hello I had a question for openssl

koolkat332: have you tried asking it, or would you like to wait for one of our psychics?

Better ask your question directly. Then someone who can answer it will be able to.

RuntimeError: OpenSSL 3.0's legacy provider failed to load. This is a fatal error by default, but cryptography supports running without legacy algorithms by setting the environment variable CRYPTOGRAPHY_OPENSSL_NO_LEGACY. If you did not expect this error, you have likely made a mistake with your OpenSSL configuration.

koolkat332: looks like an openssl question/issue, not an freebsd one to me...

setting environment variables?

koolkat332: what's throwing that error?

certbot

weird.

koolkat332: you need to upgrade py-cryptography and py-certbot

Title: 254853 – security/py-cryptography: Update to 41.0.3

helloo

I reinstalled those packages did not solve the issue

I can't pass bhyve_options in vm config using vm-bhyve

I mean I set the bhyve_options but seems like it's not being used when launching the vm

anyone experienced something similar?

I'm trying to passthru a usb device to the vm

also py-openssl

also py-openssl

koolkat332: what os version and where installed from?

huh irc lagged a bit there

i installed via poudriere, ot built directly from ports

freebsd 14 release 3 or 4

installed from not sure the question

pkg install most likely then

in that case then I dunno. hte change meant to accomodate/fix the 'legacy' thing is in that fault report but thats from a while ago..

how do I set the environment variable to open ssl?

two options either load the legacy provider or set the environmental variable cryptography_openssl_no_legacy this is an error that ensued as a result of upgrading to freebsd 14

I think it's a make variable. But I never used that. I rebuilt those three ports locally and it worked

env CRYPTOGRAPHY_OPENSSL_NO_LEGACY=1 certbot …

i use certbot on 14 and don't have this problem

what's going on

fold: you can only pass pci devices to bhyve; so are you trying to pass the usb controller?

koolkat332: i think the fact it broke after upgrading to 14 is relevant. how did you upgrade to 14 and did you reinstall all your ports?

also what does freebsd-version -kru report

there were issues I had someone help me after

so if I'm using ZFS on this install, if I don't want to bother with full-disk encryption, but would like to have portions of the disk encrypted, say passwords or stuff in ~/.cache so that stuff like website login cookies wouldn't be able to be hijacked if my laptop is stolen, do I have to plan for that ahead of time during disk partitioning?

14.0-RELEASE-p3

14.0-RELEASE-p3

14.0-RELEASE-p4

I know ZFS basically doubles as a logical volume manager, so I'm basically wondering if geli can operate on those logical volumes or if it can only operate on the actual block devices underlying them

ok - did you reinstall all your ports?

I believe so

pkg upgrade -f ?

tm512: you could create an encrypted volume using a volume of zfs

it's important because the api changes with major version change

zfs create -V 20G zroot/sekret; and then horse around with geli and ufs on top of that.

does anyone here get some segfaults with these? paste.mozilla.org/preuMHCB paste.mozilla.org/MJB1rc32

Title: Mozilla Community Pastebin/preuMHCB (C)

rtprio: thanks. I'm still a complete noob when it comes to ZFS, but I guess now isn't a bad time to learn, since it's also a popular choice over here in Linux-land

for the most part I have mostly avoided these "next-gen" filesystems, just used ext4 on Linux and UFS on BSDs

that's not 'using zfs' as using a chunk of the disk for othr things;

rtprio: well even if it's UFS on top it's still using ZFS to create the logical volume?

and I just mean I'm completely unfamiliar with working with ZFS in any capacity

well, "using"

tm512: yep; without encryption it'd be zfs create -V 10G zroot/myufs; newfs /dev/zvol/zroot/myufs; and mount it

seems kinda similar to loop mounting a disk image

creating a virtual block device in both cases

hm, didn't zfs gain native encryption with 14.0? then you could encrypt only some datasets you want to be encrypted and really use zfs

guess for disk setup I should just go with the guided ZFS option? I'm used to guided disk setup doing stuff I don't like, like splitting the disk up into many partitions

Guided ZFS worked for me for both 12.x and 13.x. YMMV if using 14.0 or -current.

at least on other systems, it ends up being really annoying being pushed into a rigid partition layout, I always just want almost everything on / so that I don't have to make guesses about how much space certain parts of the filesystem will need

I'm guessing that guided ZFS is not going to divide the disk into a rigid partition layout, and will instead just use logical volumes that can be resized on the fly without risking data loss

Well, guided ZFS solves that. You end up with a single zpool, out of which several filesystems are created, sharing the same space.

yeah, I figured it was something like that. so I won't end up in a situation where, say, I need 1GB of space somewhere, but the only free space on the disk is allocated to a completely separate partition?

and the separate filesystems under a zpool are only given the appearance of being separate partitions? with the option of disk use quotas presumably

hmm, "Error while extracting lib32.txz: Can't unlike already-existing object"

s/unlike/unlink/

this is in the FreeBSD 14-STABLE installer, the jan 25th snapshot

I'm seeing suggestions that this can happen if you're trying to install FreeBSD over an existing install, but I'm overwriting the Windows 11 install this laptop came with

also other suggestions that it might be due to a lack of disk space, but that also doesn't really make sense. this is a 128GB SSD

how do I remove a line from env ?

Yes, quota are possible with ZFS. (Have you read docs.freebsd.org/en/books/handbook/zfs ?)

Title: Chapter 22. The Z File System (ZFS) | FreeBSD Documentation Portal

V_PauAmma_V: haven't gotten that far yet

can't even get through the installer evidently

well my memstick.img.xz checksum does match, but I wonder if something went wrong in writing it to USB. might try rewriting it

well the installer even checks the checksums of each set before extracting so that probably isn't the issue

it installed fine this time around, though it gave me a warning about the zroot pool already existing, asking me to confirm that it should be reused. and now it's talking about multiple FreeBSD EFI boot entries

should I have nuked the disk before attempting the installer a second time?

Maybe. I've never installed twice on the same hardware or VM.

now unsurprisingly, it's unable to initialize the wifi (rtw88). guess I have to set that up manually after I'm finished with the installer, since I have to have compat.linuxkpi.skb.mem_limit=1 set in order to force it into a 32-bit address space or something like that

make: "/usr/ports/Mk/bsd.port.mk" line 1204: UNAME_r (14.0-RELEASE-p3) and OSVERSION (1302001) do not agree on major version number.

what can I do about this?

koolkat332, was that host upgraded from 13.2 to

koolkat332, was that host upgraded from 13.2 to 14.0?

yes

Looks like either the upgrade is incomplete or you had jails that weren't upgraded.

I'm not sure

there is one jail in the system but it's not active it's only 2gb cloud system that I spend a lot of time with

tm512: each 'zfs' is a partition. a zpool is a collection of disks

tm512: the datasets can have quotas and optionally inherit quotas

any hints?

serious? any way to make it functional?

koolkat332: did you `pkg upgrade ` again? anything need updating?

no it's not even the issue

the libssl30 issue? what's the issue now?

packages are all up to date

ok the issue is the same

certbot does not work, so I tried to do the setting, apparently that's with /etc/make.conf and then I tried to rebuild openssl but that also doesn't work

what's the paste of the error message ?

because I have two different version numbers

er, where

i have two version numbers too. 111 and 30

Title: dpaste/XQ17 (Plain Text)

would be nice if the FreeBSD installer let me go back a step, accidentally hit enter instead of space

koolkat332: and the contents of make.conf ; and again, why are you doing this?

because when I do the command certbot I get the following error

Title: dpaste/qogG (Plain Text)

tm512, at what stage of the installer are you? I think you can revisit most stages where you enter data or make choices before the install is finalized.

and it's not unreasonable to want the ports to work , etc..

koolkat332: what version of py39-openssl do you have?

koolkat332: obviously.

py39-openssl-23.2.0,1 Python interface to the OpenSSL library

V_PauAmma_V: I was at the point after configuring the network where it was asking me about services to start up automatically. I know I could just go in after the install and put this in /etc/rc.local or whatever, but *shrug*

that's the same one i have

apparently going and doing gpart destroy on the disk wasn't enough to get rid of the "zroot is already taken" warning

You'll get an option to revisit that in rhe installer.

oh lookie there; i have the same error message

The services startup part.

koolkat332: what was wrong with setting th environment variable ?

I'm confused about where the zpool info is being stored if not on the disk I'm trying to install to. I guess gpart destroy wasn't enough to scrub that?

tm512: you dlted all parts?

deleted

I thought gpart destroy deletes the entire partition table

so it does

but it might be caChed if you did not reboot

rtprio the certbot command doesn't work

rtprio: I'll try rebooting

I use it to get ssl certificates

why do I need that, why do I need to do anything really I mean I don't know

I agree but I am trying to learn something

its a pretty clear message

I'm trying to do something

but yes I could say why do I need a computer , etc.. but it's supposed to be what I am doing for a lot of years, I need to do something I think

rtprio: still a message about the pool name already being used. I don't even know if it's worth worrying about, like it might be fine to just proceed anyway

for fucks' sake man, just run CRYPTOGRAPHY_OPENSSL_NO_LEGACY=yes certbot

like it tells you to

I wonder if the pool info is stored in EFI stuff anywhere

command not found

rtprio command not found

I'm still seeing a FreeBSD boot entry in the UEFI boot selection, and it's not an entry for the USB flash drive, since it shows up even without the flash drive plugged in, hrm

koolkat332: what shell do you use?

sounds like csh still

yes

setenv CRYPTOGRAPHY_OPENSSL_NO_LEGACY=yes ; certbot

tm512, what I think happened is: After the "gpart destroy", the data in the partition that says "this is a ZFS partition with a zpool named zroot (or whatever) on it" was still present. Then, the installer recreated a partition table with the same partitions as already existed, and that made the parition reappear as ZFS.

V_PauAmma_V: in which case I guess I could probably safely ignore the warning and continue with it named zroot?

I think so. Worth a try anyway.

I just don't want to run into any weird issues from ignoring warnings and not fully understanding the implications

Fair enough.

same error rtprio

guess the alternative would be to dd garbage to the SSD and use up write cycles, which doesn't seem appealing

or maybe I could get by by just making some UFS partitions on the drive

That may also have potential to cause issues down the road. I've never done that, so I don't know.

well, install appears to have been successful, gonna try to get rtw88 working, otherwise I'll get wifibox set up

seems like one of the things I was worried about (sleep) was a justified worry. the screen doesn't come back on waking from sleep

rtprio: yes I'm trying to pass the usb controller

I mean the only one that have usb when running pciconf

so, after waking up this laptop from sleep, I can ssh in, and I don't see anything weird in dmesg even though the screen doesn't turn back on on the laptop itself :/

nothing in /var/log/messages either

not sure where I should look

is there a list of planned features for the next release? for the curious?

tm512: i think you'd have to press a key

on they keyboard

rtprio: I do, and it wakes up, I can ssh in, but the screen stays off

the laptops page on the wiki has this under the suspend/resume section for the ThinkPad E14: (./}

guessing that's supposed to be a checkmark?

why do you want the screen off when you ssh in?

I don't think you're understanding. I want the screen on, because this is a laptop, and the screen is a vital thing to have

I'm ssh'ing in just as an attempt to debug what's going on

you're saying the screen doesn't turn on when you press a key

but I'm not seeing any indication that the i915 driver is having issues regarding the screen

or after sleep

after waking up from sleep

what's your `sysctl hw.acpi.lid_switch_state`

you already browse wiki.freebsd.org/SuspendResume ?

Title: SuspendResume - FreeBSD Wiki

yeah, I've looked at that page. I can check the lid switch state sysctl momentarily, though I dunno how relevant it is. I'm suspending with zzz

so does resetting the videocard work from ssh ?

is the laptop in x, wayland or at the console?

rtprio: sorry, was eating. this is at the console, I haven't installed any packages yet, so no GUI

I'm not sure how I'd reset the video card

it's in that wiki page

Try "sysctl debug.acpi.suspend_bounce=1" followed by "acpiconf -s 3". This tests SUSPEND/RESUME methods of all loaded device drivers.

not sure what this is going to do to help given that suspend/resume already appears to function, just the screen doesn't turn back on

like I need to know where to look for debug information regarding the screen

well the screen stays on with acpiconf -s 3 while suspend_bounce is set

network connection seems to drop though, which it wasn't doing earlier

hopefully I don't end up having to resort to Linux. super reliable, I run it on my desktop, but it gets boring

gonna try this reset_video sysctl, see if that has any impact

this isn't like my desktop where I already leave it on basically 24/7 so it doesn't really matter whether suspend works properly, whereas on a laptop it's kind of a dealbreaker

it almost sounds like you already convinced yourself to go back to Linux for your laptop

And "boring" is usually a good thing, "if it ain't broke, don't fix it" :)

kenrap: I mean, Linux was always the backup plan, but I want FreeBSD to work

so there don't seem to be any kernel modules loaded that are related to i915

is this built-in to the GENERIC kernel, or...?

as in, not something that would show up in kldstat

oh, they are an external package? that seems odd to me, but hopefully with i915kms actually loaded, this will start working again

probably should check this out: docs.freebsd.org/en/books/handbook/x11/#x-configuration-intel

Title: Chapter 5. The X Window System | FreeBSD Documentation Portal

yeah, I found that

let's see if this fixes the resume issues

yes, yes it does

what's weird is that I ran FreeBSD on my desktop for over 2 years and I have absolutely no recollection of needing to install kernel modules through pkg

not out of the woods yet though because I also need to get wifi working, and ensure that the wifi (whether natively through rtw88 or through wifibox) survives a suspend/resume

now I'm hoping that compat.linuxkpi.skb.mem_limit=1 (to get rtw88 working, supposedly) won't break stuff with the GPU since that appears to be using linuxkpi as well

seems that sysctl limits DMA to a 32-bit address space and that feels like there's the potential for a performance hit at least

does anyone here get some segfaults with these? paste.mozilla.org/preuMHCB paste.mozilla.org/MJB1rc32

Title: Mozilla Community Pastebin/preuMHCB (C)

Under FreeBSD 14.0, I've a VM running as a mail server, and my host machine has a single physical NIC with public IP address. The VM is connected to a subnet of 192.168.2.0/29 as vm-vmswitch. How can I open VM to the public internet, so that the mail server could update packages from time to time, via PF without NAT?

Without NAT, redirections are working fine; such as; an SSH redirection from VM's local IP to my PublicIP:2222, like; "rdr on $ext_if inet proto tcp from any to ($ext_if:0) port 2222 -> 192.168.2.4 port 22"

However, the only thing remains is; how could I give my VM outgoing Internet access?

Any advices please? Much appreciated.

tercaL: that should just be a matter of routing

what does your routing look like, on the host?

meena: On the host, (a remote dedicated server with no physical access), I have a "re0" interface with a public IP provided by my server provider; inet 135.181.116.26 netmask 0xffffffc0 broadcast 135.181.116.27

And one more interface named "vm-vmswitch" (and tap0) with: inet 192.168.2.1 netmask 0xfffffff8 broadcast 192.168.2.7

I can't read hex, I'm sorry

my VM has; 192.168.2.4 and gateway as; 192.168.2.1 (IP dedicated to vm switch), and subnet as; 255.255.255.248

what does netstat -nr look like?

meena: Here's the output; pastebin.mozilla.org/5qy41ZpT

Title: Mozilla Community Pastebin/5qy41ZpT (Bash)

My server's public IP: 49.12.126.150 - my VM network: 192.168.2.0/29

And this is the output of 'netstat -nr', within the VM itself: pastebin.mozilla.org/OKQ0fKE6

Title: Mozilla Community Pastebin/OKQ0fKE6 (Bash)

tercaL: 192.168.2.0/29 192.168.2.4 UCn 1 0 - 4 vio0

what's that for

meena: This might give an idea? pastebin.mozilla.org/5cCH3AH6

Title: Mozilla Community Pastebin/5cCH3AH6 (Bash)

tercaL: can you get rid of that route and see if anything still works?

meena: How can I try that?

I mean getting rid of the route

man route, see del

it's usually the same as add

meena: Got it, in the host machine, right?

no, on the machine it's on

the vm

meena: Did "route delete -inet 192.168.2.0/29" in the VM, and it seems the port forwardings (from VM to Public IP) are stopped. (a sample rule, like; rdr on $ext_if inet proto tcp from any to ($ext_if:0) port 2222 -> 192.168.2.4 port 22)

cool cool cool cool cool

wait

what does your pf rules look like, in general?

here's mine codeberg.org/pkgbase/website/src/br…/main/howto/jails.md#network-config

Title: website/howto/jails.md at main - pkgbase/website - Codeberg.org

meena: Here it is: pastebin.mozilla.org/Ymv5kRXc

Title: Mozilla Community Pastebin/Ymv5kRXc (Bash)

And it works. Just trying to give outgoing internet access to VM, without NAT. So just simple port redirections and outgoing internet in VM.

i don't think you can do that without assigning it an external IP to give to to the vm, and if you don't have one, then NAT it is

meena: I see, so, without NAT, a public Internet access in a VM is not possible?...

unless you have more IPs to spare, no, it's not possible

what's your aversion to NAT?

I can request additional IP from my server provider for that VM, but would you suggest having two different PF on the same box? (one for host, and other for VM itself)?

you're probably sitting behind a CGNAT yourself right now

meena: I feel like under NAT, if the number of my VMs increase in the future, the host machine would remain portless, no?

what

number of the connections

from public Internet to those VMs

how many thousand VMs are you planning to run? how busy are they gonna be?

are we talking one couple thousand VMs or couple hundred thousand VMs?

meena: I've been reading a warning on that kind of networking with NAT. Oh, not really more than 15-20 VMs I think..

So it's not that easy to get out of ports..

I would like to know what you read that have you that impression

and secondly, in terms of network performance, I thought NAT would be slower than direct stuff.

as it's always translating and translating things.

but as you said, assigning VMs public IPs are the only way it seems.

meena: I really don't know, big, bold, red letters with that warning.. :) "don't exceed number of VMs on a single box"

this kind of translation has been optimised for decades, so i would say you shouldn't notice

what was the number? what was the context?

those things matter

Well, I definitely see your point now. Great ideas and tips! Much appreciated.

and see my mistake and mistakenly thoughts.

I'd request additional IPs and implement PF within each VM then.

that sounds expensive but fun

anyone know what happened to the bsd devroom at fosdem?

ascreen: there isn't one

they didn't accept anything BSD this year

Hello

Can i temporarily pause one of HDD in mirror?

of the HDDs

in ZFS

since 3.5" is very noicy

noisy*

V-T60: depending on what you want to do, you could offline it, detach it, or split it. sounds like you want the first

lw: i wanna offline like during night

or tweak for my needs

but how do i do that? and how do i revert?

you can put zpool-offline(8) in cron or something, but i don't think i would want to do this, i believe zfs will refuse to import the pool if the other disk fails while one is offline

this seems like a really bad idea, if you're not going to use the second drive in the mirror for most of the time then just use that drive for backups.

i want to replace later

ok

yeah, sounds like a bad one

V-T60: if you want to keep a copy of data, you could set up something to use zfs send | zfs recv to copy one drive to the other at whatever interval you like. with incremental sends it doesn't need to copy the entire filesystem every time

meena: so it was rejected and not just no interest i guess

hi folks. i'm trying to use mod_openssl in lighttpd. i keep getting this message: 2024-01-31 15:30:35: (/wrkdirs/usr/ports/www/lighttpd/work/lighttpd-1.4.73/src/mod_openssl.c.3010) SSL:openssl library version is outdated and has reached end-of-life. As of 11 Sep 2023, only openssl 3.0.0 and later continue to receive security patches from openssl.org

i assumed freebsd-update fetch + freebsd-update install would fix it, i also updated from 13.2 to 14.0, but getting the same error again

ascreen: I don't know. That's the amount of information I have, and it's only from FreeBSD, not from FOSDEM itself.

i don't use ports. it's a pretty standard machine with lighttpd installed (as a normal package)

Dieterbe, what version do you have installed?

14.0-RELEASE should have 3.0.12

uh oh. seems like my day is about to get worse: pkg --list says this: ld-elf.so.1: Shared object "libssl.so.111" not found, required by "pkg"

Dieterbe, did you run freebas-update install after you rebooted?

freebsd*

Also, you need to look at docs.freebsd.org/en/books/handbook/…ng-edge/#freebsdupdate-portsrebuild as well to upgrade your ports/packages afterwards

Title: Chapter 26. Updating and Upgrading FreeBSD | FreeBSD Documentation Portal

Dieterbe: you should upgrade pkgs before the last freebsd-install step that removes old libraries (it should prompt you to do that). you can try 'pkg-static update; pkg-static upgrade' and see if that works to install the 14.0 packages

i followed freebsd.org/releases/14.0R/installation, i'm pretty sure i ran 'freebas-update install' , i just ran it again to be safe. it says 'No updates are available to install.' and running pkg --list still breaks with the same error. i skipped the ports stuff because i don't use ports

you run freebsd-update install three times, the third times removes old libraries. you need to upgrade your packages before the third time or this will happen. ('pkg' itself is a pkg, so you can't skip that step if you want to use pkg)

oh, i never ran an explicit command to upgrade packages as such

you can't skip that step just because you don't use ports, the page says 'ports installed from the ports tree or packages installed by pkg(8)'

ok, my bad. so how screwed am i now?

try the pkg-static command i suggested

pkg-static is statically linked specifically to recover from this situation

The commands are in that handbook link I provided and it explains all you need to know.

thank you. i will try 'pkg-static update; pkg-static upgrade' now

that worked! any point rebooting at this time?

my .so error for pkg is gone, it works. and also mod_oppensl in lighttpd is happy now :-] thanks so much for your help folks

Anyone an idea for when OPNsense 24.1 hangs in pkg update at fetching the package lists? How to force upgrade to 24.1_1?

2 rules can make freebsd reboot

pass out route-to $tun_if all ; pass in fastroute all ; then run the tun... boom...

Heòòp

Hello

<nasmConnoisseur> I'm installing FreeBSD on my new thinkpad

Any tips?

help

nasmConnoisseur: Tell us what problems you encounter and we can help.

... i need somebody... help... not just anybody... help... you know i need someone... help!

lol

@mason, well first of all: how do I know if my hardware is compatible?

Google "hardware here" +FreeBSD

freebsd.org/releases/14.0R/readme notes the hardware compatibility list, but it's not a link, so half a sec.

Title: FreeBSD 14.0-RELEASE README | The FreeBSD Project

Hm, github.com/freebsd/freebsd-src/blob…583caf1/sys/dev/isp/Hardware.txt#L4 seems not as useful for laptops

Title: freebsd-src/sys/dev/isp/Hardware.txt at e6b8c1d780b490283dd378aa560591a14583caf1 · freebsd/freebsd-src · GitHub

nasmConnoisseur: what kind of hardware?

thinkpad e16 gen1

Which wireless card?

03:00.0 Network controller: Realtek Semiconductor Co., Ltd. RTL8852BE PCIe 802.11ax Wireless Network Controller

That may be your only issue.

Everything else looks supported

In case anything goes wrong I have a USB adapter that is compatible xa0z

nasmConnoisseur: worth looking into setting up wifibox, as long as Linux has proper support for the RTL8852BE

Linux does have support for it

basically proxies your traffic through a Linux VM, with the VM having control over the wifi card

What a hassle. When my Asus ROG's wireless card didn't work, I bought one on eBay for $15 that did work and swapped it out.

anyone know why sysutils/cpu-microcode is to be removed by a "pkg autoremove" ? I can't figure it out

because it's no longer a dependency of something else that has been, or is being removed.

yes, except in this case it's *the* package I installed

no wait, sorry

weust: iirc it was split into separate packages for intel and amd, then a separate rc script

see what 'pkg search microcode' says

it shows -1.0, -amd<date>, intel-<date> and -rc. but pkg autoremove shows inly -intel and -rc. this is a intel based machine

so mark -intel and -rc as manually installed then remove cpu-microcode

(pkg-set -A0 cpu-microcode-intel cpu-microcode-rc)

unless you really want to keep all of them i guess

I guess not, no need.

I also read about locking(wrong term?) them, but why is it listing in the first place? error in the pkg?

package*

I also know with plg I can list the dependencies of a package, but not the reverse. for exmaple, which package pulled in perl5? because I didn't do it myself.

try: pkg query '%rn' perl5

no results

i'm not sure what exactly changed in cpu-microcode to make it do that but i assume it's somehow related to the package split, maybe it's missing a depdency or something

well, does pkg remove perl5 list any dependencies? if not, nothing depends on it :-) i'm not sure how you would work out what installed it in that case though

Think I will email the maintainer

I did do a pkg all-depends on each package listed by pkg prime-list. nothing for perl5, or pcre or pkgconf. those are the other three I don't really get atm

thanks lw. will see what breaks in a short term, or not :-)

might want to check it still works the next time you reboot, but i'm fairly sure you don't need the base cpu-microcode package if you have -rc and -intel

I checked, I was wrong on the meta package. it's not installed. nor is the amd one.

just -intel en -rc

weust: pkg does log installs to /var/log/message, perhaps grep for perl5 and see if anything else was installed at the same time... but that won't help if it was over a week ago

I amsure it's was way over a week ago

But it's OK. if something break I will find out soon enough, I hope

famous last words

yup

luckily it's my home server for Samba and Poudriere. Samba won't go kaput, Poudriere maybe. will see

This is probably a long shot, but do any of you know how (if it's possible) to make pipewire on freebsd emulate a jack server? I figure it'd need wireplumber, which we don't have yet -- i did manage to compile wireplumber with some minor build system tweaks, and while it does seem to run and connect to pipewire, i still can't quite get the jack server bit going

Now you're just making shit up.

me? no :) pipewire has a built-in jack server, normally.

i suppose i could do with just jackd on oss, if there were a way to make a virtual oss device that combines multiple real devices.

i do believe that's the point of virtual_oss