fsmithred | hagbard, try new isos next monday | 00:53 |
---|---|---|
hagbard | I have no more unused laptops left to try it. Got it to work with a bit of manual fiddling. But thanks for the fix. | 00:55 |
rrq | ls | 01:13 |
* rrq bah! wrong window | 01:13 | |
av6 | ham5urg: have you already figured out festival+pipewire? if not, pipewire creates (or can be asked to) an alsa device that aplay can use | 08:16 |
av6 | but pipewire also hogs the actual alsa devices, so trying to use hw:x,y in aplay directly while pipewire is running will result in such "busy" errors | 08:17 |
av6 | aplay -L | fgrep -i pipe | 08:18 |
av6 | if it exists, try using aplay -D pipewire | 08:20 |
av6 | or maybe festival supports pulseaudio protocol? if so, it can talk directly to pipewire | 08:22 |
av6 | it seems there's a noticeable delay when using -D pipewire... not sure why | 08:28 |
ham5urg | av6, I used this command in festival-interactive (Parameter.set 'Audio_Command "aplay -D pipewire: $FILE") | 11:19 |
ham5urg | A (SayText "Hello") run through but just makes noise | 11:20 |
hagbard | Hi, how do i disable suspend/standby on lid close (in deadalus)? Im trying to use an old laptop as an headless server. | 13:03 |
ShorTie | look to power management i'd guess | 13:07 |
phogg | hagbard: has it changed? I believe I have set this in UPower.conf in the past (IgnoreLid=true) | 13:23 |
hagbard | thx, i will have a look there | 13:23 |
rrq | elogind likes to deal with the buttons; see /etc/elogind/logind.conf | 13:29 |
rrq | though for a server you might prefer to purge elogind | 13:30 |
phogg | ah, quite so. I have mine set to HandleLidSwitch=ignore | 13:32 |
hagbard | ah, thanks | 13:32 |
hagbard | Yes, that did the trick. | 13:36 |
furrymcgee | devuans minimal server installation is a pretty good even for desktop users | 14:54 |
impostor192[m] | I'm a bit disappointed with my devuan installation, after stripping out everything that I did not need I'm having ~570 MB of memory consumption with XFCE. I expected to get something around 300-400 MB for this kind of minimal installation. | 15:02 |
impostor192[m] | Does anyone know why that happens? | 15:02 |
djph | impostor192[m]: stuff that was loaded before you deleted it all? | 15:08 |
djph | impostor192[m]: also, that's excluding buffers/cache, right? | 15:08 |
impostor192[m] | hmm yeah, that number is probably incorrect - that's what XFCE task manager shows | 15:10 |
impostor192[m] | I have about 400 MB shown as used in htop | 15:10 |
djph | wonder what I use right after I boot, hmmm ... | 15:15 |
djph | meh, I'm not that interested | 15:15 |
Vanfanel | Hi there! | 15:34 |
Vanfanel | I have noticed that, on Devuan for Raspberry Pi, devtmpfs is automounted on /dev by the kernel. Is that a Pi-kernel-only thing, a Devuan thing or is it common practice? | 15:36 |
Tenkawa | normal | 15:37 |
Tenkawa | mount | grep devtmp; cat /etc/os-release | grep PRETTY | 15:37 |
Tenkawa | udev on /dev type devtmpfs (rw,nosuid,relatime,size=3856380k,nr_inodes=964095,mode=755) | 15:37 |
Tenkawa | PRETTY_NAME="Debian GNU/Linux 11 (bullseye)" | 15:37 |
Tenkawa | udev on /dev type devtmpfs (rw,nosuid,relatime,size=3856824k,nr_inodes=964206,mode=755) | 15:38 |
Tenkawa | PRETTY_NAME="Devuan GNU/Linux 4 (chimaera)" | 15:38 |
Tenkawa | 2 diff pi 4's I have setup | 15:38 |
Vanfanel | Tenkawa: Thanks! Is that mount done by the kernel? I boot with my own minimal init script and I am NOT the one doing it... | 15:39 |
Tenkawa | its udev | 15:39 |
Tenkawa | ii udev 247.3-6 arm64 /dev/ and hotplug management daemon | 15:40 |
Vanfanel | Tenkawa: I don't have udev running at all, I never lauch it on my minimal init script | 15:40 |
Tenkawa | its a daemon | 15:40 |
Vanfanel | Tenkawa: I don't have any daemons running here, I use busybox's MDEV, not UDEV/EUDEV | 15:40 |
Tenkawa | if you don't you need to have something handling device management (ie bb or makedev) or devices will be null and void | 15:41 |
Tenkawa | mdev works, eudev too | 15:41 |
Vanfanel | Tenkawa: I don't understand that phrase.. do you mean "your devices will be null and void"? | 15:42 |
Tenkawa | the os will have no way to map them | 15:42 |
Tenkawa | there's no handler for them from the kernel to userspace | 15:42 |
Tenkawa | without the /dev entries | 15:43 |
Vanfanel | Tenkawa: that's not true, as far as I know at least. DEVTMPFS will create the nodes in /dev, and DEVTMPS is part of the kernel. | 15:43 |
Tenkawa | yes... but it only handles "staticly" compiled enties | 15:44 |
Tenkawa | entries | 15:44 |
Vanfanel | Tenkawa: do you mean it only handles the entries for devices that are present at boot? | 15:44 |
Tenkawa | in my understanding yes.. I'm testing it now | 15:45 |
Tenkawa | (this is why having 7 of these machines is nice) | 15:45 |
Vanfanel | Tenkawa: that's my understanding too, don't worry. For new hotplugged devices, I use MDEV, which is called by the kernel when new devices are connected. | 15:46 |
Tenkawa | (personally I think the whole subsystem is a mess still) | 15:46 |
Vanfanel | Tenkawa: Well, I like it :) Kernel does it's thing, MDEV does his own thing... it's easy to understand. | 15:47 |
Vanfanel | Tenkawa: the probem is that I DON'T KNOW who is auto-mounting DEVTMPFS in /dev | 15:47 |
Tenkawa | meh I liked it in the old days.... you tell it what you wanted in there and it did what "you" told it to... no guessing lol | 15:47 |
av6 | ham5urg: noise, huh? i wonder if it's because of audio format... can you, like, save that file and open it in a player that you know works fine? or look if it has the audio format set with e.g. ffprobe | 17:32 |
av6 | also, maybe you using pw-play instead of aplay will give better results | 17:33 |
av6 | (pw-play lives in pipewire-bin) | 17:36 |
ham5urg | av6, I tried (Parameter.set 'Audio_Command "pw-play $FILE") | 17:36 |
ham5urg | Got an "Format not recognized" | 17:36 |
ham5urg | I'm using Daedulus | 17:37 |
ham5urg | Is the sampling rate wrong? | 17:37 |
av6 | yeah, i think festival uses some strange format for the audio | 17:37 |
av6 | it's probably more like bitness and endianness, not sampling rate | 17:38 |
av6 | e.g. ffprobe /usr/share/sounds/alsa/Front_Center.wav says pcm_s16le | 17:38 |
av6 | so it's 16 bits signed integer per sample, little endian | 17:39 |
av6 | that's a format both alsa and pipewire have no problems understanding | 17:40 |
av6 | i don't know what festival outputs, but it's probably something different? possibly simply unrecognized by pw-play, that's why i suggest running ffprobe on that file, just to check | 17:41 |
ham5urg | av6, I tried (Parameter.set 'Audio_Required_Format 'aiff) but now I get just an #<Utterance 0x7f258ac26510> error from festival. | 17:42 |
av6 | https://help.ubuntu.com/community/TextToSpeech suggests that the file has no header | 17:44 |
ham5urg | raw format | 17:45 |
av6 | see how it does aplay -c 1 -t raw -f s16 -r $SR to specify channels, sample rate and format | 17:45 |
av6 | it's probably raw, try doing this for aplay | 17:46 |
ham5urg | I tried (Parameter.set 'Audio_Command "aplay -q -c 1 -t raw -f s16 -r $SR $FILE") | 17:47 |
av6 | you need -D pipewire | 17:47 |
ham5urg | Still a utterance 0x7... error | 17:47 |
av6 | remove aiff | 17:47 |
av6 | oh, and that wiki page uses snd? argh | 17:48 |
ham5urg | av6, it's not worth it. Festival looks a bit strange. | 17:48 |
ham5urg | I killed pipewire to get to alsa | 17:48 |
av6 | and does festival work now? | 17:49 |
ham5urg | no | 17:49 |
av6 | if you want to figure out what festival outputs, try 'Audio_Command "cp $FILE /tmp/" or something, then look at the file with ffprobe | 17:50 |
ham5urg | ok | 17:50 |
av6 | i don't know anything about festival, but it doesn't look good to me with all those cryptic utterance errors | 17:51 |
ham5urg | av6, ffprobe got me: /tmp/123.bin: Invalid data found when processing input | 17:54 |
av6 | hmmm | 17:55 |
ham5urg | maybe crapware | 17:55 |
av6 | you sure it's the output from festival? | 17:55 |
ham5urg | yes, the filename is unique | 17:55 |
av6 | also, what does `file /tmp/123.bin` say? | 17:55 |
av6 | just "data"? | 17:55 |
ham5urg | I created the file with | 17:56 |
ham5urg | (Parameter.set 'Audio_Command "cp $FILE /tmp/123.bin") | 17:56 |
av6 | ah | 17:56 |
ham5urg | I hoped to get a usable tts to transform ebooks to audiobooks. | 17:57 |
junicchi_ | is chimaera equivalent to the debian 10 or debian 11 | 18:52 |
Guest-12345678 | hi there ...complete new user..installed devuan from netinstall and chose runit...at the moment to start a service I used "doas /etc/init.d/iwd start". If soone could point me in the right direction to enable runit services I would be most grateful eg my "/etc/sv/* -> http://ix.io/3L8H | 19:02 |
* Guest-12345678 *soone -> someone | 19:03 | |
av6 | junicchi: it's 11 according to https://www.devuan.org/os/releases and https://en.wikipedia.org/wiki/Devuan#Version_history | 19:13 |
fsmithred | Guest-12345678, this discussion has links to sample run scripts and also has some posts from the runit maintainer. https://dev1galaxy.org/viewtopic.php?id=3716 | 19:34 |
Guest-12345678 | fsmithred super thxs!! | 19:34 |
fsmithred | I made a live-iso with runit, but I guess you don't need that since you already have it installed. | 19:35 |
Guest-12345678 | I pretty much copied / pasted the cron example -> https://salsa.debian.org/runit-team/runit-services/-/blob/master/debian/README -> however I still end up with "update-service --add /etc/sv/cron" -> "update-service: fatal: /etc/sv/cron does not exist, or is not a directory" | 21:24 |
gnarface | Guest-12345678: i dunno much about runit but that error is self-explanatory. the directory along with whatever was supposed to be in it is clearly missing. it might be in another package or you might simply have to create it yourself. (i doubt there's much to it so the difference may matter little) | 21:33 |
gnarface | Guest-12345678: maybe try apt-file search | 21:33 |
gnarface | assuming they fixed apt-file finally. that might be a stretch too | 21:34 |
Tenkawa | gnarface: ouch did apt-file get broke? | 21:35 |
Guest-12345678 | I get the error msg...I was rather hoping copy/paste to test runit with given example would work | 21:35 |
gnarface | Tenkawa: it was broke in unstable from like 2013-2018 and i never checked again after that | 21:36 |
gnarface | Guest-12345678: did the example include example files? | 21:37 |
Tenkawa | gnarface: just unstable? because its working in chimaera just fine. | 21:37 |
gnarface | Tenkawa: good to hear | 21:37 |
Tenkawa | I rely on it a lot | 21:38 |
gnarface | Guest-12345678: i don't see it included with the cron package i'm using here but there's more than one cron | 21:38 |
gnarface | Tenkawa: can you do a quick search for /etc/sv/cron? | 21:38 |
Tenkawa | not in there | 21:39 |
Tenkawa | let me modify the search | 21:39 |
gnarface | Tenkawa: i did check in stable frequently, initially... whatever was stable back then, i forget exactly when i stopped checking it | 21:39 |
Guest-12345678 | I did ls previously ....In any case I made the assumption when installing a pkg the runit files would be available | 21:40 |
Tenkawa | this is the only package with anything close | 21:41 |
Tenkawa | apt-file search sv | grep cron | grep etc | 21:41 |
Tenkawa | bcron: /etc/sv/bcron.... | 21:41 |
gnarface | Guest-12345678: yea, by the packages including other files in that directory it seems that the cron package should have supplied something itself, but even if upstream killed it off like a mafia informant, i assume it should be something fairly simple to recreate or copy from an older version | 21:41 |
Guest-12345678 | yeah tbh I am not sure if I have the patience to hunt down all the runit files for my particular setup | 21:43 |
gnarface | the sysvinit support is much more complete | 21:45 |
gnarface | even still, some init scripts have gone missing under suspicious circumstances recently ... | 21:45 |
Tenkawa | gnarface: has there ever been any discussion of looking at using s6? | 21:46 |
gnarface | Tenkawa: not by me, but it is also present in the repos and has a small following. i couldn't say how much more or less complete it is than runit right now. i think openrc is the one that's closest to sysvinit in completion but that might be because it's set up to use the sysvinit scripts | 21:47 |
Tenkawa | nod.. personally I like sysvinit and am happy with it however I know some want a smaller footprint | 21:48 |
Guest-12345678 | Its kind of frustrating because I signed up to the mailing list and searched the web to get more info on devuan / runit to see how polished it is...unfortunately information is severely lacking...In any case I have learnt my leason | 21:48 |
gnarface | Guest-12345678: sorry it sucks, but i'm sure you can see this is a sort of self-perpetuating mindset | 21:49 |
gnarface | Guest-12345678: nonetheless, we all sympathize with just wanting to get work done. nobody will disparage you for giving up and installing sysvinit | 21:49 |
gnarface | Guest-12345678: (i dont' think it should require a full reinstall, i think you can just install the packages) | 21:50 |
gnarface | Guest-12345678: if you hang out here longer maybe someone can give you a tip on where to get those files more easily. for all i know they're still in the ascii or beowulf packges... | 21:51 |
Guest-12345678 | tbh I really should know better but at least I tried...right now I am too frustrated...will calm down :) and try sysvinit tomorrow | 21:51 |
gnarface | sysvinit isn't perfect either but it's better known around here so more people can help you readily | 21:53 |
Tenkawa | its far from perfect indeed... however as you said.. been around forever and very well used. | 21:55 |
Guest-12345678 | anyway thxs guys...wish you a good night (already 10pm in Germany) o/ | 22:01 |
Tenkawa | cheers | 22:02 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!