the problem here does not originate in zfs, this is a hardware failure due to a poorly supported SoC (vendor says PCIe works ‘normally’ since kernel 7.1, this system ran 6.19) and bad electricity supply

do you ever get a new shiny thing and it looks like it works very nice and coherently and then you dig a bit deeper and realize that it is just a bunch of semi-broken shit held together by glue and duct tape? that’s the state of android, that is nicely obscured on pixels by google customizations

haha, no, depending on your use case zfs is mostly set and forget, you just have to get a set of sane defaults, which, at least for me, is way easier and saner than stuff like btrfs. Arch wiki is a good source for that.

scrub is a command that reads data from disks and verifies checksums, it can automatically heal the corrupt data if some redundancy is available. 0 bytes repaired means nothing was repaired, 5 errors mean that there are 5 instances where the recalculated checksums of data did not match the expected values that were calculated when initially written. Normally the command would output ‘scrub repaired 0B with 0 errors’, that’s how you know that the pool is healthy.

shprutzling the cache vdevs sideways is generally required only for obscure optimizations, I’ve never seen a case where a wrong setting would lead to data loss.

The gotchas come from a single thing: the stupid CDDL licence, which is not compatible with GPL. That means that zfs is not included in linux kernel and most of the normal linux userspace filesystem stuff do not work with zfs.

Also, keep in mind that zfs is generally almost impossible to fuck up, what I’ve done here is a result of extreme dedication and like 8 years of experience.

I’ve seen a calculation somewhere for drive speed, size and scrub frequency where optimal number of extra drives was calculated. On fast storage and with weekly scrubs one extra drive should be mostly fine.

My philosophy is any raid is not really a data contingency solution - it just saves some time when the drive fails.

that’s no fun. also, it turned out that this is not a disk error, the zfs metadata got corrupted and now I get a panic when trying to mount this.

This is raidz1. Three 4TB NVME drives. I’ve built this machine as the primary server for services like immich as a fun project that enables me to say that my photo service runs faster than all the commercial stuff and is on ARM64. The downside is that the SoC is not supported by upstream very well so vendor patches had to be applied and there was some instability between that and my shitty electricity supply.

I feel very sorry for you. I have the same problem (panic on mount) with this one, but at least it works when mounting read-only. I’ve been moving data with zfs send the entire day

my achievement here is that this is a proper raidz setup with multiple disks and ecc ram

I was using a mix of debian servers, arch thinkpads and macos macbooks at the point where I moved everything to NixOS (macbooks did not survive this transition and were donated due to never working properly because of t2 issues).

Nix is extremely hard. I mean, don’t get me wrong, initial setup is easy, adding and updating packages is the easiest experience between all package managers out there, nixpkgs is overpowered as it has almost all the software that exists since there are no conflicting dependencies in nix. Getting something that is not in nixpkgs to run is impossible for a beginner (nixpkgs has steam-run to partially solve this, but results may vary). Getting to the point where a tornado may pass through your home and destroy all the hardware inside, and the whole recovery process from that point is about 20 minutes of your time, including the trip to the store for new hardware, requires significant effort, both time and mental, that most of us don’t have. But the entire point is that it is possible with Nix!

If Debian is like owning a 90s Corolla, NixOS is like owning a factory that can make any car you like. You can choose to get a 90s Corolla, you can choose to get a newer Corolla, you can skip corollas entirely and make something new and fast. When the new and fast thing breaks, you can make a new car or go back to the Corolla. The Corolla comes completely ready to use by you from the production line. If you need to adjust the mirrors, you just make a new car with the correct mirrors setup. All the cars you make after that will come with the mirrors setup how you like them, because that is now a part of the manufacturing instructions the factory has.

The only caveat is that factory gets updates every 6 months and then your car building pipeline may break when you get the update. In three years I’ve been using nixos fixing those took about 5 minutes and usually boiled down to stuff like ‘previously we had a list of car colors, passed as a string, now the factory expects to get a hex color code’.

Judging by the commit history, I moved most of my machines to NixOS after about 6 months building my config, then, 8 months later I rebuilt the entire thing from scratch when I understood nix enough to make everything truly modular.

I really doubt that once you’ve embraced NixOS you’d be tempted to use anything else or try LFS. NixOS is beautiful. NixOS just works (after 6 months of setup).

also, machines wipe themselves at each boot! service state data is distributed via zfs send! zfs that actually works!

I’ve been hosting grist for a while and it is quite nice. Wasn’t able to move all the stuff from classic spreadsheets though

deleted by creator

Please share your experience when you get to it! I try to avoid iot stuff though and thought of something simpler like dimmable lights that get warmer when dimmed - like the Ikea ones, but with 2700k as the highest point it does not really work.

I thought that maybe the ideal solution would be to wire DC lightning with relays controlling groups of different temperature LEDs. Maybe glueing LED strips to the ceiling and a translucent film under them to diffuse it a bit? I feel like ordering a giant bin of random LEDs should lead to the best possible CRI.

Also each of the 8 bulbs I ordered had Ra>95 written on the box and they just lie because I can instantly tell that the light is wrong. Bad CRI is so prevalent that friends come to my house and think my lights are incandescent.

Yes, we only turn the lights on after the sun’s color temperature matches our 2700K lights. During the cloudy winter days we spend the entire day in darkness to avoid mismatched temperature.

Sometimes I really want to get adjustable LEDs for winter, but it is hard enough to find warm ones with a high enough CRI. I once ordered and returned about 8 different bulbs which had price points from €2 to €100, before going to Ikea and buying bulbs there.

After about a thousand commits in my config I no longer know how to do stuff the normal way. A few days ago I spent 20 minutes trying to figure out how to run python with modules without resorting to shell.nix

But are you truly using nixos, if you don’t have a swarm of 10+ machines that do distributed builds?Can you really say that your system is deployable, if you never tested it on an obscure ARM board and a laptop from 2005?

steam deck as bluetooth controller:

You can use the Steam Deck as a Bluetooth controller by enabling Bluetooth in the settings and pairing it with your PC or other devices. This allows you to utilize its controls for gaming on compatible platforms.

Complex things that someone has already done are infinitely easier in nixos - stuff like having zfs as root filesystem is literally two lines in the config (and the magic is that it is very, very hard to break).

Complex things that are your own edge case will make you want to pull your hair out - I wanted to run immich on a raspberry pi 5 with native 16k page size, long story short, I still don’t have immich.

On the other hand, if by “normie” you mean “running a browser and some flatpaks”, nixos is likely the best distro that will work right out of the box - the graphical installer will generate a good config, the out of the box hardware support is the best in my experience, breakage is almost impossible. Automatic updates will not work though and there’s no gui that will prompt you to do so at all.

Who cares if it breaks? You can always just boot a previous generation! Need to rebuild without the breakage? You surely must now how to add a package from an earlier commit via flakes by now, right?

Well, if you can’t figure out how to integrate the flake in 30 seconds by month 6, you clearly have a skill issue. Or a “sleeping at night instead of writing nix” issue. Better use a noob-friendly distro like arch.

Seriously though, despite all the flaws, there is no other packaging system where I can as painlessly use random forks of packages. I absolutely love how I’m able to run gnome-mobile on my x64 tablet. True to the NixOS way, I found the overlay on someone’s GitHub, there were only the files, no further instructions.

I also have a USB with live debian at all times, because you never know when you stumble upon a thing that just can’t work with NixOS