Is… Is this your first time on the internet?

I think we’re talking past each other here: You are talking about blocks as physical delineations in an ssd, containing many pages with erase cycles. Op is talking about the filesystem view of blocks, which is different altogether. ZFS, for instance, should have its block size aligned with what a disk presents to the HAL, or you end up with lots of wasted space or terrible performance.

For SSDs (you’re right, outside op’s problem), the internal block size is usually much bigger than the emulated block size, often 128K. But we don’t use filesystems with 128K block sizes, or we couldn’t reasonable run databases or store small files.

There was a brief time in the wild west of SSDs being new where some shipped with 8k blocks, but this caused a bunch of problems for storage drivers because no one had yet decided if the translation of FS blocks to should be handled by filesystems or by disk firmware. We now let manufacturers decide how they should r/w to media, and trust the firmware’s presented block size.

That block size is now 4k for most disks, spinning or ssd. We still see some weird layouts in high-performance or high-endurance disks, but they are exceptions.

so if a 512e SSD doesn’t receive the rest of the page in time (e.g. with IO scheduling, 512 byte fs sectors, unaligned partitions, or bad luck)

I’m not sure what this means, pages are a concept in an ssd’s physical structure, they don’t mean anything to a filesystem (except in volatile memory).

Except for leaving things default, almost none of that comment is true.

4k is default for most disks today. Some transition-era ssds used 8k blocks.

Do not worry about emulated block size, because ssds only use that to report to the filesystem, no sectors exist on solid state media.

Erasing the partition table and/or uefi metadata space.

Not a productive comment.

Ansible certainly works here, most implementations I’ve seen are terraform with ansible or cloudinit.

Most of my stack is containers, so I personally prefer cloudinit and a handoff to ansible.

It’s better to think about it as though you are making a queue of changes that need approval to be permanent. Rather than just accepting that you just installed a ton of dependencies for no reason because you didn’t really want the software, you can cleanly reverse any of the changes.

In a sense, it accomplishes a similar thing as nix in a different way.

it’s not like sD devs forced them to adopt their option

It kind of is, though. You dont typically select your preboot environment, it’s selected for you.

Yeah, but things like human readable logs are almost all gone because of journalctl changes that went into the systemd ecosystem without anyone really running by folks who look at logs everyday.

It’s not the functionality of systemd I don’t like, it’s the Poettering way of just making changes to stuff without bothering to learn how any of the rest of us professionally use things.

Mir was canonical’s own fault for trying to pull a (blatantly obvious) licensing move where Mir would have been nearly impossible to compete with in a future Ubuntu.

And that’s setting aside the absolutely insane statements canonical’s Mir team were making about Wayland at the time.

When you fumble something so badly that Intel stops supporting your efforts on it in protest, you messed up.

Don’t run before you can walk.

Fstab is how it works in the background, best to learn that first.

Mint is pretty polished and mostly stays out of the way. I think that was probably a good choice, looking at your style of shifting over components of your workflow.

Safer with bs=4M on older flash.

Page faults are memory-related and random reboots are either CPU or memory related, typically. I would run a memtest.

The best case scenario is every one gets an equal slice of the pie so as no one player becomes the dominant force because everyone is fucking crazy.

One of the reasons I love FOSS is that it’s really difficult to corner the market and then try to screw people over, cause they’ll just leave. At the corporate level, it take a bit longer to leave because of contracts, but the principle still works.

Not that Sun was the best or anything, but when Oracle happened, it spawned projects we all know and love well.

On select machines.

Yeah, it’s very handy. The switches are a bit different, depending on what you’re trying to identify, but

pgrep -l <term>

Will usually get it done.

Yes, this exactly.

Packages are installed and removed from disk, but unless you kill the offending process(es), it might remain active in memory, depending on the quality of the install/remove script.

You can use ps, pgrep or any number of utils to figure out which process is the one to get rid of, and just kill it.

Gaben stands on the shoulders of giants.

Back in the day, we all wanted Google to fill that role…