how do i make messages ( kernel and similar ) appear on latest tty? so that it does not pollute ttyv0, ttyv1 ... it is annoying me to death

angry_vincent, so you're upset that messages are appearing on the console?

I don't see how 'annoying' = 'upset' and it reads more like they should appear somewhere, but on the last one, so that working on the first ones isn't interupted by the messages

His name is angry_vincent, clearly he's angry about this. :3

First of all, once you're on the network and have sshd enabled, why be on the console at all?

CrtxReavr: for reasons

dnp1, brilliant.

wait, why would I have sshd running on my laptop?

so you can remote into it from your other laptop

I run sshd on everything. .

But okay, a laptop then. . .

Setup X or Wayland then.

I guess the hint didn't work.

well, no idea about vincents use case, but on one system I have tty1 for system managment tasks that just need a shell and tty2 for my normal user session, so I just use it to start a graphical session

I mean. . . I've had to work on a system console before. . . usually to fix something (usually netowrk related) or because I was in a datacenter, and didn't have a workstation or whatever. . . but hit alt-F2 and work there. . . let the system messages go to /dev/ttyv0

Prolly no good reason to use a VGA console, unless maybe you're on a braile monitor.

On a regular, non-maintenance basis, I mean.

Running sshd on every machine is kind of a mono culture, security wise

I tend to avoid it

So you mix telnet in to mix things up?

OpenSSH is probably the single most scruitinized and reviewed piece of code ever.

Speaking of, have you seen this?: probablymarcus.com/blocks/2026/06/2…native-graphical-shell-for-SSH.html

TL;DR - click the 'YouTube mirror' link.

Being the most audited (got any sources for that?) would still make it a mono culture

Well, it's maintained as an offshoot of the OpenBSD project, which is kinda big on their code review/audit process, and while no, they're not perfect, they make a grand effort, plus OpenSSH gets audited by many other projects that rely on it.

I don't have a hard statistic to quote, but I think you'd be hard-pressed to establish that it isn't the most-trusted piece of network software in use.

But I can also tell, that you're going to argue this point 'til you're blue in the face, and I honestly don't care enough to continue. Believe what you want.

i usually use ttyv0 and ttyv1 for tasks and i do not want system/kernel message appear there as nimaje said

from i can see in /etc/ttys, messages are sent to /dev/console, is it ttyv0 then?

CrtxReavr: grant effort is done how? Using a layered approach. If one layer fails, the next should hold. Not using ssh on every machine can also be considered that.

You accuse me of arguing this "till you're blue in the face" -- look here, I just made a remark, you tried to ridicule me by mentioning telnet. Looks like you're the one not wanting to have a discussion

"You're going to argue this till you're blue in the face" is not an argument for or against anything, it's just cheap rhethoric

wordswordswords

angry_vincent: Are some of those messages not logged? iirc they all should be logged, no?

also guys

i tried to run xfce on freebsd

but my mouse was like REALLY laggy

it happened in the installation menu

too

i have no idea why the mouse is enabled there

yeah, they should all get written to some log file, but how does that matter for wanting them to be printed to some other tty? you could disable printing to tty if having it in a log file is enought for you

i set to log them into /var/log/console.log so i am happy now

burntsoda: I'm running xfce4 as my daily driver, and not running into the symptoms you describe.

Happening on the cli installer as well, is odd. What kind of mouse by chance?

do you have moused enabled? While it is nice to have a mouse in tty, afaik it affects performance on Xorg, as it can't use the mouse directly, but has to go thru moused for that, if it is enabled

burntsoda: what exactly is your mouse^H^H^H^H^Hpointing device ? is it some sort of an I2C HID touchpad ?

It happens on both trackpad and a wireless USB mouse

burntsode: use evtest tool in pure CLI to see if it's responsive (pkg install evtest).

burntsoda: could be some USB issue (missing interrupts?)

again, it happens on my laptop's trackpad too

I doubt it's a USB issue

burntsoda: how trackpad is connected ? isn't it a USB device ?

no

it's literally on the laptop itself

right below the keyboard

(That doesn't mean it's not a USB device.)

burntsoda: mostly likely it's a USB HID, same as your laptop keyboard, check usbconfig.

What's usbconfig?

It's a command?

burntsoda: usbconfig is a system tool, yes.

is it only on freebsd?

burntsoda: yes

I guess I'll have to fire up a freebsd vm then

but I'm not sure if that'll work

burntsoda: errr.. are you trying to use FreeBSD as a guess OS (in VM) ?

no

I had to uninstall FreeBSD recently

I switched back to LInux

But

I will be using it again

So I need to make sure the same thing doesn't happen

I'm not willing to reinstall my entire OS yet again

So I'll run FreeBSD in QEMU this time

burntsoda: running FreeBSD in QEMU will likely give you a different set of issues.

I did with a .qcow2 image once and it went fine

it was with zfs

albeit the freebsd i installed on my real hard drive used ufs

burnrsoda: you can boot FreeBSD from USB stick to check things if you don't feel like installing

like live install?

burntsoda: I won't suggest ZFS for a desktop (laptop), use UFS.

I know

the qemu vm was for server testing

burntsoda: like live CD, yes. you can have a fully working FreeBSD system on a USB stick, btw.

burntsoda, this is a screenshot from the Windows Device Manager set to "view devices by connection," and you'll see my ThinkPad touch pad, mounted inside the laptop case, is indeed a USB device: pasteboard.co/fUSu-lVoMTR_.png

my usb is only 16 gigabytes though

or

wait

how much storage does base freebsd ufs+xfce take up

CrtxReavr: from what I can see it's I2C HID actually. ;)

so laptop keyboards aren't usb based?

burntsoda: it fits well into less than 8GB.

alright

then I guess I'll be trying freebsd on my usb stick some time soon

burntsoda: laptop keyboards and pointing devices can be anything of: serial, USB HID, I2C HID.

I don't know about serial. . .

But they are often PS/2.

yes, PS/2 as well.

What's the official name for PS/2 ports/

I specifically use an asus zenbook 13

i84prt or some such. . .

I borrowed it from a friend who didn't need it

i8042

burntsoda: it's totally possible to have working FreeBSD on zenbook 13, yet it will take you some time to investigate and resolve issues.

Some useless trivia: A PS/2 port provides 5V * 300mA of power.

CrtxReavr: that's really cool.

Encountered a number of devices over the years that used PS/2 passthrouch plugs for power.

Parallel port ethernet adapters, modems. . .

*passthrough

CrtxReavr: excellent telnetd-trolling. you walk around construction sites unnecessarily a lot just because you've got a trusty hard-hat too?

The last thing I had that still did that was an old serial GPS receiver, but it got killed with one of the rollovers.

I wish I had one of those - I'd use it to run a first stratum ntp server.

CrtxReavr: fyi lectronz.com/products/gps-ntp-server

jgh, intriguing.

To do that nowadays you'd generally find an IC with a decent PPS output and run that into the serial port (or a GPIO pin with interrupts on an embedded system if we're being realistic about modern times)

...holy hell some of these are expensive now. If I tried this again I might as well find some new rubidium tubes and rebuild the standard

Yeah, better rubidium than GPS.

or just buy a cheap SDR receiver for the HF band and set up a code to parse the WWV broadcasts. You could even set it to adjust freqency as the day progresses.

Actually, that sounds like a pretty neat project.

Not really a project. It's an hour or two of work.

I was thinking more like HS science project for one of my neices if they ever ask.

I have a schematic somewhere for a DCF77 reciever that could be built from scratch. I'll post it here if you'd like to base the project on it...if I can find it.

Already have an SDRPlus in storage, but thanks. I'm kinda thinking the project would be learning how computers deal with and set time, plus learning some programming and DSP with GnuRadio.

That's probably enough for otherwise not-very-sciency (that I know of) kids.

Might get them interested in ham radio, or not.

IME, it's hard to get kids that grew up with the Internet into ham radio.

unfortunately true, it's hard to get them to put the phone down for 5 minutes

I find it's usually something they can appreciate later in life.

I didn't even get my license till about30 years after I first encountered it... although for two reasons. One, I didn't care to learn CW, and two, couldn't afford it even as kit radios in the 80s.

Removing the CW requirement was a mistake IMHO

Well, I wouldn't have ever gotten my license if it'd remained *shrugs*

Fair enough, but it's not hard to learn, and arguably the one mode that'll always get out.

I'd say that if the sun ain't cooperating, nothing's getting out, CW or digimode or phone.

It'd be sad to see it die out.

CW has a better chance

But sure, if K=9, everything's going to suffer

Rare DX is mostly CW or SSB for me

I would say that computer produced modes like digimode might actually work better than CW because a computer can "hear" a signal below the noise floor. A human requires additional assistance to wash the noise away. Actually I'd love to see an actual study on whether CW human only, versus digimode would be better in "getting the signal out" - and copied.

Even digimodes with FEC will lose data if QSB is bad enough. Human brains, on the otherhand, are quite good at interpolating lost word/sentence components.

There are digimodes designed for poor signal conditions though, and if you go to digital-only signals with error correction, there might be some surprises for "conventional wisdom"

Automated CCW/OOK decoding is more of a solved problem than CW.

Yeah, I use those modes a lot: ft{2,4,8} as a propagation measuring tool, and js8 for actual QSOs.

I used Olivia a lot. Mostly for text ragchews. My shack isn't assembled atm though.

I specifically don't use olivia because my only option is fldigi.

I really hate fldigi.

I do occasionally hop on psk or rtty though.

Heh, I like it.

It's awful to use. I'm working on a replacement TCI client. I'll gradually fold in various modes as I get around to implementing them.

I'm not overly fond of WSJT-X and the new hotness FT-8 though.

I get it. FT8 is just a tool to measure propagation for me.

You might like js8call; it's FT4/8 but an actual QSO, not just a signal report.

I played with JT-65 for a while, but I decided I'd rather talk to someone, so outside of PSK when the bands are exceptionally good, I started using Olivia because of the error correction.

VarAC is the new thing there to replace olivia, but it's closed.

If you're into chasing rare DX, I can only emphatically recommend CW.

Yeah, not really. I'm terrible with learning new languages, and I've never really cared about OMG someone's transmitting from North Korea... but it's only a contact report. *shrugs* just me.

Only if it's a pileup

For someplace like NK when isn't it a pileup?

But even still, OMs like an occasional ragchew to break up pileups. It gets monotonous for them as well.

If you hear them before they get spotted on a cluster!

heh, yeah, but I'd need a really good antenna system or extremely lucky. That's not to say I wouldn't, it's just not really my interest.

Right now, If I hooked my station up I'm effectively that cliche "100W and a wire"... which you can do a lot with, but not bust major pileups.

Especially when those guys with the legal limit bazookas come along.

I q

Actually, on the HF digimodes I also get a lot less of those far right MAGA jerks pushing their favorite conspiracies.

With 100W and a wire, you can work the world on CW!

Yeah, I can work the world on digimode too, and have. Scandanavia, Eastern and Central Europe, almost caught Australia once, but the path faded too quick.

From a vertical wire on my 3rd floor balcony at the time on 10m

Lately my antennas have been more suited for NVIS so mostly it's the US I catch now.

i build my system from src with edited /etc/src.conf. however, i want to make it more stripped down ( for the jails ) but i do not want it to interact with my system's customization. would SRCCONF= /path/to/new/src.conf and OBJDIR = /path/to/new/obj ( to not interfere with my system /usr/obj ) be sufficient?

you want MAKEOBJDIRPREFIX rather than OBJDIR

oh, right, i remember this var, and i have this set for my system. but it is said that it must be set in /etc/src-env.conf and I have that.

man pages mention SRC_ENV_CONF but is not described anywhere

unlike SRCCONF

yeah, it's just the equivalent to SRCCONF

we love naming consistency

yes. it would be nice SRC_ENV_CONF be also desribed

SRC_ENV_CONF, SRCCONF, __MAKE_CONF

then, i can test

i meant that i can already test ( without SRC_ENV_CONF described as i understand it is the same as SRCCONF )

yeah, it's just 'th path to src-env.conf'

we have this notion of using src.conf from the top of the tree if it exists, I wonder if we should do similar for src-env.conf

symmetry is good

symmetry is what rules the world. though broken one is the magic souce

thoughts on what's the preferred storage for bhyve on zfs? image file or zvol? this would be freebsd inside freebsd, FWIW.

and the vm would be using zfs as well