Hence the "continued" part in the title.
Example: OpenBSD and NetBSD both will install X11 and a window manager by default from the installer, if you want. FreeBSD doesn't. They're working on it, but they've been working on it for years.
You can integrate a fleet of Linux storage servers with a collection of K8s clusters more easily than a you can BSD storage servers. Getting BSD and Linux talking smoothly needs Real Unix Skills.
Colleagues in K8s shops tell me that their k8s guys come for help with complex tasks like.... deleting files. They know to construct and deploy containers, but not how to delete a file without using a GUI (meaning Windows or at best WSL).
I am told that some companies are now demanding "bash" among the list of skills for would-be container-herders. And here, "bash skills" means you should know to to create, edit, and then delete a file -- over an ssh connection.
[0] https://forums.truenas.com/t/clearing-the-air-on-build-scrip...
At work the trend is only adopt open source on a paid plan, adopting or relying on open source otherwise is just waiting for the rug to be pulled. Essentially treat open source as closed source.
How did we end up with this state of affairs?
Which path you go down depends on a lot of factors of that particular piece of software.
Can I trust FreeCORE to be a stable, secure project that hangs on to my NAS data if it’s an unfunded fork? Or would it be better to give up some openness for something like TrueNAS SCALE?
When it comes to stuff like Proxmox and TrueNAS I think a lot of what keeps those good products for homelab use is the fact that they have commercial customers paying for the development. I can count on TrueNAS being around and getting updates because they have a proven revenue stream.
That said, it’s a balance, because sometimes that commercialization becomes a rug pull. A lot of the viability of those projects for homelab use revolves around the companies that put those products out recognizing that they have a customer base that will not pay and accepting that fact without deciding to try to squeeze out every last person who just might pay by compromising the free edition.
I like the way Proxmox and OpnSense handle this: you get the community patch channels which are “not for production use” which scares the corporate businesses into paying but has zero negative impact for home users.
To be clear I’m also not saying FreeCORE specifically is a fly-by-night project or unstable or anything. I’m genuinely unfamiliar with it. I’m just saying, sometimes projects like that end up not being the right choice for certain use cases.
One must have principles onesself to see this as imortant though.
A lot of people use words like "practical" or "pragmatic" to avoid inconvenience.
I've already been using zVault for a year which is a fork of the last truenas core, but it never got any further updates and it's about to become impossible to run up to date software in jails on the old host. The pkg repos that support the host version are about to go eol and the binaries will be gone. So I've already gone through the "what if this doesn't go anywhere?"
My point: It's a pain but I won't relieve that pain by running trunas scale.
I'd rather try switching to xigmanas or dump the whole concept of the front end and just run plain freebsd and learn a few commands to manage storage and file services directly and a maybe a seperate dedicated jail manager like bastille. At least that way I can never be stuck in this position again where I can't update my own software because someone else stopped updating their software.
This entire time you’ve been swapping between NAS distros fiddling with your NAS box my TrueNAS SCALE installation has required approximately zero of my time and has never lost the ability to be updated.
To me, the NAS aspect of my homelab is the least interesting and is something I just want to install once and forget about.
Did they violate some principles of the open source ethos? I guess so; they deprecated build scripts, but the source code of the product is still available and GPLv3. The practical impact to the end user is effectively zero. I’m not even sure I would personally need the source code to be available when it comes to a NAS solution — I did consider unraid as a potential alternative before settling on TrueNAS.
If TrueNAS/iX whatever turns around and sells to Bending Spoons, then I’ll of course owe you a beer. Until then, their leadership is just trying keep the project alive - and funding other open source efforts at the same time.
Maybe AI fucks all of this up, who knows ….
> At work the trend is only adopt open source on a paid plan, adopting or relying on open source otherwise is just waiting for the rug to be pulled.
"Won't do free labour for our commercial enterprise to profit off of forever" is a rug pull now? It's incredible that you can type these words with a straight face.
People act like you're saying Disney should be able to hold a copyright on Mickey Mouse for over 100 years.
Calling it a "rug pull" when the creator of an open source project decides to stop giving away future work for free is a delusional level of self-entitlement.
(1) The previous releases are still open source. If you don't like it, fork the last free release and maintain it yourself.
(2) People deserve whatever the market is willing to pay for their work. See above.
Frankly, if D. Richard Hipp decided to stop giving away future versions of SQLite for free, I would be the first person to congratulate him for finally having the courage to put his foot down and say "my fellow committers and I deserve at least a few million dollars for 20 years of work making something that billions of machines rely upon."
The issue with rug-pulls like this is that they are deceptively masquerading with this philosophy as a calculated business strategy to adopt users. Then in this case they start using deceptive tactics like removing build files so they can continue appearing to be fully open source, even though they are violating the spirit of that term. Finally, they drop the facade when it is deemed valuable enough. Even if you are a strict capitalist, there is nothing virtuous or meritocratic about these kinds of business strategies, they rely on tricky, concealed intentions, and rightfully deserve criticism.
I'll also add that there is nothing "courageous" about deciding to extract more value from your labor at the expense of public accessibility. True courage is found in acts of selfless sacrifice.
No it isn't. Open source isn't inherently political in any way. Maybe that's what it's about to you, but I open source things because I want to give to others out of a general sense of wishing to build up the whole community. I don't have anything against capitalism.
And with BSD licenced code this is their right.
And it's my and anyone else's right to observe all of this and criticize it.
Fuck that noise. They are the ones breaking the social contract, not the folks criticizing them for it.
Shameful of you.
This is literally why BSD exists in the first place. To allow companies compete freedom with the code, including freedom to not release changes.
I hate to break it to you: basically every commercial storage vendor on the planet bases their appliances on either BSD or Linux and not a single one is sharing their proprietary code with you. There’s nothing shameful about it, they built a company to make money.
I would phrase it slightly differently. BSD licensing exists to allow everyone complete freedom with the code, not just businesses. It is a gift to the commons without restriction on who may use it or for what, which is why I love BSD (and other permissive licenses).
They're allowed to stop publishing new work whenever they want to. That's part of the free, open-source software game, wherein: We're all free to provide as much as we wish, and also free to consume as much as we want. Nobody necessarily owes anyone else anything and it's really very beautiful in these ways. This is enabled in large part due to the very low cost of distributing copies of intangible software (which can be very close to zero).
But when money gets involved, then things become different. Contracts can be had. Support requirements can be determined. If a contract requires 5 years of support, then it can be reasonably expected for support to continue for at least 5 years.
It's OK to want that, and to never use software that doesn't involve money. It's also OK for an author/company/org to not want that, and to never agree to those terms.
These are different goals, but one party doing it one way doesn't detract from another party doing it some other way. It's all OK.
I explicitly stated in my comment that we pay for open source (via vendor support). That is not the issue.
It is now a common business strategy to commercialise an open source project. FreeNAS was commercialised as TrueNAS, which is now having some of the source pulled.
TrueNAS doesn’t exist in a vacuum. It’s built on source projects.
Your statements clearly took issue with the fact that your company has to pay open-source devs whose work it uses for reliability. You presented the options as "pay them, or else get rugpulled", and then went on to say "how did we end up in this state of affairs [where we must pay to not get rugpulled]". There is no possible way to interpret your comment other than lamenting the "state of affairs".
> TrueNAS doesn’t exist in a vacuum. It’s built on source projects.
As long as it isn't violating the license of those projects, they still don't owe anyone anything. If it is violating the license, complain about it on those terms. But your complaint was about open-source in general, not specifically only this project.
But, at this point, AI makes code cheap enough that I would just fork the open source software and maintain it, instead of paying for a third party. Heck, even with saas, I may clone internally.
The entire reason to use it is because it's trained on huge amounts of expertise.
It's like those SOHO router vendors that release incomprehensible tarballs of their GPL components; lotsa luck building it.
If I'm paying for closed-source ZFS, I'll go with the real thing and buy a Solaris license.
I think scale is also BSD licensed (atleast the gui and middleware)
> The “Corresponding Source” for a work in object code form means all the source code needed to generate, install, and (for an executable work) run the object code and to modify the work, including scripts to control those activities.
But he's very pro-ai, though, and is vibe closing other projects like https://www.phoronix.com/news/Linus-Torvalds-Vide-Coding
Linux is definitely going to speed up as it starts to get vibe coded. In a few years, I bet that it's all going to be built with agentic loops, and nobody will be looking at the code at all.
And why do you act like it's an insult? For most problems, reading and writing code by hand is dead. For the remainder, it will be soon.
> "Also note that the python visualizer tool has been basically written by vibe-coding"
He'll keep ramping up his vibe coding, because he can see it's the future.
As far as I see it, he is vibe coding tools for himself but (hopefully) still manually monitoring crucial bits of software like Linux.
But AI is already better than most people at coding, so obviously Linux is going to end up vibe coded.
I suppose the key is whether you consider vibe coding a bad thing or a good thing. Because if it's good, then that's good news, right?
How long has it been around? Anyone tried it and can give an impression of what it's like?
I gather now there are two flavors: CORE which continues to use FreeBSD under the hood, and SCALE (aka Community Edition) which is based on Debian Linux. If I understand correctly, the former is only receiving security and stability updates, and iXsystems is developing all new features only on SCALE.
If you were to install fresh today, would you prefer this (FreeCORE), BSDnas, or the official SCALE version - and why?
No MacOS clients, so don't need SMB Spotlight.
When I finally became determined to give ZFS a serious go around 2016, I looked at all of my options -- including throwing some hardware together for FreeNAS, and running FreeNAS in a VM.
I decided to just use an existing Linux box that was already doing storage stuff and add ZFS to it.
It's been fine ever since. Keeping ZFS going isn't deep voodoo at all; of all the things this box does, ZFS has probably been the most hands-off aspect of it.
It's a lovely filesystem and it can work in very stodgy ways with zetabytes on Serious Hardware, but it can also work with relatively small devices on silly hardware.
Another example: I've got a ZFS scratch disk that has seen dozens of terabytes written. It's an old ~120GB Intel-branded SATA SSD that is connected with a very cheap SATA to USB3 adapter cable. I've been using it this way for years.
All of the Internet lore has told me that this must result in resolute failure. Whether because USB is broadly not a serious data bus, or because ZFS doesn't like non-serious things, or because 1 copy is 0 copies, or because SSDs are fragile creatures that hate write cycles, or because I've created a portal to hell, or something else: The lore states that it must result in failure, and usually stipulates that ZFS would amplify this concern.
But it's been great. Really. Zero issues -- except when I unplugged it on accident while it was hot...but even then, it came back fine. (Of course it did. ZFS is intended to be robust in the face of unplanned failures regardless of cause.)
And because ZFS: If it ever has any issues, then they'll be reported and logged either in the normal course of doing things, or during a periodic scrub. Every read is checksummed (even in normal operation), and every discrepancy is reported. It's been clean.
It is my intent to keep it as a scratch disk until it becomes problematic, and my further hope is to be able to analyze those problems once they occur.
It won't last forever, but it's lasted for quite a long time so far. (A little bit of inquiry with smartctl and some math just now shows that it has endured writes equivalent to ~43,000% of its capacity.)
If you need your hand held to file share, that's why Windows Server still exists and is the gold standard for SMB.
you also get a whole community testing the stack, fixing it too.
And don't be that pissi... when I started 20 years ago, stuff like this would have been the gateway to linux world and is for sure still today for plenty of people...
If you prefer driving towards Linux all that is necessary is to migrate your TrueNAS CORE instance to TrueNAS Scale which is Debian based. (It is reportedly fairly painless).
This specific project is for people that didn’t like being pushed towards Debian and wanted to stick to BSD.
Lets hope FreeCORE will make it longer.
Some comparison between Linux based TrueNAS SCALE and FreeBSD based TrueNAS CORE here:
- https://vermaden.wordpress.com/2024/04/20/truenas-core-versu...
Regards,
vermaden
One nitpick is that while the site looks clean, and there seems to be some good amount of documentation, i think there was a lost opportunity for a simple, yet more universal description of *what this is*! I mean, i get that this descends from FreeNAS and TrueNAS CORE...but why could they not add a small, single line of text that also helps explain it to other folks who don't know fully what the means? Here is a crappy but at least slightly more descriptive option:
"FreeCORE is an operating system and associated software for your network attached storage (NAS) needs. FreeCORE descends from FreeNAS and TrueNAS® CORE, and builds on FreeBSD and OpenZFS..."
I'm not going to win the Pulitzer prize any time soon, i get that. But at least most folks will get a little more of the gist of what this FreeCore thing is or could help them. I'm only making this suggestion because i want more folks to participate and enjoy free and open source software. I'm not expecting someone's grandparents to understand this, but i betcha other folks who for example might be migrating from Windows NAS-like offerings might have an interest...so making the intro at least a tiny more descriptive helps.
Again, even with my comments, i do wish the team well, and only want them to succeed. So please take my comments as merely a suggestion for improving things for all to benefit. In fact I'll be playing around with this to set it up and test on my homelab.
The migration path went smoothly but I preferred the FreeBSD setup if only because it forced me to do something a bit more interesting in my spare time. I do quite a lot of Linux work in my day job and it was nice to occasionally work on something that was surface-level similar but subtly and not-so-subtly different.
[1] Over 11 years I have replaced every component (even the CPU/motherboard) at least once except for the case. Migrating from MBR to UEFI was "fun".
Usually, for JBOD file archiving people want to avoid cutting-edge anything. Sometimes Linux kernel less-popular fs updates add or remove data-corrupting fringe filesystem bugs and features. People usually discover this on their own after the first SSD/HDD starts to wear-out, upgrade breaks compatibility, and the solution is re-install/copy-over everything.
Accordingly, I was surprised people now deploy btrfs on linux / for some distro, as your options when things go awry was limited at one time. Would be cool if it was finally production ready, and actually reliable on degraded hardware.
Too late for me, started my own mini vanity NAS/automated-testing project last year with a few built in features similar to https://guacamole.apache.org and Proxmox. If the OS build stops sucking over winter break, than I may share the build-script someplace for people to pull off parts they find useful, and a USB-bootable portable rtx GPU image for test-drive/recovery scenarios. =3
This was a while ago so it may have changed but then changes can introduce new bugs...
ZFS does have bit-rot checksum detection/repair, but the encrypted file feature was a bit of a YOLO feature for awhile. Not sure where it is at these days, as ddr5 ram prices skew away most of ZFS best key features. =3
IMO use LUKS unless you really need the features of ZFS native enc.