r00ty

I’d say the ideal situation is that tools are developed library first, then cli or gui as preferred allowing others to pick up the slack and make the other tool (or tools) using the functions in the library.

One of the reasons automation is so much easier on linux than windows is because there are many more cli tools to do things. On windows many tools are gui first and cannot easily be automated.

That’s weird. Whenever I’ve had gpu drivers fail the environment didn’t come up and I would be left at a terminal.

But, they shouldn’t need rescue. The issue is no nvidia driver, but you can still login from the text terminals. Ctrl + Alt + F3, F4 etc etc. In fact when the window environment fails to load it should drop back to terminal.

I have a 3080, so 590 is fine for me. But, I’m sure the legacy one is a dkms. But the process of installing that should be done as part of the install. E.g. you install, reboot

What does lspci -k show for the card in terms of Kernel driver in use, and kernel modules? Also what does dkms status say?

If the module is installed and showing in dkms status and showing as used in lspci -k, it should be available for desktop environments.

I do agree in terms of effort when things go wrong though. I remember when I was a lot younger and I had no problems just sitting in front of my keyboard finding whatever the latest problem is. Now, I want to be doing things with my PC.

But, a bit of debugging might be worthwhile before doing a new installation.

Did you also uninstall all of the components of the new driver as per the arch site?

https://archlinux.org/news/nvidia-590-driver-drops-pascal-support-main-packages-switch-to-open-kernel-modules/

Otherwise it’s investigate from the tty as to what driver, if any is in use for the gpu pci device.

Are you sure it was dot pitch and not dot clock?

Dot pitch on a crt might make the image look bad (trying to draw onto the shadow mask) but I doubt it would damage it.

Setting an invalid dot clock could damage some crts. But most of the modern (read from mid 90s on) would just go to the power save mode when they got a clock they couldn’t use. The warning did still remain on the xfree86 configuration guides though.

Showing my age perhaps.

More than 51 years if there’s one of those updates that will randomly decide to overwrite the UEFI removing your bootloader entirely :P

I would say, now it’s learning that actually sticking your head in the sand is only ever a delaying tactic. But, if it DID learn that, it’d mean it has surpassed us already.

Of course not. As the merovingian in the matrix says. French is a fantastic language, especially to curse with.

Then I suggest they use an XNOR pointer instead! Checkmate patent trolls!

Huh. I am sure you could search for individual books. For sure you could do it by goodreads ID I think? Yes, adding an entire author as the primary way to do things is a bit much for some. I know for sure I have managed to do individual books before now.

It’s a real shame because Readarr did work and they really just needed to fix their own metadata servers. No? Or were there other problems I’m not aware of?

I mean, I have to say I’ve hastened my own demise (in program terms) by over-engineering something that should be simple. Sometimes adding protective guardrails actually causes errors when something changes.

Yes, had the same happen. Something that should be simple failing for stupid reasons.

Yep. It seems they haven’t changed a thing about the format. Probably a script much older than mine on their end is generating it too.

I have a tool that I wrote, probably 5+ years ago. Runs once a week, collects data from a public API, translates it into files usable by the asterisk phone server.

I totally forgot about it. Checked. Yep, up to date files created, all seem in the right format.

Sometimes things just keep working.

With IPv6 for most use cases there’s actually more security. With privacy extensions (pretty sure it’s enabled on windows by default), when you make connections from your device, it uses a “private” IP. That is a randomly chosen address inside your network’s prefix, that changes regularly.

These addresses don’t accept incoming connections. You have a main address that doesn’t really change that you accept connections on. Firewall that for ports you want to allow and then hackers need to port scan 2^64 or 2^80 address space to find your real IPs in your prefix. If they capture your IP from a connection to a web server etc, they won’t have luck scanning you.

Again as per my post above, the biggest risk right now is bad default configurations on many home routers.

The “firewall” features are called connection tracking and, a firewall. With IPv6 I have my firewall setup very similar to NAT. Established and outgoing new connections are allowed (this is done using connection tracking). Incoming new connections are not allowed unless I open up a specific port.

Home firewalls SHOULD be setup the same for IPv6, a lot are not and IMO is the main problem right now.

Just lay down, and pretend you’re compiling.

Does /dev/null support sharding?