As to how I did it, I’m afraid I don’t remember, but it can’t have been that difficult. There may have been some kind of raw mount option in the virtualisation software.

The other potential obstacle is the fact that things have moved on since I did it. Newer Windows / NTFS might be not be as easy to fool into accepting a drive over weird virtualisation pathways. Or the virtualisation software might not allow it as easily or at all.

Hopefully that’s not the case.

$ locate -r '\.so$' | wc -l
4468
$ locate -r '/lib[^/]*\.so$' | wc -l
2488

We’re going to be here a while.

Fun fact: The Windows BSOD colour was as easy as adding a couple of lines to a .INI file for a long time. Then, as they tend to do, they made it more difficult, but it was still possible. Third party tools were written to do the work.

Very recent MS Windows I have no idea about. My search-fu is failing me.

Anyway, my point is that the “two lines in a config file” method would be nice.

Knowing systemd though, it’ll be “send some kind of message into a /proc pseudo-file”, or a sub-sub-sub-command of one of the many systemd* commands which ultimately does the same thing.

It’s rare that I get to feel anything remotely comforting about not being able to afford new hardware, but if I understand correctly, my BIOS-only dinosaur can’t be exploited.

Still vulnerable to thousands of other exploits no doubt, but not this one.

Pipewire? It’s very new to me and can’t say I know much about it, not that I knew much about its predecessors either.

The pipes in the diagram are obviously named pipes, but they’re not Linux pipes. There seems to be not only multiple types (which is disturbingly Microsoft), but often multiple by the same name (which would confuse most sane OSes, if not the insane ones too.)

It’s almost like they’re instances of a subroutine object all running in parallel…

Makes sense to me.

My only concern is that pipe c is shown as having two different shapes: straight and slightly curved.

Based on the fact that the design requires that a and b be different, there would undoubtedly be the same situation for the four slightly curved c pipes. That is, there would need to be two “c2” pipes and two “c3” pipes in the set rather than just four more of the same c pipe.

That makes me think the diagram at the bottom was made before a decision to cut costs and/or simplify. Four regular c pipes will undoubtedly be cheaper and logistically simpler to manage for both shipping and user construction than having those two extra pipe types.

It was, of course, relabelled to match the supplied parts, but the hints of the original design still remain.

To distinguish two Firefox profiles that I run simultaneously, I use different themes on each. For Firefox this might actually be the best way.

For a file manager (I assume the Dolphin you’re talking about is the file manager), the closest I remember seeing is a red toolbar on the unrelated Nemo file manager when it’s run as root.

If Dolphin is per-user theme-able, then you could do what I do with Firefox. If it supports other kinds of plug-in, then maybe there’s one that does what you want already.

To my knowledge, windowing systems can’t override the title of an application’s window, and even if they could, the application could change it back again at any time, creating a race condition, or a very ugly situation where the system picks and chooses which windows are allowed to modify their titles and which ones aren’t.

Therefore, I think you’d have to write your own plug-in (if they’re a thing and the API permits title modifications), modify Dolphin’s source code yourself or submit a feature request to Dolphin’s developers, cross your fingers and wait.

“UNEXPECTED_EOS” is almost certainly “unexpected end of stream”, that is, the file is missing the end or there’s data corruption and the unpacker has interpreted the bad data as meaning the file should be longer than it is.

Redownload the file, or try to download it using a different tool (e.g. wget or curl rather than a browser). If that still gets a truncated file, try a different source / mirror.

That looks like it might be the monitor’s own on-screen display rather than anything Puppy related. My guess is that the monitor hasn’t been detected properly and Puppy is putting out a resolution that the monitor can’t deal with.

Since the message says 1280x1024, either the monitor is 1280x1024 and can’t deal with anything else, or it’s not 1280x1024 and is being sent 1280x1024 resolution and is complaining about it.

(Or worse, it’s a clock frequency error which was a real problem back in the early days of Linux.)

As for how to fix, the answer is going to be different depending on the age of the base Linux under Puppy and the graphical subsystem.

For X/X11/Xorg it’s probably going to need use of the xrandr shell command, perhaps to delete the mode that is causing the problem. For Wayland, it appears that each window manager has its own xrandr equivalent. I see talk of a gnome-randr, for example.

To get to a shell in the first place, try the Ctrl+Alt+F1 key-combo. If the computer isn’t frozen, that might get a text-based console login prompt. (Puppy might do things differently here though. Not sure.)

Alternatively, look up how to boot to a single-user shell by modifying GRUB options, that is, if no such option is there already.

Caveat: I am no expert. Take this under advisement. Also try web-searching some keywords. It might be there’s a really simple fix for this that I don’t know about.

Never trust a corporation. It will almost always do whatever makes the most money for C-levels, shareholders and end-of-year profits, and when it doesn’t, we should be even more wary of its actions. Occasionally these unspecified actions and choices align with the preferences of people outside the corporation and this makes the corporation “one of the good guys” for a while.

Corporations have no right to complain about being called out on this. In fact, they’d do better to acknowledge it. All it needs is one change of CEO and the whole corporation can change direction in a heartbeat. Twitter is an example of this.

Also see: The fable of the scorp(orat)ion and the frog.

With Microsoft, any love shown could well be the Embrace part of the strategy that will lead to Extend and then Extinguish just as soon as they can figure those parts out. They might already have a plan.

The fact they’ve been able to turn things to their advantage so far does not mean they don’t have such a plan. Or won’t ever have one.

Recursive chmod (or chown) has been breaking things since before systemd was a thing, so even if systemd is now responsible for stopping things from working, it can’t have been that previously, especially at the time I might have done something silly.

As for repairing permissions only, I suppose it would be possible, assuming the system still works (or can somehow be encouraged to do so) to copy only the permissions (or at least infer them) from a backup or something rather than the whole files.

Reminds me of the test server shenanigans I had at an old job versus a colleague. All in fun. Nothing in production.

One was the faux Bash shell that kind of worked OK until you pushed it or tried to do anything fancy. It was the default shell for the user called “root”, but that wasn’t the UID 0 user. It had been, but I renamed it. Then created a new “root” with a different UID. Of course, the faux shell would tell “root” that it was UID 0.

The other was the simple background loop that would detect any rival admin sessions and SIGHUP their shell process. First user on the box to run that pretty much had free reign, and everyone else was logged off instantly.

Instant, sure, OK, but pick a slightly less evil brand. I’d say Kenco maybe, but that doesn’t seem to be as well known outside Britain. Maybe Folgers for the North Americans.

But, how can it say anything without audio?

Just put MOUSE.SYS in your CONFIG.SYS or run MOUSE.EXE from your AUTOEXEC.BAT

… wait, where am I again?

Well, it definitely works when shifting people politically.

Coming soon: Mockrosoft Overton Windows™

Theoretically yes, but yes, in that order.

I’ve worked with Linux for decades at this point and I’m still not 100% sure exactly what breaks; it’s a mistake you make once, if at all, and you’ll only get a little way into even trying to figure out how to fix things before you throw your hands up in disgust and reinstall / restore the OS (or whatever subdir was affected).

If I was to hazard a guess, it’s the kernel itself that balks, but there are other, almost as fundamental things (lib*.so files and the like) that may also be deliberately fussy.

Dude’s in bed. ZZ is clearly the superior choice for the sleepy vi user.

TL;DR Maybe FreeBSD?

Back in the day, I was very interested in (the now extinct) PC-BSD, but it really didn’t like the unusual HDD setup I had (Third IDE channel maybe? The details are fuzzy now.)

Never got to the stage of trying gaming on it, but I think I might have been planning to dual boot?

Anyway, it must have been a while ago because that was my previous PC, which I donated to a relative the better part of a decade ago.

So, given that it was a FreeBSD, I guess that’s what I’d be looking into, but I can’t say I know enough right now.