BSD Future/Roadmap

[CJ]

I don't know how much folks are paying attention to TrueNAS and what they're doing now that they have both a Linux and BSD version but with the current state of affairs regarding BSD I wanted to bring it up here and see what Deciso and the maintainers where thinking regarding the future of BSD.  Here is what Kris Moore of iX Systems had to say regarding their thoughts on it.

https://www.truenas.com/community/threads/what-is-the-future-of-truenas-core.116049/page-2#post-804807

His whole post(and that whole thread as well) is worth reading, but I wanted to highlight one particular part.

Take a look at all the work Intel is pouring into Linux for up and coming technologies like CXL, which will be highly relevant to all of us very soon. Does a vendor like iX want to duplicate that effort into the FreeBSD stack, often at a very high cost of time and resource that could have been used for other efforts? Or if we do for that one tech how about others? Pick any other random feature which already has a perfectly viable equivalent on the Linux side of the fence with a very active user and developer base. You have to know where to best spend your (very limited) time and resource, and re-inventing wheels without some very tangible benefit (not just "caching up") isn't a good strategy. That said, I'm not one of those "FreeBSD is dying" doomsayers, I think it won't just "die", its going to just become more and more of a hobbyist and academia focused OS, and that's not necessarily a bad thing. But I think the odds of it having a serious resurgence at this point in wider industry relevance is very very low. Of course it could still happen, and maybe I'll eat my words someday, but just looking at the facts right now in 2024, I'd be hard pressed to place any real bet on that.

I will say that I'm very happy with OPNsense and I don't plan on moving away anytime soon.  But it makes me wonder and I'd like to know what those in charge can share regarding where they see things headed.

[Patrick M. Hausen]

"BSD is dying" has been proclaimed since 1994. I call bull* on this particular statement. Deciso delivers appliances that route and filter 60 Gbit/s. Netflix delivers 800 Gbit/s of video from a single box running FreeBSD. The FreeBSD foundation sponsors developers targeting specific areas where we might be behind, like e.g. the complete reworking of the bridge code.

If Realtek interfaces don't work as well as they seem to work on some Linux platforms that's because Realtek provides binary drivers for Linux but they are not open source. They don't publish the specs of their silicon without you signing an NDA. And their hardware is still crap. Just avoid.

I'm tired of these discussions. Does OPNsense fit your needs? Great. A move to Linux is plainly impossible, because there is no "pf" packet filter on Linux. It would be a completely different product. For those who want Linux under the hood for whatever reasons there are plenty of alternatives.

Kind regards,
Patrick

[cookiemonster]

I do love freeBSD for infra but if I am honest, I struggle to understand why routing and filtering seem to take so much work and many new and old users get frustrated.
If we look at how many new users install OPN and immediately have to ask: help, why am I getting only X speeds  when my line is Y bandwith. Followed by answers of the style a) have you tried a, b, c tunables ? ; b) you need at least {insert value} of CPU clock speed and/or single core thread performance.
All these aren't normally required under linux.
Then we all have heard of what Netflix achieves _but_ I've never seen if that is routing or filtering, or as I imagine, the capability of their tuning to deliver as output from a single host. That's to say, the filtering and routing for it is maybe done on another type of device. I'd love to know that.

[Patrick M. Hausen]

Netflix simply deliver that streaming bandwidth. No filtering involved and the FreeBSD box just pushes out the data to its default gateway, which is probaby some big Cisco, Juniper, Arista, ... gear at these speeds.

I have never had a problem meeting the 100, 250 Mbit/s or now 1 Gbit/s fiber at work with OPNsense and PPPoE. No tunables, no messing around. A Deciso DEC650 appliance. I consider something like this a low end system.

[passeri]

Despite having negligible posts to back me up I differ from some of the views.

For years I used a couple of Draytek routers. They worked, I was happy, then my needs and topic interest grew so (after reading about several more) I tested four firewalls: (alphabetically) Firewalla, IPFire, Mikrotik and Opnsense. Result is I use Opnsense as my primary and one secondary router and Mikrotik mostly for some low-cost wireless functions.

Ask me as a relative novice to stand up a router for SOHO use with a VPN, DNS, NAT, sub-net isolation and some sort of IPS and I will pick Opnsense first, confident I can do it quickly and probably accurately. It's pretty clear, highly flexible, good documentation and foum support, and it works.  Other firewalls encounter the same or related problems as mentioned by cookiemonster — read their forums. 
In fact I have installed Opnsense on the Firewalla box in place of its Linux-based proprietary firewall and am using it internally in my network. It now generates less heat (system and thermal camera measurements) which touches on the question of efficiency.

I have no control over the future of BSD itself. It still seems to be used, seems highly suitable for a firewall, and even I can do what most users seem to need. I hope it and Opnsense continue.

[franco]

The future of BSDs has always been "troubled" but they always managed to stay reasonably modern as a whole even in the face of Linux being "the best of the best of the best" with vendors making really sure it works with all of their hardware before it is even released (looking at you, Intel).

In hindsight BSD is not the best choice, but you have to understand that if it weren't BSD it may have been become a big company axing the free open source products years ago that produces something resembling "OPNsense" today becoming just another vendor offering a firewall product competing with the likes of Cisco and Juniper (good riddance!).

If you are happy with the project "OPNsense" it may very well have to do with the BSD it's been "stuck" with.

FWIW, a reorientation of OPNsense on Linux is not what I'm personally intending to look forward to. :D


Cheers,
Franco

[mimugmail]

Amen :)

[CJ]

I think the intentions of my starting this thread may have been misconstrued.  Currently I use two BSD systems, OPNsense and TrueNAS Core.  I've been very happy with the experience of both of them.  I assumed that would have been known by the amount of time I spend here trying to help people figure out their issues.

The future of BSD and the discussion of the path forward with TrueNAS has come up a lot in that community.  Reading through Kris' post made me wonder what the maintainers of OPNsense thought and what they saw as the way ahead.

I'm not attempting to convince anyone to switch to or from anything.  Just curious if it had been talked about as Kris mentions that there has been a lot of discussion among vendors of FreeBSD based solutions, although he doesn't mention any in particular.

We (and many other vendors of FreeBSD based solutions, some still keeping quiet) have shared a concern for years now about its long term viability in the wider marketplace of solutions being based on newer and faster paced technologies.

I'm trying to get an idea of how things might look in 2030+ so that I can keep an eye out and plan/react accordingly.  Obviously, no one knows for sure but everyone has opinions. :)

I hope that helps explain things.

[Patrick M. Hausen]

I have not seen Kris Moore or any iX representative at a FreeBSD vendor or enterprise users summit. Neither in one of the regular bhyve/jails production users and developer calls scheduled by Michael Dexter.

Nor - if they indeed have technical difficulties with FreeBSD and specific hardware - has anyone contacted the FreeBSD foundation or the relevant developers like Kristof Provost (networking) or Warner Losh (NVMe/PCIe).

Nor have I seen any one from iX at a EuroBSDCon in years. They might attend the "domestic" (i.e. US based) conferences, I don't know.

I have not heard from any other vendor about "concerns" and certainly not from Netflix who are one of the most prominent and most intensive users of the technology.

[CJ]

I have not seen Kris Moore or any iX representative at a FreeBSD vendor or enterprise users summit. Neither in one of the regular bhyve/jails production users and developer calls scheduled by Michael Dexter.

Nor - if they indeed have technical difficulties with FreeBSD and specific hardware - has anyone contacted the FreeBSD foundation or the relevant developers like Kristof Provost (networking) or Warner Losh (NVMe/PCIe).

Nor have I seen any one from iX at a EuroBSDCon in years. They might attend the "domestic" (i.e. US based) conferences, I don't know.

I have not heard from any other vendor about "concerns" and certainly not from Netflix who are one of the most prominent and most intensive users of the technology.

Interesting.  Thanks for the information.  Outside of OPNsense and TrueNAS Core I'm not particularly familiar with the BSD world so I wouldn't even know where to look.