systemdlete | gnarface, onefang: The results are in. When I boot the host hardware with beowulf, the mic works. When I boot with the live chimaera ISO, the mic does NOT work. | 07:23 |
---|---|---|
systemdlete | So, some good thinking there guys. Thanks! | 07:23 |
onefang | At least you didn't break it. | 07:24 |
systemdlete | (for once) | 07:24 |
systemdlete | (yeah) | 07:24 |
onefang | Well, next time. B-) | 07:24 |
systemdlete | I'll try harder, yes. | 07:24 |
systemdlete | gnarface, you suspected a kernel regression | 07:24 |
systemdlete | and it does sound like one | 07:25 |
systemdlete | Yeah, I guess this little device is pretty hardy after all. | 07:26 |
onefang | In my new hobby as a digital music producer, I've been messing with all the audio hardware I got. Including USB device, two audio chips on the motherboard where one is a USB front end to the other, and those six HDMI audio devices on my graphics card I mentioned before. | 07:28 |
systemdlete | I'm wondering if pulseaudio is required for all sound stuff now. | 07:28 |
onefang | I use ALSA for the HDMI speakers on my main monitor, JACK for everything else, and no pulseaudio installed. | 07:28 |
onefang | Oh I think Firefox-ESR is using apulse. | 07:29 |
systemdlete | with chimaera? or beowulf? | 07:29 |
onefang | So might just be drivers for your particular device? | 07:29 |
onefang | Chimaera. | 07:29 |
systemdlete | must be then | 07:29 |
systemdlete | maybe the kernel people don't think that maintaining old fart devices is worth it | 07:30 |
systemdlete | (remember the alsa vs OSS wars?) | 07:31 |
onefang | Oddly enough, the motherboard is only a couple of years old, but the audio chips on it are crap, but the USB device is great, and it's ancient. | 07:31 |
systemdlete | hmmm. | 07:31 |
systemdlete | This cam is only a few years old... I bought it about 5 years ago I think | 07:32 |
rrq | systemdlete: does your beowulf have a .asoundrc file? | 07:32 |
systemdlete | But I think it was introduced c. 2012 or so, from the posts I read | 07:32 |
onefang | Not sure how good the HDMI audio is, the speakers on my main monitor are tiny things with no bass. I didn't buy it for the speakers, they just came along for the ride. | 07:32 |
systemdlete | hmmm. | 07:32 |
systemdlete | good point. | 07:32 |
systemdlete | rrq: Only if the beowulf live ISO has a .asoundrc file | 07:33 |
onefang | I use a .asoundrc for the ALSA stuff. | 07:33 |
systemdlete | I always forget about configuring the asoundrc file for some reason. | 07:33 |
rrq | I thought USB sound equipment need some special declarations in .asoundrc | 07:34 |
onefang | Not mine. | 07:34 |
systemdlete | rrq: It could be the case. | 07:35 |
systemdlete | what I meant was, I ran the live beowulf ISO on hardware. I made no modifications. | 07:35 |
onefang | Hmmm, might be a difference in the order of devices between beowulf and chimaera, and it's the first ALSA input on one, but not on the other, thus not the default. | 07:35 |
onefang | cat /proc/asound/devices | 07:36 |
onefang | cat /proc/asound/pcw | 07:36 |
onefang | aplay -l | 07:36 |
onefang | May shed some light on things, then dig deeper. | 07:37 |
systemdlete | But you know what. On my regular work machine, I get the same problem, and it is running chimaera. | 07:37 |
onefang | OOps, missed the one you should try first - cat /proc/asound/cards | 07:37 |
systemdlete | no pcw file | 07:38 |
rrq | pcm | 07:38 |
onefang | Might be a typo in my notes. lol | 07:38 |
onefang | Ah yep, pcm. | 07:39 |
* onefang corrects my notes. | 07:39 | |
systemdlete | https://pastebin.com/UxZzrkFC | 07:40 |
systemdlete | (sorry I did not separate the outputs :( ) | 07:40 |
rrq | there might also be a difference in "file mode" for /dev/bus/usb/yada/yada between beowulf and chimaera live iso's | 07:41 |
systemdlete | I need to switch to my other PC... | 07:41 |
* rrq sees VX800 as card 2 | 07:42 | |
onefang | SB will be the ALC887-VD, which looks like a motherboard type chip. Listed first so it'll be the default. | 07:43 |
onefang | It and the VX8000 both have capture devices, so my guess is the SB capture is your default unless you tell ALSA otherwise. | 07:44 |
onefang | And the other computer / other OS might have them in different order. | 07:44 |
systemdlete | still looking for my old asound config files | 07:44 |
systemdlete | (right, I got you) | 07:44 |
onefang | Oh and SB has two capture, microphone and line in on your mobo? | 07:46 |
systemdlete | yep | 07:46 |
systemdlete | that's if you want to sing a duet with your sweetie? | 07:47 |
systemdlete | or something else entirely? | 07:47 |
systemdlete | j/k | 07:47 |
rrq | so alsa needs a "default" setup that uses VX800 for mic and SB for output ? | 07:47 |
onefang | Well now that I have my brand new bedroom music studio, with two microphones, .... | 07:47 |
systemdlete | interesting that when using, say, zoom, it recognizes both sound devices and lets me pick which one to use. But it still doesn't hear my mic | 07:48 |
onefang | Though it's been many decades since I last sang, voice is rusty. So I'm trying to find open source synthesized singing. lol | 07:48 |
systemdlete | but I'll keep looking for my config files. | 07:48 |
onefang | Or compose one from that info you just found. | 07:49 |
rrq | what do you get from "amixer -c 2"? (regarding the mic) | 07:49 |
onefang | As for something else ... guitar in line in, microphone in microphone. | 07:49 |
systemdlete | rrq: https://pastebin.com/s4MqP0dP | 07:51 |
rrq | hmm "79%" ... should be audible | 07:53 |
systemdlete | nothing, not even if I tap on the unit | 07:55 |
systemdlete | (arecord | aplay) | 07:55 |
rrq | I wonder if the live ISO user (root?) has .asoundrc file(s)... to explain the difference | 07:55 |
systemdlete | for beowulf, right? | 07:56 |
systemdlete | I'll need to get off here to test that... | 07:56 |
rrq | well it would be the non-working system that would exclude the card 2 input | 07:57 |
onefang | ZMY next suggestio nis to compare the cat /proc/asound/cards output on Beowulf and Chimaera, and your regular work machine to. My guess is different order in the first case, and no SB in the second. | 07:57 |
onefang | And my suggestion after that is for me to learn to type better. | 07:57 |
rrq | random web advice is to have "defaults.pcm.dsnoop.card 2" in .asoundrc for the testing user | 07:58 |
rrq | https://bbs.archlinux.org/viewtopic.php?id=271966 | 07:59 |
rrq | (it says a bit more, but I took that as decoration) | 07:59 |
systemdlete | /proc/asound/cards has same order on both machines | 08:00 |
systemdlete | I think rrq has rung a bell in my head | 08:00 |
onefang | Though one of the reasons for /proc/asound/cards is that you can use the device names instead of guessing which order things will boot up this time. So SB, HDMI, ond VX800. | 08:00 |
systemdlete | Istr needing to do something like that with "snoop" | 08:00 |
onefang | Snoop Dawg is that sweetie you sang a duet with? | 08:05 |
* rrq turns back to his flute | 08:07 | |
systemdlete | onefang: :p | 08:07 |
systemdlete | rrq: arecord or aplay (or both) don't like my .asoundrc file | 08:08 |
systemdlete | if I remove the "dsnoop" part, it is ok, but still doesn't work | 08:08 |
rrq | hmm maybe it also needs that pcm.!default ... block | 08:10 |
systemdlete | oh, I tried it with and without the pcm block | 08:10 |
rrq | ok hm | 08:10 |
rrq | have you "verified" the mic h/w somewhere else? | 08:11 |
systemdlete | Let me show you what I put there so far: https://pastebin.com/LnXei0rV | 08:11 |
onefang | https://wiki.archlinux.org/title/Advanced_Linux_Sound_Architecture#Set_the_default_sound_card might be useful. It's in a comment at the top of my .asoundrc. | 08:11 |
systemdlete | rrq: Yes, on beowulf. It worked fine. | 08:11 |
systemdlete | I tried removing both blocks, but no luck | 08:12 |
systemdlete | so, apparently, yes the hardware works. But not on chimaera (so far) | 08:12 |
onefang | Try "card VX800" | 08:13 |
rrq | ah; it needs the other pcm.!default, the "type asym" which adds "dmix" on output and "dsnoop" on input | 08:13 |
rrq | doesn't need the hint block | 08:14 |
systemdlete | sorry I am not expert on the syntax | 08:14 |
rrq | ok; page https://bbs.archlinux.org/viewtopic.php?id=271966 | 08:14 |
systemdlete | output works fine | 08:14 |
rrq | post #2, second code block | 08:15 |
rrq | but use card 0 instead of card 1 of course | 08:15 |
rrq | fwiw, I'm guessing; alsa is a mystery for me :) | 08:16 |
rrq | too | 08:17 |
systemdlete | strangest yet is that, iirc, the /etc/asound.conf file has a comletely diff syntax | 08:18 |
systemdlete | *completely | 08:18 |
systemdlete | for the example, the ctl should be on card... 0? 2? | 08:18 |
onefang | ctl is to control the card you mentioned in the pcm stanza. | 08:18 |
systemdlete | see, I am confused whether ctl is for the input or the output | 08:19 |
onefang | For the volume control I think. | 08:19 |
systemdlete | but there is a "stanza" right after the ctl stanza also | 08:19 |
onefang | cat /proc/asound/devices has "control" devices. | 08:20 |
systemdlete | and arecord/aplay are complaining that "card" is not an integer. | 08:20 |
systemdlete | (this is NOT at all intuitive) | 08:22 |
rrq | may need double quotes if not numbers .. except keywords | 08:22 |
rrq | allows double quotes or keywords though | 08:23 |
systemdlete | rrq: WHAT may need double quotes--see this is where I get really confused | 08:23 |
onefang | https://pastebin.com/j3tkXVQS is what I use. | 08:23 |
onefang | That sets Mobo0 is the ALSA device, which is in and out. But makes it easy for me to switch which device is used. | 08:24 |
onefang | At one point I was mixing devices in ALSA, but I ended up doing that in JACK instead. | 08:25 |
rrq | yes the syntax is all too confusing | 08:25 |
onefang | RIG is the name of my external USB sound device. | 08:26 |
rrq | eg the Mobo0 within the pcm.dmixer block does have double quotes | 08:26 |
onefang | Yep. | 08:27 |
rrq | but the pcm.Mobo0 block intro is without... although could have (maybe?) | 08:27 |
rrq | and there the card valu are literals without double quotes | 08:27 |
rrq | ... value is a literal ... | 08:28 |
systemdlete | I wasn't even using any strings/literals, only iintegers like 0 and 2 | 08:29 |
systemdlete | and why the exclamation before "default" ? Does that mean negation, like in 99.9% of the rest of software? | 08:30 |
onefang | The advantage of using the names is that sometimes the numbers are different on your next boot. | 08:30 |
systemdlete | or is that an escape to tell asound that it is about to eat a literal? | 08:30 |
systemdlete | or maybe a predefined item? | 08:30 |
rrq | the exclamation mark says "drop any previous 'pcm.default' sub tree and use this instead" | 08:31 |
onefang | "!" -- override | 08:31 |
systemdlete | I'm reading the docs on asound at the alsa website. But it is not exactly clear. | 08:31 |
systemdlete | ah, I see. | 08:31 |
onefang | So it'll override /etc/asoundrc. | 08:31 |
systemdlete | I would think that merely defining the item would automatically override any previous notion. | 08:32 |
rrq | yes one should think of the succession of directived as they are read by the config loader a stream of "tree editing" commands | 08:32 |
rrq | directived=directives | 08:33 |
systemdlete | I'll try to think that way. | 08:33 |
rrq | without "!" it means "add to what there is" | 08:33 |
onefang | There are different operation modes for parsing nodes, the default mode is merge and create. If operation mode is either merge/create or merge, type checking is done. Only same type assignments can be merged, so strings cannot be merged with integers. Trying to define a simple assignment in default operation mode to a compound (and vice versa) will also not work. | 08:33 |
onefang | Prefixes of operation modes: | 08:33 |
onefang | "+" -- merge and create | 08:33 |
onefang | "-" -- merge | 08:33 |
onefang | "?" -- do not override | 08:33 |
onefang | "!" -- override | 08:33 |
rrq | without "!" it is a silent "+" | 08:34 |
rrq | so it will only replace the leaf level, unless that is a plural value component | 08:34 |
systemdlete | Sorry, I really don't work with sound very much. So every time I run into sound issues, I have to go through the whole process of trying to understand the syntax. | 08:34 |
systemdlete | I get that now, thanks. | 08:35 |
systemdlete | But I don't recall having to do anything too esoteric to make my webcam work. | 08:35 |
onefang | I only know these things coz I was setting up my bedroom music studio last month. Lots of fancy hardware and software to try out and figure out what I prefer. | 08:36 |
onefang | Still don't know what the hell Mobo3 is supposed to be, or why my graphics card with two HDMI and two DIsplay Port has six HDMI audio devices. lol | 08:38 |
rrq | systemdlete: did you see the capture control on alsamixer? | 08:41 |
rrq | see=have | 08:41 |
systemdlete | rrq: Yeah. Checked all that a while back | 08:42 |
systemdlete | Yes, I enabled all the capture devices | 08:43 |
rrq | and it's "root" that wants to use it? | 08:43 |
systemdlete | made sure they are ON ("00") and green | 08:43 |
systemdlete | no | 08:43 |
systemdlete | not root | 08:43 |
systemdlete | just a regular user, why? | 08:43 |
systemdlete | I don't normally run programs as root unless they have to be | 08:44 |
rrq | righ. does that user have access to the usb /dev node and the audio device? | 08:44 |
systemdlete | and most of those are run out of cron and other means. | 08:44 |
systemdlete | rrq: The user can hear sounds. | 08:44 |
systemdlete | The computer cannot hear the user. On chimaera. On beowulf, all works fine. | 08:44 |
rrq | ok that's the card... the /dev/bus/usb/???/??? | 08:45 |
systemdlete | IOW, output works fine. | 08:45 |
rrq | and ? at the end of that :) | 08:45 |
systemdlete | It's just input on chimaera that seems to be the issue. | 08:45 |
systemdlete | sure, those files under /dev/bus/usb/... are all there. | 08:45 |
rrq | and the user has access? | 08:46 |
systemdlete | I can see the cam device at its bus and device location in that tree, based on lsusb output | 08:46 |
systemdlete | user has access... you mean to open files under /dev/bus/usb/... ? | 08:46 |
systemdlete | idk | 08:46 |
rrq | rw mode | 08:46 |
systemdlete | no, user wouldn't. The file is /dev/bus/usb/001/003 and the perms are 664 | 08:47 |
systemdlete | root-root | 08:47 |
systemdlete | so the user would not have write on it | 08:47 |
systemdlete | but... hold on | 08:47 |
systemdlete | This worked under beowulf, and I never tooled with those files at all | 08:48 |
systemdlete | (not directly at least) | 08:48 |
rrq | something to look at; maybe there's an "upping of security" for chimaera ... | 08:49 |
rrq | does it work if you change it to rw for user | 08:49 |
systemdlete | hmmm. | 08:49 |
systemdlete | I'll try that | 08:49 |
rrq | (rw for other I suppose) | 08:49 |
systemdlete | Nope. | 08:50 |
rrq | pity :) | 08:50 |
systemdlete | I removed the .asoundrc file | 08:50 |
rrq | yes if the capture control is still there in alsmixer then the .asoundrc file was a herring | 08:51 |
systemdlete | selinux? | 08:52 |
systemdlete | sometimes that has been a real pain | 08:52 |
rrq | oh? is that enabled? | 08:52 |
systemdlete | not sure | 08:53 |
systemdlete | I have not fiddled with it since installing chimaera on any of my systems here, hardware and VM | 08:53 |
systemdlete | I don't even see the se* utilities | 08:54 |
rrq | is there some "complaints" in /var/log/syslog ... about audio or capture or VX800 ? | 08:54 |
systemdlete | the only complaints are about packets coming from 209.25.233.253 | 08:55 |
systemdlete | (from my firewall) | 08:55 |
systemdlete | an occasional smartd message, cron jobs, etc | 08:56 |
systemdlete | nothing unusal | 08:56 |
rrq | are you comfortable with strace? ... to trace some recording program | 08:57 |
systemdlete | sure. I've used it several times. | 08:57 |
systemdlete | strace arecord, with a sprinkling of options perhaps? | 08:57 |
systemdlete | this is a test box, so we can muck around as much as needed | 08:58 |
rrq | ok; hmm not sure what to look for though | 08:58 |
rrq | audio has #channels and sample rate | 08:58 |
systemdlete | I can experiment with various options | 08:58 |
systemdlete | But the error messages I saw seemed to be syntax errors. | 08:58 |
rrq | "arecord -L" ? | 08:59 |
systemdlete | Sorry, it is late here and I was busy today. I think I'll keep poking around and see what I can find... but tomorrow. | 08:59 |
rrq | nw | 09:00 |
onefang | No worries. | 09:00 |
systemdlete | yeah lots of subdevices | 09:00 |
onefang | SNAP! | 09:00 |
systemdlete | including snoop | 09:00 |
systemdlete | I just heard something. But not from the VX800. | 09:00 |
rrq | "arecord -D VX800" should work | 09:00 |
systemdlete | it came from... onefang? | 09:00 |
rrq | :) | 09:01 |
rrq | it has mono, so "-c 1" | 09:01 |
systemdlete | nope. It says it knows vx800 not | 09:01 |
onefang | How close to Woolloongabba are you? lol | 09:01 |
systemdlete | uh... very far | 09:02 |
rrq | so VX800 is not in the output of -L ? | 09:02 |
systemdlete | it is in the output | 09:02 |
systemdlete | but that syntax doesn't work | 09:02 |
systemdlete | no | 09:03 |
rrq | how about "arecord -D hw:2" ? | 09:03 |
systemdlete | sorry, I mean passing those args don't work | 09:03 |
systemdlete | ok, a little better, but now, as with the presence of my .asoundrc file, it complains about the input | 09:04 |
systemdlete | format | 09:05 |
systemdlete | only one format is available | 09:05 |
rrq | for input or output? | 09:06 |
rrq | (tell us when you fall asleep) | 09:06 |
systemdlete | arecord, so input | 09:06 |
rrq | ok maybe "-c 1" would help? (telling that it's mono) | 09:06 |
onefang | I should fall asleep soon. Been up way too long. lol | 09:06 |
systemdlete | aplay complains of course because it wasn't fed any intelligible output from arecord | 09:07 |
systemdlete | ALSA lib pcm.c:2660:(snd_pcm_open_noupdate) Unknown PCM VX800 | 09:07 |
systemdlete | arecord: main:830: audio open error: No such file or directory | 09:07 |
systemdlete | aplay: read_header:2839: read error | 09:07 |
onefang | Unless the neighbours don't stop being loud. | 09:07 |
systemdlete | (the other errors are similar. And sorry for flooding the channel...) | 09:07 |
systemdlete | But onefang, why would I want to fly all the way to Brisbane? Are you anxious to show off your new sound system? :P | 09:08 |
systemdlete | It's almost as if the kernel knows about my VX800... there is evidence all around. But alsa doesn't ??? | 09:09 |
rrq | how about "arecord -D 'USB Audio'" ? | 09:10 |
systemdlete | The error "No such file or directory" could mean anything--the same lousy few error codes are used over and over again throughout the kernel. | 09:10 |
systemdlete | Same error, but this time it says it does not know "USB Audio" rather than not knowing "VX800" | 09:11 |
onefang | Well next time I volunteer for the local reggae festival, I'll borrow all their sound systems to test my sub bass sound. Then you wont need to come to Brisbane to hear it. | 09:11 |
systemdlete | lol | 09:12 |
systemdlete | There wouldn't be much left to see, I think, if I am reading you | 09:12 |
systemdlete | (you haven't been hanging around syria and turkey lately, have you?) | 09:13 |
systemdlete | onefang revives the original intent of the brand "Sound Blaster" | 09:14 |
onefang | lol and g'night. | 09:14 |
systemdlete | ok, all. My brain is timing out | 09:14 |
rrq | night | 09:14 |
systemdlete | thanks for all the help. I will try with strace tomorrow. I'll follow up and let you know what I found out. Probalby something really dumb. It usually is. | 09:15 |
joew1 | hi the kernel dosent pick up my usb wifi adapter (tp link 722n) and when i run lsusb it dosnet show up | 14:09 |
gnarface | joew1: run this real quick: dmesg |grep -i firmware | 14:12 |
joew1 | gnarface https://paste.debian.net/1271736 | 14:16 |
gnarface | joew1: looks like you're missing more than one firmware package. the relevant one for this is probably the last 3 lines (i think those should come from the firmware-realtek package in non-free... or non-free-firmware now i guess if you're on daedalus or later) | 14:18 |
nemo | grrrrr synaptic and synaptic-pkexec both fail for me. | 16:43 |
nemo | sudo synaptic fails due to x session relay (even when I try to pass DISPLAY to sudo and set xauth) | 16:43 |
nemo | and synaptic-pkexec fails with a "no session for cookie" error | 16:44 |
nemo | *rage* | 16:44 |
rwp | Sounds like the best answer is to use "sudo apt-get install ..." then since that will work regardless. | 16:44 |
nemo | rwp: yes.. I just wanted a nice search interface | 16:46 |
nemo | which is helpful sometimes when hunting for packages | 16:46 |
nemo | which is why I probably only just noticed this problem now | 16:46 |
rwp | I like having a nice search capability too which is why I use "apt-cache search xxx | grep yyy" for the best interface for me. :-) | 16:52 |
systemdlete | So still looking at the problem with my VX800 webcam. I was finally able to get the mic to work on devuan live--I had to push the capture up almost all the way to get anything, but it does play back. However, the mic still does not work on installed chimaera systems. One thing I notice is that the LED goes out soon after plugging the | 23:56 |
systemdlete | device into a USB port on either machine, for those instances when arecord does not work. | 23:56 |
systemdlete | When arecord does work, the LED stays on consistently before, during, and after from first inserting the USB device into a port on the machines. | 23:57 |
systemdlete | When arecord does work, I can see the vumeter changing. This does not happen when the LED is off, that is, when the device is not working. | 23:58 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!