rwp: okay.. i will give that a try.. this is only one of those disks.. i as not swaping things out

Thanks rwp i got it figured out

rwp: old age

that does not sound good

it passed but got a few Pre-fail and old-age

i take it that means the drive is getting long in the tooth and may possibly fail

rwp: pastebin.com/YSL7MgJD

Title: To check the status of drives, use the following: /usr/local/sbin/smartctl - - Pastebin.com

im getting some kde errors about dbus after i uninstalled it. how do i capture the error and show it here ?

jb1277976: i usually use pastebin.com if you have lots of txt to share with people..

well 417 minutes to run a long test on a hard drive just seems like a very odd duration

meena: With cmounts?

Ah forgot that doesn’t work on irc. Wonder what that looks like lol.

meena you have to build with cmounts after changing mount_helper.fs to mount_rclone in order to symlink it to mount_rclone and use it in fstab.

I was able to build it just fine without fstab / helper support :P

I have a mountpoint /xyz on /ssdpool/xyz. How do I change it to another pool in another drive which is /zroot/xyz? This xyz has Postgres data. I believe if I change it, the data will migrate to another pool ?

got answer. nvm

Macer: building with cmounts requires fusefs-libs: freshports.org/net/rclone

Title: FreshPorts -- net/rclone: Sync files to and from various cloud services

14

why use rclone instead of rsync?

polyex: can rsync sync with s3 or Dropbox?

dunno

"no"

ah

So let's try again: why use rclone instead of rsync? because rclone does a completely different thing

ty

you're very welcome

hello

Hi.

what backends are there for discover

i have no idea what this mean

richard_: do you mean kde?

yeah

they stopped calling apps Ksomething ?

babz: since plasma, i guess?

I would've expected at least Diskover

Hi. I am using acme.sh to manage LE certs using freedns hook. I am getting errors, when i try to renew the cert. I would like to ask for recommentation which other tool i could use instead, that supports freedns. Thanks!

FaFaS: what are the errors?

1st error: Can not find dns api hook for: dns_freedns (to fix it I created TXT record manually)

2nd error: Please refer to curl.haxx.se/libcurl/c/libcurl-errors.html for error code: 3 (to fix it, i modified /etc/ssl/openssl.conf to use MinProtocol = TLSv1.2)

Title: libcurl - Error Codes

meena I had it installed. I wasn’t using the ports tree to build it though. I pulled it from github

Wouldn’t come the header for it. (fuse.h). It doesn’t matter anyways. I already installed Debian on it. I’ll just keep FreeBSD on the nas where it doesn’t have to be a smb client (just a server)

*find

I was just tinkering to finally see if I could use fbsd as an end user but I need smb and if I have to work that hard just to use smb/cifs then the better solution is to just use Debian.

Macer: i did it from GitHub, too

and in that process did, rclone/rclone #7435

Odd. I had fuse libs installed and it did not want to build for me.

Title: Makefile: use POSIX compatible install arguments by igalic · Pull Request #7435 · rclone/rclone · GitHub

but haven't done your patch yet

Well I want to say upstream rclone had it enabled but disabled it because issues.

There was a big related to this somewhere even though building with cmounts is supposed to work.

mmm does an ssh user certificate contain any secret info? it seems it requires you to have the equivalent private key loaded to get access anyway

so I'm speculating wildly that it doesn't

it contains a key to match the pubkey

Passwordless keys could be pretty dangerous.

Macer: that doesn't really clarify it for me.

1. cert won't work without corresponding key loaded

2. does cert contain any secrets?

No. It doesn’t. Only the secrets required to authenticate against the pubkey.

I guess the pubkey is typically made with a source reference at the end of it (user@host)

If you consider that private.

sorry I'm not following here... "Only the secrets required ..." what do you mean by that?

I can see what's in the cert via `ssh-keygen -Lf id_ed25519-cert.pub` for example

and it contains private info.

sure

but does it contain any *secrets*

private things like my address is not the same thing as a *secret*

if that makes more sense

Public certs and keys do not contain secrets.

thanks!

The nature of public certs and keys is that they're distributed to "the world." (Usually as needed, as all the ones in use would take up too much space for everyone and everything to have all of them.)

So, they're kind of anti-secrets.

see cvsweb.openbsd.org/src/usr.bin/ssh/PROTOCOL.certkeys?annotate=HEAD

Title: src/usr.bin/ssh/PROTOCOL.certkeys - annotate - HEAD

oh boy.. the smartctl -t LONG /dev/da0 was STILL at 10% complete after 6 hours..

rwp: i would say that the run of smartctl was NOT successful as i manually aborted it

trying a short test to see if result of noncompletion is the same

Have you checked the smarts attributes yet?

smartctl -A

babz: yup

the short online test completed without issue

now we wait for the 2 external hard drive enclosures and see if this is also the case for the other two.. cursing my power backup procedures.. back-up to onedrive, synology but never backup teh backup.. all backups are failures until you can restore !

I'd say that if you haven't yet seen the error counters explode, it's already a good sign.

thats great to know, i am still in the "kick myself in the butt phase" of this whole situation.. but we are getting there.. even went down an openzfs rabbithole.. i really like the compressed feature, had no clue that even existed

Ok so, time to try and game dev on FreeBSD

Godot

it was in repo

So you haven't been waiting for Godot?

No? Wat?

it's already in repo what should I wait for?

zyxer: en.wikipedia.org/wiki/Waiting_for_Godot

Title: Waiting for Godot - Wikipedia

we wanna promote removing the ipv4 stack again

do what?

the ipv4 stack, the part of the kernel that does ipv4 packet processing between userspace applications and the underlying network transport (usually ethernet or wifi)

we do not need the ipv4 stack. we have a perfectly good ipv6 stack. we should use it.

wuuuut

I don't have IPv6

I want IPv6, but I am not getting it from ISP

SoniEx2: kame to my mind ;)

In the best world we never standardised IPv4 to begin with, it would have died just as fast as 3G, e.g. not even properly supported before dropped for new version

if a communication stack has been used for x number of years ... its going to take of x*2^2 to deconstruct it ....

i still see v4 being used at least internally 50+ years away

unless something changes in the hardware that is ipvNEXT centric

CmdLnKid: or software

yeah yeah

zyxer: you don't need real IPv6, you just need a NAT64 prefix

(or well, stateless packet translator)

You mean I can get IPv6 without ISP providing IPv6?

SoniEx2?

and then what ... get a HE service... if ya ask me its plumbed stupid

no, you can get rid of the ipv4 stack without the isp providing ipv6

for adoption only and should have ended long ago

SoniEx2: So how I get internet?

soniex2, how so

waiting for the transport howto

zyxer: ipv4 packet <-> stateless IP/ICMP translator <-> ipv6 packet <-> ipv6 stack <-> libc <-> legacy ipv4-only APIs

there's a compatibility layer ?

oh

I get it

note how the stateless IP/ICMP translator (SIIT) is not an ipv4 stack, but merely a packet translator

and it could be implemented in userspace (ipv4d, using raw sockets and a virtual network interface with the well-known /96 assigned to it)

show me a system where this exists ?

CmdLnKid: the idea is to make a system where this exists. this is a proposal.

ah

oh

I thought that existed

voy4g3r2, The SMART data on that disk did not look concerning to me. It's a disk in the middle of its life. The power cycle count and the load cycle count were both on the high side for the age but it is not clear those are strong indicators not really. It has had a history of SMART selftests all without any error. All seems okay.

zyxer: SIIT exists actually

basically all of the parts exist, but nobody's put them like this yet

well, libc would need changes

usually SIIT is implemented by calling out to the ipv4 stack, which we're trying to remove here

SoniEx2: normally i would be extremely on board with such a bold proposal, but i now live in a country where there's no infrastructure outside the capital. that includes things like trains, and buses, and doctors and IPv6.

meena: but it doesn't require upstream v6

well, downstream, we still don't fully support IPv6.

SoniEx2: until we get dhcpcd imported

rwp: thanks! for the sanity check.. i should know a little more between 330-630 when the enclosures come for the other drive...

did 14.0 come out yet ? I didn't see anything on the website.

as for the power cycle, not 100% sure i have an answer.. but i did at one point.. have it goto sleep WAY too early and was doing testing.. but good to know

memset(*, *, *)

poor connection rob4226?

mns: not yet... probably is already around the corner (rc4 is running with no major issues)

i seeded the 14.0 rc torrents (only for i386 and amd64) all weekend

At least some architectures have release images available. I've got one system running that, but am holding off on pkg upgrade as the UPS package isn't currently available for 14.

I asked this on discord about an hour ago but no answer

I bought this amazon.com/dp/B08XB66YYH?psc=1&ref=ppx_yo2ov_dt_b_product_details and this amazon.com/dp/B00NMXY2MO?psc=1&ref=ppx_yo2ov_dt_b_product_details Plugged everything into the right port and sound is working as far as the headphones but the microphone isn’t working. Any ideas ? I turned the volume up on the mic and

everything. All this is via usb. Should be working right ?

Title: Amazon.com: SOLMIMI Headphone Mic Splitter, TRRS Headset Splitter with Separate Headphone and Mic Jack, 3.5mm Combo Audio Adapter for Dual-Plug Headset Compatible to Xbox One, Smartphone, PC, Ps4-10inch : Electronics

oh if i switch the headset from mute to unmute adacity and a zoom test call can pick that up

could it be the headset ?

Well can you record audio ?

cybercrypto: thanks. I'd seen some talk of it being released last week, since release builds had started.

babz the microphone didn't with at all

Work*

did you try to record sound with anyhing else than this hadset ?

head

Nope. I did a zoom test call and mictest.com

Nothing picked up the microphone. Also audacity

Well, if you have the same problem with two different drivers

... The problem could be this headset

Or your configuration ^^

Yep I'm going on Amazon right now

do you have a gaming console?

maybe can plug it into a gaming console such as xbox or ps

if it is NOT working there.. will DEFINITELy isolate

Or a smartphone, or most PCs, or...

My son plugs the header to his PS4 controller and talks all the time. But it could be that the controller is has a microphone

Headset*

jb1277976: first thing first. PS4 controller does not have a built in mic. If he can hear and talk = this means the 'plugable' headset is working properly (mic and speakers)

Hmm

It's my setup

jb1277976: ps4-controller=only has speakers built-in ps5-controler=built in mic and spearkers. I suspect it is your setup.

Dan so buying a new headset and mic won't do anything?

Dam*

Probably something with device.hints will work unless it doesn't work on USB devices

voy4g3r2, Here is on of my older drives for your comparison. 108,935 hours and still going strong! That's 11+ years. bsd.to/qu6v/raw

Title: qu6v

ohhh

nice! and i am "stealing" bsd.to also

holy moly has pastebin become an ad nightmare

I always run uBlock Origin as a security measure. But look at the channel topic. That's the channel recommended pastebin. :-)

Happy 14.0-RELEASE everybody! 🎊

yay

yay

now the waiting of press release...

i updated last week

release announcement is published main site, i see 8-)

oh wow, that was a few seconds delay

i just checked a few seconds ago..

thats great release, Congratz! Amazing job for everyone who has contributed.

I contributed a new bug

Congratulations everyone!

pstef: the best thing about freebsd is even the unstable versions (including nightly snapshots) are way more stable than most so called "stable" OS'ses around 8-)

time to download the RPI image.. and give this bad boy a try

will be done in 45 seconds :)

well, the thing I referred to has little to do with stability and more to do with interface (I fixed mount not to lie about one case, but now it can lie about another case I didn't foresee). And now that's released, can only be fixed in the next release

voy4g3r2: 👍

This is my first time being on a os that had a major release. to much distro hopping in the past. i'm here to stay

and it is writing to flash storage :)

Oh, neat, I found a repo with a more updated rpi gpio implementation in python: github.com/ethanlmiller/python-GPIOfbsd, gonna do what voy4g3r2 is doing and try this out.

pstef: I see. sry to hear that.

No updates needed to update system to 14.0-RELEASE-p0.

WARNING: FreeBSD 14.0-RELEASE is approaching its End-of-Life date.

lol

danger zone righ there.. flash card, stick in raspberry pi, try to login ssh for first time.. no monitor/keyboard in sight

kenrap: do you notice that your "lights" go off on raspberry pi when it gets to freebsd

mine always were on in linux land

FreeBSD generic 14.0-RELEASE FreeBSD 14.0-RELEASE #0 releng/14.0-n265380-f9716eee8ab4: Fri Nov 10 08:59:18 UTC 2023 root⊙rnfo:/usr/obj/usr/src/arm64.aarch64/sys/GENERIC arm64

woohoo

voy4g3r2: I was able to reproduce that. That red led would turn off mid way into booting the system.

yeah same here

It might be worth making a bug report. I could make one later but I need to get some rest for now.

And I would to see if I can play with neopixel LEDs with that repo's gpio wrapper in FreeBSD

That would be badass

Oh joy: editing the diffs between 13.2-RELEASE and 14.0-RELEASE invokes vi. I've hated vi since 1995...

At least, it seems I only have to delete some lines.

I guess it is possible to choose a different editor

How do you do that? Set VISUAL for root?

not sure what is the parent process, is it etcupdate?

VISUAL or EDITOR

whatever it is, I'm almost sure setting EDITOR should work

and not for root, just for the shell that will run the update

i'm doing a zfs send into a zfs recv on the same system (actually syncoid). mirror disk with one nvme and one ssd. just doing this w/ a recvoption of -o recordsize=1M because it's all for music files, so i would rather them be in a dataset with this property. I need to limit the bandwidth to 5M/sec because the SSD is getting overwhelmed.

gstat shows the SSD choking after even 10 min limited to 10M/sec. the cpu's are sitting around doing nothing, but the system is practically halted if it needs to write to disk. had to control+c a few times to get to a comfortable bandwidth

interesting exercise. i look forward to zfs destroying the original dataset once this completes and i confirm everything is good to go

still.. half hour, and i had to abort the process. i think the SSD has some smart cache that is getting overwhelmed, or heat, or who knows. it's relatively inexpensive SSD though. tempting to replace. wish money grew on trees

the disk remains +/- 100% busy minutes after aborting the send/recv

rwp: you were wondering yesterdy about "weird" things the hard drive was showing.. raid.wiki.kernel.org/index.php/Timeout_Mismatch explains a lot.. i have a "red" drive.. western digital looks to have gone to "shit"

Title: Timeout Mismatch - Linux Raid Wiki

Well, adding an FTL to spinning rust doesn't sounds like the greatest idea...

14.0-RELEASE is still being built??

Ah wait no it was released yesterday super

6 day building time is a lot, wow

I wouldn't do the leap to 14 before patchlevel 1

YMMV

flatdog: I've been on 14 since it branched in some capacity

(and I'm now on 15 on my dev vms)

She was on 14 before it was cool

Fair enough. My nick says "flatdog", not "guinea pig"

I like guinea pigs

I'm also on latest

To each his/her own

I'll just zfs snapshot in case of issues

Time is irreversible, sometimes so is data

And there's bectl too of course

Oh all my data is backed up anyway

Well, you have a pretty much well designed policy

Borne from many mistakes

Sounds familiar

Like that time I had my DOS drive mounted on Linux somewhere in the late 90s and I did rm -rf /<space>something

And the space was nonintentional

:)

i just learned, you can now buy freebsd: aws.amazon.com/marketplace/pp/prodview-4csbncujmsgbk

Title: AWS Marketplace: FreeBSD 14 (ZFS)

It took me a few seconds before thinking "umm why does it take so long for a few files"

.oOo0(excuse me, uhm, ...!?)

crazy new world...

And then I spent 2 days with norton undelete guessing the first characters

Well in the breakdown it says the software is $0

Not a fan of AWS tho

But to each their own like flatdog says

me neither :)

wildeboskat: boer?

You asked for it

and got it :)

cheers

When doing a freebsd-update fetch on the freshly upgraded FreeBSD 14.0-RELEASE machine I get: WARNING: FreeBSD 14.0-RELEASE is approaching its End-of-Life date.

This doesn't seem right :D

bendodroid: perhaps you missed a reboot and another freebsd-update?

I did a freebsd-update install right before that and it said: No updates are available to install.

So I don't think I missed one. I strictly followed section 26.2.3 in the Handbook

(and the following applicable parts about reinstalling packages etc.

Guinea pig is my (unpaid) job.

Don't you have to do -r 14.0-RELESE ?

Ideally with the A

wildeboskat: I already am on 14.0-RELEASE, that's why I'm confused