| rustyaxe | pipewire is pretty nice these days | 00:00 |
|---|---|---|
| gnarface | fonky: yea you can't get rid of libpulse without rebuilding the actual package but as was mentioned above, it's vestigial if pulseaudio itself isn't installed, so you don't really have to | 00:00 |
| gnarface | buZz: uh, no idea | 00:00 |
| buZz | rustyaxe: yeah amazing, especially on slower machines | 00:00 |
| fonky | ill find something else than audacious, as long as it plays flac and has an equaliser | 00:01 |
| buZz | really cuts cpu time down from pulseaudio at least | 00:01 |
| gnarface | fonky: probably everything is gonna have that libpulse dependency built in | 00:01 |
| fonky | dedicated sound card does it better :) | 00:01 |
| fonky | haha :) | 00:02 |
| fonky | everything new you mean gnarface | 00:02 |
| buZz | you dont really have to write pulseaudio clients anymore to get 'pulseaudio like benefits' , when having pipewire as host audio setup | 00:07 |
| buZz | alsa clients, pulse client, native pipewire clients, they're all the same kind of 'citizen' inside pipewire's routing | 00:07 |
| buZz | they're all fully routable, per application/source/sink volumes, its great | 00:08 |
| buZz | and finally nearly carefree bluetooth audio again :D | 00:08 |
| onefang | Jack2 does that for me. Though I don't use Bluetooth. | 00:16 |
| rrq | and I stay pure alsa, with bluetooth (and automatic device discovery) | 00:21 |
| AlexLikeRock | LOL https://xmms2.org/ | 00:28 |
| buZz | onefang: jack works natively on pipewire too | 00:31 |
| buZz | fully transparant, and takes less cpu than running jackd :D | 00:31 |
| onefang | I have 64 cores that run at over 4GHz, and I use music production software. I can afford to throw CPU at my music. B-) | 00:40 |
| onefang | Most of the pro music stuff uses JACK. | 00:40 |
| AlexLikeRock | yes | 00:40 |
| rrq | yes, lock-in is a bitch :( | 00:48 |
| rustyaxe | jackd is awful, thankfully pipewire lets you avoid it | 01:04 |
| * rrq starts to think about steak knives... | 01:06 | |
| onefang | If you use jack2 with qjackctl, it's great. | 01:09 |
| onefang | But sure, if you are only consuming media, not making it, the others might be better. | 01:12 |
| rustyaxe | ditched all that mess for pipewire+vcvrack for my audio processing for live audio. Immensely more stable and less latency. pulse was the worst of the lot tho | 01:14 |
| * AlexLikeRock installing LUTRIS ; Lutris is a video game preservation platform aiming to keep your video game collection up and running for the years to come. | 01:16 | |
| AlexLikeRock | Over the years, video games have gone through many different hardware and software platforms. By offering the best software available to run your games, Lutris makes it easy to run all your games, old and new. | 01:16 |
| buZz | there is #steamlug btw | 01:16 |
| buZz | and #vronlinux | 01:16 |
| buZz | onefang: you can use qjackctl as-is with pipewire too and not run jackd | 01:17 |
| buZz | :) | 01:17 |
| buZz | and do all the same things | 01:17 |
| onefang | Jack won me over, I have no need to change. B-) | 01:18 |
| buZz | :) i'm telling you, you dont have to change | 01:18 |
| buZz | all the same tools will work, all the same workflows | 01:19 |
| onefang | Ripping out jack and replacing it with pipewire IS changing. ;-P | 01:19 |
| fonky | \o/ mplayer does not require libpulse | 01:19 |
| fonky | oh it does :\ | 01:19 |
| buZz | onefang: not from a usage viewpoint ;) | 01:19 |
| onefang | I still have to change SOMETHING. Pffft | 01:20 |
| buZz | likely just skip 'launch jackd' button :) | 01:20 |
| buZz | or it might die silently 'already running a jack server' | 01:20 |
| onefang | My audio setup is complex enough as it is. It works. Not changing anything. | 01:21 |
| fonky | which dac are you using onefang? | 01:21 |
| fonky | daw sorrz | 01:21 |
| onefang | MusE, which I compile myself. The developers added a few features for me that haven't hit the repos yet. | 01:22 |
| fonky | oh | 01:22 |
| onefang | I have a MIDI turture test, most of them failed it. lol | 01:22 |
| AlexLikeRock | hahahahahahaha | 01:22 |
| * onefang used to be a MIDI developer. | 01:22 | |
| AlexLikeRock | i now , its basic ! midi for an studio | 01:23 |
| fonky | midi controller torture test_ | 01:23 |
| fonky | ? | 01:23 |
| onefang | A MIDI file of Mike Oldfield's Tubular Bells. It's the shere number of instruments and tracks that tend to make things fail. I renamed it Torturous Balls. | 01:24 |
| fonky | wow nice | 01:25 |
| fonky | collaboration | 01:26 |
| onefang | Some software just plain fell over straight away. Some would ask me to assign instruments, when others could figure that out for themselves from the MIDI data, which is a pain whith that many instruments. Others had other problems that quickly showed. | 01:26 |
| fonky | it just like having a band, but on the other side of the world | 01:26 |
| buZz | onefang: how does fluidsynth handle it? | 01:26 |
| onefang | There's MIDI protocols for playing live with band members on the other side of the world. | 01:26 |
| buZz | also, can i have it? :D | 01:26 |
| buZz | there is 'netpd' to sync puredata live (coding) performances over the internet | 01:27 |
| onefang | I use Fluidsynth as a MusE plugin for some of the instruments on the music I'm making. | 01:28 |
| fonky | what genre? | 01:28 |
| buZz | i use fluidsynth with dosbox through mpu401 passthrough ;) | 01:28 |
| onefang | https://untalenz.rocks/ so we don't fil the channel with offtopic crap about my music. lol | 01:30 |
| onefang | A work in progress. | 01:30 |
| buZz | onefang: but can i have thhe midi file? :D | 01:30 |
| buZz | i'd love to torture some synths | 01:31 |
| onefang | Muahahaha! | 01:32 |
| onefang | Guess I should add it to that web site some day. | 01:32 |
| onefang | Ah yes my notes did mention that some software played it entirely through Fluidsynth with no problems. Likely my first attempt with MusE did. | 01:34 |
| onefang | What I should do, eventually, is to have a page for that torture test, run throughd the latest versions of everything testing them all again, and write copious notes to publish there. Then I'll have an excuse to put up that MIDI file. | 01:37 |
| buZz | hahah yeah, and then try older hw synths | 01:37 |
| buZz | i think we have some rolands and yamahas at the hackerspace | 01:38 |
| buZz | and some more dubious ones | 01:38 |
| AlexLikeRock | init 0 | 01:58 |
| * debdog shits down | 01:59 | |
| debdog | hrrhr | 01:59 |
| debdog | *shots | 01:59 |
| debdog | *shuts | 01:59 |
| debdog | there's an *shats missing | 01:59 |
| debdog | wait, wrong channel | 01:59 |
| gnarface | onefang: if you're an old MIDI developer, perhaps you can help me with a dosbox quandry: i swear i remember at one time i'd managed to offload MIDI processing from dosbox to timidity... somehow. nobody in #doxbox has any idea. | 02:10 |
| gnarface | i may have asked this in here before, but the vague memory is from over 12 years ago | 02:11 |
| onefang | Timidity I have used, dosbox I have used, not together though, not even in the same decade. | 02:12 |
| gnarface | yea, i've found no mention of it even being possible, but the memory of succeeding at it is so clear... i'm starting to feel like someone changed the timeline on me | 02:13 |
| onefang | WE should get back to that good timeline. | 02:13 |
| onefang | So you play MIDI in some dosbox application, but you want it to route the MIDI to timidity on the host? | 02:14 |
| gnarface | yea, basically i was just trying to supplement the MIDI quality regression with extra software | 02:15 |
| gnarface | nothing sounds as good as old hardware wavtable cards from the 90's | 02:16 |
| gnarface | but timitidy sounds close, if you use the right soundfonts | 02:16 |
| onefang | When I was a MIDI developer I was mostly dealing with Amiga's and Atari's. So I'm not familiar with how MIDI worked under MS-DOS. | 02:17 |
| onefang | And maybe dosbox managed to inherit some of the more modern MIDI over network stuff? | 02:17 |
| gnarface | not sure... i also have a clear memory of someone at one point suggesting there was a way to get dosbox to directly load the same soundfonts timidity uses, but the memory suggests dosbox would accomplish this by parsing timidity's config file on startup... which doesn't even make sense. | 02:18 |
| onefang | IBM PCs used joystick ports to plug MIDI stuff into back in the day, if I recall. Though MIDI itself was just a modified serial communications. Setting your serial port to the odd MIDI baud rate and some level changers did the trick. | 02:19 |
| gnarface | hmmm... so, in the memory where i managed to offload to timidity, i do recall figuring out some magic device address to put in the dosbox config, which would make dosbox think the timidity daemon was just my regular soundcard's midi port... couldn't recreate that recently when trying though | 02:20 |
| onefang | RS-232 serial ports, for those who'se beards are not grey enough. Ancient communications standard. | 02:20 |
| gnarface | i also recall discovering timidity and dosbox had to launch with the same user or at least same group permissions for this to work... which doesn't seem secure and also seems nonsensical, if i'm in both groups anyway | 02:21 |
| gnarface | anyway, several clear memory fragments with no resolutions, and partially in conflict with each other | 02:22 |
| onefang | Yay! | 02:22 |
| gnarface | but i feel like this must be possible and also obvious but just dead knowledge now... | 02:22 |
| gnarface | like if i could go back to 1995 everyone would just know | 02:23 |
| gnarface | because midi was still hot back then | 02:23 |
| onefang | Sounds like the various databases I've had to deal with during this last house move. Several clear memory fragments, some in conflict. lol | 02:23 |
| onefang | https://linuxmusicians.com/ might be useful. | 02:26 |
| gnarface | hmm, thanks | 02:26 |
| onefang | I use the same name there. | 02:27 |
| gnarface | noted | 02:27 |
| rrq | gnarface: is this any good? http://macilatthefront.blogspot.com/2010/05/timidity-and-dosbox-in-ubuntu-1004.html | 02:34 |
| gnarface | rrq: um... well it really suggests this should be possible | 02:42 |
| gnarface | must have a typo somewhere... | 02:42 |
| gnarface | rrq: do you happen to know if timidity has to be launched as the audio user instead of the timidity user for this to work? | 02:50 |
| gnarface | basically that puts me back at the last roadblock; if timidity is running already, dosbox claims it can't use the soundcard | 02:52 |
| gnarface | but if it were a group permissions issue... | 02:53 |
| gnarface | it's like timidity is either blocking dmix or just not providing it? | 02:53 |
| rrq | don't know. but it might need dsnoop capture as well; (like firefox which otherwise locks out itself) | 02:55 |
| rrq | I've got my default having an asym with a dsnopp capture and dmix playback | 02:56 |
| gnarface | do you have any alsa config lines for midi or timidity specifically? | 02:57 |
| gnarface | i've had to enable dmix/dsnoop for other things on this driver before... | 02:58 |
| rrq | no. right now using fluidsynth to "default" | 02:58 |
| rrq | (with an asym slave ... ) | 03:00 |
| gnarface | rrq: are you still running timidity as the default "timidity" user the debian packages configure, or are you running it as your own user or the audio user or something? | 03:05 |
| rrq | sorry; I don't use timidity at all. Maybe I should :) | 03:06 |
| rrq | I run fluidsynth as me, but then reniced it -3 as root | 03:07 |
| rrq | no dosbox either | 03:08 |
| rrq | (not bad at web search though :) | 03:08 |
| gnarface | rrq: well, that tip was good, and it seems to have helped me find the syntax error in my config that was preventing dosbox to output to timidity, but it doesn't solve the primary issue that using timidity this way blocks regular soundcard access for sound effects, so now i have Dark Forces with beautiful midi but no laser pew-pew sounds, like it's some silent film | 03:11 |
| gnarface | but the dosbox errors relevant to that are: | 03:11 |
| gnarface | ALSA lib pcm_dmix.c:999:(snd_pcm_dmix_open) unable to open slave | 03:11 |
| gnarface | MIXER:Can't open audio: No available audio device , running in nosound mode. | 03:11 |
| gnarface | ^ these strongly suggest that either dosbox or timidity is simply failing to activate dmix, and if it wasn't, this would work | 03:12 |
| rrq | I solved that with making default an "asym" plug | 03:12 |
| gnarface | hmmm | 03:12 |
| rrq | the capture slave is a dsnoop setup and the playback a dmix setup | 03:13 |
| gnarface | i wonder why mine isn't working for this, it's working for everything else | 03:13 |
| gnarface | i have a asym plugin there, but it's 2 levels down | 03:13 |
| rrq | aligned to 48000 stereo | 03:13 |
| gnarface | so basically mine goes from type plug to type asym | 03:14 |
| gnarface | i wonder if that's the issue | 03:14 |
| rrq | wget https://transfer.rrq.au/5RQ9fanQDn/.asoundrc | 03:15 |
| rrq | (it's a bit noisy config with some extras) | 03:15 |
| rrq | you can weed that out I guess | 03:16 |
| gnarface | alright, i'll look at it | 03:16 |
| gnarface | maybe i'll figure something out | 03:16 |
| gnarface | but i'm worried this might turn out to be another stupid snd-hda-intel bug | 03:16 |
| rrq | my playback is via an "@func refer" so I can change output for running programs | 03:16 |
| rrq | wget https://transfer.rrq.au/LEp9fAOJeO/.alsa-pcm-select | 03:17 |
| rrq | again, a bit noisy :) | 03:17 |
| rrq | and hmm it has another asym (?? didn't remember that one) | 03:18 |
| gnarface | what's that second one? like just a partial config something else loads? | 03:19 |
| gnarface | oh, it's an include, i see | 03:19 |
| rrq | must have been a late evening experiment... I think you could avoid that and select pcm.mixy directly... I'll try | 03:20 |
| gnarface | here's what i was working with currently: http://paste.debian.net/1320623/ | 03:22 |
| rrq | hmm works but it causes some setup disagreements ntoted by fluidsynth | 03:22 |
| gnarface | just recently discovered that "@func refer" trick myself | 03:22 |
| rrq | yes, me too; very useful :) | 03:25 |
| gnarface | hmm, it doesn't help testing that now when i don't try to use timidity, and just use my soundcard's regular midi address in the dosbox config, i get no errors but also no music at all now | 03:25 |
| gnarface | i remember that before, i at least got midi with the soundcard's stock features, it just wasn't goodi | 03:26 |
| gnarface | "good" i mean | 03:26 |
| onefang | MIDI is GOODI. B-) | 03:27 |
| rrq | would it help to use just 2 channel output? | 03:28 |
| gnarface | you think that would be the issue? i'd expect a particular error from alsa if it were, that i don't see here | 03:29 |
| gnarface | with other stuff, this setup does still properly accept stereo sound, just doesn't use all speakers unless i use that "stereoupmix" block | 03:30 |
| rrq | nothing from arecord? | 03:30 |
| gnarface | what do you mean? arecord works normally with mics and stuff | 03:30 |
| rrq | yes, my meory suggests that asym gets confused about channel count imbalance | 03:31 |
| gnarface | hmmm | 03:31 |
| rrq | I remember having difficulty combining stereo out with mono in | 03:32 |
| gnarface | HOLY FUCK! | 03:34 |
| gnarface | sorry | 03:34 |
| gnarface | rrq: thanks, that blogspot link helped me figure out the missing trick | 03:34 |
| rrq | great :) | 03:35 |
| onefang | YAY! | 03:35 |
| gnarface | still not quite sure what's going wrong here, if it's a permissions issue or just something alsa/dosbox/timidity is doing wrong by failing to initialize dmix in the right contexts, but... | 03:35 |
| gnarface | ... the trick from that blog entry which is more important than it's presented as at face value is... | 03:35 |
| onefang | And I was just about to say that you can now go back to knowing everything gnarface. B-) | 03:35 |
| gnarface | you can launch timidity with your own user, instead of as a daemon | 03:36 |
| gnarface | not clear exactly why, like i said, whether it's a alsa/dmix implementation problem or just a mundane permissions issue, but when i run timidity in the background from my own user's shell like that blog says to, instead of launching it as the timidity user using the system init scripts, midi and sound works | 03:37 |
| gnarface | and it sounds great | 03:37 |
| gnarface | (his buffer settings cause static for me but it's fine to just omit them) | 03:37 |
| gnarface | so, i think what must have happened, is i discovered that core issue way back then and just switched the timidity config to launch as the user "audio" instead of the user "timidity" and for whatever reason that simply doesn't work anymore | 03:38 |
| gnarface | but launching it from the command-line as my own user, simply backgrounded with "&" from the same shell i start dosbox with next, that works fine | 03:38 |
| rrq | that timidity user has the same .asoundrc ? | 03:39 |
| gnarface | ... once i also changed midiconfig=128.0 to midiconfig=128:0 in my config too XP | 03:39 |
| gnarface | hmm | 03:39 |
| gnarface | rrq: no, the timidity user wouldn't have the same ~/.asoundrc, that's a good point. | 03:39 |
| darwin | gnarface, yes: http://p.bsd-unix.net/?27c8c792fdeebe45#Fs4K9Z8uR4NNFiYMBaSQ2ttVu5Ctqj38u12NxDKmzXys | 03:40 |
| gnarface | dmix/dsnoop are supposed to be on by default for all users though... but every time they fix that it regresses with almost the very next kernel update, so maybe that's the issue | 03:40 |
| rrq | yes dsnoop seems to be missing | 03:41 |
| gnarface | rrq: /etc/passwd says timidity's home directory is /etc/timidity/... could i just copy put my ~/.asoundrc into there and expect it to get used?! | 03:41 |
| darwin | when amdgpu is installed, it changes he driver then claims the monitor disconnected (which never does without amdgpu) | 03:41 |
| darwin | 'the' | 03:41 |
| rrq | gnarface: yes that should work | 03:42 |
| gnarface | rrq: oh man... talk about cursed things one is likely to forget doing and never figure out again | 03:42 |
| * gnarface wonders if that night, 12 years ago, after 3 straight days with no sleep and no food but alcohol, if that's what he actually did, then accidentally deleted it purging unused configs during a later update | 03:43 | |
| onefang | Take notes this time. And eat. And sleep. | 03:44 |
| gnarface | this time maybe i'll put a file in there named YOU_PUT_A_DOT_ASOUNDRC_IN_THIS_DIRECTORY_DUMBASS | 03:44 |
| onefang | lol | 03:44 |
| rrq | :) here's some soothing funk: http://www.jsbach.net/midi/midi_goldbergvariations.html | 03:44 |
| gnarface | hmm, no, no good | 03:50 |
| gnarface | timidity seems to ignore the ~/.asoundrc | 03:50 |
| gnarface | i'll need another method | 03:50 |
| gnarface | but for now it's fine to just launch it as my own user, i can use a wrapper script or something | 03:50 |
| gnarface | thanks for helping me recover some progress with this guys | 03:50 |
| rrq | nw.. I think darwin's note got drenched.. that'd be for you, gnarface | 03:54 |
| gnarface | rrq: yea, sorry, just got distracted, thanks | 03:55 |
| onefang | No worries, sometimes we get to help you gnarface. | 03:56 |
| gnarface | darwin: uh... two things, 1) can you use paste.debian.net or just /msg it to me? ... and 2) can you sanity check this with the daedalus desktop live iso to make sure it's got the same problem? starting to look like a critical kernel/driver bug here but that's almost the same card i have, and i'm using 2 monitors just fine | 03:57 |
| darwin | the paste bin I used doesn't have ads so should work | 03:58 |
| gnarface | just a new face, don't like new faces. please? | 03:58 |
| darwin | i no longer have the log and removed that and rebooted | 03:59 |
| gnarface | it's fine, it's not important. if you're sure it was not showing any connected monitors, i believe you. what's weird though is i have the 5500 and we should be using all the same relevant software here now, and that's almost an identical card, and it's not new enough that you should be having this type of problem | 04:00 |
| gnarface | so my best deductive suggestion now is to try the live iso to make sure it's not something weird with your install | 04:00 |
| darwin | i'm making the liveCD to try | 04:00 |
| gnarface | but if it's broken on the live iso too, i think that means it's time to escalate a bug report about this to upstream | 04:00 |
| darwin | the DVD is rated for 4x but growisofs said it's doin 4.1x1352KB/s | 04:02 |
| darwin | doing | 04:02 |
| onefang | Which graphics card is this? | 04:03 |
| gnarface | radeon 5700 XT, think? | 04:03 |
| darwin | yes | 04:03 |
| gnarface | what does lspci call it? | 04:04 |
| gnarface | let's sanity check that too | 04:04 |
| onefang | Ah, I have a 5600 XT, and no problem-s with multiple monitors. | 04:04 |
| gnarface | Advanced Micro Devices, Inc. [AMD/ATI] Navi 14 [Radeon RX 5500/5500M / Pro 5500M] (rev c5) | 04:04 |
| gnarface | i've got this one, no problem with multiple monitors | 04:04 |
| onefang | Advanced Micro Devices, Inc. [AMD/ATI] Navi 10 [Radeon RX 5600 OEM/5600 XT / 5700/5700 XT] (rev ca) | 04:04 |
| darwin | http://paste.debian.net/hidden/dec6defc | 04:04 |
| gnarface | [ 8.801] (WW) AMDGPU(0): No outputs definitely connected, trying again... | 04:06 |
| onefang | That's the one part of this super desktop I cheaped out on, coz all the better cards at the time had RGBling. | 04:06 |
| gnarface | darwin: don't suppose you've got any different cables to try? maybe this could be some stupid HDMI cable version issue? i've seen it before... usually on stereo equipment though | 04:06 |
| darwin | it's DVI-D | 04:08 |
| darwin | i don't see how it can be a cable issue when it's always working without firmware-amd-graphics | 04:09 |
| gnarface | oh, i do | 04:09 |
| gnarface | especially with DVI-D involved... | 04:10 |
| gnarface | that's the one where it's trying to use one port for two monitors at once, with a splitter cable, right? | 04:10 |
| darwin | no | 04:10 |
| gnarface | you sure? | 04:10 |
| darwin | it uses one port for one monitor | 04:10 |
| onefang | My second monitor is plugged into a HDMI port, but goes through an adaptor to DVI. | 04:10 |
| darwin | might be called DVI-DL | 04:10 |
| darwin | it's a monitor between '1080p' and 4K resolution that needs the most advanced DVI connector | 04:10 |
| darwin | DP doesn't work to convert these | 04:11 |
| gnarface | hmm | 04:11 |
| gnarface | might be related | 04:11 |
| onefang | Mine is running at 1080p. | 04:11 |
| gnarface | i'm just using regular cheapo max 1080p monitors with DP->DVI cables (because for whatever stupid reason the HDMI ports are slower) | 04:11 |
| gnarface | i know the HDMI ports worked though | 04:12 |
| onefang | I use LXRandR to make sure each monitor is in tho proper place, res, refresh rate. | 04:12 |
| onefang | DP is my main monitor. | 04:13 |
| gnarface | darwin: well, i think we're narrowing down on the variable that's probably related to the bug anyway... any luck with that live iso? | 04:13 |
| darwin | i'm making the liveCD to try. The DVD is rated for 4x but growisofs said it's doing 4.1x1352KB/s | 04:13 |
| gnarface | i'm not really familiar with growisofs, i was still using cdrecord to burn dvds last i did it, i think... | 04:14 |
| darwin | so did I but it only worked for CDs | 04:23 |
| onefang | This is the video card I have with the 6 audio devices on it according to ALSA. Surprised me, there's only 4 actual video sockets. Just had to figure out which one my main monitor with the crappy little speakers is connected to, and I now route my system beep and timer bells there. | 04:25 |
| onefang | Motherboard has two audio chips in it, I don't use either. lol | 04:26 |
| gnarface | darwin: if the live iso also doesn't work, one thing to circle back on is that DVI-D thing, i specifically remember a bug with regards to those... something about either needing to use a actual DVI-D cable or not using one... | 04:36 |
| gnarface | should be searchable as a linux bug specific to DVI-D somehow though | 04:37 |
| gnarface | and it was definitely a bug that didn't affect Windows | 04:37 |
| gnarface | if you have multiple ports on the card, try identifying which are DVI-D and DVI-I, and if you're not using splitter cables, just use the DVI-I ports only, if there's enough... | 04:45 |
| gnarface | i do think i had run into that issue at some point, but couldn't locate any actual dual-output DVI-D cables | 04:46 |
| gnarface | but that was several cards ago, and a NVidia one at that | 04:46 |
| darwin | there's only one DVI port on the card, an it has more holes than less advanced ports | 04:47 |
| darwin | there is definitely no splitter | 04:47 |
| gnarface | they were hard to locate because i guess they were only used for like 1 generation of doomed VR headsets | 04:48 |
| gnarface | and the DVI-D ports were supposed to work fine with DVI-I (single-output) cables, but iirc that was the actual bug, that it wasn't working | 04:49 |
| darwin | the liveDVD appear to have no problem | 04:49 |
| gnarface | iiinteresting..... | 04:49 |
| gnarface | nothing special on there that's not in daedalus | 04:50 |
| gnarface | so it's gotta be your install or your config | 04:50 |
| gnarface | i'd start by checking versions on stuff | 04:50 |
| darwin | i just installed the entire base system plus shell & SSH utilities/servers and some other stuff I had in a list but wasn't anything unusual | 04:50 |
| gnarface | could it just be really out-of-date? when is the last time you updated it? | 04:51 |
| darwin | today, and installed the day before. It's clearly the firmware, not my installation, because works fine for the monitor for hours/days without firmware | 04:51 |
| gnarface | yea, but the live iso should be loading the same firmware... only thing is it might be just slightly older if there's been a recent update in the repos | 04:52 |
| darwin | only after I installed the firmware was there no GUI. This happened twice | 04:52 |
| gnarface | yea, but WHY? the live iso should be using that same firmware already | 04:52 |
| gnarface | check it | 04:52 |
| gnarface | check the version, and the kernel too | 04:52 |
| darwin | installed the firmware, got no GUI, removed it got GUI, installed it, got no display, removed it, got GUI | 04:52 |
| gnarface | yea, i'm clear on that, not refuting it... but something doesn't add uphere | 04:53 |
| gnarface | *up here | 04:53 |
| darwin | kernel 6.1.0-21 | 04:53 |
| gnarface | compare it to the live iso | 04:53 |
| gnarface | also the firmware-amd-graphics package version | 04:54 |
| darwin | the liveDVD was months ago so it's newer if anyhing | 04:54 |
| darwin | anything | 04:54 |
| gnarface | yea, so it could have been a very recent regression | 04:54 |
| darwin | oops, no I said this is a Vega 64... think I said that last night or earlier today | 04:54 |
| gnarface | which means it should be easy to prove too...] | 04:54 |
| gnarface | Vega had compat issues but even that one should be well supported by now afaik | 04:55 |
| darwin | firmware-amd-graphics 20230210-5 | 04:55 |
| gnarface | same version as here.. what does the live iso have? | 04:55 |
| darwin | i don't know how to check anything on that | 04:56 |
| darwin | i have liveDVD 5.01 | 04:56 |
| gnarface | you just run "dpkg -l |grep firmware-amd-graphics" while you're booted into it | 04:59 |
| gnarface | shouldn't be any different than the desktop install in that regard | 05:00 |
| gnarface | if it's got a slightly older kernel or firmware, maybe try those versions | 05:00 |
| darwin | yes... what about kernel? | 05:00 |
| gnarface | "uname -a" for the kernel | 05:00 |
| gnarface | same as the desktop | 05:00 |
| gnarface | (that one is distro agnostic) | 05:01 |
| gnarface | brb | 05:01 |
| darwin | well I don't have it booted, just in my CD/DVD/BD drive | 05:01 |
| gnarface | darwin: uh... off the top of my head i'm not sure how you'd find it without booting it. i'm sure there's some way to mount it and look inside but i'm not clear on the structure of the live isos | 05:22 |
| gnarface | when you get a chance though, ...these versions checks could be the key to your fix | 05:23 |
| gnarface | fsmithred: you didn't happen to put anything like a config fix specific to AMD Vega 64 graphics in the live isos, did you? | 05:24 |
| darwin | all the Devuan DVDs had serious problems on our PCs. When it got to the black & white screen, usually the graphics (even though it's just text) garbled due to something went wrong with display mode... sort of... except when I pressed <TAB> before that and then vga=ask and selected the mode, it gets through that screen fine | 05:27 |
| darwin | that was with everything from Radeon RX Vega 64 (with installer but not liveDVD) to RX 6900 XT and GeForce RTX 4090 | 05:28 |
| darwin | by all I mean some older versions like 4.n, maybe 3.n, and might've tried liveDVD and seen this on some but not all. I'll check that kernel version soon | 05:28 |
| gnarface | well, you gotta be careful here, because "garbled text" might be a symptom of burnt video ram, (one that's actually very common with old NVidia cards) and in general we really don't have that much trouble with our shit around here, so it strongly suggests you're doing something else weird or wrong too, if it's not just bad video ram | 05:31 |
| rrq | sn't that black screen an issue with the monitor more than graphics card? | 05:31 |
| gnarface | ...but that said, we don't really know and bugs don't get fixed without troubleshooting, so some amount of pedantry would be appreciated | 05:31 |
| gnarface | rrq: there's no way to tell at this point. i was even starting to suspect a rare bug specific to DVI-D/DVI-I cable compatibility, but he says the live isos work fine so there's still an unknown variable here | 05:32 |
| gnarface | darwin: if you can take a camera snapshot of a screen with the "garbled text" symptom you describe, i can usually visually diagnose whether it's a case of fried video ram | 05:34 |
| rrq | I think live and installer isos have different vga boot settings | 05:35 |
| gnarface | also, in general, when you see screen corruption it's a good idea to try to take a screenshot with software running on that system, ... for further disambiguation (some sorts of corruption will not show up in a screenshot, so that's a valuable tool for deduction) | 05:35 |
| gnarface | rrq: that's a good tip | 05:36 |
| gnarface | darwin: while you've got that live iso up, also copy down the contents of /proc/cmdline | 05:37 |
| fsmithred | gnarface, firmware-amd-graphics is installed in the desktop-live isos. | 05:44 |
| fsmithred | I don't think there are no special configs for graphics. | 05:46 |
| fsmithred | sorry, didn't mean the double negative | 05:46 |
| fsmithred | remove the one of your preference | 05:46 |
| * fsmithred goes back to sleep | 05:46 | |
| rrq | live iso boots with gfxmode=640x480 and gfxpayload=keep for grub | 05:50 |
| rrq | that'd correspond to vga=785 | 05:54 |
| rrq | actual boot parameters don't include vga setting though | 06:05 |
| rrq | live cdrom legacy boot uses isolinux, otherwise grub | 06:06 |
| darwin | using the kernel the liveDVD has didn't fix the error of no GUI | 06:14 |
| rrq | it comes up with black monitor after booting? | 06:16 |
| rrq | does ctrl-alt-f1 lead to a tty1 login prompt? | 06:17 |
| darwin | no | 06:17 |
| darwin | it's all black | 06:17 |
| rrq | before booting? | 06:17 |
| darwin | what do you mean before booting? | 06:18 |
| rrq | I mean you don't see any boot notices | 06:18 |
| darwin | i already said I did | 06:18 |
| darwin | it shows text then goes blank and <CTRL><ALT><Fn> doesn't do anything | 06:18 |
| darwin | boot notices aren't before booting | 06:19 |
| rrq | right. that indeed suggests that X has started but there's something with graphcs... you are using X ? | 06:19 |
| darwin | first you boot the BIOS, then (maybe bootloader then) OS but before that of course it's blank | 06:19 |
| darwin | yes | 06:20 |
| rrq | so not wayland? | 06:20 |
| darwin | not unless that's default on Devuan 5 | 06:20 |
| rrq | which DE ? | 06:20 |
| darwin | XFCE | 06:20 |
| rrq | with slim? | 06:21 |
| darwin | lightdm | 06:21 |
| rrq | mmm should be ok, but there's a possibility it's configured towards wayland; I'm not sure.. | 06:22 |
| rrq | can you boot with S on boot command line? | 06:22 |
| darwin | what does that mean? | 06:22 |
| rrq | called "single-user mode" I think .. means it starts with rcS.d only... no X | 06:23 |
| rrq | so doing that would log you in as root on a command line (if I remember right) | 06:24 |
| darwin | can't I just turn X off in /etc/inittab ? | 06:24 |
| rrq | not easily | 06:24 |
| darwin | so Devuan/Debian deviated from that standard? | 06:25 |
| rrq | which standard? | 06:25 |
| darwin | inittab runlevels including command-line only and GUI/X/etc. | 06:25 |
| onefang | You can easily disable X from starting by disabling the display managers init script. | 06:26 |
| rrq | X usually starts towards the end of rc2.d by starting the login manager | 06:26 |
| rrq | but all that requires you to have command line access | 06:26 |
| onefang | Display / login, whatever it's caled today. | 06:26 |
| darwin | which one is that? | 06:27 |
| rrq | I have xdm ... you said you have lightdm | 06:28 |
| darwin | i think I should use something that has proper inittab runlevels | 06:30 |
| rrq | proper? | 06:31 |
| onefang | I suspect darwin has /etc/inittab confused with sys-v-init scripts and how you disable them. | 06:32 |
| onefang | Or the "some distros use specific run levels for specific things". | 06:33 |
| rrq | https://en.wikipedia.org/wiki/Runlevel | 06:34 |
| rrq | looking at those tables, I think I got schooled with Gentoo runlevels. Haven't really thought about them much | 06:35 |
| rrq | or "System V runlevels" perhaps | 06:36 |
| rrq | yes, that's with sysvinit implements | 06:37 |
| rrq | ... so the S on the boot command line makes it select runlevel S | 06:38 |
| rrq | though since that's supposedly implemented in the initrd init scripts things may have changed | 06:43 |
| onefang | Is systemctl OK to install? | 07:02 |
| onefang | Trying to figure out how to reboot from Linux into the computers firmware / BIOS setup. | 07:03 |
| onefang | Web says use systemctl, which I didn't have installed. | 07:03 |
| onefang | My basic problem is that the Del key no longer does that. rEFInd has an option to reboot into the BIOS, which does work. I want to switch to syslinux to do both BIOS and EFI boots. | 07:05 |
| onefang | Maybe syslinux has a module for that? | 07:05 |
| rustyaxe | systemctl is a systemd thing im pretty sure | 07:06 |
| onefang | The description says it's for systems where systemd isn't available, but it could be. | 07:08 |
| darwin | so I'll need instructions how to boot without GUI. If inittab isn't like *BSD/OpenSolaris UNIX or the only strictly UNIX-like GNU/Linux, Slackware, then I'm lost how you disable GUI for the next boot | 07:32 |
| darwin | but multi-user boot without GUI would be just as good, right? Can I also use telinit to change to one of these non-GUI modes? | 07:32 |
| darwin | when I originally used Debian sometime 1998-2002 it automatically didn't boot GUI, which is preferable | 07:37 |
| gnarface | darwin: disable or uninstall the graphical login manager to boot to text mode. you don't have to change inittab at all for anything you've reported doing so far, but i seriously think you should spend some time trying to figure out why your install isn't working like the live iso | 07:42 |
| gnarface | you can disable any service, including the graphical login manager, using the update-rc.d tool, the sysv-rc-conf tool, or just editing the symlinks in /etc/rc*.d/ manually (i'm assuming you used the default init, sysvinit, which has a very predictable behavior that hasn't changed in decades) | 07:45 |
| onefang | Think I'll install both syslinux and rEFInd. Boot with syslinux, if I need the stuff rEFInd can do, then chain load it. Ugly, but might work. | 07:47 |
| CueXXIII | onefang: i can run the efi setup from grub by calling fwsetup, os-prober generates a menu entry for it automatically | 08:14 |
| CueXXIII | ah, not os-prober, just /etc/grub.d/30_uefi-firmware which is part of grub-common | 08:15 |
| CueXXIII | https://paste.debian.net/plain/1320631 (exception from the file) | 08:16 |
| darwin | removing desktop managers didn't bring back the display (multi-user) though I was able to in single-user mode | 08:16 |
| rrq | assuming it means that you are able to boot in single-user mode, then do so, mount the disk as needed an review the Xorg.0.log file | 08:18 |
| CueXXIII | onefang: rEFInd also has a "firmware" tool to boot into the systems firmware (= uefi settings) | 08:22 |
| darwin | i don't think X runs in single-user mode | 08:24 |
| onefang | I know, that's what I'm talking about, and why I still need rEFInd, even though I want to switch to syslinux. | 08:25 |
| rrq | darwin: that's right. you would use "less" to review the file | 08:25 |
| rrq | that log file would be from the last Xorg run | 08:25 |
| rrq | which supposedly would be from an attempted but locked-up login manager run | 08:26 |
| rrq | you may also want to review syslog rollback and perhaps kern.log rollback, so as to make a picture of what did happen | 08:27 |
| darwin | well, I got that already | 08:27 |
| darwin | i posted it earlier twice | 08:27 |
| darwin | the Xorg.0.log | 08:27 |
| darwin | without firmware-amd-graphics, the monitor works fine. With firmware-amd-graphics (because it's broken) Xorg.0.log changes drivers then keeps saying the monitor disconnected. It was used normally for months, and then I installed firmware-amd-graphics which made it blank on next boot, then installed it again and it happened again (and got log) then removed it and it worked fine again | 08:29 |
| darwin | it worked fine in between... I've done this several times now | 08:30 |
| darwin | the log was http://paste.debian.net/hidden/dec6defc | 08:30 |
| rrq | hmm only complaint I could see is for missing /dev/fb0 | 08:32 |
| rrq | plus no monitor ?? | 08:34 |
| rrq | which port would it be on? | 08:35 |
| rrq | which port was it on when things worked? | 08:36 |
| rrq | why is /dev/fb0 missing ? | 08:43 |
| darwin | on the DVI port. I don't know | 08:49 |
| rrq | is there a /sys/module/fb/ ? | 08:50 |
| rrq | a /sys/module/fb/uevent | 08:50 |
| darwin | yes | 08:52 |
| rrq | ok; "mknod -m 660 /dev/fb0 c 29 0 ; chgrp video /dev/fb0" | 08:53 |
| rrq | would get you a /dev/fb0 device node | 08:53 |
| darwin | okay, I did that from ssh | 08:54 |
| rrq | ok. so is the host running in runlevel 2 at the moment? | 08:55 |
| darwin | it's in a multi-user runlevel | 08:55 |
| darwin | the standard one | 08:55 |
| rrq | if so, as root: service lightdm restart | 08:56 |
| rrq | i.e. make the login manager restart, and thus restart Xorg while /dev/fb0 is present | 08:57 |
| darwin | still nothing happened other than (in log): 'AMDGPU(0): Output DVI-D-0 disconnected' (which isn't true) | 08:57 |
| darwin | it's a 2560x1600 monitor | 08:57 |
| darwin | that's why it's on DVD-D | 08:58 |
| darwin | which is just one cable | 08:58 |
| rrq | do you have another monitor to test with? | 08:58 |
| darwin | not like this one which the users want to use. They're hard to find in decent condition and expensive | 08:58 |
| darwin | because it's 16:10 | 08:58 |
| rrq | that may be fine, but it might then also require an explicit Monitor configuration for Xorg... | 08:59 |
| rrq | unless the monitor responds well to EDID reuqests | 08:59 |
| darwin | it's never required it before. All Debian derivatives we used it with didn't (Neon, Mint) | 09:01 |
| rrq | it'd be good to know wether it's "just" a problem driving that monitor, or a deeper graphics problem | 09:01 |
| rrq | ti may well be that the Devuan software, indivdually or in combination, takes issue with driving that monitor | 09:04 |
| rrq | so when Xorg restarted it ressulted in the very same log file (apart from timetamps I suppose)? | 09:06 |
| darwin | looks similar | 09:11 |
| darwin | i didn't compare them though | 09:11 |
| rrq | including not finding /dev/fb0 ? | 09:11 |
| darwin | i have no idea | 09:11 |
| darwin | yes | 09:12 |
| rrq | and "ls /dev/fb0" agrees with that? | 09:12 |
| darwin | well, I restarted since then | 09:12 |
| darwin | but I created it when you said | 09:13 |
| darwin | someone said the particular display/video/graphics card (Vega 64) had compatibility problems... that may be it. The problem is newer ones can't do 2560x1600 which requires DVI-D (not DP) | 09:13 |
| rrq | I need to go I'm afraid; right now it looks like /dev/fb0 does not get set up as it should, but maybe that's caused by graphics disagreeing with monitor. | 09:17 |
| gnarface | rrq: not sure on this one, but he reports that the live iso works, so at this point it looks like it might be a specific bug in the login manager, but i think two have been tried so... | 09:42 |
| gnarface | darwin: you didn't actually edit your /etc/inittab did you? | 09:42 |
| gnarface | like commenting out all the gettys or something? | 09:42 |
| darwin | no | 09:43 |
| darwin | why would I do that? | 09:43 |
| darwin | only a fool would comment those out | 09:43 |
| gnarface | dunno. if you really disable the login manager, you should get a text prompt not a black screen. if the live iso works the only difference here seems to be the login manager, so you should be able to still start xorg with "startx" from the text prompt | 09:43 |
| gnarface | (it's in the package "xinit", which may not be installed by default) | 09:44 |
| darwin | should I switch login manager? | 09:44 |
| darwin | well, I removed lightdm, sddm, slim... didn't change anything | 09:44 |
| gnarface | haven't you tried two already? at this point you should first see if startx works. hell, you should first see if you can even get a text prompt. nobody has these types of problems... | 09:44 |
| gnarface | you think it's possible the text prompt is rendering offscreen due to miscalibration? i've seen that on older monitors commonly.... nothing recently though, but it's possible | 09:45 |
| darwin | i haven't tried two and there was no text multi-user prompt after I removed them all... only single-user | 09:45 |
| gnarface | maybe if you're sure you disabled the login manager and you still get a black screen, try just logging in blind to see if it works | 09:45 |
| darwin | ok | 09:46 |
| gnarface | or maybe go into the monitor's OSD menus at that point, and dig around for an "auto-adjust" feature | 09:46 |
| gnarface | sometimes you have to run it once to get it to calibrate | 09:46 |
| gnarface | (VERY common to see this on very old monitors, the CRT type, but i remember seeing it even on some of the early LCDs, so i think in theory it could be a problem that crops up again) | 09:47 |
| darwin | the monitor doesn't have a menu | 09:47 |
| gnarface | the live iso isn't the same thing as a regular install, but there's only a few things that could be different here. logical deduction should work | 09:47 |
| gnarface | if you can at least get a text prompt in mult-user mode, "startx" should give you a desktop | 09:48 |
| darwin | most/all 2650x1600 monitors didn't have menus | 09:48 |
| gnarface | well, if it's something weird, which i have had before here, i know that sometimes you have to set a Modeline in your xorg.conf specifically to make it work, but that wouldn't explain why the live iso works | 09:49 |
| darwin | nor why this installation works without firmware-amd-graphics | 09:49 |
| gnarface | and both your desktop and and the login manager should be using the same xorg, so there shouldn't be a difference but still... | 09:49 |
| darwin | how long should I wait before I login blind | 09:50 |
| gnarface | no, basic VESA functionality without firmware-amd-graphics is perfectly expected with this hardware. what you won't get is hardware opengl acceleration, which is what i initially thought was causing your login manager to give you a black screen (seen that bug before with slim) | 09:50 |
| gnarface | till the harddrive stops running and the cpu load goes to 0 | 09:50 |
| darwin | well HDD is just /home, not OS | 09:51 |
| darwin | someone said they could hear SSDs | 09:51 |
| gnarface | wait, you mean you don't have text output even during boot? | 09:51 |
| gnarface | it's black screen the whole way? | 09:52 |
| darwin | i said two or three times I see the boot text then after a lot of that it just goes blank | 09:52 |
| gnarface | oh, yea, then you can probably log in as soon as it goes black | 09:52 |
| gnarface | but if you have boot text but it still goes black after that, that suggests you did not succeed in disabling the login manager | 09:53 |
| gnarface | hmm... or it's rendering offscreen somehow | 09:53 |
| darwin | i completely removed them | 09:53 |
| gnarface | like off the top fo the screen? | 09:53 |
| gnarface | when it gets to the black screen, can you adjust vertical positioning or scale? | 09:53 |
| gnarface | *top of the screen | 09:53 |
| gnarface | if you run "ps aux --forest" over ssh after it gets to the black screen you can be sure gettys are running and nothing else is getting in the way | 09:54 |
| gnarface | after that it has to be a display issue | 09:54 |
| gnarface | adding the firmware-amd-graphics package would probably dramatically increase the default resolution auto-detected, so that might explain it working without it | 09:55 |
| darwin | i did manage to login but it's still blank | 09:56 |
| gnarface | you can get in with ssh too, right? | 09:56 |
| darwin | yes | 09:56 |
| darwin | so I saw I ran bash and a command | 09:57 |
| gnarface | "ps aux --forest" doesn't happen to make anything scroll on to the screen does it? | 09:57 |
| darwin | no | 09:58 |
| gnarface | can i see? | 09:58 |
| darwin | i don't know how you'd do that--fly out here, LoL? | 09:59 |
| gnarface | i just want to sanity check the output of your "ps aux --forest" to make sure there's nothing weird about it you missed | 09:59 |
| darwin | oh, like from ssh? | 09:59 |
| gnarface | yea. it should be basically nothing running except a couple daemons and your gettys | 09:59 |
| gnarface | at a glance i could tell you for sure if something running was blocking your display somehow | 09:59 |
| darwin | there was a lot of stuff | 10:00 |
| gnarface | well, the ones that start with \_ are kernel things | 10:01 |
| darwin | paste.debian.net wouldn't let me paste it: 'do not spam' | 10:01 |
| gnarface | you can ignore those if you're running the stock kernel | 10:01 |
| gnarface | make sure your email address isn't in it | 10:01 |
| darwin | it's not | 10:02 |
| gnarface | scroll down past the last "\_ [kworker/blah:1]" thing, and just paste the regular processes after that | 10:02 |
| darwin | i removed the first few pages of kernel stuff and the same happened on my next try there | 10:02 |
| gnarface | you can just /msg it to me | 10:02 |
| darwin | i can do that if you're sure you don't want it on p.bsd-unix.net --GNU/Linux was inspired by UNIX anyway... | 10:03 |
| gnarface | some other time maybe | 10:03 |
| gnarface | darwin: yea, hmm... you disabled the login manager though, disable seatd too. to be honest, this is a lot of stuff for a base install... i almost feel like the thing to do really looking at this would be to do a minimal install and then put stuff back one at a time until you find the culprit | 10:05 |
| gnarface | darwin: ... but disable seatd first to make sure it's not that | 10:05 |
| gnarface | i'm sure xorg will still warn you it's not running but i can't imagine a situation you'd need it for without a login manager to use it anyway | 10:06 |
| gnarface | (and so far i'm getting by fine without it here) | 10:07 |
| darwin | i did '/etc/init.d/seatd stop' then typed 'clear', 'reset' on the blank screen (nothing happened) | 10:07 |
| gnarface | no, that's not how you do it | 10:07 |
| darwin | i don't know how to disable things in there since they don't seem to have that option, or if you have to reboot after | 10:07 |
| gnarface | well, i'm not sure but i think for this you might at least find it easier to reboot, but let's make sure it's disabled for real first. run this: ls -l /etc/rc*.d/*seat* | 10:08 |
| gnarface | and before you reboot make sure the word "quiet" doesn't appear anywhere in /etc/default/grub | 10:09 |
| gnarface | is your grub prompt rendering fine? if so, try also setting GRUB_GFXPAYLOAD_LINUX="keep" in there | 10:10 |
| gnarface | (make sure "nomodeset" is also not present anymore) | 10:10 |
| darwin | there was no /etc/rc*.d/seat* | 10:10 |
| darwin | yes, GRUB is fine | 10:10 |
| darwin | quiet was in there | 10:11 |
| gnarface | yea, add GRUB_GFXPAYLOAD_LINUX="keep" and remove quiet from GRUB_CMDLINE_LINUX | 10:11 |
| gnarface | it might in fact be worth trying a reboot just to see what that changed, but disable seatd first | 10:12 |
| darwin | that pisses me off they made the default quiet | 10:12 |
| gnarface | and don't forget to run update-grub after changing that /etc/default/grub file | 10:12 |
| gnarface | yea, it pissed me off too, but that was a good few years ago | 10:12 |
| gnarface | ok, but before rebooting | 10:13 |
| darwin | should I also remove 'splash'? | 10:13 |
| gnarface | you want to symlink K01seatd to "../init.d/seatd" in each of those /etc/rc*.d/ directories | 10:13 |
| darwin | i read otherwise the text will be over some sort of 'splash screen' maybe hard to read | 10:13 |
| * gnarface facepalm | 10:13 | |
| gnarface | good god yes remove splash | 10:13 |
| gnarface | i thought about mentioning that but i assumed you were getting enough text that couldn't be the issue | 10:14 |
| darwin | but where is K01seatd? | 10:14 |
| gnarface | do you know what a symlink is? | 10:15 |
| darwin | yes | 10:15 |
| gnarface | K01seatd is the name of the symlink you're creating | 10:15 |
| gnarface | in each of those directories make one named the same thing, and point them all to, literally: ../init.d/seatd | 10:15 |
| gnarface | that's how you disable a daemon in sysvinit | 10:16 |
| gnarface | though there are several tools to automate the process, it's impossible to make this conceptually simpler by adding more abstraction | 10:16 |
| gnarface | lrwxrwxrwx 1 root root 15 Mar 1 2023 /etc/rc0.d/K01seatd -> ../init.d/seatd* | 10:17 |
| gnarface | lrwxrwxrwx 1 root root 15 Mar 1 2023 /etc/rc1.d/K01seatd -> ../init.d/seatd* | 10:17 |
| gnarface | lrwxrwxrwx 1 root root 15 Mar 1 2023 /etc/rc2.d/K01seatd -> ../init.d/seatd* | 10:17 |
| gnarface | lrwxrwxrwx 1 root root 15 Mar 1 2023 /etc/rc3.d/K01seatd -> ../init.d/seatd* | 10:17 |
| gnarface | lrwxrwxrwx 1 root root 15 Mar 1 2023 /etc/rc4.d/K01seatd -> ../init.d/seatd* | 10:17 |
| gnarface | lrwxrwxrwx 1 root root 15 Mar 1 2023 /etc/rc5.d/K01seatd -> ../init.d/seatd* | 10:17 |
| gnarface | lrwxrwxrwx 1 root root 15 Mar 1 2023 /etc/rc6.d/K01seatd -> ../init.d/seatd* | 10:17 |
| darwin | K01seatd already exists like that in in rc0.d, rc1.d, rc6.d | 10:17 |
| gnarface | woops, that didn't all go to the right window, sorry | 10:17 |
| gnarface | yea, that's fine, if it's already there leave it there | 10:18 |
| darwin | i got it in 0-6... not rcS.d? | 10:19 |
| gnarface | hmm, nope, not that one, no K's in there, not sure why | 10:21 |
| gnarface | you could try it. maybe you can actually just put ONLY a K01seatd in there and you wouldn't have to put one in the others | 10:21 |
| gnarface | ah, the README says it's for single user mode | 10:22 |
| gnarface | /usr/share/doc/sysv-rc/rcS.d-README | 10:22 |
| darwin | ok | 10:23 |
| gnarface | er, it says it's for booting, even in single user mode | 10:23 |
| gnarface | so maybe everything in here gets run regardless | 10:23 |
| gnarface | so yea, maybe you could just put it in that one, but i'm not sure | 10:23 |
| gnarface | i've always been safe and just put it in all of 0-6, never had a problem with that | 10:23 |
| gnarface | i think S was added... later | 10:24 |
| gnarface | maybe not, maybe i'm just remembering it wrong, but if you put it in all of 0-6 it will definitely work | 10:24 |
| gnarface | and that will work for cups too | 10:24 |
| gnarface | and avahi-daemon | 10:24 |
| gnarface | and a probably a bunch of other gremlins you should kick overboard | 10:24 |
| gnarface | but maybe you like that stuff, who am i to judge | 10:25 |
| gnarface | then reboot it and... if nothing changes make sure you logged into the right computer | 10:26 |
| darwin | in various older UNIX/GNU/Linux with SysVInit (and some BSDInit) you just make /etc/rc.d scripts non-executable | 10:27 |
| darwin | i can disable a lot of the extra stuff | 10:30 |
| darwin | a lot of it was just dependencies and shouldn't have been running without permission | 10:30 |
| gnarface | i see you have colord installed, which was probably a recommendation... you might prefer to do updates with recommends disabled | 10:31 |
| gnarface | maybe you like the features colord provides, i dunno, but for practical use it's usually unnecessary unless you're doing something fancy | 10:32 |
| gnarface | avahi is an enduring pain in the ass if you're not expecting network-level auto-configuration | 10:32 |
| gnarface | some stuff looked like it was installed as windows compatibility stuff... you using windows on this network? | 10:34 |
| darwin | ls: cannot access '/dev/fb*': No such file or directory | 10:34 |
| darwin | i saw the kernel booting stuff now until about the time it changes to framebuffer mode, then went blank again | 10:34 |
| gnarface | is that after a reboot? | 10:34 |
| darwin | yes | 10:34 |
| darwin | rrq talked to me earlier and had me recreate it (didn't change anything) | 10:34 |
| gnarface | apt-get update && apt-get --no-install-recommends install eudev | 10:35 |
| gnarface | what does this do? ^ | 10:35 |
| gnarface | saned ... only for scanners, do you own a scanner? if not you don't need to run this | 10:36 |
| gnarface | smbd < only for windows file sharing | 10:36 |
| gnarface | elogind-daemon < you should have ditched this with the login manager | 10:37 |
| gnarface | winbindd < also only for windows | 10:37 |
| gnarface | bluetoothd < you got any bluetooth stuff? | 10:37 |
| darwin | yes, we have a scanner | 10:38 |
| darwin | actually you can use SMBD on various OS | 10:38 |
| darwin | we have some bluetooth stuff | 10:39 |
| gnarface | sure, macos but macos can do better stuff with linux | 10:39 |
| darwin | i don't know why winbindd is there | 10:39 |
| darwin | we even used SMBD to share stuff on GNU/Linux | 10:39 |
| gnarface | well, whatever works but nfs is faster | 10:39 |
| darwin | eudev is already the newest version (3.2.12-4+deb12u1) | 10:39 |
| gnarface | it's really weird your fb0 isn't being created, but it's clearly a symptom of another issue | 10:40 |
| gnarface | ditch elogind-daemon and reboot again, see what happens | 10:40 |
| darwin | if I remove that it's going to install consolekit... is that okay? | 10:42 |
| darwin | it's also removing KDE | 10:42 |
| darwin | we don't use it anymore, bu some KDE software | 10:42 |
| darwin | but | 10:42 |
| darwin | i'll be back & forth from IRC and helping my dad | 10:43 |
| rrq | is it possible that the graphics firmware makes the monitor be mis-classified? | 11:47 |
| rrq | i.e. that without firmware, the monitor is recognized generically, and with firmware it's recognised specifically but wrongly? | 11:48 |
| fonky | a.f.a.i.k. the os should display the right screen no matter if fw is installed or not | 11:49 |
| fonky | oh | 11:49 |
| fonky | i may be wrong, i get vga monitor if no firmwar | 11:50 |
| fonky | even if installed vga monitor | 11:51 |
| rrq | yes, and darwin's worked without firmware but fails with it | 11:51 |
| fonky | this is over my head, better someone qualified help you | 11:51 |
| fonky | possibly it is the cable | 11:52 |
| fonky | let me try | 11:52 |
| fonky | argh i would have to reboot | 11:55 |
| fonky | using vga. tried hdmi | 11:55 |
| fonky | oh now it shows dvi monitor | 12:02 |
| fonky | with hdmi | 12:02 |
| fonky | and lets me switch via software, cute | 12:02 |
| darwin | after I disabled seatd and removed elogind, SSHD no longer ran so I had to reinstall. The first thing I installed was firmware-amd-graphics but I still got no GUI after that. Had nothing to do with my installation, of couse | 13:16 |
| darwin | course | 13:16 |
| gnarface | darwin: maybe the live iso works because of a kernel command-line argument | 13:28 |
| data41201 | systemd-tmpfiles, deleting /home | 18:46 |
| data41201 | https://x.com/DevuanOrg/status/1802997574695080067 | 18:47 |
| rustyaxe | hmm twatters new logo looks almost copyright-infringely similar to the Xorg logo :O | 19:05 |
| AlexLikeRock | soup! | 19:22 |
| mason | data41201: That did make me wonder if there's an equivalent Fediverse account. | 19:54 |
| data41201 | mason: I got the link from here: https://mastodon.social/@nixCraft/112637213238431183 | 20:40 |
| mason | data41201: Yeah, but that's not Devuan. It mentions a Devuan Twitter account. My curiosity is if there's a Devuan Fediverse account. | 20:41 |
| data41201 | mason: https://dev1galaxy.org/viewtopic.php?id=5955 | 20:47 |
| mason | Ah, thanks. | 20:47 |
| mason | https://toot.community/@devuan | 20:47 |
| onefang | This is an interesting result from my divide and conquer RAM gremlin hunting strategy. The firefox in the host hung up so badly even killall -KILL can't kill it. The one in the 128 GB RAM VM guest kept running fine. | 23:09 |
| onefang | On the other hand, the VM one doesn't have all my open tabs, that was the next step to copy them across. lol | 23:10 |
| onefang | But now I have to reboot the host, which is where my IRC client is. So BRB. | 23:11 |
| onefang | OK I'm impressed. That hung firefox wouldn't even go away during the shutdown. Kept things from unmounting. I half expected to spend the next hour fscking 22 TB of storage, mostly spinning rust. | 23:25 |
| AlexLikeRock | X11 no let me use keyboard and mouse | 23:28 |
| AlexLikeRock | ideas ? | 23:28 |
| onefang | Ah looks like the filesystem that firefox kept locked up was tmpfs anyway. Phew. | 23:29 |
| rrq | AlexLikeRock: daedalus? | 23:30 |
| AlexLikeRock | cat /etc/issue | 23:31 |
| AlexLikeRock | Devuan GNU/Linux 4 \n \l | 23:31 |
| AlexLikeRock | chimaera | 23:31 |
| rrq | ok.. is that USB keyboard + mouse? | 23:33 |
| rrq | and is it via a login manager (so called greeter) or via startx? | 23:35 |
| rrq | and, yes, has it worked previously or is this a new or partially new setup ? | 23:40 |
| AlexLikeRock | its laptop | 23:42 |
| AlexLikeRock | embed jeyboard | 23:42 |
| AlexLikeRock | startx by mow | 23:42 |
| AlexLikeRock | i test by OPENBOX : FAIL | 23:43 |
| AlexLikeRock | MATE : FAIL | 23:43 |
| AlexLikeRock | BOTH BY STARTX | 23:44 |
| AlexLikeRock | http://paste.debian.net/1320717/ | 23:45 |
| rrq | how about commenting out lines 26-37 (inclusive) so as to let X11 autodetect input devices? I assume you have xserver-xorg-input-libinput installed | 23:48 |
| AlexLikeRock | ok, upload log | 23:50 |
| AlexLikeRock | yes , its all ready installed | 23:51 |
| AlexLikeRock | its installed this script : /etc/init.d/keyboard-setup.sh | 23:52 |
| AlexLikeRock | http://paste.debian.net/1320719/ | 23:52 |
| rrq | good. so X11 should be able to autodetect | 23:55 |
| rrq | if your run startx as non-root user though, then that user must have group access to /dev/input (whether autodetect or not) | 23:56 |
| rrq | to all /dev/input/* devnodes | 23:56 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!