mbnunes: the os will boot fine, i've swapped hard drives many times

When mounting a filesystem, a `struct mount` is passed to vfsops vfs_mount_t function. This struct contains `mnt_data` which is the private part of the filesystem. Each filesystem has a different structure for `mnt_data`. Now my question is, how does this get populated? It is directly casted to specific FS structure as soon as it is in vfs_mount_t, which is the first function that runs when FS is

mounted, so I'm not sure where this `mnt_data` is populated. Does mount utility do it?

I notice qbittorrent does not have bitorrent v2 available on freebsd 13.1

Is there any way to fix that?

I had it on archlinux b4 i switched

concrete_houses: compile it yourself

looking at Makefile, "LIBTORRENT2_DESC= Version 2.x (unstable, has memory issues)"

yuripv: still unfixed? ( according to 268074 )

i'm running into an issue compiling cmake on a pi

/usr/ports/devel/cmake-core/work/cmake-3.24.3/Source/kwsys/kwsysPlatformTestsCXX.cxx:169:12: fatal error: 'ext/stdio_filebuf.h' file not found

nacelle, Are you compiling/building on a flash card, or a hard disk or SSD?

s/flash card/SD &/

sd

microsd in this case, its a pi3b+

that will be a lot of writing for a microsd

if you could might be with mounting one of your other systems sshfs or w/e

s/with/worth/

zdb can pull proper info from the raw dev ada0 and ada3 however the pool doesn't show up in zpool status

Is there a way to tell zfs to rescan/try again?

er zpool-import... I guess

Nothing in /var/log/messages on why that didn't auto import/mount at start up... any where else to look?

Only thing intesting that I did recently was add a new single disk pool.

Did you get any messages around that time?

Not that I know of but I wasn't looking either.

Nothing interesting from 'fgrep -e add0 -e ada3 -e zfs -e jppool /var/log/messages'

jppool is the pool name.

ada0 rather.

good morning everyone

oh I think I know what happened.

Do tell

Booted to single user mode to copy data from UFS to the new pool.

I imported one pool (I have three total including the new one).

Did all the data copy... rebooted. I guess that state is saved and since the last state didn't include that pool (jppool) it wasn't imported.

Yeah, I can see that (cache would be missing the new pool)

s/would/might/

The new pool was there and imported and mounted at start up.

It was one of the old ones that was missing post reboot.

But it was one I didn't need during single user mode.

Title: FAQ — OpenZFS documentation

angry_vincent: TBH, I failed to find any leaks description anywhere

yuripv: heh

Hi

There are some prebuild flavours of freebsd like nomad etc. What one is most easy and has almost all things taken care of these days? user accounts, login,

etc

this is for desktop use ^

and will steam run on freebsd? for gamming

what's hard about "user accounts" or "login" ? i don't know how you would "take care" of that otherwise

skered: does the pool show in `zpool import` ?

Gentlemen, gurus makers porters. Help me in this forum thread

Title: make[1]: cannot open Makefile. | The FreeBSD Forums

Sircle, unless you have several people using the same computer, you can use FreeBSD with your favorite desktop environment. The installer will guide you through creating the one actual-person user account you need in that case. (It will also let you make more than one on install, if desired.) Or you can use one of the desktop-in-a-box derivatives (NomadBSD if you're comfortable with always booting

out of a USB stick, GhostBSD, or the other whos... ah, MidnightBSD) if you're comfortable with their choice of desktop environment.

For Steam on FreeBSD, I think you need the Linuxulator as well.

if I am on my own kernel build FreeBSD atlas.barnabas.dk 13.1-RELEASE-p5 FreeBSD 13.1-RELEASE-p5 #0 releng/13.1-n250148-fc952ac2212b-dirty: Sun Dec 11 17:00:28 CET 2022 root⊙abd:/usr/src/sys/amd64/compile/MYKERNEL amd64

and I run a FreeBSD-update fetch / install it detects I can upgrade to a clean p5 version built in the repos

after having done that

(20:40:35 <nikolaj@atlas:~>) $ uname -a

FreeBSD atlas.barnabas.dk 13.1-RELEASE-p3 FreeBSD 13.1-RELEASE-p3 GENERIC amd64

seems to downgrade the kernel?

have you rebooted?

jep

> D2 in xmpp:en⊙c4c

> 2022-12-25 10:41 (GMT)

> FreeBSD is a great system, but as it turns out, the MIT license is holding back development. I have to admit that FreeBSD will die within 10-20 years

i...

just wondered if anyone else have seen this strange behaviour

BarnabasDK: userland is -p5, but as kernel didn't receive any update since -p3 it is left at that

ah

thanks for the explanation

ixmpp: well, I'm still going to enjoy the ride for the next 10-20 years

ixmpp: that's what poeple were saying 10-20 years ago. 🤷

remember the death predictions of IRC?

also, i have a hard time trying to fathom how mit would hold it back;

me too

MIT doesn't force you to contribute back. which is then why lots of people think nobody contributes. period.

as MelMalik said in the same room, i feel like GPL holds linux back more than any licensing issues would hold fBSD back, but look how well linux is doing

I'm late to the convo, but given license law is almost never enforced, I don't think licenses matter that much.

of course not, but the GPL does serve as a pretty strong deterrent in some situations

it does have some effect

True.

"Imminent death of the Internet predicted. News at 11."

god i hope so. computers were a mistake. sand was never supposed to think.

I'd rather not lose the medium for my social life and support network, if it's all the same to you. :-)

Agreed, with V_PauAmma_V

Copy-free licenses are more popular than they've ever been with strong copy-left licenses being in decline.

that's going to turn many people on the "weak" side of the protective and enforcing licenses.. which would be a net reduction of freedoms and a tighter grab for non-optional and unavoidable licensing and inability to have reduction of copyright and patents as a result influence and control

which is why those licenses (GNU and similar protective-enforcing) are favoured by commercial and corporate enterprise marketeers (big tech)

while computing is moving east to west coast for 50 years already

epony: you might wanna hold off with the zealotry in #FreeBSD

and into the public domain compatible direction

ok

beware the GNUpats ;-D

Today I am hobbying on a 6x drive zfs raidz2 array. bsdinstall sets up zfs using da0p3, da1p3, ... da5p3 types of device names.

I should have said this is with root on zfs.

But using device names is recommended against instead using descriptive labels.

Is there an easy way to convert such an array from one to the other? I would like to switch it to gpt/foo5 types of names.

rwp: I believe you'll need to ensure that kern.geom.label.disk_ident.enable or kern.geom.label.gptid.enable is set to 1, but you should be able to set one of them in /boot/loader.conf even after installing, because ZFS doesn't really care what the device names are.

The advice about device names isn't relevant for FreeBSD though, it only affects Linux.

The reasons why have to do with how Linux handles floppy drives, because it can cause devices will be arbitrarily renamed between reboots - something that won't happen on FreeBSD.

Just to add some system data to this: bsd.to/SobC/raw

Title: SobC

debdrup, I'll just start by saying the advice to use system labels comes from Michael Lucas & Allen Jude in their FreeBSD ZFS book.

But I have also myself chased down failed disks before when not having any clues as to which drive was which drive.

If you want custom labels, you'll need to do a custom partitioning using glabel(8), I believe.

It's always possible but more trouble than if you know which drive is where from the start.

I've not yet had a chance to test it, but with vdev properties being added, it should be possible to get SES information embedded into zpool status automatically.

That'll give you enclosure and bay number information, and I believe vdev properties can also contain serial numbers.

I think I know of "a" way to rethread these. I can remove a drive. I can delete all signatures from it. Then I can use it as a replace for another drive.

does the iwm(4) driver in FreeBSD have power saving support?

And walk through each of the drives one after the other. But it just seems more difficult than it needs to be to do it that way.

pertho, Sorry but I have no idea. And a quick look at the documentation does not say anything. So... No idea. :-(

rwp: the problem, as far as I know, with using glabel, is that unless you're using manual way, it needs to store information on the disk which needs to exist before the disk is used with zfs.

pertho: as in txpower?

debdrup: as in making the device use less power when networking is more... idle

I know the OpenBSD driver doesn't have it (says so in the CAVEATS section of the man page), so I was curious if it was implemented in FreeBSD.

pertho: it's been more than a decade since I took my CCIE Wireless certification, so apparently things have changed since then.

debdrup, It does seem that zfs zpool needs to know the provider name when joining the provider to the vdev.

I want to evaluate how well FreeBSD would conserve battery on laptops. Linux already does a great job at this but I don't know what it is.

hiii all

pertho: or are you talking about UAPSD?

debdrup: hmm yeah I think it might be UAPSD

pertho, Unfortunately WiFi is a soft spot for FreeBSD and it is not as good as it is with Linux. Sad. But things are getting better.

even waking up every 100-200 ms should be enough to get some decent power savings

pertho: cgit.freebsd.org/src/tree/sys/dev/iwm/if_iwm_constants.h#n86 is not a guarentee, but it's certainly better than a complete "Nope"? :P

Title: if_iwm_constants.h « iwm « dev « sys - src - FreeBSD source tree

#ifdef notyet :/

that's for the queues part, as i read it

pertho, I have not tried this myself yet but reports are good so it is on my task list freebsd.org/status/report-2022-01-2022-03/wifibox

Title: The FreeBSD Project

cgit.freebsd.org/src/commit/sys/dev…76cf6ee0e644fbfb5c0278b3a305a024d64 certainly mentions power management

Title: src - FreeBSD source tree

hmm cool.. think I will give it a try on this other laptop I'm considering.. Cheers rwp and debdrup :)

wifibox.. that seems more for devices that are supported under Linux but not FreeBSD

but neat concept

It's a pretty cool concept. And would also be for devices with better support too.

pertho, I dug out my notes and I would like to mention jrgsystems.com/posts/2022-04-20-802.11ac-on-freebsd-with-wifibox and xyinn.org/md/freebsd/wifiboxhttps://xyinn.org/md/freebsd/wifibox

Title: JRG Systems - 802.11ac on FreeBSD with wifibox

rwp: does iwm(4) do 802.11ac or just b/g/n?

ahh the man page says 802.11ac.. kewl

pertho: might wanna read the rest of the mentions of 802.11..