yeah, I found topton as the I guess original designer of those AOOSTAR boards, but I wanted a complete pre-built PC, link

auto power-on sounds like a normal ACPI thing, no? at least all my hardware does that.

USB dock is not an option due to the reason you said yourself, so yes, one box is critical.

Thanks, I see older versions of this thing with socketed RAM that are much closer to the budget. Maybe this will be the one. Port situation is strange, I’ll need to research it.

All my servers are routers, I have a mesh network. OpenWRT routers do not participate in the mesh since that is a pain to set up and maintain, that’s part of my desire to move the routers to my unified NixOS config. Since my repo is larger than the available storage on most consumer routers, you’re right, I’d need a small custom box, be it a small pre-built NUC-like thing or something I build myself.

I dislike OpenWRT because it does not have automatic updates, and updating remotely while keeping all the packages required for the network to work is extremely risky.

I can’t just ignore the updates due to all the zero-days we are getting almost weeky with the LLMs and my ISP or someone on the network is trying to actively get into my stuff. The routers behind CG-NAT are getting almost the same kind of obnoxious port scanning and ssh login attempts as my servers with public IPs, drastically less volume though.

USB adapter sounds extremely flaky, I agree, and that would never be something I’d use here, but! my first NAS was a windows 8-era tablet with an HDD connected via USB 3.0 adapter. It was way more reliable than you’d think.

I also got the impression that you think that I’m messing with people’s routers just for the fun of it. That is not the case, the problem is that there’s no internet without VPNs provided by my routers (that’s a simplification of the situation, but just accept this explanation). JetKVM would also not work for that reason. My goal is to be able to monitor that the heating and water supply still work and my 80 year old grandmother will not freeze to death because her CNG supply ran out and she has too much pride to ask someone to do something about that.

I like small things and I actually recently built a gaming/workstation PC with mITX ans SFX PSU. The gpu is probably taking about the same footprint as two 3.5" disks will. The complete PC is at least five times larger than what I’d like to have as a router and I promise there’s absolutely no space inside that goes unused. I liked micgerne’s suggestion since that board does not require an additional power supply.

thanks, the trap worked as intended

hmmm this is very nice and I guess I can just print a generic-looking case for this. can it power the HDDs though?

edit: I see it can. this is amazing, thanks!

another edit: but it has the same issue as the AOOSTAR one - the H4 plus is out of stock. The newer H5 can’t provide SATA power.

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

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