Have you looked at https://zfs.rent/ for cloud storage? You have to provide your own drive, but the monthly cost is relatively low.

Yep. The 4 Samsung SSD 990 PRO 4TB NVMe SSDs are $1100 apiece. Yikes!

https://www.samsung.com/us/memory-storage/nvme-ssd/990-pro-p...

> So now I have a large script which debootstraps a Debian based NAS into being.

Is part of this publicly available? After seeing, - https://words.filippo.io/frood/ - https://0pointer.net/blog/fitting-everything-together.html

I've been looking at setting something up with mkosi/systemd for my home server/NAS - it's interesting seeing how different setups are solving the problem for inspiration.

I'm also planning a new home server build now and the prices are definitely relatively ass. So far I've spent 820€ on two 22 TB WD Elements HDDs, 375€ on 2x16 GB DDR5 kit, and 520€ on two 2 TB M.2 SSDs (cache). About 1700€ and I still have no server to show for it. Doesn't help that I've been procrastinating on picking the CPU and motherboard.

> and while this made sense when drives were expensive

I don't have the answer to the latency question, but HDDs have shot up in $/TB over the last couple years too. They are once again kind of expensive.

It's complicated, but https://jro.io/capacity/ actually walks you though how it works, towards the middle of the page.

``` Unlike traditional RAID5 and RAID6 implementations, ZFS supports partial-stripe writes. This has a number of important advantages but also presents some implications for space calculation that we'll need to consider. Supporting partial stripe writes means that in our 7wZ2 vdev example, we can support a write of 12 total sectors even though 12 is not an even multiple of our stripe width (7). 12 is evenly divisible by +1 (3 in this case), so we don't even need any padding. We would have a single full stripe of 7 sectors (2 parity sectors plus 5 data sectors) followed by a partial stripe with 2 parity sectors and 3 data sectors. This will be important because even though we can support partial stripe writes, every stripe (including those partial stripes) need a full set of p parity sectors. ```

bcachefs is slowly working up to what you're describing - describing what you want and letting the filesystem sort it out. We're basically there for local storage, and other people have been building some nice reporting on top.

Next year (post Rust, because networking code is so much nicer in Rust) will be send/recv, and I think we should be able to make some nice improvements over the state of the art there.

USB/SATA adapters are incredibly flaky. Some don't support TRIM, some have major firmware bugs, many overheat during long data transfers and disconnect from the host. I'd strongly advise against using them for any important data.

10 months ago I've built a Raspberry Pi 5 mini NAS with 3 SSDs connected through RADXA Penta Hat. It runs OpenMediaVault, system is on the SDcard, other data are on the SSD ZFS pool. I have autosnapshots enabled, and I tune compression methods on specific datasets (lz4 for almost everything, zstd for file backups).

We use it for storing (backups, media files), playing (PaperMC), and watching (Jellyfin). I can only complain about the lack of hardware decoding in Raspberry Pi 5. Jellyfin loads CPU much if I enable transcoding so it's always disabled. If I knew this, I'd consider a cheaper and faster, but less popular, RADXA machine. Storage is fast enough for me, rsync and Samba speeds are usually limited by my network. PaperMC also runs without a hitch, thanks for asking!

As I didn't have high requirements for the machine, I also considered USB Bays at first but they wouldn't go well with ZFS.

The problem with RHEL is that the free version exists at the sufferance of IBM, a company not well known for being motivated by keeping tech enthusiasts happy. I would use Debian, personally. Not as long of a support cycle as RHEL, but still quite stable and no possibility of corporate rug-pulls.

Or rocky, which is a 1:1 clone of RHEL that doesn't require licenses. https://rockylinux.org/

> I have a Broadwell dual socket Xeon workstation that I'm going to upgrade to Proxmox. Would it be reasonable to run something like this as an LXC or VM or would you put it in the base kernel?

I've got a small HP Z440 workstation (small in that it's 10 years old and only single-CPU) and a 14 cores / 28 threads Xeon e5-2680 v4 in it. It's a little workhorse.

I run Proxmox (basically "Debian + ZFS out of the box + easy to spawn VMs / LXCs") on it.

And you can totally put something like FreeNAS in a VM. If you plan to really serve lots of files and need lots of perfs, you should get a tiny bit fancy and passthrough to the FreeNAS VM an entire disk controller (device passthrough to a VM in this day and age of LLMs that are there to find the relevant parts of docs for you in a split second is easy). IIRC the FreeNAS documentation itself explains how to run FreeNAS in a VM so I don't think it's "unsupported".

There's nothing wrong with running it bare either.

Depends on what you are after.

if you are after quietness and "power" then an old workstation is a great bet. They normally have space for at least 4 fullsized HDDs, the more modern ones have lots of PCI space for nvme-ssd cards (for space) and some have lots of lanes for speed

They also tend to come with SAS/whatever remote nvme is called/SATA expnasion options

The down side is they are not as space efficient, they also tend to have 60-120watt CPUs, so expensive to run

and checksums!

It's a horrible idea likely, but I have an ancient old Dell PowerEdge R510. Probably sucks way too much energy, but it does what it does and the price of SSDs have skyrocketed so I'm not touching it.

A lot of them can, yes.

The models that are a crappy PC can, almost like any or other crappy PC. Some have a very tiny DOM that might or might not be enough for a recent system. Of course, once the system is minimally up, it can use the big disk for software. No idea if any have special weaseling on them, but you can always search for the model and e.g. Debian (that tends to have best hits). Running with read-only root or just initrd is also quite possible.

The ARM models tend to be very very very slow and small and have architectures that are obsoleted in many Linux distributions combined with hardware that's weird enough that NetBSD doesn't support it (e.g. LAN chip or SATA, which you might need in a NAS). And yes, looking at NetBSD pages and build configs is one avenue for finding well supported devices. Although the documentation isn't great or complete. Some models may also have weird boot sequences or rescue modes that are extra hassle.

I keep wondering if a neat little buildroot config would work with some of these devices. Other odd options would be https://armbian.com/boards and https://openwrt.org/supported_devices .

Apart from better (luck with) enclosures, cables, and power, probably not. USB-SATA bridges are notoriously flaky.

Try a different host machine to rule out it being a DAS enclosure.

I'm hardly a sysadmin and I think Unraid is great. My usecase is some Docker containers and media backups. The UI gets the job done well, whereas Synology tries to dumb things down and make things pretty.

Android is to Unraid as iPhone is to Synology.

> usb adapter kept getting io problems there too with zfs, drive gets disconnected during scrub or high load

That's usually caused by the adapter chipset overheating. Don't use USB adapters for important data.

I’m running an 8-drive ZFS RAIDZ2 pool. I’m wondering if you know — are the free space recommendations around ZFS cargo or real?

Like I’m already giving up two full drives for redundancy (which saved my ass - I recently had two drives fail on me in quick succession — both SSDs from what looks like an identical batch) but then the advice is kinda saying I need to keep at least another drive worth of space free for the pool to perform well and not crap itself. That hurts with current prices for sure.

> combined with LUKS, TPM keys

Does it work? Server can reboot and use TPM to unlock rootfs? What about /boot - encrypted and tamper proof? Resists evil maid attack?

Rabbit hole, I know, but so fascinating if you solved it all :)

> Not a great time price-wise to be building a NAS

That under-states the matter. It is a terrible time, price-wise, to build a NAS.

I'd almost rather have no AI whatsoever and have storage 1/10 the price of pre-AI times.

(If there were a magical choice between having AI and significantly more expensive storage, and having no AI and some program to dump that investment money into getting and somehow leveraging significantly more available storage, that is.)

I've been thinking about building my own NAS as well. Mind sharing how much did you pay for those hard drives, and what motherboard did you choose?

For Samba, add a separate entry for the user home directory because the homes directory default triggers some sort of OCD.

You assert that ECC RAM being necessary for ZFS is just a myth but provide no justification for why that is untrue.

Is it not the case that if you don't have ECC memory, ZFS could end up writing a checksum that does not match the data if you get a bitflip in just the right (wrong) spot?

> 4x the 14 TB WD Elements Desktop.

I recommend staying away from SATA drives (huge consumer rush) and look for SAS drives on eBay, particularly HC520 Helium drives from HGST/WD. Need a SAS3008 PCIe adapter ($20) and a SFF-8643 splitter ($30). No backplane is required. Huge lots of decommissioned drives frequently appear. I bought Qty 10* of HC520 (12TB) SAS drives for $1000 about 3 weeks ago, avg age is about 2.5 years, still well within its rated lifetime.

Yea may be some of the stuff is fear mongering and cargo culting. I was told ECC is necessary for ZFS. When the article was written, it was cheap af (2024) to buy ECC RAM so not much consideration was given to it.

-- (*) I have no idea what to do with it. Anyone has any good idea for using 120TB space? I have about 40TB unused bandwidth in the datacenter, may be host a Debian mirror? Donate storage/bandwidth to Internet Archive? Please contact me, appreciate it.

> ThinkCentre M910q

I love these machines. I have 3 of them and an M720Q. Needs changed so only one of each is actually in use, but they are so low profile and quiet that I'm considering deploying the other two at home just in case I ever get the urge to play with K8s or something. The only thing I'm missing is a splitter so I can use one plug socket instead of three - not sure if something like that exists.

One nice little fact about the M910Q is that they are rated to maximum 32GB but I have two of them with 2x32GB no problem. Recent RAM prices and forever regret is the only reason the third one doesn't have the same. The M720Q only takes 32GB though.

You can also run 2 SATA off of the M2, and someone created a 3D printable enclosure for them: https://makerworld.com/en/models/1280680-thinknas-2x-hdd-enc...

I run zfs as the storage pool for my incus (next lxd) services. It is the ideal fit. Here is a list:

- Instant, zero-copy container cloning from images via Copy-on-Write. If you boot a new image like the existing ones it's seconds.

- Atomic, millisecond-level instance snapshots regardless of storage size

- Block-level container migration using native 'zfs send' and 'zfs receive', very short command lines and seems to work perfectly.

- Granular dataset nesting (every instance, image, and custom volume gets its own ZFS dataset). You can see every filesystem even on the host.

- Transparent, inline data compression (LZ4/ZSTD) enabled automatically per dataset. For services that don't change much, you might as well use a compressed image to make them even smaller.

- Mirroring / Raid

- Sub-volume sharing and direct management via native ZFS administration tools. If my home directory has a build area and a million files, I can just save time and put my home, pre cooked into a new machine and not copy or even rebuild on my new machine.

- Dedup keeps blocks with the same data as a reference. This costs a lot of memory and has not saved much for me as a lot of my images are similar and already shared I think, but it's cool.

I'd probably just buy a cheap prebuilt PC with a big case, or surplus enterprise tower. You can get cheap LSI PCI cards and do miniSAS to SATA if needed.

The biggest concern I'd have with USB is power delivery to the hard drives, but I haven't even done the napkin math so maybe it's fine. The SSDs seem like they might be a waste of money. USB hard drives have a poor reputation, but I don't have a ton of experience to say how much of that is deserved. On a practical level, I'd also be concerned about knocking the cable out.

The RAID5 write hole is not present in RAIDZ1 I believe.

It also surprised me that the author said "4 Cores, Xeon Server CPU can be had for cheap".

But the specs also said ECC RAM and I don't think the N100 supports that.

It's pretty minimal and straightforward written in Haskell. Probably can write one from scratch with Claude now.

Can you explain a bit more about this. Would this help in a single server to facilitate backups? Or does this need the the NAS and the Proxmox Backup Server on a different system?

I'm sure there's better options now but the HP ProLiant MicroServers (used).

They support ECC ram, 4 caddies, one extra PCIe slot, and to my knowledge you're not cpu limited for a zfs file server usecase.

Keep in mind though, all you need is linux* support, iDRAC, ECC if you're a snob, and drive bays ... and that's basically any free server.

In my extremely opinionated opinion I would only get used enterprise server gear, because a zfs file server will just work unless hardware fails. And a UPS.

*ZFS will be a more natural choice on FreeBSD. It's better documented, and will meet Linux 1:1 in hardware compatibility for this.

Not strictly a recommendation, but Terramaster is a good brand to look at if you want Synology-shaped hardware which can run TrueNAS or Proxmox or any flavour of Linux you want.

Along with various other devices (including a large Synology which I wouldn’t buy today), I run Proxmox on a small two bay+two nvme Terramaster. I have a bare bones Ubuntu LXC running Samba configured for Apple Time Machine, an VM running Scrypyed, and PBS for Proxmox backups. Nothing on it is critical so I don’t bother with any storage redundancy.

its probably dumb but what i do is use proxmox with a debian VM with docker installed and each docker container gets a virtual disk i make on the proxmox host. obviously more involved than running unraid or truenas but i think the complexity underneath is simpler.

Where does that 8GB requirement come from?

LVM does have snapshots.

I've lost more data to "proven reliable systems" than to my homegrown hacked-up stuff (which may be because I trust the latter much less).

> I would rather use a proven reliable system that’s been used and tested by millions of other users

You're describing ZFS.

Same, using NixOS as a NAS, though it kept growing and now I'm trying to share it with other people. I use BTRFS on MDADM though. It's ended up being an all-in-one home server and router now because I was tired of the noisy power hungry rack I had in my closet.

https://HomeFree.host

This is what I'm hoping to do. I haven't done much research yet, so I look forward to reading about your learnings. (Also to the person who responded to you with their experience and link.)

Thanks for saving me probably some hours...

EDIT: I'm also coming from an old Synology NAS from the ~2015 era.

It's more the single drive storage layout that is an issue in your case. No resilience to a single disk failing and you are risking all your data if your hardware starts silently corrupting data. This is assuming all your backups are done using the data on the single hard drive.

If you want to keep your setup simple I would consider the following:

- Keep a list of file hashes and check your files against this list periodically.

- Use some type of backup program that supports snapshots. Restic is a good choice.

- Get a hard drive the same size as the one you have and use SnapRAID to manage file integrity or set up a two disk mirror using btrfs.

Once you add multiple machines reading and writing simultaneously, you'll hit a limit. But that's fine, build what you need now, no need to over-optimize.

> What about /boot - encrypted and tamper proof? Resists evil maid attack?

> Rabbit hole, I know, but so fascinating if you solved it all :)

Encrypting `/boot/` is the wrong thing to do. One can just replace it with something that logs the encryption key instead (and sends it off to somewhere). You actually need to verify that what you expect to be there is actually there and a encryption isn't gonna do that for you.

A reasonably secure boot chain looks roughly as follows: bootloader is checked by the UEFI (read secure-boot), bootloader chain loads a signed UKI which also uses secure boot verification (or just skip the bootloader if you only have a single kernel with no fallback), this then loads a signed dm-verity `/usr/` which brings you to the login screen. Root is just data (e.g. mounted `noexec`) and is encrypted using e.g. PCR 11 so that it only unlocks if the UKI is the one that is expected (with some other backup unlock method).

This is basically #2 of the "Design Goals" of Poetterings "Fitting everything together"[0]. It's a good read if you are looking to designing your own Linux image and are looking for some inspiration. There is ParticleOS[1] which somewhat dogfoods this kind of system, though I don't think it is anything remotely considered for a production system.

[0]: https://0pointer.net/blog/fitting-everything-together.html

[1]: https://github.com/systemd/particleos

I run my cheap hosting box with a cleartext boot setup that uses ssh to automatically grab the key for the real root from my home server (or an alternate at my MILs house). Using FreeBSD, but similar concepts.

A previous hoster once gave me someone else's drives without wiping them. I don't want random customers snooping around on my data if a similar mistake happens with my disks.

For home use, I run without disk encryption. If I ever need to do data recovery, it's not going to be possible with encrypted disks and one point of a centralized NAS is to have stable long term storage.

Looking at current hd prices, I keep kicking myself for giving away several 3tb and couple of 4tb drives I had in my unraid server few years ago when I started upgrading my array to 8 and 12tb drives. At the time, a drive slot in my server was more valuable to me than a mere 3tb drive. i could have kept them for emergencies if any one of my array drives fail.

Yeah, the default [homes] share is really not necessary on a home network where you really just want a common share. On my home storage server, I just have a “files” user, and a corresponding, anonymously readable/writable share:

  [public]
     vfs objects = catia fruit streams_xattr
     fruit:encoding = native
  
     comment = Public share
     path = /home/files
     force user = files
     force group = files
     guest ok = yes
     browseable = yes
     read only = no
     create mask = 0644
     directory mask = 0755
  
     readdir_attr:aapl_max_access = no

The default [homes] share is still there in case I need it for anything, but it's “browseable = no” so it doesn't confuse visitors.

It was €329 per hard drive (through a reputable store), and I chose the ASRock B650M-HDV/M.2 mATX motherboard combined with a Ryzen 5 8500G. Stock CPU cooler, replaced the Jonsbo case fans with Arctic P12 Pro PST LN. I only slightly regret the PSU (MSI MAG A650GL), which could be quieter. Not that it's very noisy, and it's a great PSU otherwise, but I should've chosen one that just shuts the fan down at low power usage.

You're telling me.

I had a drive go bad in my 4 year old home-built NAS a few weeks ago, and it cost 2x what it originally did to buy the same capacity drive, and that was going to a grey-market importer on eBay, it was more on "reputable" sites.

> - Dedup keeps blocks with the same data as a reference. This costs a lot of memory

To get around this you can add an SSD mirror as "Special" vdev, the dedup table will then be saved there, freeing up your memory.

So it has the potential to even look and work like an actual NAS! That is so cool. Sad state of affairs that 64GB seem to be around €400 nowadays, that's actually double of what the whole machine cost. I'll wait for the AIpocalypse and come out later to buy some sticks.

Had it happen to my PBS backups running on ZFS without ECC just the other day. Turned out to be heat related, memory was fine but during long backups heat was causing bit flips.

> You can get cheap LSI PCI cards and do miniSAS to SATA if needed.

People say those generate a ton of heat because they're normally in a rack mounted server with plenty of airflow. Probably a significant drawback for a "cheap prebuilt PC".

RAIDZ1 will add more stress to the consumer ssd's as to enterprise one due to surprinsingly different hardware. Consumer ssd's life times are calculated on a 8 hour shift (the tbw) and enterprise are calculated on 24/7.

Please - dont - use - consumer ssd's - with zfs raidz1

The big USB racks all have separate power.

What's next!? Let Claude analyse clinic visits!? Oh well...

I'm running PBS on a separate mini pc, it is the perfect forever near instant incremental backups that sold me. I did perfectly recover in 20 minutes after a full hardware failiure. Best recovery experience ever.

> But the specs also said ECC RAM and I don't think the N100 supports that.

If I remember correctly it can in theory but in practice I have never seen a N100 with ECC.

Agreed. I have used a HP N40L ProLiant MicroServer since 2013 as home NAS and Time Machine backups via samba. Rock solid hardware, incredibly expandable, and today runs FreeBSD 15.1 with ZFS. Additional hardware modifications include; CD-ROM replaced by two 3.5" HDDs on mounts (now six HDDs of 10TB each in ZRAID1), a SAS card to add two mirrored bootable SSDs underneath CD-ROM drive space, a 2x 2.5G NIC (limited to 4GB/s slot) for dedicated NFS link, while additional internal SATA and external SATA ports unused. Next: replace PSU fan with quieter Noctua fan.

I'm still running an old Gen 8 MicroServer. Modern drives can actually saturate the SATA controller, and because it only has a single PCIe slot I can't add both a 10Gb NIC and a storage controller - I went with the 10Gb NIC.

It works well enough though and has lasted me over a decade at this point. 16GB DDR3 ECC, an old 4 core/8 thread Xeon, 4x14TB drives and the Mellanox NIC.

The N100 supports "In-band ECC" (IBECC), which uses regular non-ECC RAM at the cost of less available memory and a 10-20% performance drop. It’s unclear how well it works, and almost nobody uses it.

I have the F8 Plus, great little unit. It did need a BIOS update when I first got it to enable Proxmox/other OSes to work properly.

Recently replaced the internal USB boot drive with a small NVMe USB enclosure; using a 90-degree USB connector and using a dremel to sand away an opening for the cable to come out so I could mount the enclosure externally.

I got a gen10 plus microserver and liked it so much that I got a second one.

Throw FreeBSD on it and add a couple lines to /etc/exports and rc.conf and it's a NAS right out of the box

Thats perfectly reasonable. Double isolation but that isn’t too costly.

I’m talking more about people doing the same as you except they’re linking the VM storage to another vm (truenas) over network despite it all being on same host. Think it’s mostly because people don’t want to deal with zfs via command line

I'm also using NixOS for my NAS, you should look into quadlets (docker containers managed by systemd) to manage containers with nix.

Is snooty a autocorrect for Synology or some other product?

That is kind of exactly how zfs works though. The guide isn't complete, sure, but "rebuilding" the array is just replace the disk and run a single zfs command.

ZFS handles drive failures more robustly than anything else. There's a reason synology uses mdadm under BTRFS instead of the built-in BTRFS RAID features, and mdadm operates at the device level. That means that to replace a drive, mdadm has to rebuild the entire drive while zfs will only rebuild what's actually in use.

> Also, any guide like this that doesn’t guide you through “disk 3 failed, this is how you safely replace it” is imho incomplete, even if it doesn’t go through telling you how you know a disk has failed.

`zpool replace my_pool disk3 newdisk`

ZFS makes this completely trivial except for the "beeps and lights a led" part

First: Do you have backups of your single high-capacity HDD? That's my biggest worry. What's your plan for when that HDD fails one day, as it will?

The main reasons people go for ZFS and a "NAS" is checksums and data integrity protection, as well as maybe not wanting to keep their primary machine awake all the time (I personally don't).

Then there are useful features like snapshots, which means I don't have to worry as much about accidentally deleting, or over-writing a file and losing data.

I don't see anything wrong with using a main machine that's up 7x24 as a NAS, don't buy things for the sake of something, but I'm worried about your reliability and bitrot protection. (Yes, it happens, I've seen it first-hand thanks to ZFS).

> and I can even run shell commands on phone without ssh.

How do you do the authentication to cockpit, basic HTTP auth in front or something? I've always ended up making my phone SSH-capable rather than dropping down to anything else but public/private keys for authentication to my servers, guess I have a bit of conspiracism when it comes to that.

For home usage, if you have backups raidz1 is fine (just do an incremental backup at the first sign of trouble). If you don’t have backups, then you probably shouldn’t be running a NAS in the first place.

ZFS will auto-degrade performance if there isn't enough headroom. In addition, if you use SSD, you also want the headroom, because otherwise you end up writing and rewriting on the same small empty space which kills SSD (unless you have enterprise SSD, which have built-in headroom).

> I’m running an 8-drive ZFS RAIDZ2 pool. I’m wondering if you know — are the free space recommendations around ZFS cargo or real?

I'm not entirely sure, but it seems to me that free space (and the 20% reservation) is mostly a proxy for fragmentation, and you can therefore better look at fragmentation directly. That would mean that if you mostly store large files, there shouldn't be a lot of fragmentation even at high utilization. The whole "ZFS changes allocation algorithm from 80% usage on" is something of 10+ years distant past, and lots of things around the allocator have been improved. It's also something that probably isn't too different from the performance of other filesystems at high utilization, so it shouldn't be exaggerated.

I don't think they are. I've already filled a ZFS pool almost to the brim with no adverse effects.

If you are worried about someone breaking in to your house and replacing the bootloader while leaving your drives in place I probably wouldn't use the TPM auto unlock even if in theory secure boot should be able to handle this.

But in reality that will never happen and the only actual attack you need to be worried about is junkies breaking in and flogging the drives on facebook marketplace. For which, this level of security is fine.

I would kill (figuratively, anyway) to have no AI whatsoever. No slop machine threatening to replace my job, or turn my job into babysitting its stupidity, and hardware would be reasonably priced? That would be awesome. AI has brought me nothing but downside.

> I'd almost rather have no AI whatsoever and have storage 1/10 the price of pre-AI times.

Almost? ALMOST!?!?!

If you handed me a button that would make it like LLMs had never existed I'd be slamming that button so hard Sam Altman's clothes would spin around.

Return memory and storage prices to normal, undo the sloppification of ~everything, remove all these annoying "features" that are so useful they have to force them upon people, and make scammers actually have to put in a slight bit of effort, all at the "cost" of real human developers, artists, writers, etc. getting paid for their work....

If this is a hard decision for you, you are the problem.

I'm doing it anyway and I've found some secondhand deals. Maybe COVID homelabbers are offloading gear?

I built a 24 TB HDD NAS in a 36 bay chassis for around $1000 all-in: mb, ram, rails, chassis, rack, disks, hba, nic

Awful Watts/TB but the plan is to run it and the GPU rig on solar.

https://diskprices.com/ is great for this

You have SATA or SAS to pick from. The CPU requirements for a storage server are not high. On a typical ATX board you have motherboard SATA and can put SAS controllers in the spare PCIe slots.

My first "NAS" was two 22TB hard drives in a ZFS pool on my motherboard SATA

> You assert that ECC RAM being necessary for ZFS is just a myth but provide no justification for why that is untrue.

ZFS without ECC is no more risky than any other file system / software RAID without ECC.

As no one owes you an explanation, it would take you five seconds to Google this and discover:

1. It's been disproven, with one of the original ZFS developers chiming in.

2. The original source of the rumor was a forum post that somehow became canon.

> I bought Qty 10* of HC520 (12TB) SAS drives for $1000 about 3 weeks ago, avg age is about 2.5 years, still well within its rated lifetime.

That's cheap indeed. Enough headroom for some failing disks too. How is the noise and power usage? I didn't look at SAS drives at all, because my impression was that they're very noisy. I can place my NAS in a closed off room, but it's not too far away and I was afraid SAS drives would be audible through the wall. At the same time, the shucked drive I'm using presents itself as an WD Ultrastar, which comes very close to a SAS drive, and isn't very noisy.

Yes, indeed. ECC RAM is better than non-ECC RAM, also for ZFS.

The myth, popularized by a notorious thread on the TrueNAS forums [1], is specifically that ZFS requires ECC RAM, and will do worse than other filesystems without it, because scrubbing will multiply a single bitflip into a failed pool.

A ZFS core developer says that that isn't the case [2]. Here's some more reasoning [3], also about many other myths.

[1] https://www.truenas.com/community/threads/ecc-vs-non-ecc-ram...

[2] https://news.ycombinator.com/item?id=18480016

[3] https://kldload.com/zfs-wiki/myths

"It is important to regularly monitor the snapshot’s storage usage. If a snapshot reaches 100% of its allocated space, it will become invalid." [1]

ZFS snapshots use Copy-on-Write differently and have no such allocated space limit. Thus, we can do interesting things like snapshotting a file system after an OS is installed, and then roll back to that snapshot upon an OS upgrade failure, or even clone a new file system from that snapshot to have a different copy of the OS.

Essentially, the nature of the snapshots are different in both.

[1] https://docs.redhat.com/en/documentation/red_hat_enterprise_...

From the official documentation.

But why? Why not the standard oci-containers module?

I've seen the quadlet-nix project but don't get what additional it gets me.

Indeed. Didn’t notice and it’s too late to edit now.

To be fair, the "beeps and lights a led" part is very important in case of a dead disk in an array. The only time, I disrupted a service in production for longer than a few hours was when I didn't get the email of a degraded raid 1 array and the second disk died a few weeks later.

The middle ground I go with is to structure the pool using datasets and zvols, and then periodically backup the critical ones using zfs send. For example, I can live without my movie collection, but I cannot afford to lose personal photos.

This isn’t black and white. You might have huge amounts of non-essential data that aren’t worth the cost of off-site backups, but worth the cost of an extra disk of redundancy to lessen the risk. Even when you do have backups, it will reduce the risk of extended downtime (and possibly egress costs) caused by having to restore large amounts of data from backups.

My experience has been escalating fragmentation past 90%, along with lower performance owing to the fragmentation.

> It's also something that probably isn't too different from the performance of other filesystems at high utilization, so it shouldn't be exaggerated.

Well, other filesystems can defragment.

> junkies breaking in and flogging the drives

For this you dont need TPM. Just a LUKS key in rootfs /etc. He said TPM :)

Same, I'd smash it like it was offering me a million dollars. I cannot think of a single positive thing that has come out of the AI boom. It has been a net negative for all of society.

I don't really agree with the specifics but I can identify with the emotion. To me the problem with AI isn't AI, it's that investors are idiots and put hundreds of billions of dollars into AI companies that have near zero chance of being profitable enough to pay them back and those companies took that money and blew up computer parts supply chains that are vital to the rest of human existence. If we didn't have this massive build out that has no chance of repaying itself and instead had a build out that only doubled ram/gpu/nvme prices we would be far better off. As it stands, we're going to have a ton of gear that will be of questionable utility once any of the disruptors get their 100X+ per watt improved/different hardware going and to get that hardware we broke everything else. This ai stuff will also likely plateau and have the openweights models and opensource tooling catch up and at that point the companies are basically GPU REITS with diminishing demand as the stuff ages out and the disruptor's gear gets built for much better performance per watt. They've made it really ugly for us now and it will get really ugly for them in the midterm future imo. Not a good outcome for anyone.

Same. Every time I think about this, I come back to the fact that I was excellent at my job before AI, and even enjoyed aspects of it, so I would be fine if tomorrow it all blew up.

That’s comforting to know; that wasn’t true in the past for ext4 over lvm (is it true now?).

But what is that command? And how do you know which disk has gone bad?

I am sure I can get an answer from Google / Claude / ChatGPT, but a guide is incomplete without it - and the failure report should be active like a beep or flashing hardware light - I typically log into my NAS only a few times a year. A motd or other banner isn’t sufficient.

smartmontools makes this trivial and can email you which might be better than beeping and lights if you’re not always near your server.

I use that plus smartctl_exporter so I can do metrics and alerting off that.

The worst part of either approach is: how do I find disk number 39 out of 48.

In the Linux ISO community the math is different because none of the data is essential, however there is a lot of it. Backing it all up even one time to an off-site location would be a 2x expense for a hobby. So 3-2-1 is hard to justify.

Instead the math can be different e.g. it's ok to lose some data but not all. Therefore you might prefer the unraid approach over zfs where losing more than parity doesn't kill the whole pool.

I have external parity for everything with rsync. I don't have any storage that isn't matched with external backups. Fortunately I've never seen any bitrot or had a drive fail on me since I tend to replace them before they do. And yeah, I could add in some extra HDDs for RAID, but this is more a personal choice because of how much I hate SATA.

Filesystems like ext4 are because I value boring and rock solid stability over semi-experimental status of modern filesystems with all the features.

For most people snapshots are enough backup. It is still useful to have an offsite backup, but realistically fire is rare enough that you can risk it, and that is about the only risk most people have (you can have your NAS in a location likely far from where a fire might break out to reduce the risk farther).

The risk of z1 is that if you get a read error during resilvering, that data is permanently corrupted. The odds of this happening go way up the larger your individual drives are. This is why I chose RAIDZ2 for my NAS. I've had to resilver 2-3 times over the past 10 years, and never lost a byte of data.

Not sure why I'm being downvoted for being factually correct (and also, why I can't edit my above comment).

ZFS will definitely degrade write performance gradually from 90% utilization, and will hit a stronger cliff at 95%+. Same with SSD, using a consumer SSD above 90%+ utilization would rapidly degrade its lifetime. The effect will be smaller for very large pools and very large files, but the effect is stil there.

I'm repeating this so that people who set up these drives know what is actually going to happen.

AI has freed most people I know from the tedious job of writing actual emails and birthday greetings - leaving then more time to e.g. wash dishes or clean the floor.

Seriously, if arts and creativity is what sets humans apart from other animals, then AI has almost completely displaced our capacity to even consider doing these activities ourselves. People reach for AI when they should be composing a birthday greeting themselves.

But we now have trillionaires, aren't you happy about that? It means you might be the one some day too /s

This. My synology nas got pwned so many times i cant remember.

> You're describing ZFS.

..running on illumos.

What an incredibly rude comment

The only hangup with the myth "debunking" is that the point is that the corruption doesn't happen to a per-disk buffer, but to the in-flight data before it's persisted to your stripe.

Which means all copies of the data will be corrupted. This can be anything from an irrecoverable file to complete filesystem corruption.

But generally, yeah, not any more dangerous than any other filesystem, and ECC used to be cheap so it was a no-brainer, you should have backups anyway TBH if we're being honest about storage resiliency.

What has been debunked is the "scrub of death" issue, on a scrub a bad bit flip would cause an error, which would be copied over with good data -- well it was technically good before. It would be statistically difficult to have a fault on a read, then a clean read, then a second bit flip destroying the data.

Updated link to the post from Matt Ahrens here[1], which is one of the ZFS creators[2] not "just" a core developer.

[1]: https://arstechnica.com/civis/threads/ars-walkthrough-using-...

[2]: https://en.wikipedia.org/wiki/ZFS#2004%E2%80%932010:_Develop...

> five seconds to Google this and discover

...(after much longer) that it's a rabbit hole with nuance. https://news.ycombinator.com/item?id=18481910

It's more of a recommendation.

> 8GB+ of memory for the best performance. It’s perfectly possible to run with 2GB or less (and people do), but you’ll need more if using deduplication.

Snapshotting before OS upgrade is something that has saved my system more than once. It's not to be underestimated how much time is saved over having to restore from backup or reinstall.

Plus there is sending/receiving snapshots across systems if you want to.

Not really, zfs is part of it, but not the whole picture, you as a user you need more than just a file system.

Truenas for example add replication, external cloud sync, gui (yes it’s important not to wipe your data with wrong command), HA, custom caching, containers to extend ot with other apps, raid-z expansion with openzfs, and other for monitoring, user management and a lot more.

Other solutions also add more than just a zfs with ssh access.

Oh come on! I'll take a few years of expensive RAM in order for humanity to get wide access to something near as makes no difference to the Star Trek Ship computer.

Which is 100% what this has felt like to work with, all spring and summer.

So we're in a slop phase, it'll pass. The first few years of youtube gave no hints that we'd get stuff like Veritasium.

Ignore the slop and use the tools to create something you never thought you could do. That's what it's for!

Same here, the usefulness of the slop machines is quite limited compared to the downsides.

I obtained a used Dell PowerEdge and just stuffed a bunch of drives in there. It's old enough that the RAM is not overly expensive (DDR3), so it has 192GB of RAM to effectively cache the data, and then 32TB of raw storage. (Of course I regret not spending the extra money to take it to 384GB now.)

The nice thing about the array being big is that I can just RAID 1 it instead of worrying about tinkering with RAID 5, and leave a drive as a hot spare.

PCIe 2.5G NIC for the uplink, and then it can serve over SMB or iSCSI. The main use case for this thing, incidentally, is it just is a caching proxy that holds Docker images, models off of Huggingface, and so on.

What size discs are you using? This post is kind of hilarious, 24TB is achievable in literally a 2-bay NAS nowadays.

There should be more sites like diskprices. So refreshing to see no styling, no cookie banners, just pure information.

Usually if you’re using a NAS you don’t want to lose data, ZFS is not significantly more sensitive than everything else.

But everything is actually quite sensitive.

We’ve accepted lack of ECC because Intel decided it would be a product line differentiator, and serious customers who didn’t want random crashes or to lose data would buy chips with ECC.

It’s actually less of an issue these days because DDR5 has (by spec) some in-line ECC; won’t help with multi-bit errors but its an improvement on what came before.

> How is the noise and power usage?

I think they're same as SATA drives, just different interface. AFAIK they have the same physical dimensions and same internals.

You can do those things with LVM snapshots too, that's what they are used for.

Hey everyone, stop progress! Wolrah has decided that November 29, 2022 is as far as things should go. Because everyone knows before that day, there were no scams, every artist lived lavishly from commissions, and software was flawless and without any extraneous features.

36 bay? Why so many?

36 bays and only 24T of storage?

Can't you get that in a single disk that costs less than a kilodollar?

ATA (SATA), SCSI (SAS), and NVM (NVMe).

> in the past for ext4 over lvm

I used to look after storage arrays for VFX places.

everytime I saw lvm I inwardly sighed. The docs were terrible, almost as bad as MDADM. snapshots were for a long time unrecoverable. You'd then have to work out what pattern of LV you had, was it a suprise raid0? or a misaligned raid1?

zfs is a night and day compared to LVM/mdadm, two tools, rich help, the man pages are reasonably good, and once you understand zfs vs zpool, you're usually good to go.

zpool-replace(8)

https://openzfs.github.io/openzfs-docs/man/master/8/zpool-re...

ZFS got this right. MDADM on ext4 somehow did not.

EDIT: ext4 on MDADM.

How ZFS gets this right: zfs(8) on zpool(8). So, basically, ext4 and mdadm can talk to each other.

man zfs, man zpool, cron, your mta of choice (I like dma) :)

schedule a zpool scrub every month, send an email if it finds errors. zpool status will show the errored drive. zpool replace will initiate the drive replacement.

That’s awesome. But a guide that does not include a howto for alerting (such as the one linked) is incomplete.

By using /dev/disk/by-id when adding them to the pool - and looking at the physical label on the disk itself. All disks I ever worked with had that ID printed somewhere on there.

I'm actually in this boat right now. I have 2x26TB drives in a 4 base nas (two empty, plus 3x M2 empty) and I'm trying to decide the best way to set them up. I have around 3TB of backups that I care about as backups (but they don't need to be online, just archive backups at this point), and around 5TB of media that can be replaced.

I don't want to lose 26TB just for a mirror, and I have a spare 8TB USB HDD. I'm torn between unraid and just Debian, and I'm torn between just two separate devices and one RAID 1 partition one RAID 0

Or like me where you get two SSDs in your pool that happen to likely be from the same batch and both decide to permanently disappear at roughly the same time :(

People were Googling these things. Those who were good with words, didn’t do that before AI and won’t do it today. They don’t have to.

someone at work unironically said they use ai to compose "thoughtful responses" to people, and my immediate thought was that what they are doing is the literal opposite of thoughtful

DDR5 on-die ECC is to achieve acceptable yields in the face of denser process nodes that decrease the reliability of RAM cells. It’s not clear how much of an improvement that is to what we had before, other than allowing for higher RAM speeds. It doesn’t replace side-band ECC.

I know that LPDDR5 has ECC, and not just single bit AFAIK, but if you enable it you lose some memory capacity (the 128 bit bus minus 16 bit for ECC error correction, making it effectively 112 bits)

This reply is exactly right about using enterprise drives (hdd), and ideally SAS. Note the power usage of SAS vs sata for an identical drive make/model is very small sometimes 0.5 W to 1.5 W at idle (idle power draw is what matters for spinning HDD’s as underload it doesn’t increase much and idle is what you’ll be at 98% of the time). Also enterprise SSDs are much better than consumer SSD‘s for ZFS, and be aware that when you get above the 4 TB size SSD drives frequently use more power than HDD drives (i’m mainly referring to enterprise non-consumer level drives as that’s all I run for 100s of disks across many zfs systems over past 15 yrs).

Also something people forget for home nas or figuring out cost for a nas, the power draw of the entire system times your electricity price you then have to multiply this times 2x to 4x times if your climate requires cooling air conditioning of any kind

I immigrated wanting to become a millionaire. Never found time working my ass off. Now that I get pension and tokens from Musk and Zuck, I can finally try!

That's quite uncharitable. What are you trying to achieve here?

They (probably) don't want to stop progress (especially unqualified like this) in general. They'd like a world where LLMs didn't come to exist.

And whether LLMs are progress at all remains to be proven.

That will be a single point of failure, though. (Which you might not care about depending on the backup strategy.)

The 36 bay was all I could find on fb marketplace.

I fell in love with the silly idea of squeezing 10G NIC + 10G writes (to a 2x12 RAID 10 HDD array) on the thunderbolt bus of my 12 year old macbook pro.

Then I got distracted with the server mb and truenas and ZFS.

smartmontools misses data corruption at the drive layer, as it trusts the drives to report faithfully.

"In hashes I trust"

How do you know you haven't seen bitrot? What rsync arguments do you use to error out when an old, untouched file suddenly changes?

What do you use for external parity? Par2?

If you were to replace your current drive with a new larger one and your RAM or your SATA controller silently corrupted the files how would things play out? Would these corrupted files propegate to the backups?

Ext4 is battle tested so I understand your reasoning. I think you just need to figure out a way to detect silent corruption and a way to snapshot your files in case they do get corrupted.

From experience I can tell you that it is an absolute pain having to manually sort through a bunch of files trying to detect which ones are corrupt and which ones are good.

rsync does not protect you from bitrot unless you have some script that specifically collects the rsync log and tracks when existing files change, and you somehow know every file that should not have changed and get alerted that a file changed when it shouldn't have.

Even with that, that is still just 2 equally authoritative copies, without any way to know which one suffered the bitrot.

Ok the tie-breaker data could come from history. Yesterdays log and the day before gives you the extra data points to say which side changed. As long as the log itself is somehow above reproach.

Except this script is a myth anyway. You would still need to have something where you tag files as "this file shall never change again, so if it ever appears to, tell me so I can ok or reject the change." and you would have to actually do that tagging and reviewing.

What do you dislike about SATA?

Fire, malware, accidental deletions, capricious RAID controller (pre-ZFS). And that’s only the stuff that happened to me. Add power surge, theft, correlation in SSD failures (eg power on counter overflow firmware bug), damaging the array while moving, etc.

> The odds of this happening go way up the larger your individual drives are.

Technically true.

Practically? There are several other major factors - such as the quality of your drives, whether you periodically run zfs scrubs and/or disk read tests, and how long a resilver takes to complete.

Behind those are the eternal duo - size of budget and cost of failure.

> The odds of this happening go way up the larger your individual drives are.

This is a common claim, but honestly, citation needed.

We can't just apply bit error rates from the datasheet that haven't been updated in 15 years. I'm sure a 2-day rebuild is a little more risky than a 4-hour rebuild, but I'm not convinced it's by all that much. Especially if you had a monthly scrub going to prevent disk rot and disks secretly getting super fragile.

> I've had to resilver 2-3 times over the past 10 years, and never lost a byte of data.

And how many times did you have a read error on one of your other disks during a resilver?

> Not sure why I'm being downvoted for being factually correct

Maybe because your statement about SSDs is incorrect. Consumer SSDs have spare blocks too, and wear leveling prevents the scenario you're describing.

Are you being sarcastic? There's no way you believe people couldn't wash dishes or clean floors because of all the emails and birthday greetings they had to write, or that people have almost completely lost the ability to consider making art.

> We’ve accepted lack of ECC because Intel decided it would be a product line differentiator, and serious customers who didn’t want random crashes or to lose data would buy chips with ECC.

AMD has been allowing ECC on lots of regular hardware for a long time.

People don't tend to buy ECC for desktop use because it costs significantly more (used server ram is/was often cheap... but it often doesn't work in desktop boards), and the performance specs are poor.

My home servers are mostly retired desktops, so they get my old desktop ram and I don't want to pay premium prices for jedec speed ecc ram on my desktops, thanks.

Since DDR5 doesn't include reporting on bit errors (afaik), it likely means much fewer single bit errors, but most experienced errors will be multi-bit. Although, I dunno what proportion of bit errors is on the ram chips and what's on the bus... there's no protection from bus errors.

If there were reporting, you could replace chips with high error rates, but without reporting you'll keep running them until they fail enough to notice.

When we allocate space to an LVM snapshot, the COW data is written to that space and then when that space is used up the LVM snapshot must either be given more space or must be deleted.

When a ZFS snapshot is marked, then changes to existing blocks in the snapshot as well as new writes all go to free space in the disk. There is no pre-allocated space within which these COW changes are written to.

Thus, we could have a single ZFS snapshot marked today followed by years of data being written to. This is not the intent of an LVM snapshot and the warning that I had quoted above is worth reviewing once.

Is it a genuine birthday greeting if you just let AI do the job?

There's a difference between standing in the rain and being hit by a tsunami, even if it's both just water.

no. just sata & sas. There is no nvme spinning drive and flash loses bits just from sitting, it's not archival. flash is good for working not for storing.

I suppose I couldn't guarantee I haven't, but I keep a total backup, and other backups based on file lists. I keep a record of every file of importance that I've ever written/modified/interacted with, and those lists themselves are also git managed.

I compulsively look at the rsync backup running to see what's going to transfer, and typically do a dry-run first. That's no perfect solution, but it works for me. If I see something odd I don't quite remember, I check the hashes between both drives, and every time it turns out it was just a path change since I try to keep my data as organized as possible.

rsync arguments are just the plain `--archive --acls --xattrs --verbose` and depending on which backup I'm doing, `--recursive`, `--delete`, `--files-from=`. Nothing other than vanilla at all.

Yeah, for that I use periodic scrubs and zfs event daemon

Misspoke. I didn't mean that type of parity, but just hashes, and a matching drive (in make and capacity) externally for an exact copy.

Does Synology with its beeps and bops catch that?

> rsync does not protect you from bitrot unless you have some script that specifically collects the rsync log and tracks when existing files change, and you somehow know every file that should not have changed and get alerted that a file changed when it shouldn't have.

You're right and that's the big thing most are missing. Name your files and write your scripts accordingly I'd say.

My rsync backup'ing script first does a dry run and then serious analyses on the result of the dry run is done. If anything looks out of the ordinary, the script refuses to go ahead and warns me.

rsync silently copying bitrot to backups is the stuff nightmares are made off.

> Even with that, that is still just 2 equally authoritative copies, without any way to know which one suffered the bitrot.

I personally have a sizeable percentage of my files (like, say, family pictures) that have part of the checksum in the name of file:

    dsc003784-b3-79a7208c51.jpg

This means that file should have it's Blake3 checksum beginning with 79a7208c51. And these have been vetted, so in case there's a mismatch, the authoritative file is the one with the proper checksum.

And the system/disk that gave back a bogus checksum has explanations to give ; )

> Except this script is a myth anyway. You would still need to have something where you tag files as "this file shall never change again, so if it ever appears to, tell me so I can ok or reject the change."

I don't think it's a myth. My script does that with the checksum as part of many filenames: if during the dry run the script notices a file with such a checksum is present but different on both systems, I know immediately that something is off.

+100 to this. I don’t want to be overtly paranoid but you genuinely are unlikely to notice real data corruption unless you actually checksum.

It is a matter of time when you’re storing a lot of data — not if.

It's a paraphrasing of the popular 2024 Joanna Maciejewska quote: "I want AI to do my laundry and dishes so that I can do art and writing, not for AI to do my art and writing so that I can do my laundry and dishes."

I believe it was rhetorical. They were trying to illustrate that we've automated tasks that people might find to be fulfilling (drawing, writing) but are still a long way off from affordably automating drudgery.

> or that people have almost completely lost the ability to consider making art

They haven't lost it, no, but there's a lot of financial incentive to stop paying people to do it.

This is true, but if you use a file system that does scrubs (and you should use such a file system for a NAS), and if you keep the NAS on 24*7, this won’t be an issue.

that was their point

A NAS is plugged in. Flash works fine for it.

For 90% of birthday greetings, are they ever genuine?

Like seriously, outside of some close friends and family, are you sitting there deliberating over a message for your coworker Steve in the Slack chat or a cousin you barely talk to outside of a birthday and Christmas card?

I used to write birthday emails to my friends years ago. I quit when calendar systems started automatically promoting me when birthdays happened. Remembering someone's birthday no longer seemed like a significant effort on my part and so it wasn't worth the bother since it didn't say as much.

Malware, accidental deletion, RAID controller issues - all things ZFS with snapshots are immune to. You can decide if the other worry you or not - the risk is not zero, but it may be acceptable. Your backups are all subject to similar issues.

The cables and how they have to be managed/positioned in a case. I've had panics over a drive failing when it was just the cable after I'd already replaced the drive, restored data, and ran tests on the old drive that turned out to be fine each time.

> This is a common claim, but honestly, citation needed.

This is based on the data I had available 10 years ago when I set up this array… If you have data that points to drives having fewer read errors per GB in recent years, I'd love to see that citation.

> And how many times did you have a read error on one of your other disks during a resilver?

I don't know, because I have RAIDZ2, so that read error would probably be corrected without it being reported to me.

I appreciate the reply, but my warning consumer SSD is not incorrect. Perhaps I should have clarified that While they typically have a small overprovisioning of 7%, enterprise SSDs have close to 4 times as much, exactly to allow more writes (which is very relevant if running a ZFS raid on the drive). Keeping 10-20% headroom will prolong your SSD's life significantly, and increase performance of ZFS.

That would be a thinly-provisioned snapshot in LVM.

>> The odds of this happening go way up the larger your individual drives are.

> This is a common claim, but honestly, citation needed.

FWIW, here's the quick & dirty argument for that:

Assume:

- We're working within a given generation of disk technology, with comparable drive and interface tech.

- The drives' I/O speed is relatively constant, not scaling with the drive capacity.

- The chance of any given bit (of data on a disk) going bad, in a given time interval, is relatively constant.

- The drives, regardless of size, will be some fixed-ish percentage full. (If nothing else, the boss won't sign off on overspending for unneeded capacity.)

From there - the time needed for resilvering scales linearly with the drive size. As does the number of non-redundant bits that could go bad in any given second of the resilvering. Linear time linear - so the chance of data loss during resilvering rises with the square of the disk size.

> AMD has been allowing ECC on lots of regular hardware for a long time.

Often unsupported or untested by the motherboard manufacturer, because the precedent set quite a while ago has stood so strongly.

> because it costs significantly more [...] and the performance specs are poor

Because it's niche. If a large share of the desktop market was using ECC memory, the extra cost would have been 10-12%, and a typical kit would push the clocks and timings at least as much as non-ECC memory.

> Since DDR5 doesn't include reporting on bit errors (afaik), it likely means much fewer single bit errors, but most experienced errors will be multi-bit. Although, I dunno what proportion of bit errors is on the ram chips and what's on the bus... there's no protection from bus errors.

Yeah that sucks. I'm really hoping that DDR6 standardizes a light form of end to end ECC. (Bursts are 24 bits wide and 24 bits deep, so on top of the 512 bits of payload there's a bunch of extra bits that can be used for ECC. If memory chips would just store 4 extra bits per 128, that's enough for reasonable ECC on a per-burst basis.)

Was it a genuine birthday greeting when you bought a card? Or used an online card? I don’t really see the difference. The point is you thought of them, not the actual words

I have multiple home servers running ZFS and the first thing to check when encountering errors is the SATA cables. Those blue foil fan-out types are the worst from my experience.

Snapshots don’t protect you from malware. I bet 99% of home users use the same credentials on all their machines, once a malware compromised one, the others are compromised within seconds.

[edit] also snapshots aren’t really workable for large files. Remux a movie file and now it occupies twice the space.

Go ahead and let them try consumer SSD's. It seems to be a lesson everyone has to learn with ZFS once.

> - The chance of any given bit (of data on a disk) going bad, in a given time interval, is relatively constant.

We can't just assume this, though. Especially for data that's being checked regularly, because the drive can see how many pre-ECC bits have gone bad and rewrite iffy sectors.

And even if it's true, it's normal to go years without errors. If you have a 15% chance of getting an error each year, and your degraded array has 4 drives without redundancy, then the chance of getting an error in 4 hours is .02% and your chance of getting an error in 48 hours is .3%. Having that risk show up once every few years is pretty negligible compared to everything else that could go wrong with your array.

On top of that, if you don't have a hot spare and it takes 5 days to get a new drive, the difference in resilvering speeds is now 5.2 days versus 7 days and that barely matters.

The main situation where you have to worry about extra drives failing is when you think the resilvering process will greatly accelerate drive deaths. And that depends very much on the specific numbers you have in mind.

Thank you. I went back to the link I shared and saw that there is section just below on thin provisioning via snapshots.

> This is based on the data I had available 10 years ago when I set up this array

Data you collected or data you found? Any idea where it was?

I've only seen oversimplified math, not real data, and different versions of the math make different assumptions.

> I don't know, because I have RAIDZ2, so that read error would probably be corrected without it being reported to me.

Zpool status gives you a per-disk error count. I kind of assumed someone doing a resilver would use that command.

> the precedent set quite a while ago has stood so strongly.

I wasn't really around for it, but I appreciate the early penny pinchers that developed fake parity ram. The IBM PC required a parity bit per byte?, but you could add an xor chip to calculate it at read and save out on the expensive ram chip.

I wouldn't mind paying a little more for ECC, what with the extra chip and a little more circuitry, but desktop ECC often starts at 50% more, which I'm way too cheap to pay for.

With AI you don't have to think about them.

My favorite parity trick was how the N64 was designed around ECC RAM chips but didn't bother using ECC, so it has a ninth bit at every address that the CPU ignores and the GPU uses as extra storage.

> Data you collected or data you found? Any idea where it was?

I think it was a lengthy blog post citing manufacturer data, found via the FreeNAS forums. The added safety of RAIDZ2 was a good tradeoff. Someone else on this thread reports they lost two SSDs in rapid succession, which I hear happens sometimes with drives from the same batch. In any event, I felt no need to launch an independent investigation of the math and underlying assumptions.

> Zpool status gives you a per-disk error count. I kind of assumed someone doing a resilver would use that command.

Yes, I did indeed use zpool to do the resilver… I don't recall if there were any read errors on other drives at the time of those failures, but probably not, as the scrub probably would have flagged it. They've all been replaced with higher capacity drives, so at the moment I see "No known data errors"

Great business opportunity there. Make a free service that sends AI generated birthday notes to people. All the user has to provide is the other persons birthday.

People cannot be relied upon to provide accurate birthdays but nobody would suffer the social faux pas of an incorrect birthday congrats note. Nor would they send it to a spam catching email, but rather are guaranteed to send it to a regularly checked address. They know the persons birthday after all.

You can then sell high quality birthday information correlated to contact information to ad agencies.

Fuck this current internet. So dystopian.

Okay, interesting. Also to be clear I do think Z2 makes sense in a lot of situations, but I suspect that it's not very dependant on drive size like a lot of people think. If I wanted high reliability I'd use Z2 even with small drives.

Also to reply to your other comment: Personally one extra drive is a small price to pay for the peace of mind.

I achieve peace of mind via backups. For each copy, once I get the AFR below 2% I'm not too fussed about pushing it lower.

Then the receiver AI can thank the sending AI before deleting.