26[00:22:03] <EdePopede> i tried to use something that looked
like alsamixer for that modern sound system for some time because
this gtk tool just sucks.
27[00:22:43] <somiaj> EdePopede: it shouldn't. You can
talk to pulse with different clients.
28[00:22:46] <EdePopede> now when i unplug and replag the
loudspeakers i don't have to switch manually from digital to
analog like i did in the past, i have to kill the sound server
instead.
29[00:22:51] <somiaj> there are clit tools for pulse (if you
are talking about pulse)
30[00:23:00] <EdePopede> right, pule
31[00:23:02] <EdePopede> pulse
32[00:23:27] <EdePopede> but forget about their cli thing,
it's unusable for non-technicians
33[00:24:26] <EdePopede> and i can't even find any trace
of that tool. either i installed it with apt, so i should have
something in the logs and a deb. or downloaded, so it should be in
/usr/local/bin. nada.
34[00:25:22] <EdePopede> lol, found it. in the download area of
course^^
50[00:33:57] <n4dir> not saying it makes sense, but i would
also look if something is to be found in alsamixer
51[00:34:15] <EdePopede> somiaj: a cookie file, 3 blobs, then 2
more: $HASH-default-{source,sink}
52[00:34:25] <n4dir> else there is #lau and
#opensourcemusicians.
53[00:34:35] <EdePopede> 1 line each, that's
alsa_output.pci-0000_00_14.2.analog-stereo and
alsa_input.pci-0000_00_14.2.analog-stereo
54[00:34:47] <somiaj> EdePopede: you could move all that stuff
out of the way, and see if pulse restets back to default settings.
Might fix the issues.
55[00:35:11] <EdePopede> right, could have thought of it
already. usually it's just what i'm doing
56[00:35:30] <somiaj> since some app could have modified files
in there if you think the app you ran caused this new issue
57[00:36:07] <EdePopede> odd, right now it is on Line Out
(unplugged), has the same entry for Headphones. nothing digital,
first time ever afaik Oo
66[00:40:18] <EdePopede> nice. back to normal it seems. killed
the server, started the controller, had only digital. plugged out
and in, analog was back. even with (plugged in)
67[00:40:54] <EdePopede> n4dir: i tried to get through the
architecture, got stuck every time.
68[00:41:21] <n4dir> tell me about it. Once sound goes south it
gets very confusing. Glad you got it sorted
81[00:42:48] <somiaj> there was another few generations of
sound servers before jack/pulse, gnome use to have its own one,
esound or something like that.
83[00:43:04] <EdePopede> knew a guy who built his own mini
moog. even this....
84[00:43:04] <somiaj> kde had their won sound server.
85[00:43:19] <n4dir> EdePopede: Say what ? How cool is that
86[00:43:39] <somiaj> though mostly sound servers are used due
to the fact that cheap hardware is cheap and doens't have
hardware mixing, though the switching sound streams on the fly, and
better per application control of pulse is things you can't
easily do at the hardware level.
87[00:43:41] <n4dir> i found a how-to build a mini-moog with
vcvrack, but haven't dared yet
88[00:43:53] <EdePopede> n4dir: indeed. he was proud of it and
had a reason to be :)
89[00:44:55] <EdePopede> i'd really like someone with A
LOT of time and enough knowledge of the topic to write a long paper
on the whole topic
90[00:45:26] <EdePopede> are sound cards even popular these
days?
101[00:46:46] <EdePopede> i remember i tried to understand the
whole codec stuff in win3.1 already. which of course came with a
book of 800 pages :P
102[00:47:00] <somiaj> EdePopede: problem is dmix is a poor
software mixer, and most sound cards can't mix, but I agree,
for most alsa and its dmix should be enough.
103[00:47:26] <EdePopede> tbh the reason why i HAD TO get to PA
was skype. and i didn't want to drop it back then
106[00:48:34] <EdePopede> what i don't get with alsa how it
handles different hardware combos. i don't implant new things
on a regular base, still i had no idea what some of the entries were
meant to be
108[00:49:38] <EdePopede> one thing i really liked tho: plugging
in the headphones didn't mute the speakers, so it has to be a
software thing in PA i guess
127[01:08:34] <edufmass> Hello, I don't speak english very
well hehe, I've read a lot about partitions and there are many
recommendations. I have one ssd 240gb m2 disc and one hdd sata. I
would like to have some virtual pc (with kvm) for php dev, another
java dev and win c# dev. which partitions are recommended? which one
in ssd and hdd ?
145[01:15:47] <coruja> partitions depend on your needs and
customs, i'm fine with one partition for each os, one /home,
and some more for data, audio, video which i mount into $HOME
146[01:15:49] *** Quits: phunyguy (~blaahchm@replaced-ip) (Remote host closed the connection)
174[01:24:56] <ratrace> EdePopede: heads up, no more console
scrollback in buster :/
175[01:25:54] <EdePopede> i didn't really like it anyway.
no chance to switch to another tty and keep the scrollback. not to
talk about the lack of erm... what C-o thing in midnight commander.
alternate screen or w/e.
176[01:26:41] <EdePopede> i was just surprised to see that i
could use mpv in it while running X :o
178[01:29:12] <sorcerer> so i normally been ubuntu never really
did debian head on, got it installer on an older laptop to play with
now though
179[01:29:37] <sorcerer> anything i should know coming over from
ubuntu?
180[01:29:43] <sorcerer> tips/tricks?
181[01:29:55] <EdePopede> forget about PPA and the like
182[01:30:17] <EdePopede> if you need newer things, build or
look for backports first.
183[01:31:12] <sorcerer> i see it doesnt come with sudo setup
for the first user like ubuntu
184[01:31:21] <EdePopede> !sudo
185[01:31:21] <sorcerer> no problem, it was just weird when it
wouldnt let me sudo
186[01:31:21] <dpkg> [sudo] (Substitute User DO) Not enabled by
default on Debian systems; configure /etc/sudoers first, ask me
about <visudo>, <sudoers>. sudo can give limited super
user privileges to specific users, or allow you to do silly things
like run X apps with root permissions, or good in scripts with
"username ALL = NOPASSWD: /some/program". See also
<sudo path>.
replaced-url
187[01:31:24] <sorcerer> lol
188[01:31:48] <sorcerer> i wasnt expecting it
189[01:32:07] <EdePopede> oh, does #ubuntu have such a helpful
bot?
190[01:32:30] <sorcerer> if you trigger it right
191[01:32:41] <sorcerer> doesnt trigger just from one word like
that did
209[01:34:55] <EdePopede> srsly! 2GB RAM with a swap partition
on a 80GB barracuda. and a GT630 graphics card, can you imagine how
faaaaaaaaaast this system is?!
221[01:36:36] <ratrace> sorcerer: 68.12 is, but the next esr,
78.3 is coming
222[01:36:45] <sorcerer> oh ok
223[01:37:11] <ryouma> are there settings one can do on debian
that make it more user-responsive in times like when the system
takes minutes or hours to respond to mouse movement or keyboard?
224[01:37:11] <sorcerer> whats the esr mean? i never seen esr
before, or never noticed it (keep in mind this is my first actual
debian install lol)
225[01:37:14] <EdePopede> from 68 to 78? did they drop their +7
scheme?
226[01:37:55] <ratrace> well they do have money issues..., and
now big chunk of the team is gone
227[01:37:57] <ryouma> (please assume "add more
memory" and "purge firefox-esr" and "don't
run any firefox extensions ever" are not in this case options
for my above question.)
228[01:37:59] <sorcerer> i need some more usb flash drives
229[01:37:59] <EdePopede> sorcerer: extended support release or
so
230[01:38:21] <sorcerer> okay
231[01:38:24] <sorcerer> question
232[01:38:26] <sorcerer> its booking now
233[01:38:29] <sorcerer> but
234[01:38:35] <sorcerer> how do i do firmware stuff
235[01:38:43] <sorcerer> it says some stuff is being disabled
cause of it
236[01:38:43] <ryouma> sorcerer: suggestions for those are
welcome if you are willing to pm me
237[01:39:03] <ryouma> (referring to --- 16:37 <sorcerer>
i need some more usb flash drives)
238[01:39:23] <EdePopede> ryouma: in my case it is usually
firefox (RAM) or copying tons of files (or just really big files)
(disk i/o, maybe also in team with swapping at the same time, see
the firefox mentioning)
240[01:40:16] <EdePopede> unfortunately it uses around 30% soon
after start already, even with noscript :/
241[01:40:39] *** Quits: luuuciano_ (~luuuciano@replaced-ip) (Remote host closed the connection)
242[01:40:57] <ryouma> EdePopede: yeah i think it was firefox,
downloading a big download, plus just you know firefox, and swapping
to a swap file (which arch wiki says is as performant as partition)
on the same hd. and i had 100% swap used and 89% memory used or so.
but in such times, i'd like much more priority given to
keyboard and mouse and display. can this be done?
245[01:41:36] <genr8_> try using ZRAM for memory and swap
compression
246[01:42:02] <EdePopede> ouch, that's a lot. right now i
have 1.18G/1.96G on RAM and 1.70G/5.00G on Swap
247[01:42:05] <ryouma> oom killer went on a spree killing
irqbalance first and then emacs (i'd have preferred firefox,
i.e. whatever was using and requesting memory)
248[01:42:41] <EdePopede> i was thinking already to define a
hotkey to start something that kills firefox in that case
249[01:42:42] <ryouma> genr8_: is that an odyssey to set up?
253[01:43:51] <ryouma> can oom killer be set to kill what is (2)
alraedy using, and (2) freuently requesting memory? can kernel
priorities be set to favor user actions?
255[01:45:06] <ratrace> the upcoming systemd-oomd will improve
that situation
256[01:46:13] <EdePopede> i hope with Rust the situation will
get better. as what i've read one of its advantages is you
don't leak memory all the time.
257[01:47:33] <EdePopede> maybe there will be a linux release of
these one day :)
replaced-url
258[01:47:37] <ryouma> firefox will use rust?
259[01:47:48] <ryouma> upcoming meaning soon to be in stable?
263[01:49:43] <ryouma> ok so upcoing to be in testing
264[01:49:52] <EdePopede> ryouma, this looks like a starting
point
replaced-url
265[01:50:52] <ratrace> ryouma: with systemd v247 I think
266[01:52:15] <ryouma> i am systemd-ignorant but will just hope.
what is very strange is that everything perfoprmed smoothly on my
previous machine with only 4gb memory with 4gb swap partition; this
one has 6 and 1 swap file and is much worse.
267[01:52:41] <ryouma> swappiness makes no difference whtehr 60
or 1 on this machine while on the old machine 1 was really good
268[01:52:43] <EdePopede> swap *file*?
269[01:53:00] <ryouma> yes, arch wiki says it is just as
performant
270[01:53:30] <EdePopede> never used one i think. could test it
on the other one... some day :)
275[01:57:57] <Mazhive> guys i am having dificulties to get my
pxe to work again i had a unexpected rebbot whil configuring , at
had bein working for days my setup is router dhcp and a pxe server
diffrent then the dhcp
294[02:15:27] <dpkg> Unofficial <live> system and
<netinst> and DVD images containing non-free Debian
<firmware> packages are available from
replaced-url
421[03:28:06] <RadoS> ratrace, on system A "quiet"
makes disappear both kernel messages and service startup, while on
system B it affects only the kernel messages but keeps service
startup showing.
422[03:29:01] *** debhelper sets mode: +l 1083
423[03:29:17] <RadoS> Since the OS install is identical, it must
be something hardware related...
424[03:30:49] <RadoS> On A I had to change the default video
driver from glamor to intel, because of glamor didn't funtion
performantly.
443[03:58:25] *** Quits: Immanuel (~Manu@replaced-ip) (Ping timeout: 265 seconds)
444[04:00:08] *** Joins: Immanuel (~Manu@replaced-ip)
445[04:00:26] <somiaj> jim: kde doesn't come with its own
compositor? xcompmgr is the one I know of, but looks comption is a
fork which has some additional features (that is what I've used
in fvwm)
450[04:06:55] *** Quits: Tom01 (~tom@replaced-ip) (Remote host closed the connection)
451[04:07:52] <jim> somiaj, actually I'm given to
understand that the window manager kwin -is- the compositor (which
kinda means I dunno what a compositor is, so my intent is to try ed
effects in the DE)
452[04:08:41] <jim> somiaj, need to pause for about 30 mins
453[04:09:18] <somiaj> yea some window managers are compositors,
others are not, xcompmangr/comption are for window managers that
aren't
459[04:19:15] *** Joins: metro (~metro@replaced-ip)
460[04:20:01] <Mazhive> it seems there is no answer yet i had it
working very strange... i am getting pretty frustrating... now
461[04:23:05] <somiaj> Mazhive: ghose here with experience with
preseed files are a bit low, and I agree some documtation is a bit
sparse on them too. Don't they produce a log? That is where I
would start debugging.
462[04:23:14] <Mazhive> i have search the net like crazy and i
am getting a headache why it worked and now it is broken
463[04:24:24] <Mazhive> the preseeed works great but i had some
body shutdown the computer while configuring the pxe server and it
doesn't work any more
465[04:24:34] <somiaj> are you using a different install image
since before it worked and now that it doesn't work?
466[04:25:05] <somiaj> So there is an issue with the pxe server
after this?
467[04:25:27] <somiaj> again this is out of my
expreience/knowledge, so not much real help I can do except maybe
talk about it to see if something comes up you didn't think of
yet
468[04:25:34] <Mazhive> let me explain my pxe server deploys
images that pxe serve does not work any more what ever settings i
throw on it it just does not work
469[04:26:40] <Mazhive> i am pulling my hair out at the moment
and getting pretty bald right now
470[04:27:01] <somiaj> And you said the pxe server was shutdown
while you were modifing it?
471[04:27:37] <Mazhive> yes so i had to revert to a an old file
which ofcourse was a previous one not working
473[04:27:50] <somiaj> is the pxe server some service, have you
looked at the logs for this service? Checked that it opened up a
port (or some method to talk to) on the network?
475[04:29:53] <Mazhive> i cant remember pxe server has logs it
hands over the services to tftp-hpa after initial dhcp requests
476[04:29:57] <Lady_Aleena> Hello. What do I run to find out if
my system is picking up the microphone's connection?
477[04:31:00] <Mazhive> and in between that its not working
478[04:31:15] <Lady_Aleena> I ran lshw, but no microphone
listed.
479[04:31:19] <somiaj> Lady_Aleena: do you have pavucontrol ran
480[04:31:25] *** Quits: ChiLLabiS (~jimmy@replaced-ip) (Disconnected by services)
481[04:31:31] <Mazhive> pxe client gets ip but stops after that
482[04:31:32] <somiaj> Lady_Aleena: what type of mircophone is
it? Is it a usb microphone or attached to your sound card?
483[04:31:48] <Lady_Aleena> somiaj, yes, it doesn't have it
on the input list. It is attached to my sound card.
484[04:32:09] <Mazhive> it seems the handling of who has to do
what gets lost
485[04:32:13] <Lady_Aleena> somiaj, and it is a cheap microphone
that came with one of my old old old computers a long time ago.
486[04:32:27] <somiaj> Mazhive: Sorry can't be of much more
help, maybe someoen with experience with pxe experience can help.
Checking back during the week (monday) may also get more users with
said exprience.
496[04:37:02] <somiaj> seems to suggest if you add load-module
module-alsa-source device=hw:2,x, where x is 0-4 (I'm seeing
four different devices there, I would assume 0, but unsure which
audigy device you plugged it into)
497[04:37:40] <somiaj> but that seems to be suggestions to make
pavucontrol (pulse) see the mic if it doen't
498[04:39:50] <jmcnaught> "pacmd list-sources" should
say if pulseaudio recognizes the mic is plugged in (under ports it
will say available: yes)
505[04:49:29] <jmcnaught> Lady_Aleena: weird on mine the
built-in card does. Are you also using pavucontrol? What do you have
the Audigy set to on the Configuration tab there?
506[04:49:30] <Lady_Aleena> Should I use `pulseaudio -k ;
pulseaudio -D` to restart pulseaudio or something with `systemctl
restart ...`?
507[04:50:04] <jmcnaught> I use "systemctl --user restart
pulseaudio" (as user, not as root)
529[05:01:53] *** Lord_of_Life_ is now known as Lord_of_Life
530[05:01:55] <ryouma> i noticed there is an xbindkeys amybe it
can do it?
531[05:01:58] <Lady_Aleena> somiaj and jmcnaught, thanks for
your help. I still can't get my mic to work, but it is probably
tied to all my other audio problems that have followed me since
wheezy.
546[05:05:15] *** Quits: metro (~metro@replaced-ip) (Quit: Leaving)
547[05:05:39] <jmcnaught> Lady_Aleena: well now it says
"available: unknown" under ports, maybe your sound card
doesn't to plug detection
548[05:06:12] <Lady_Aleena> jmcnaught, probably not, the card is
ancient.
549[05:09:21] <Lady_Aleena> jmcnaught, I have so much that
doesn't work with my sound system. I have a speaker that
won't play, I can't turn the volume below 20%, controlling
the volume with my keyboard volume keys locks the button on my
taskbar for several minutes, and now the mic won't work.
551[05:10:17] <ryouma> somiaj: thank you. is there any
intelligence on what mod1...modn do? i get that mod1 is alt, is all.
i think i want to assign ctrl alt f12
552[05:11:04] <Lady_Aleena> But XFCE doesn't come with its
own sound controller like KDE did, so I have to make do with what
little sound control I have with the very lackluster pavucontrol.
553[05:13:12] <ryouma> hmm emacs captures some of these
keystrokes... i have to choose something it does not capture?
554[05:13:39] <nvz> triggerhappy is much lower level, as I said
already
555[05:13:56] <Lady_Aleena> Maybe when the next version of
Debian comes out, I will pull the current hard drive out and install
a fresh Debian on a new hard drive.
557[05:14:46] <nvz> in fact if you need something more global
and lower level than triggerhappy you're probably going to have
to go directly into the kernel itself
558[05:14:53] *** rf-n00b_ is now known as rf-n00b
559[05:15:56] <jim> Lady_Aleena, ltns... how are you?
560[05:18:14] <Lady_Aleena> jim, hello. I haven't been
using the command line for much lately, so I've stayed out of
Linux and Debian. Right now I am frustrated/annoyed with my system,
but when am I NOT frustrated/annoyed with my system? How are you?
562[05:19:47] <jim> taking a somewhat easy piano class, and a
-hard- english class...
563[05:20:17] <Lady_Aleena> jim, shall we go to offtopic?
564[05:20:19] <ryouma> nvz: does it need a daemon? there is no
man page, so i am just looking at th3e readme and it is not clear
what to do to assign e.g. ctrl shift f12 to a command
565[05:20:33] <jim> sure
566[05:20:35] <nvz> ryouma: its thd is the executable, and yes
its a daemon
567[05:20:50] <nvz> ryouma: look in /usr/share/doc/ for example
configs I believe are included
568[05:24:04] <ryouma> the 1 and 2 stuff in the examples seem to
say that this controls what type of keypress it is. but when you use
+ to combine them, you still only have one of those. so it is not
clear what one would do with one modifier key and one function key.
569[05:24:23] <ryouma> i read the readme and the examples
613[06:00:59] <nvz> if whatever you specify in the first field
is held, its 2
614[06:01:39] <ryouma> i could do 1 then but i have to hold for
what3ever i have set for xset r?
615[06:01:39] <nvz> yes if you want it to happen on press
616[06:01:48] <ryouma> 2*
617[06:02:12] <nvz> !tias
618[06:02:12] <dpkg> TIAS is "Try It And See".
619[06:03:07] <ryouma> ok so sounds like i have to create a
file, souce it using an option, and run the daemon, and ... nohup
it? and then KEY_LEFTSHIFT+KEY_F12 2 ado kill-firefox in the file?
620[06:03:48] <ryouma> maybe i can do some kind of <() deal
on the command line instead of a file to not keep everything in one
place
621[06:03:59] *** Quits: leorat (~leorat@replaced-ip) (Remote host closed the connection)
632[06:06:32] <nvz> might wanna learn at least the basics
633[06:06:44] <ryouma> so now it is even more confusing. the
daemon is started upon installation, but i didn't specify the
config file for it to read from
634[06:06:51] <ryouma> which it says is necessary (or so it
seems)
635[06:07:09] <nvz> yes, well thats why you might want to
restart the service or look at the status
636[06:07:15] <ryouma> oh ok in etc
637[06:07:25] <ryouma> which is reasonable
638[06:07:37] <nvz> its a GLOBAL hotkey daemon
639[06:07:43] <nvz> it works outside X or any user login
640[06:08:35] <ryouma> well here goes nothing i will try that
line there
641[06:09:59] *** Quits: b (coffee@replaced-ip) (Quit: Lost terminal)
651[06:23:19] <c_nix> netinst of testing cannot locate iso
652[06:23:26] <c_nix> i see its a recurring error
653[06:23:30] <c_nix> over the years
654[06:23:31] <somiaj> ryouma: look at xmodmap, they can be
configured to be any key you want, but default alt(meta), shift and
ctrl are not mod1-5, though one of them is the windows key by
default
655[06:23:40] <c_nix> wonder why
656[06:23:48] <ryouma> nvz: i tried putting
"KEY_LEFTSHIFT+KEY_F12 2 /root/sh/ado kill-firefox" in
triggers.d/kill-firefox, then in case it helps"systemctl
restart triggerhappy", then had that ado script write to
/root/triggerhappy to show the result, then held down shift f12 for
a few seconds. the file was not created.
657[06:23:52] <somiaj> c_nix: how do you copy the iso to your
usb drive?
658[06:23:57] <c_nix> dd
659[06:24:01] <c_nix> bs=1M
660[06:24:07] <c_nix> too big?
661[06:24:31] <somiaj> I just use cp, it has nicer defaults, but
that shouldn't be an issue. That is a common error that one
gets when they use tools like unetbootin
663[06:24:57] <somiaj> anyways two comments 1) testing support
is in #debian-next on irc.oftc.net, and 2) testing is usually best
installed using the stable installer then upgrading
664[06:25:00] <somiaj> !testing installer
665[06:25:01] <dpkg> Repeat after me: the testing installer is
for testing the <installer>, not for installing
<testing>. To install testing, do a minimal installation using
the <stable> installer and ask me about
<stable->testing>. Installer bugs should be filed against
the debian-installer pseudopackage.
replaced-url
666[06:25:12] <c_nix> so devs failed to make proper img?
667[06:25:39] <nvz> ryouma: and what do you suppose the
permissions are on /root/ ?
668[06:25:40] <somiaj> could be, the testing installer is know
to be buggy, though it isn't as buggy as it use to be, it has
never been the suggested way to install testing
680[06:27:43] <ryouma> you are fond of socratic dialog :/
681[06:28:18] <nvz> ryouma: since you seem to lack the basic
knowledge I had given you credit for, lemme let you in on this bit
of info.. typically dameons like this that start via systemd have a
file to override their defaults.. in /etc/default/ you may want to
have a look at /etc/default/triggerhappy
682[06:28:21] <c_nix> i dunno how to answer it in more details
683[06:28:56] <ryouma> nvz: doesn't seem to work for /tmp
either
696[06:31:43] <somiaj> it might just be bad testing images, but
the install guide recommends cp as it has nicer defauts, but this
shouldn't be an issue. Also make sure you run sync when done
697[06:31:46] <nvz> ryouma: in fact, think of anything you can
do to someone, and try do that to nobody and see how that works out
for you
698[06:32:00] <nvz> its called nobody for a reason
699[06:32:50] <nvz> ryouma: take my advice, have a look at
/etc/default/triggerhappy
700[06:33:31] <nvz> ryouma: if you want to do something that
requires permissions of /somebody/ then dont run it as _nobody_
701[06:33:41] <somiaj> nvz: they really don't need that if
running fluxbox, and probably should just use fluxbox to make
keybindings
702[06:34:21] <somiaj> ryouma: anyways, I would just use
fluxbox's key bindings, no need for a thrid party app just for
keybindings
703[06:34:44] <nvz> somiaj: idk, all I know is I suggested a
hotkey system I know to be very flexible and that supercedes most
everything.. wms, xsessions, userlogins..
704[06:34:59] <nvz> they seemed to indicate that hooking x
layers was not sufficing for their needs
705[06:35:33] <somiaj> Might be nice if you wanted something
that is independent of the wm, but in most cases wm bindings give
you access to the wm features more directly
706[06:35:38] <ryouma> i changed tghe /etc/default settings to
retain root and it still doesnot work
707[06:35:54] <nvz> ryouma: did you restart the daemon?
708[06:36:11] <somiaj> independent: don't message people
links.
716[06:40:59] <ryouma> somiaj: i prefer something that will work
when the system is overloaded and keys are almost not working at
all, and i like the idea of working in the console also in case i am
there and not in x, and also if i switch wm some day, and i like the
idea of holdingit down for a bit before it will work (for safety).
BUT i will use whatever works at this point. i decided to forget
about the to-me-inscrutable modn syntax
717[06:40:59] <ryouma> and use shift instead, so shift f12 will
work (dangerously). i will possibly kill firefox by accident as a
result, but it is a start. if i figure out which mod is control, i
will add that! thank you for the polite and useful set of tips.
718[06:41:26] <ryouma> nvz: the status does not include logs. it
does have the error message i reported above.
719[06:42:27] <ryouma> namely Device /dev/input/event5 not
suitable.
727[06:50:29] <nvz> I don't know enough about systemd to
tell it to regenerate from the init.d script, but I do know you can
manually edit the service file
728[06:51:02] <ryouma> so it is a bug?
729[06:51:09] <nvz> ryouma: you can edit
/lib/systemd/system/triggerhappy.service and do systemctl
daemon-reload and systemctl restart triggerhappy
730[06:51:24] <nvz> ryouma: might be, not sure.. I'd file
it against triggerhappy either way if you care to, yes..
731[06:51:33] <ryouma> and then upon a dist-upgrade i would need
to edit again?
732[06:51:38] <c_nix> just avoid systemd - problem solved
733[06:51:39] <ryouma> i think i will pass if so
734[06:52:02] <nvz> ryouma: however I tested writing to /pressed
after editing the service file and verifying the daemon was actually
running as root, and it worked
735[06:52:13] <nvz> ryouma: I simply used touch /pressed
737[06:52:41] <nvz> ryouma: the key still had the effect of the
focused app in X, but triggerhappy got it first and did what it was
supposed to
738[06:53:07] <nvz> ryouma: i.e. the key still did what it
normally would with whatever was on the screen, but triggerhappy did
its part
739[06:54:39] <ryouma> somiaj: control shift might work; i will
try that at some point when i am not exhausted. again thank you.
740[06:56:51] <jmcnaught> Shouldn't be editing files in
/lib/systemd/ they are owned by packages and will be overwritten
during upgrades. Files in /etc/systemd/ take priority over files in
/lib/systemd/. There's also "systemctl edit
foo.service" which opens an editor and saves the file as
/etc/systemd/system/foo.service.d/override.conf ("man
systemd.unit")
741[06:58:28] <ryouma> jmcnaught: thanks. too exhausted to try,
but have made note that maybe i can edit to get it to work.
742[06:58:30] <jmcnaught> that override.conf file is a
"drop-in", you can use it to override just certain
directives instead of replacing the whole file from /lib/systemd
743[06:58:48] <somiaj> (the only thing I wish systemctl edit did
was show the default options so you knew which ones to change -- end
up having to read the service in another terminal as well, but the
override idea is nice)
744[06:59:08] <somiaj> maybe have them all commented out so you
only uncomment what you want to override
745[06:59:50] <jmcnaught> yeah I think it would be more
intuitive if it opened the unit file in the editor. Saving that
(edited) file in /etc/systemd/system/foo.service.d/bar.conf would
still work I'm pretty sure.
746[07:00:46] <somiaj> like only save a diff (though that might
be limiting). I understand the power of the overrides, allow system
updates to take effect if they aren't overriden, but I find it
confusing to use in practice.
747[07:01:10] <somiaj> hence why I was thinking comments, the
editor could strip them out before saving the file.
748[07:02:13] <jmcnaught> Well you can also use drop-in
overrides for systemd conf files like logind.conf, but I just copy
the original default logind.conf to
/etc/systemd/logind.conf.d/local.conf and edit it there.
749[07:02:30] <somiaj> jmcnaught: the only reason they
don't save the whole file is it allows system updates to take
effect if only some small part was override (my understanding)
750[07:03:14] <ryouma> so it's in sections or something?
751[07:03:21] *** Quits: Ericounet (~Eric@replaced-ip) (Remote host closed the connection)
754[07:04:10] <jmcnaught> So that your local override drop-in is
not blanket overriding everything including options you didn't
intend to be different from default that have changed in the new
version.
755[07:04:14] <somiaj> ryouma: basically you can only override
part of the unit file instead of have to matain your own full unit
file. It is a useful idea, just a pain to actually use without
knowing systemd inside and out (for knowing what overrides to write
out of the box)
756[07:04:48] <somiaj> not having a copy of the unit file while
you are editing just means you need another terminal for most users.
760[07:06:37] <nvz> afaik other daemons respect /etc/default/
when run from systemd #971064 fwiw
761[07:07:00] <nvz> judd won't be aware of it for awhile,
but thats the number in my configurmation email
762[07:07:44] <nvz> seems like this just wasnt fully updated to
work when ran from systemd as it just sorces that
/etc/default/triggerhappy from the init.d script which is
superceeded by the service file
763[07:12:46] <jmcnaught> looks like the triggerhappy.service
unit file is missing EnvironmentFile=-/etc/default/triggerhappy
under the [Service] section, which could of course be put into a
drop-in.
773[07:47:48] <ryouma> i am still unclear on what /etc/default
does (at least, why it isn't just something in
/etc/whateverprogram). i understood it at one point, i think.
774[07:48:20] <ryouma> also, i noticed that there is a magic
sysrq facility that is enabled by default on debian there. i wonder
if that can be useful in low memory cases?
775[07:48:43] <ryouma> the url that it points to is a 404
however which is not encouraging:
replaced-url
776[07:48:49] <nvz> the files in /etc/default are suppose to be
global defaults for things that are started automatically by the
system
777[07:49:01] *** debhelper sets mode: +l 1083
778[07:49:30] <ryouma> why not just have
/etc/triggerhappy/global-default-for-daemon.txt?
779[07:50:07] <nvz> jmcnaught: I wasnt sure how something like
that could be achieved in systemd, if you could append the bug
report I filed.. if not I'll try remember to test it and look
into it later
780[07:50:22] <nvz> #971064
781[07:50:46] <nvz> apparently judd still isnt aware of it yet
782[07:51:33] <nvz> its been 45min now .. idk what the deal is
with judd not seeing newly filed bugs
783[07:51:40] <somiaj> files in /etc/default were kinda a hack
to put configurable stuff for sysv init.d scripts without having to
edit the whole script. With systemd there really isn't a need
for it, just make a local overide to the unit file.
784[07:52:10] <somiaj> judd probably only updates the database
it uses once a day or so.
785[07:52:29] <nvz> scraping rather than looking up on demand
seems wasteful
786[07:53:07] <nvz> and the local override thing seems far more
complicated than the maintainer putting somehing in /etc/default
where people can easily find and modify it
787[07:53:36] <somiaj> only because that is what many older
users are use to, 'systemctl edit unit' and then put the
right info is quite simple
790[07:54:37] <somiaj> the only reason to really keep files in
/etc/default is if you want a way that works with both sysv and
systemd (or other init systems). Some packages will use this, but
not all.
791[07:54:40] <nvz> the nice thing about the stuff in
/etc/default is it often has a template and commentary for the kinds
of things you'd want to change
792[07:54:54] *** Quits: leorat (~leorat@replaced-ip) (Remote host closed the connection)
808[08:32:11] <bigfluff> forgive me debian, as i have sinned
809[08:33:02] <bigfluff> for some reason my clean install
process involves: 1) backing up files, 2) installing a new
distribution, 3) regretting #2 and re-installing debian
918[10:12:07] <Guest6262> Debian buster has a GUI-configuration
for wake on lan. Technically it should work if i check the magic-box
under "Ethernet" in "Editing
<connection-name>", right?
919[10:12:34] <Guest6262> (with etherwake on the remote machine)
925[10:19:51] <genr8_> your BIOS has to support it and have the
option enabled too
926[10:21:07] <Guest6262> genr8_ ok... i'm looking for a
manual for my mainboard right now. Are my machines on OSI Layer 2
when they are connected over a modem/router on the same subnet?
927[10:21:32] <genr8_> um
928[10:21:43] <genr8_> you mean the router is in between them ?
929[10:22:23] <genr8_> a router is layer 3, if thats what
you're asking.
930[10:22:36] <genr8_> a switch is layer 2
931[10:23:01] <flayer> if the cable is plugged in on both ends
they are connected
932[10:23:22] <Guest6262> genr8_ so if my modem has a built in
switch they are on OSI 2?
936[10:25:17] <Guest6262> hmm... Scince the only relateable
BIOS-Option (Power on by PCI Devices) is enabled i guess it's
the router that blocks the magic packets somehow
1045[11:50:01] <Jemt> Hi. OpenDKIM fails on reboot, but works
fine when restarting it using 'systemctl restart
opendkim'. /var/log/mail.err and mail.info contains no errors.
Can someone give me a couple of ideas on how to identify the cause
of the problem ?
1046[11:50:35] <Jemt> It "feels" like OpenDKIM is being
started too early - perhaps something is not ready yet
1052[12:00:44] <Jemt> Weird. I was able to resolve the problem
with OpenDKIM failing on boot by making it start after MariaDB. I
simply altered
'/etc/systemd/system/multi-user.target.wants/opendkim.service'
- i added "mariadb.service" to the After parameter:
After=network.target nss-lookup.target mariadb.service
1130[12:56:26] <reith> Hi. if I want install some script in my
package that is not intended to be executed directly by user, but
shipped services i put in my package, where should i install them?
the closest thing i find is to include them in examples which goes
to /usr/share/<pkg>/doc/examples but mine actually is not doc
related :)
1243[14:37:43] <RadoS> Just for comparison, can anyone see the
service startup messages on boot on console with "quiet"
boot param to drop the kernel messages?
1253[14:40:26] <RadoS> What would be the correct setup to drop
the kernel messages but keep service startup/systemd?
1254[14:41:36] <ratrace> RadoS: you sure you have that flag in
correct place? should be on the line linux /boot/vmlinuz/... ro
quiet in the grub menu entry, when you edit it with 'e'
1258[14:42:51] <RadoS> And it is there at boot time.
1259[14:43:26] <RadoS> When I removed it manually once for
testing with 'e', the messages appeared (along with
kerne).
1260[14:43:27] <ratrace> RadoS: you can silence only the kernel
messages with sysctl kernel.printk but I don't know if
that's effective within intramfs from the beginning
1261[14:43:44] <ratrace> RadoS: yes, removing "quiet"
would start showing kernel and systemd messages on boot
1262[14:43:57] <ratrace> having quiet and still seeing them
should not be possible
1263[14:43:58] <RadoS> Maybe I should file ticket.
1264[14:44:38] <RadoS> But I don't want to _lose_ that, but
have it "back" for the failing system. :-(
1265[14:45:08] <ratrace> not sure I understand. you want kernel +
systemd messages displayed?
1266[14:45:19] <RadoS> No, just systemd, not kernel.
1267[14:45:44] <RadoS> kernel is too noisy, scrolls too fast to
be useful to notice problems.
1268[14:46:31] <ratrace> then I think you can remove
"quiet" and adjust dmesg/printk log level with loglevel=N
kernel command line option
1271[14:47:27] <RadoS> ratrace, even if that's the
"right" way to achieve what I want, still it would be nice
to have all 3 systems behave consistently. ;-)
1272[14:47:32] *** Quits: banisterfiend (~textual@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1273[14:48:01] <ratrace> RadoS: and all three systems have
exactly the same grub menus + sysctl.conf ?
1274[14:48:18] <ratrace> exactly the same, save UUIDs of course,
I'm talking kernel options
1275[14:48:19] <RadoS> ratrace, yep, & thanks for the lin.
1276[14:48:53] <ratrace> RadoS: maybe you think you do but you
don't really? updated grub and kernel initramfs after changing
relevant files?
1277[14:49:04] <ratrace> compared /boot/grub/grub.cfg between
them ?
1278[14:49:39] <RadoS> Yes, after you suggested to me removing
the "quiet" for crosschecking.
1284[14:52:57] *** Quits: JohnML (~john1@replaced-ip) (Remote host closed the connection)
1285[14:54:49] <RadoS> all are 32-bit legacy boot, no software
difference except for hardware dependencies, which primarily is the
gpu: 1 ati OK, 1 Intel(R) 945GM OK, 1 Intel(R) GM45 Express
1298[15:00:06] <reith> i'm trying to use substitution
patterns in my packee's install file but neither of these
directives expand: var/lib/${PKGNAME}, var/lib/${env:PKGNAME},
var/lib/${DEB_HOST_MULTIARCH}. install have `#! /usr/bin/dh-exec`
and I exported PKGNAME in rules.. i just see files being installed
in /var/lib/${<whatever>} :\
1309[15:07:08] <RadoS> That's my problem, I don't know
how to isolate it...
1310[15:07:54] <RadoS> I don't know enough about the
interaction between kernel, gpu, systemd at boot.
1311[15:08:22] <RadoS> gpu being the only relevant difference as
far as I can tell.
1312[15:09:50] <ratrace> without that the filing of the bug
is.... much less effective
1313[15:11:24] <RadoS> I know, but maybe those maintain each
component have a better grasp of which part their component plays in
it.
1314[15:11:30] <RadoS> +ing
1315[15:11:42] <ratrace> sure, just saying if you can find a
reproducer, that'd be great
1316[15:12:04] <RadoS> That's what I hoped to find here, a
clue for it. :-)
1317[15:12:11] *** Quits: Night-Shade (~TimF@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1318[15:13:22] <ratrace> RadoS: so to be 100% sure what
you're saying, is that with "quiet" on the kernel
command line options, you still see systemd and/or kernel prints on
boot?
1319[15:13:40] <RadoS> No kernel, just systemd.
1320[15:14:15] <ratrace> RadoS: are those just errors or ALL of
them?
1321[15:14:20] <RadoS> ALL.
1322[15:14:49] <RadoS> You know the lines where it says
"service name: OK" where OK is in green.
1323[15:15:47] <ratrace> and if you add systemd.show_status=false
alongside quiet ?
1333[15:19:33] <RadoS> ratrace, hmm, that killed it. Will it make
sense to try on the "offending system" to explicitely set
it to true?
1334[15:20:29] * RadoS will do next time.
1335[15:20:35] <ratrace> RadoS: systemd.show_status is
"auto" by default, meaning it observes the existence of
"quiet", thus with those two it's possible to show
one (kernel or systemd) or both or none, on boot
1336[15:21:02] <cskr> rclone in Debian repositories is 7 versions
behind upstream. There is already a bug reported for this at
replaced-url
1339[15:21:36] <ratrace> RadoS: now you could look into your
initramfs and see if systemd is started with a --show-status=
override somehow, though I don't know which one has the
priority, if both commandline and kernel option are added... by some
logic, --show-status (explicit) should override implicit
1344[15:22:37] <oxek> cskr: unlikely to receive a new version in
debian-stable
1345[15:22:40] <ratrace> cskr: you could contact the maintainer
and ask them
1346[15:22:56] <ratrace> but yes, that's never gonna reach
buster, only bullseye
1347[15:23:12] <cskr> oxek: The version in unstable is 7 versions
behind upstream.
1348[15:23:17] *** Quits: earthundead (~earthunde@replaced-ip) (Remote host closed the connection)
1349[15:23:22] <ratrace> !debian-next
1350[15:23:22] <dpkg> #debian-next is the channel for
testing/unstable support on the OFTC network (irc.oftc.net), *not*
on freenode. If you get "Cannot join #debian-next (Channel is
invite only)." it means you did not read it's on
irc.oftc.net. See also
replaced-url
1351[15:23:31] <ratrace> ^^ also good place to ask about .....
next.
1358[15:25:17] <ratrace> RadoS: by looking through
/usr/share/initramfs-tools/{hooks,scripts} if any package added
something to override, but wait, I think there's a simpler way
to determine it, just a moment
1359[15:25:57] <cskr> The latest upstream version of rclone is
1.53.1, but the one in unstable is 1.50.2.
1360[15:26:36] <ratrace> cskr: we got that. your optoins are to
askn in #debian-next on OFTC, and/or contact the maintainer asking
how you can help with a version bump
1361[15:27:05] <cskr> ratrace: Thanks. The channel is
invite-only, how do I get invited?
1362[15:27:15] <dvs> cskr: on OFTC!!
1363[15:27:24] <ratrace> RadoS: is there an init= override in the
kernel command line?
1365[15:29:25] <cskr> dvs: I misread the bot message and thought
the channel is invite-only on OFTC.
1366[15:29:35] <ratrace> RadoS: well I don't know then.
short of dissecting the initramfs init script (which I'm not
willing to guide you through as it's time consuming), I guess
you could just file a bug report and at least state that you can
override that behavior with systemd.show_status kernel option
1367[15:30:02] <RadoS> ratrace, that already helped a lot, thank
you for your patience.
1411[15:58:22] <cskr> ratrace: I'm talking about
Debian's rclone repo. I didn't even ask. Either it was
just an awesome coincidence or the maintainer is here. The bug got
updated after about 3 months just when I happened to ask about it
here.
1535[17:54:08] <Mazhive> oke a fresh vm install just works out of
the box yet my configuration is not working i suspect i have a
running service that interferes with the dnsmasq not able to hand
over the path of the pxelinux.o
1571[18:18:53] *** Quits: fxp (~fxp@replaced-ip) (Remote host closed the connection)
1572[18:19:55] <Waxhead> Trying to figure out how to remove grub2
from a system with many disks. I am trying to debug a boot problem
and on the server I am using the BIOS is a bit nuts e.g. I
can't enter the setup screen and choose boot order. Therefore I
would first and foremost like to uninstall grub on all disks except
one before trying to restore my beloved mess to it's former
glory. So how do I remove grub (and not the partition table) ?
1598[18:34:20] <jelly> dd if=/dev/zero of=/dev/sdXXX bs=440
count=1 # UNTESTED, presumably GNU dd is smart enough to first read
the whole sector, and then put back a copy with new values in 440
bytes, and rest unchanged
1621[18:46:32] <EdePopede> live always includes nonfree?
1622[18:46:47] <jelly> velix, I don't think
packages.debian.org is behind a CDN, so something is interfering
with your http(s) access. If you need to grab a .deb manually you
can also try snapshot.d.o, eg.
replaced-url
1629[18:49:18] <dpkg> Unofficial <live> images - containing
non-free Debian <firmware> packages - for Debian 10
"Buster" as a live OS are available at
replaced-url
1630[18:49:53] <EdePopede> ah, just found it. on the jigdo page
there's an explicit link to nonfree+live. went up the tree from
there a bit.
1641[18:54:36] <velix> jelly: hehe, faketime uses
/usr/$LIB/faketime/libfaketime.so.1 in LD_preload ;)
1642[18:54:40] <jelly> faketime is also useful if you have a
Debian 6 system still installed, the only way to verify and apt-get
install packages without completely ignoring gpg sigs
1643[18:54:59] <velix> I'm curious what happens on 2039 :D
1664[19:07:52] <jelly> which particular file fails to exist?
1665[19:07:57] <EdePopede> i tracerouted some of the secondary
servers in the country before, the next i could identify from its
domain has values around 20ms also
1666[19:08:04] <EdePopede> 1st jigdo
1667[19:08:22] <EdePopede> but works with http, so probably it
has only a problem with the protocal
1707[19:27:13] <jelly> EdePopede, oh, if jigdo only tries plain
http and the server redirects you, breaking the download,
that's either a bug for
replaced-url
1727[19:32:56] <Mazhive> oke i need a debian specialist ....
concerning services or protocols
1728[19:33:10] <EdePopede> so i'd say it may be a protocol
issue
1729[19:33:22] <jelly> EdePopede, sounds like a clear-cut bug in
for jigdo package
1730[19:33:27] <transhumanist> hi is there any file system that
can run, even as a third party add in for windows that would allow
me to access rw to by both windows and linux that supports unix
permissions? I was looking at ext2 but unfortunately those windows
programs wont run on Windows 10, figured I would ask here to get
some more abscure ideas, I have exhausted all other avenues
1733[19:34:05] <EdePopede> jelly: a case for debbugs or oftc?
1734[19:34:53] <jelly> transhumanist, windows has much better and
sadly very different ACLs than Linux. Perhaps you can use a network
share like samba?
1751[19:39:37] <Casper26> Anyone had problem with firefox
speedtest being very slow? i upgraded to gig service and speedtest
using cli show 960 down in firefox only get 310 down subsequently
chrome gets 960 down?
1752[19:39:48] <transhumanist> jelly, do you know when the drive
shows up in windows explorer , do they mean windows explorer as a
disk?
1753[19:40:08] <transhumanist> 20 bucks seems like an ok price to
pay if it does
1754[19:40:25] <EdePopede> done. so roundabout half an hour for
ISO-1.
1755[19:40:45] <jelly> they also work differently, NTFS ACL have
consistent inheritance. unix permissions only inherit set[gu]id from
parent dir as a special case
1760[19:42:13] <transhumanist> for me as a development
environment using OpenNAS-ISCSI-Windows
Initiator-WSL2-Debian-Networking- ext4 disk mount, I think preseving
the basic permissions would be ok
1761[19:42:31] <transhumanist> the problem is the programs in
linux balk at lack of ability to set permissions
1771[19:44:40] <Casper26> but speedtest are accurate on
everything except firefox works in chrome?
1772[19:45:48] <jelly> transhumanist, used to be "for
Linux", and they're the same company that made FUSE-based
ntfs-3g, I can't find a better page for Linux than that now
1773[19:46:16] <transhumanist> thank you very much jellly, I
think because its just data transfer, that I should stick with EXT4
mount on windows and Linux native , so I will use the first one but
thanks for the second link anyway
1786[19:48:56] <transhumanist> I have had good luck with iscsi
and I know it well , so nfs would be a good adventure, but with what
you just said , its good to know
1787[19:49:02] <jelly> Casper26, maybe avoid reliance on browser
performance completely, and use speedtest-cli ?
1791[19:50:35] <Casper26> jelly cli works fine but since firefox
is my default browser i would like for it to work correctly chrome
works out of the box
1797[19:55:57] <sponix> Casper26: odd, because I have had the
speedtest-cli with the broken stats, and the browsers posting
accurate ones - you claim the opposite
1798[19:56:13] <Casper26> yup
1799[19:56:51] <Casper26> but like i said only happens in firefox
chrome works as expected but i dont like chrome
1800[19:57:00] <jhutchins> transhumanist: Once you have gigabit
networking, local hardware/performance has much more effect on
speeds. My old spinning-rust core2 duo gets nowhere near the speeds
my wife's newer SSD system gets.
1801[19:57:07] *** Quits: coruja (~coruja@replaced-ip) (Remote host closed the connection)
1886[20:43:51] <Mazhive> jezus.,... i got it ... it seems the
firewalld rles got corrupted and had to manually add the ports again
i figured it out because a fresh install worked out of the box which
does not use a firewall damn.... that toke me some time to figure
out ...
1887[20:44:43] <genr8_> What am I supposed to do when I
"cat" a Binary file to my xfce terminal, and it corrupts
all the text from then on...
replaced-url
1888[20:45:03] <joepublic> typing `reset` can help
1964[21:17:23] <somiaj> is nextcloud even in debian. That does
sound like a targeted fix issue that could be fixed in a point
release.
1965[21:18:01] <somiaj> alexbh: file a bug report with the debian
BTS about that issue, that does seem like a serious enough bug that
if a targeted fix can fix it, it might make it into a point release
(but only the fix, not a new version)
1966[21:18:08] <jmcnaught> Nextcloud itself is not in Debian but
the desktop client is, and some integration packages for caja, nemo
etc.
1970[21:19:45] <alexbh> lets say... I use 2.5 which is not
working... but there is a fix on 2.7 and 3.0 (bullseye).... so I
dont want to wait until the package 3 is released with bullseye
1971[21:20:02] <somiaj> alexbh: you can build your own backport
1977[21:21:25] <somiaj> debian helper is in backports already, so
that part should work, unsure about dh-sequence-sphinxdoc though,
but that might come with the debian helper from backports, so yea
looks like it is possible you could compile a newer version
1980[21:22:50] <somiaj> alexbh: there seems to be some bugs
reported about issues, but I can't seem to find one exactly
releated to that issue on github. Something like that might be fixed
in buster provided they can do a targeted fix.
2093[22:25:11] <merethan> Hi all. Some unnamed failure occured
buring my installation (latest netinst img) of preset package
collections (Desktop/Gnome, sshd, Utils). I could redo the step in
the installer menu. Any chance files are now duplicated on my disk?
2094[22:26:50] <merethan> Can I grep for an extension, or run a
script to check for disowned (not belonging to any package) files?
2095[22:27:05] *** Quits: fax (~quassel@replaced-ip) (Remote host closed the connection)
2098[22:27:43] <teo7> hi, i'm trying Debian stable, but
installing the drivers for my Nvidia GeForce 950m (with the command
"apt install nvidia-driver"), the screen sporadically does
graphical glitchs. These consist of flickering, until I move the
mouse abruptly. I tried to migrate from xorg to wayland, but the
latter is blocked by default when installing proprietary drivers.
how can i have a hybrid system running on debian stable? This guide,
for Debian 10
2099[22:27:43] <teo7> recommends using either the integrated GPU
or the dedicated one, but they don't seem optimal solutions to
me ->
replaced-url
2104[22:29:39] <somiaj> merethan: Nope, debian packages will just
overwrite the old files when a newer (or same) version is installed.
2105[22:30:34] <somiaj> merethan: dpkg -S path/to/file will let
you know what package owns a file, but finding file that are not
owned by any packages is a bit more work (as in many cases people
will have packages outside the debian system in /usr/local, /opt,
/home, etc
2106[22:31:00] <merethan> I can omit certain paths such as /opt
2107[22:31:13] <merethan> Install is fresh though, so none are
there.