Just a thoughtโฆ Wouldnโt it be nice if capsicum in #FreeBSD could be used in such way that you didnโt need to alter binaries, but from e.g. daemon(8) which would jail your binaries with the restricted capabilities
@jpmens@max@abimelechbeutelbilch When introducing a pipe here you spawn at least one process more than you would when redirecting to /dev/null. Nice trick for interactive sessions though!
The other day I have also committed an update to the sync(8) manual page, which documents that running sync three times in a row hasn't been necessary since the 80s.
I started a simple project called checkrc that validates the /etc/rc.conf file on #FreeBSD. I'm still a beginner in C programming, but I'm giving it a shot with something usful:
The project is still a work in progress, right at the beginning, so don't expect too much yet. But I'll be working on it every day. :blobfoxcofecute: :blobfoxcomputerowonotice: One day I'll be great!
If I use NFSv3, then all my shares are full of #AppleDouble files (i.e., with the "._" prefix).
If I use #NFSv4, then "git fetch" just hangs forever and never finishes.
If I use #Samba, then either 1) everything is 755 but I cannot delete files xD or 2) (after applying https://askubuntu.com/a/1126633/413683) the permissions are correct, but something is wrong with my .git: ad_convert: Failed to convert [.git].
@owa@darkpatterns It's in the options of every browser app. Also, you get a huge pop-up forcing you to select the default browser with the recent iOS update. It seems quite reasonable to me.
EDIT: Actually, you only get the pop-up if you open Safari.
Oh what joy. Seems like #Mail.app on #iOS no longer allows sending uncompressed 2.2 MB JPEGs.
My #SMTP server advertises 250-SIZE 157286400 (accepts messages up to 150 MB) and base64-encoded 2.2 MB binary would weigh roughly ~2.86 MB.
What non-3rd party options are there to transfer pics from #iPhone to #FreeBSD laptop nowadays?
This is strange: last October I created a FreeBSD VNET jail and defined a wireguard interface and a vxnet interface. It was working. Today I started the jail and it doesn't work anymore:
If you are upgrading to #FreeBSD14 right now, make sure to read the late announcements from the current #FreeBSD Release Engineering Lead, Colin Percival.
Update was smooth, if a little time-consuming on the Pi, and everything seems to be working correctly, which is awesome! Let's do more testing and find out :)
It's the sysutils/nix package... er, "port" or whatever. This could be awesome, because I would be able to have reproducible builds an an operating system that provides #ZFS support, which on #Linux is possible but still a bit experimental. And also you can make use of #BSD "jails" or "zones" as app containers, which on #Linux is not possible at all. (And no Docker doesn't even come close to providing the isolation and security features that BSD jails provides.)
But still, so many questions...
so when you install stuff using Nix, does it still build it using the Clang compiler that FreeBSD uses for the rest of it's stuff? Or does it force you to use the #GCC from Nix Packages?
Is there an easy way to configure which compiler you want to use for building all of your stuff?
Or (and I hope this is not the case, but) does it make use of some kind of BSD comparability layer for Linux kind of like what WSL2 does on Windows? I wouldn't think so, but it is possible that this is how it works.
Is it possible to build Nix packages into BSD jail? I know Nix provides support for creating Docker containers, but is there support for jails?
Is there a Nix expression that can construct a Nix derivation from a BSD "port" itself? I would assume so, but then, not necessarily.
@mntmn Do you know if anyone from the community is working on porting #FreeBSD to #MNTReform? I guess it is too early to ask about #MNTPocketReform as the first batch is not out yet AFAIK.
@mntmn Thanks! I was thinking if porting #FreeBSD to #MNTPocketReform could be a nice project to learn more about the #ARM architecture (and #kernel programming in general).
@mntmn Thanks to the common ancestry, the code usually ports nicely from one #BSD to another. It does require some work though. Which is what would make it interesting :)