gyptazy.com is a Fediverse instance that uses the ActivityPub protocol. In other words, users at this host can communicate with people that use software like Mastodon, Pleroma, Friendica, etc. all around the world.
This server runs the snac software and there is no automatic sign-up process.
The new way of selling 'AI' seems to be to push it as a bugfinder. The latest example being waved around as of yesterday includes, as one of its non-embargoed examples, what is patched by this #OpenBSD patch.
https://ftp.openbsd.org/pub/OpenBSD/patches/7.8/common/025_sack.patch.sig
The problematic thing is that this sort of bugfixing hasn't changed the commentary in the code, which stated that p points to the last linked list entry at the point of the added null check and can never be null. But it actually can be, if there was a sole linked list entry that ended up being fully encompassed and thus deleted.
So this kind of 'AI' use is going to give us a lot more comments-do-not-match-code maintenance headaches down the road.
(Both #NetBSD and #FreeBSD factor this out into a separate tcp_sack.c and do the linked list handling slightly differently without a 'previous' pointer.)
I've been finding a few issues in Valgrind on FreeBSD 16. Some are just the usual bit rot maintenance.
Two issues are more serious.
The first is because the servers that I was using have arm64/amd64-binutils installed on them. When they are installed clang will default to using them for linking rather than the LLVM linker. During startup Valgrind will try to read its own debuginfo (which it will use if it aborts to print its own callstack). FreeBSD does some funky things that when loading the ELF RW of an exe. I had added a nasty hacky hard coded bodge to handle that. This bodge assumed that if Valgrind is built with clang then the LLVM linker is being used. In this case that assumption was not satisfied, resulting in an assert firing. I've rewritten that code to be compiler (and linker) neutral without any hard coded values. It is still a nasty hacky bodge of course.
The second problem is a bit less serious. FreeBSD 16 switched to using the LLVM toolchain to produce split debuginfo files. Previously GNU ld.bfd was being used. The change has resulted in the split debuginfo files now containing ELF program headers. This broke some code that assumes that split debuginfo files do not contain ELF program headers. Fortunately this only triggers if you run Valgrind double verbose (-v -v) or higher. I'm working on a fix for this,
The 4 minutes notes about running #FreeBSD on the #Raspberry Pi Zero 2 W. Honestly not my primary / expected choice but it seems to be doing the job. So I'm gonna go further on this path.
https://www.tumfatig.net/2026/freebsd-14.4-on-raspberry-pi-zero-2w/
My MongoDB oneliner patch landed. 🆒
This makes it possible to remove a local patch from the #FreeBSD port. 🎉
https://github.com/mongodb/mongo/commit/2931bad02b1decd630b6da36bf7e97c2dc8a4437
#FreeBSD 15 and proprietary nvidia drivers and KDE Plasma 6 Wayland kernel-panics on me, so that kind of shuts down **that** test-scenario.
I'll swap back to AMD video for desktop use.
so i've fixed this in main by changing the package build to follow the new ABI, and i'm about to land some flua changes so we use the bootstrap flua (from src) for package builds instead of the host flua, which will avoid problems like this in future, and is also required to cross-build packages on Linux and macOS.
but building 15.0 on main is still broken... i think we're going to revert the specific libucl commit that broke this for now, since there's only ~2 users of lua ucl in the base system so it should be safe.
An overview on running #FreeBSD, #NetBSD and #OpenBSD on the #PINE64 #ROCKPro64 #arm64 board, bare and with PCIe extension cards, in the context of building a NAS system.
https://www.tumfatig.net/2026/bsd-discovery-on-the-pine64-rockpro64/
@pertho out of curiosity, what are the benefits of #voidlinux and #freebsd? Agree with #i3wm and a few other light utilities to support it. You may want to try #xlibre in lieu of xorg - it is a drop-in replacement
Latest 𝗩𝗮𝗹𝘂𝗮𝗯𝗹𝗲 𝗡𝗲𝘄𝘀 - 𝟮𝟬𝟮𝟲/𝟬𝟰/𝟬𝟲 (Valuable News - 2026/04/06) available.
https://vermaden.wordpress.com/2026/04/06/valuable-news-2026-04-06/
Past releases: https://vermaden.wordpress.com/news/
#verblog #vernews #news #bsd #freebsd #openbsd #netbsd #linux #unix #zfs #opnsense #ghostbsd #solaris #vermadenday
I took advantage of Easter Monday Bank Holiday to wipe Void Linux off my daily driver laptop and now it's running FreeBSD 15.0-p5.
Moved all the various bits and pieces back and got i3 working solidly on xorg. I'm not sure if I'm going back to KDE. The xorg keymaps work great for my Keychron K3 Pro keyboard (US int'l layout) and I have a shell script that I run that detects the external monitors and enables them as needed.
Won't have Steam on it as that seems to not work with my games but that's OK. I'm kinda burned out from gaming.
I will be making a bhyve vm for OpenBSD so I can update the ports I maintain.
The #FreeBSD is installed and running on the #Raspberry Pi Zero 2 W. And it hasn’t froze yet. Of course, I have an Ethernet Hat because WiFi is not supported (yet). Also, it is running on PoE, now that a switch port has been freed for it. Next step will be to set up my secondary DNS and failback DHCP server on this.
Both implies a post on TuM’Fatig, of course. It’s not because anybody can do it that I won’t brag about have done it myself 😬
"Call for testing: introducing the Laptop Integration Testing project" by the FreeBSD foundation: https://freebsdfoundation.org/blog/call-for-testing-introducing-the-laptop-integration-testing-project/
This is really cool, I like that they're crowd sourcing this, AND that they're paying attention to more than just the technical data. I'll be sure to submit my report on my laptop.
Call for Testing: Laptop Integration Testing Project
We’re expanding the Laptop Support and Usability Project and inviting the community to help test FreeBSD on real hardware.
-Which laptop works best with FreeBSD?
-Will my current laptop support the features I need?
-What configuration tweaks might be required?
Testing is automated, anonymized, and straightforward, and your feedback helps improve FreeBSD for everyone.
Learn how to participate:
https://freebsdfoundation.org/blog/call-for-testing-introducing-the-laptop-integration-testing-project/
#FreeBSD #OpenSource
Is Vultr basically the only non-big-tech host with #FreeBSD images? I'm trying to word this carefully, because I expect that a modern host in 2026 shouldn't require me to spend an hour figuring how their flavor of Linux-based rescue environment + custom image delivery works and do a manual install myself.
I'm aware that AWS and probably GCP have FreeBSD images, but I'd prefer to not use them for a lot of reasons.
Running Podman on FreeBSD? It’s a totally different beast than Linux.
I just published a follow-up to my previous Podman deep dive, going into the FreeBSD operational model for OCI containers. No systemd, no Quadlets, and no rootless mode, but you get native ZFS storage drivers, rc.d service integration, and the Linuxulator.
We also cover the big question: why Podman complements Jails instead of replacing them.
Read it here: https://blog.hofstede.it/podman-on-freebsd-oci-containers-without-systemd/
SYNOPSIS
bastille template [-ax] TARGET|convert TEMPLATE
DESCRIPTION
The bastille template sub-command will apply the specified TEMPLATE to TARGET.
-a, --auto : Auto mode. Start/stop jail(s) if required.
-x, --debug : Enable debug mode.
EXAMPLES
Apply www/nginx to myjail:
bastille template myjail www/nginx