| systemdlete8 | I have a vx800 webcam (I know, it is crummy, but it has worked fine in the past) which is no longer working for me. The led comes on and the system seems to detect it alright. But zoom does not show me an image (that is, me). Also, speaking does not seem to activate the audio in (the detector does not move when I speak) | 01:47 |
|---|---|---|
| systemdlete8 | lsusb shows the cam | 01:47 |
| systemdlete8 | I have an old version | 02:04 |
| systemdlete8 | upgrading | 02:04 |
| systemdlete | different problem now. I'm trying to use my usb cam/mic in vbox. Host is beowulf, guest is beowulf. (I'm aware I need to upgrade to chimaera soon.) Running lsusb in the guest shows the webcam which is a MS VX800. arecord -l shows the mic device and I can access the mic in, say, audacity. But if I try to record, the line moves, | 04:08 |
| systemdlete | but no sound is recorded. I've tried using arecord from the command line but nothing is saved to my wav file. | 04:08 |
| systemdlete | The cam is also not working, but I am pretty sure it will once the sound is working; this usb device has worked before in devuan. And I think it was beowulf/beowulf before (but not 100% sure). Do I need to do something with an .asoundrc file or the alsa defaults in /etc? Seems like I needed to in the past, but I don't remember now. | 04:10 |
| systemdlete | I want to make sure I haven't overlooked some devuan config issue before I go to the bowels of vbox forums for assistance. | 04:11 |
| systemdlete | I am NOT using pulse on the host or guest | 04:11 |
| systemdlete | the led on the cam does come on when attempting to record sound or use zoom. It's just that I can't get video or the mic to work. | 04:13 |
| gnarface | this rings a bell, but i think it is something to do with vbox | 04:20 |
| gnarface | will it work on the host outside of vbox? | 04:21 |
| gnarface | i seem to recall it's usually just a permissions issue... vbox needs access to alsa inside and outside the guest or something | 04:21 |
| gnarface | vbox isn't something i know much about | 04:21 |
| gnarface | or it might be playing to some vbox internal dummy alsa that doesn't reach the host by default | 04:22 |
| systemdlete | Yes, I think that could be right. I don't remember. This was working about 2 months ago then stopped. Maybe due to a recent vbox upgrade or something else. Not sure. | 04:22 |
| gnarface | ugh, something else that rings a bell is needing to downgrade vbox to get around a systemd requirement | 04:23 |
| systemdlete | uh-oh. | 04:23 |
| systemdlete | Recently? | 04:23 |
| gnarface | no, not recently | 04:23 |
| gnarface | if it's that, it's not a new issue | 04:24 |
| gnarface | it's something that had been plaguing us for a few releases | 04:24 |
| systemdlete | That's very possible actually. idk. In the past, there have been upgrades that broke a few things, though not very often. | 04:24 |
| gnarface | i might be getting vbox confused with a different one though too | 04:24 |
| systemdlete | a few releases of devuan? or vbox on devuan? | 04:24 |
| gnarface | a few releases of devuan i mean | 04:24 |
| systemdlete | or just sound problems | 04:24 |
| systemdlete | oh, ok | 04:24 |
| systemdlete | I am running beowulf on both host and guest. | 04:25 |
| gnarface | this vbox issue i'm recalling is years old | 04:25 |
| systemdlete | And I am pretty sure I was running beowulf on both when it was working. But again, I'm not sure. | 04:25 |
| gnarface | there's a trick to it, i wish i could remember more | 04:26 |
| systemdlete | Is this something to do with .asoundrc or the like? | 04:26 |
| gnarface | i don't think it's only ~/.asoundrc | 04:26 |
| systemdlete | Istr that sometimes asoundrc has to be configured, but I forget if that's on the guest or the host (or both) | 04:26 |
| gnarface | though edits there might be part of making it easier | 04:26 |
| gnarface | check to see what driver it is using for the soundcard inside the guest | 04:27 |
| gnarface | that might be a clue if it's something like "dummy" | 04:27 |
| adhoc | has anyony put devuan onto recent lenovo SR5xx or SR6xx rackmount servers recently ? | 04:27 |
| systemdlete | what's an easy way to check that gnarface? | 04:27 |
| systemdlete | there is a snd_seq_dummy | 04:29 |
| systemdlete | lsmod shows that, not sure if that's the one you meant | 04:29 |
| systemdlete | gnarface: Here is something interesting. When I run lsusb -v in the guest while the cam is attached to it, it shows the usual output, but it also shows 1 error per device "Couldn't open device, some information will be missing" | 04:35 |
| gnarface | hmmm | 04:37 |
| gnarface | well, i can tell you snd_seq_dummy isn't loaded automatically here | 04:37 |
| gnarface | so it might be evidence that it is | 04:38 |
| gnarface | someone who knows vbox should know this, i don't | 04:39 |
| gnarface | but yea that thing about some information missing might be a clue too | 04:39 |
| gnarface | could just be permissions, i'd check those first | 04:39 |
| gnarface | but there might be some audio routing thing too | 04:39 |
| gnarface | the user that runs vbox on the host in theory would need to be in the audio group afaik | 04:40 |
| gnarface | but in the guest you gotta make sure the user in the guest has permission too, and maybe that it's using the right driver? | 04:40 |
| gnarface | eh, my guesses are useless anyway | 04:40 |
| systemdlete | sorry got disconnected for a moment | 04:46 |
| systemdlete | I'll check groups for user | 04:46 |
| systemdlete | (I don't remember this though) | 04:46 |
| systemdlete | so both host and guest users are members of audio on their respective systems. | 04:47 |
| systemdlete | when you say permissions, do you mean on /dev/audio* or something else? | 04:47 |
| systemdlete | I am wondering if vbox has recently dropped support for alsa, or else are getting chummier with systemd or the like. | 04:49 |
| systemdlete | (it's possible) | 04:49 |
| systemdlete | That would break a lot of things, so I wouldn't think they would be that bold. But there is economy (for them) to use pulseaudio and other systemd friends in terms of support. | 04:50 |
| systemdlete | At one point, they changed their startup scripts and broke startup for us non-systemd users. They fixed that pretty quickly though. | 04:51 |
| gnarface | systemdlete: it's /dev/snd but afaik being in the audio group should take care of that | 05:25 |
| gnarface | i wonder if there's a forum thread about vbox issues | 05:26 |
| gnarface | which vbox are you using? i only see it in ascii jessie and ceres | 05:28 |
| gnarface | considering you said you're on beowulf, yea i'd suspect the vbox version | 05:28 |
| gnarface | vbox is the same thing as virtualbox right? | 05:28 |
| systemdlete | yes, just saving some keystrokes | 05:35 |
| systemdlete | virtualbox aka vbox is available on all versions of devuan | 05:36 |
| systemdlete | well, I'm waiting to get some assistance. I'm chatting with one user over in #vbox but they are useless. All they want to do is barb devuan and twist my question. So, it may be some time. | 05:37 |
| systemdlete | The other possibility is that the thing is broken. Maybe the LED still works. I did bang it a few times, though I would think that being a solid state device it would be fairly resilient. | 05:38 |
| systemdlete | Maybe I'll try this on my test box. I could try downgrading and changing some stuff around. After all, that's what a test box is for. | 05:42 |
| gnarface | i'd try it on the host machine or yea a clean test box | 05:59 |
| gnarface | even without bugs, you would expect USB devices not to be able to access host hardware from inside the guest | 05:59 |
| systemdlete | ? | 05:59 |
| gnarface | it's expected behavior, basic security | 06:00 |
| systemdlete | attaching USB devices to guests works quite well actually. I have several USB ethernet adapters attached to VMs | 06:00 |
| systemdlete | They all work. | 06:00 |
| gnarface | that's different because you're dedicating the ethernet device to the guest | 06:00 |
| gnarface | but in this case you're trying to play audio on the host motherboard; that breaches containment | 06:01 |
| systemdlete | This device has worked in VMs previously. | 06:01 |
| systemdlete | I'm not saying you are wrong. | 06:01 |
| gnarface | well i have to assume you must have taken some steps | 06:01 |
| systemdlete | Maybe the devs have tightened up the kernel lately. I see they have issued a lot of updates to the kernel in the past few weeks. | 06:02 |
| gnarface | vmware has like, guest drivers you have to load | 06:02 |
| systemdlete | so does vbox | 06:02 |
| gnarface | special guest drivers to give the guests access to the host | 06:02 |
| systemdlete | yep, same for vbox | 06:02 |
| gnarface | i think qemu has some similar thing | 06:02 |
| systemdlete | mostly for usb and some other features like shared folders | 06:02 |
| gnarface | it really could just be a regression in vbox, could you possibly get the older version? pkginfo.devuan.org doesn't show a beowulf version... | 06:03 |
| systemdlete | I'm about to try that. It will take me some time though. | 06:03 |
| gnarface | (and by "regression" i mean "feature") | 06:03 |
| systemdlete | Need to shut down the VMs reinstall the old vbox, etc. But first I want to see if it works on that machine. | 06:04 |
| systemdlete | actually you are right about the repos. I've been manually downloading the updates from virtualbox.org. Part of this is because I need to download them ahead of updates due to my routing here. | 06:05 |
| systemdlete | gnarface: Is it just how I have my systems configured, or has devuan gotten kinda fat lately? I'm noticing some noticable delays at times, long disk syncs (I can see the disk LEDs running while the system slows way down) | 06:07 |
| systemdlete | I've tried hard to get rid of anything I don't use or don't need on all my systems, hosts and VMs. | 06:07 |
| systemdlete | I wonder if upstream has been force-feeding the software repos or something. | 06:08 |
| systemdlete | bbl | 06:08 |
| gnarface | systemdlete: eh, hard to say. i haven't noticed any difference but i've put everything running newer kernels on SSD now | 07:04 |
| gnarface | and i'm not using much of the desktop stack in most places | 07:05 |
| gnarface | i wouldn't be surprised if there have been changes to the way virtual memory is handled that's affecting disk traffic | 07:06 |
| gnarface | ram usage after cold boot has definitely gone up with the newer stock kernels though | 07:08 |
| gnarface | i think i'm seeing ~261MB on ceres | 07:09 |
| gnarface | there may be also some attempts to optimize for solid state storage that penalizes spinning platters | 07:10 |
| gnarface | worth looking into kernel parameters or config settings to turn it off maybe | 07:10 |
| systemdlete | gnarface: I forgot to try this on the host! It works there. So something is wrong with vbox's usb and or sound drivers | 10:23 |
| systemdlete | heading to the vbox forums... | 10:23 |
| systemdlete | but that is only on the testbox... on the PC where I do most stuff, the host is seeing the device in lsusb, but alsa doesn't seem to see it | 10:25 |
| systemdlete | seems I have to install packages for uvc. I do not recall having to do this in earlier versions of linux in order to use the vx800 cam... | 10:33 |
| systemdlete | nvm. That wasn't it. Nothing to do with uvc. Turns out I had pulseaudio installed on the testbox. I am removing it now. | 10:53 |
| systemdlete | But it does very clearly show that webcam works if pulseaudio is there, but not otherwise. | 10:54 |
| systemdlete | I tried apulse but I did not have much more luck with that | 10:54 |
| systemdlete | aplay -l does not show the webcam devices even though lsusb shows it is there -- on both workbox and testbox | 10:55 |
| systemdlete | but at least I now know that the webcam works at all; it is not broken | 10:56 |
| ham5urg | I debootstraped an unstable and installed task-gnome-desktop. Looks like it boots fine and when I mouse-click on the user, shown by GDM (I guess) I enter the password and the GUI-session try to start but hangs. What could be the prob.? How to debug the GDM? | 14:09 |
| ham5urg | I peeked into /var/log/gdm3/greeter.log | 14:11 |
| ham5urg | Unsupported maximum keycode 708, clipping. X11 cannot support keycodes above 255. | 14:12 |
| ham5urg | I doubt is is the keycode. It's a non fatal error | 14:22 |
| Kingsy | so I have a laptop with both intel and nvidia, what packages and config do I need to do to get it working as it should? | 14:22 |
| ham5urg | gnome shell message: error starting x11 services: error fetching own systemd unit: no such proccess - gnome-shell mutter-CRITICAL failed to init x11 display unknown error | 14:25 |
| ham5urg | what kind of process does gnome expects? | 14:26 |
| ham5urg | Kingsy, I would guess at least xserver-xorg-video-nvidia or xserver-xorg-video-nouveau | 14:28 |
| Kingsy | nvidia-persistenced seems to have failed. E: Sub-process /usr/bin/dpkg returned an error code (1) | 14:29 |
| Kingsy | anything to wrry about? or should I just reboot? | 14:29 |
| Kingsy | specifically installed nvidia-persistenced package post-installation script subprocess returned error exit status 1 | 14:31 |
| gnarface | systemdlete: note that the webcam mic will only show up under arecord since it is not also a playback device | 20:23 |
| gnarface | Kingsy: i'm not sure if they ever fixed the problem where nvidia's opengl library can't be installed parallel to the mesa ones. last i checked you could only get either nvidia or everyone else's hardware acceleration working in parallel... however nvidia-persistenced has only been an intermittent dependency for whatever reason. if it's doing this to you right now chances are the answer is to just uninstall it and | 20:28 |
| gnarface | ignore it | 20:28 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!