01:30:40 hi! there is this pkg, pulseeffects (that enhances audio output) that doesnt build/exist in current. Is there some place i can find it ? i would be ok if i could create a bhyve vm with that old port just for listening music 01:38:14 i mean find the old pkg 07:43:16 hernan: is it no longer in ports? 07:44:15 looks like it's not a package because it's broken: https://www.freshports.org/audio/pulseeffects/ 07:44:17 Title: FreshPorts -- audio/pulseeffects: Audio effects for Pulseaudio applications 07:45:50 pulseaudio is crap 13:39:18 how do i find the CD/DVD device? 13:40:05 That feels like a noob question, but, hey 🤷🏻‍♀️ 13:40:38 I'm not seeing anything in dmesg, but camcontrol devlist shows it 13:41:29 Apparently there is a cd(4) driver and it should show up as /dev/cd[0-9][a-h] 13:41:30 showed it… when it was SATA, doesn't show it for SCSI… 13:41:36 ah 13:41:44 let's switch back and try again 13:42:19 external CD/DVD device? 13:42:26 kenrap: it's a VM 13:42:29 Ah 13:46:13 i just noticed that cd(4) references the soon to be deprecated bsdlabel(8) 13:49:32 meena: doesn't devlist tell you the device name? 13:49:53 lw: nope, just the bus: at scbus0 target 0 lun 0 (pass0) 13:50:59 you do have cd(4) in the kernel, right? 13:51:30 (it is in generic) 13:52:22 good question 13:52:42 it is generic, but it's generic-mmccam 13:52:48 let's reboot into GENERIC 13:53:49 it should work in GENERIC-MMCCAM, and if it doesn't i should file a bug 13:54:33 SCSI CD drives are pretty unusual, i wonder if this code has even been tested recently 13:55:12 hi! can I run a single test using kyua? 13:55:31 actually, I'm trying to debug erros I get in kyua's own test suite 13:55:33 apteryx: kyua debug ... 13:55:49 works in GENERIC 13:56:39 i don't see anything in GENERIC-MMCCAM that would stop it working 13:56:56 must be something 'options MMCCAM' does i guess? 13:58:01 what ever it does, it shouldn't have any affect on cd(4) 13:58:41 wait a minute 13:59:06 i just remembered that GENERIC-MMCCAM is based on MINIMAL on amd64 13:59:32 oh, i was looking at the arm64 one 14:00:02 why isn't it called MINIMAL-MMCCAM then? 14:00:20 re kyua test suite errors, this is what I see: https://bsd.to/Yrx6 14:00:21 Title: dpaste/Yrx6 (Plain Text) 14:00:22 or why doesn't it just include GENERIC if it's called that? 14:00:42 https://github.com/freebsd/freebsd-src/pull/798 14:00:44 quickly bat-programmers, to the bug report form! 14:00:44 Title: Mmccam vs minimal by igalic · Pull Request #798 · freebsd/freebsd-src · GitHub 14:00:56 ah you already did 14:00:57 i should add `device cd` here: https://github.com/freebsd/freebsd-src/pull/962 14:00:58 Title: amd64 & i386: enable VIMAGE in MINIMAL by igalic · Pull Request #962 · freebsd/freebsd-src · GitHub 14:01:30 oh that PR is like a year old 14:01:48 good grief what 14:02:03 oh, right. wow. 14:02:21 well, a year old if your years have 8 months 14:04:28 look, in most places that don't celebrate Halloween, once you're past August, they start putting christmas stuff in the supermarkets so you know it's just going downhill from there 14:11:23 anyway, once you actually boot the kernel with `device cd` in it, `camcontrol eject cd0` doesn't actually eject this (virtual) cd 14:13:18 rtprio: but is the old pkg available for download somewhere ? 14:15:28 hernan: probably not 14:16:01 you might be able to turn back the clock and see which older version builds, or you could try to get it to compile yourself 14:16:35 * meena switches back to SCSI to see if that will eject 14:16:45 where is the kyua database schema supposed to be installed? 14:17:04 I try running single tests of the kyua test suite like so: integration/cmd_report_test filter__ok 14:17:05 thanks meena 14:17:15 and I get: kyua: E: Cannot read database schema '/gnu/store/jay961fw6xir7hbpmf7hamw1gainj72l-kyua-0.13/share/kyua/store/schema_v3.sql'. 14:19:29 lw: aaah, that also explains why camcontrol devlist didn't show the device when it was on scsi 15:20:18 Doesn't sound like it on SCSI. 15:20:55 Sounds like it was on "SCSI" - AKA SATA. 15:36:45 mate mate-desktop-environment doesnt work i dunno ^^ seems it crashes when sddm starts it 16:33:19 if I make a change to /etc/periodic.conf, or even add one, what do I restart? If at all needed 16:34:08 weust: nothing to restart, periodic.conf is run by periodic(8) which is a shell script started from cron 16:34:13 s/run by/read by/ 16:34:33 That was a thought I had 16:34:38 thanks 18:01:38 does anyone know off hand how std::chrono;:steady_clock is implemented on freebsd? 18:48:11 lw: look at the various definitions of __libcpp_steady_clock_now() in /usr/src/contrib/llvm-project/libcxx/src/chrono.cpp 18:49:56 babz: thanks. clock_gettime(CLOCK_MONOTONIC) apparently 19:06:13 hey kids 19:48:12 can someone who has a real CD ROM / DVD drive test if `camcontrol eject cd0` actually ejects anything? 19:48:46 meena: does it have to be a SCSI CD ROM? 19:49:28 # camcontrol eject cd0 19:49:28 Unit stopped successfully, Media ejected 19:49:39 asked person near this system if the drive actually ejected, waiting for reply 19:52:36 hahaha 19:53:31 meena: answer: yes it did physically eject the drive tray 19:53:55 at scbus1 target 0 lun 0 (pass8,cd0) 19:54:34 this is a SATA device 19:56:07 then if that doesn't work on a VM, what's the bug? in cam? in the buses? or in qemu? or somewhere else? 19:56:50 what do you expect ejecting qemu cd-rom to actually do? 19:57:31 to disassociate the loaded ISO, so that a reboot won't pick it up 19:58:03 hmm 19:58:16 does that usually work in other OS? i don't think i've ever tried this 20:01:37 yup, i remember I had to re-select the iso for the virtual optical device to reboot from say the fedora installer 20:01:45 because it ejects it before rebooting, and it does work in some hypervisors 20:03:35 Motivation is this cloud-init pr, https://github.com/canonical/cloud-init/pull/4769 20:03:36 Title: feat: prefer udev's cdrom_id -e to eject by catmsred · Pull Request #4769 · canonical/cloud-init · GitHub 20:05:24 meena: maybe GENERIC-MMCCAM doesn't include ejecting support, gotta save those bytes somehow 20:06:39 (this comment is like 50% serious, i don't i ever saw a kernel option for 'ejecting support' though) 20:06:47 lw: I've been running GENERIC, like some rube, for half a day now 20:09:29 meena: aaaah cloud-init i am running away. bad memories of CoreOS 20:10:05 is CoreOS still a thing? 20:10:18 * meena can't keep up with Linux 20:11:49 idk it was back when i was laughing at the pathetic state of Kubernetes security 20:12:10 i hated that so much i just stopped turning up at work until they fired me so no idea what's going on nowadays 20:14:50 Kubernetes sounds like a pretty good reason to walk out of a job (and into the sea) 20:16:15 I have this weird feeling that a serious number of companies that use Kubernetes don't understand what it is, what it does, and how to use properly, but that sure as heck won't stop them from using more of it. kinda like violence 20:17:20 meena: so here's my favourite Kubernetes story. they have (had?) an admin panel that let you manage the cluster over the network. the 'getting started' setup instructions told you to install this. installing this meant that anyone who could make a network connection to any cluster node could gain unauthenticated root access to the entire cluster. 20:17:31 i reported this and was told it's not an issue because 'you should be running a firewall' 20:18:48 i later chatted to a google employee who was like, the one person responsible for kubernetes security and he told me he spent most of his time trying to get shit like this fixed, and the developers just *did not care* 20:19:33 this experience has convinced me to never, ever deploy kubernetes or use any service based on it 20:20:27 lol. if you don't start with X from the start, it's really gonna be hard to squeeze it in, once your product is out 20:21:22 X could be usability, speed, security, or any number of things that are hard to bolt on 20:21:59 Accessibility. 20:22:23 brrrr, yeah 20:22:37 * meena points at Gitlab 20:23:27 yeah, doubly glad i'm not running gitlab anymore 20:23:36 ugh we run gitlab internally and i fucking hate it 20:23:44 it needs like 8GB RAM for two concurrent users 20:23:44 but also it exceeded the memory of the vm 20:24:21 plus it only runs on linux which is hilarious given it's written in ruby, what could it possible be doing that only supports linux 20:24:31 (there is a freebsd port but it's perennially outdated) 20:25:56 the canonical deployment only supports Linux 20:26:14 Omnibus. 20:26:21 you dont understand 20:26:23 the thing is, you're supposed to run gitlab *in* kubernetes 20:26:29 i might try to replace it with sr.ht, although i've heard that's also a massive pita to install 20:26:50 babz: hilariously i actually tried this and it was terrible, the two are like completely incompatible 20:27:00 although that was years ago, they might have improved it by now 20:27:32 is the kubernetes in the room with us right now? 20:32:17 kevans: Kubernetes is everywhere. It is all around us. Even now, in this very room. You can see it when you look out your window or when you turn on your television. You can feel it when you go to work… when you go to church… when you pay your taxes. It is the world that has been pulled over your eyes to blind you from the truth. 20:32:20 etc 20:32:40 the kubernetes is coming from inside the server 20:33:18 meena: do you IRC with a proportional font? 20:33:45 lw: yes 20:33:49 ah 20:34:11 … looks really weird in monospace 20:34:48 haha, it looks really weird not in monospace 20:34:48 Atkinson Hyperlegible on the desktop, and who knows what on the phone 20:35:47 are there GUI IRC clients now then? 20:36:23 I use TheLounge 20:36:47 So i can use the same thing on my laptop and on my phone without a handstand 20:36:47 quassel 20:38:09 babz: yes, but, I also wanted so my chat things to be in one window, and that meant it needed to support a Web client. and the Web clients for was Quassel aren't great 20:38:43 i sometimes wonder about how often things crash on my desktop https://www.le-fay.org/tmp/30d/EHDF6B.txt 20:38:54 maybe this is just application bugs 20:40:59 shotman seems to be the worst, what do people use for screenshotting on Wayland? 20:41:18 lw: OpenBSD developer Stefan whatshisface did a good talk on building an accessible laptop for a friend who had lots of special needs, and he did a tangent on that 20:41:43 on what, applications crashing? 20:41:52 how annoying it is that there's so many crashes recorded that no one ever looks at 20:41:58 ah yeah 20:42:18 the thing is i've used Firefox, Handbrake, etc. on macOS and Windows and they never, ever crashed 20:42:37 so i don't know why they're so much less stable on freebsd 20:43:04 https://stsp.name/maurice-laptop.html 20:43:05 Title: Installing an accessible OpenBSD laptop 20:44:33 lw: because FreeBSD isn't an active target they actually develop and test on 20:44:56 it's just GTK though isn't it, what could be different from linux 20:45:06 meena: this is why i hate GUIs though, they're too complicated 20:45:20 computers were not meant to have graphical interfaces 20:46:10 i've worked on ircd that hosts 150k+ concurrent users and that's far, far easier than writing a GUI app, that isn't right 20:46:38 yeah, GUI means "a lot of ifs" 22:03:08 I know how to create a new tmux session, but I also want to create it with a specific command then not be attached to it. e.g. tmux new-session BLAH <=== but I want to come back to the command line and not be in that session. 22:06:02 dvl: new-session takes a -d to remain detached 22:06:11 cheers 22:06:33 Just what I need. thanks 22:28:30 FYI, I'm doing another of these, but with Band of Brothers. https://dan.langille.org/2024/01/14/clearing-out-multiple-drives-concurrently/ 22:28:32 Title: Clearing-out multiple drives concurrently – while Watching Fargo Season 5 – Dan Langille's Other Diary