10[00:02:41] <sney> plt: that page has instructions on it for
debian, click the 'apt-get' tab. debian does not
distribute this application, so stripe's repository is your
only option
15[00:05:02] <tom_> doing in by hand results in inet
123.123.123.123/32 scope global eth0 and doing it via iup results in
inet 123.123.123.123/17 brd 123.123.127.255 scope global eth0:1
16[00:05:18] <tom_> but it shows up as a scope global rather
than an interface?
17[00:05:28] *** Quits: voidSurfr (~todd_dsm@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
18[00:05:30] <tom_> actually you know what
19[00:05:46] <tom_> does interface aliasing even matter if
I'm not using an IPv4 gateway router?
20[00:06:12] <robi> hi. i have had the misfortune of getting a
computer with nvidia. i'm currently trying to build this:
replaced-url
21[00:06:55] <tom_> can i just assign as many private ipv4s to
a non-aliased interface as i want as long as i only use them to
communicate within the same subnet?
22[00:08:00] <sney> robi: that repository is pretty old,
it's not likely that whatever it is supports your 5.7 kernel.
also testing/sid questions should go to #debian-next on irc.oftc.net
23[00:08:10] <jhutchins> robi: What did you think you were
going to get from some random github repo? It's certainly
nothing we can support here.
24[00:08:13] <jhutchins> !nvidia
25[00:08:13] <dpkg> Where possible, Nvidia graphic processing
units are supported using the open source <nouveau> driver on
Debian systems by default. To install the proprietary
"nvidia" driver, see
replaced-url
29[00:09:22] <sney> bl seems to stand for backlight, there may
be a more current way to support your backlight controls with 5.7
and the 440 nvidia driver
30[00:09:37] <sney> but again, #debian-next on oftc for
bullseye/sid.
71[00:55:48] *** Quits: Tobbi (~Tobbi@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
72[00:57:30] <graphicsv> I just compiled a new custom kernel
and it's stuck on `Loading initial ramdisk`. I already both
updated initramfs and also created new one. How can I fix it?
It's 5.7.10
88[01:02:26] <sney> 'make deb-pkg' is still the
recommended method last I checked, but 5.7 is bleeding-edge enough
that there may be some other issues. you may want to ask in
#debian-next on OFTC where people are more familiar with
sid/bullseye.
95[01:09:26] <sney> IME make deb-pkg from vanilla upstream
source builds a working kernel, but there may be caveats I'm
not aware of. I would really go to the aforementioned channel for
qualified answers. (you may have to be patient.)
96[01:11:45] <Unit193> Yeeeah, I'm lazy so when I do build
a kernel it's using the makefile target. :3
113[01:32:46] <prp-e> Hello guys. I'm looking for CLI
live-installer. Is there any option? I used calamares for the
graphical installer in most of my projects, but now I don't
have a Xorg session or ability to run Qt. So, I'm looking for a
CLI equivalent.
131[01:42:52] <sney> does calamares do anything special based on
the live environment? I thought it just runs an installer, in which
case, the regular-ass d-i netinst is the best option for a "cli
installer"
132[01:43:02] <sney> !netinst
133[01:43:02] <dpkg> somebody said netinst was a small CD image
with which you can install Debian. If, during the installation
process you have a working Internet connection, you can install more
packages straight away, otherwise, you will have a base install and
more packages later. See
replaced-url
134[01:43:43] <sney> see also <firmware image> if this is
a laptop or otherwise needs binary firmware to support certain
hardware.
160[01:59:17] <sney> netinst! the debian netinst is the cli
installer. it's probably the most popular debian installer,
period. is there a reason that doesn't fit your use case? (or
did your crash hide that bit of the scrollback)
162[02:00:22] * dvs grumbles about the firmware iso
163[02:00:51] <prp-e> sney, I didn't explain well. I
bootstrapped a debian system, then I made squashfs out of it, then I
put it into a live disc. Now, I just want it have a live installer,
in No-X mode. I have made something before, but all old projects had
a GUI such as GNOME or XFCE, I went with calamares.
164[02:01:08] *** Quits: de-facto (~de-facto@replaced-ip) (Quit: See you around.)
171[02:04:20] <sney> I suggest asking in #debian-live on OFTC or
on their mailing list, unless calamares has a hidden option to use
ncurses or terminfo
225[02:48:48] <graphicsv> sney: It simply stuck in blue GRUB
screen forever. Completely locked. I can't even reboot the
system. 4.19 stock kernel works just fine. Maybe missing firmwares?
I'm not sure if firmware-amd-graphics files comes automatically
with custom kernel?
226[02:48:53] *** Quits: oerheks (~OerHeks@replaced-ip) (Quit: because something is patented doesn't make it any
good)
227[02:49:13] *** Quits: ijurisic (~ijurisic@replaced-ip) (Remote host closed the connection)
228[02:49:22] <graphicsv> sney: quiet is already removed but it
doesn't even get out of GRUB screen. Simply stuck at the line
"loading initial ramdisk"
229[02:50:40] <sney> if firmware-amd-graphics was already
installed on your system when the initrd was generated, it should
have made it into the initrd just fine.
231[02:51:44] <sney> were there any issues generating the
initrd? did you see any messages about i/o errors or 'no space
left on device', even if you seemed to have an initrd in place
afterwards?
242[02:59:33] <graphicsv> Maybe the problem is the name of the
kernel version. I named it `5.7.10-ryzen`
243[02:59:44] <sney> nah, custom version strings don't do
anything
244[03:00:00] *** Quits: kawaiipunk (~from@replaced-ip) (Quit: Leaving this Club)
245[03:00:02] * dvs gasps ;-P
246[03:00:12] <sney> you probably have a more fundamental config
issue. did you start from the debian 5.7 kernel config before making
your modifications?
273[03:13:36] <sney> I also thought it was oldconfig. but I
haven't built a kernel since some 4.x so I guess they might
have changed it. is that what 'make help' says to do?
274[03:14:08] <ZeroSploit> am i connected?
275[03:14:25] <sney> ZeroSploit: nobody here but us chickens
276[03:14:36] <ZeroSploit> sorry was test had a bit of trouble
setting hexchat up im new
284[03:16:04] <graphicsv> sney: Yes. I guess I will be trying
things until the end of the year. If nothing fix the problem,
I'll have to go back to Intel
285[03:16:33] <graphicsv> I already lost at least 1 year with my
Ryzen 7 1700 which had the same problem.
286[03:17:32] <sney> I never buy anything newer than about ~9
months for this reason.
287[03:17:40] <dvs> graphicsv, my Ryzen 7 3700X did
288[03:17:52] <sney> there seem to be #amdgpu and #radeon
channels on this network that may have something for you to try
289[03:17:55] <graphicsv> dvs: Worked out-of-the-box with
Debian?
290[03:17:59] <dvs> yup
291[03:18:20] <graphicsv> dvs: How long have you been using it?
I had months without any issue with mine and then suddenly it was
completely wasted
292[03:18:48] <graphicsv> dvs: wasted = often lockups
293[03:19:10] <dvs> graphicsv, I bought mine about a month ago.
I guess the difference is that I didnt actually do an install: I
copied the install from my previous machine.
294[03:19:41] <graphicsv> dvs: You know what, I also did that
and it worked perfectly until I reinstalled it.
295[03:20:06] <graphicsv> dvs: It's so weird. I've
created a bug on kernel and I'm updating with frequent log
files >>
replaced-url
298[03:22:20] <sney> I wonder if there's something in the
general-purpose initrd that the ryzen needs, that gets omitted on a
system that tries to guess at a targeted initrd
303[03:23:44] <graphicsv> dvs: I hope it keeps working
perfectly. If you can do something for me: Try a Debian 10 live DVD
with Plasma and keep using it for some time and then tell me if you
get a crash. I think that was when I got the first lockup
306[03:27:21] <dvs> graphicsv, I'm using it as a server so
I don't really want to take it down.
307[03:27:46] <graphicsv> dvs: Nevermind then. I thought it was
a desktop. It works great on my side if I use it without X for some
reason (not always, though)
311[03:29:11] <sney> hm, I have a celeron system like that.
works forever headless (it's my media server, with an uptime of
13 days since the last reboot) but if I use it with Xorg it locks up
within about 2 hours. I assumed it was a hardware bug, since
it's a cheap box that was only ever shipped with windows
312[03:29:17] <sney> I wonder if there's something deeper
348[04:11:02] <graphicsv> dvs: It was linux-config-5.4. Copied
/usr/src/linux-config-5.4/config.amd64_rt_amd64.xz to .config and
set the option that I needed false. It's working just fine so
far
349[04:11:18] <graphicsv> dvs: I wonder if AMDGPU will be used
from custom kernel or firmware-amd-graphics?
350[04:11:46] <graphicsv> dvs: I guess firmware is not the same
as kernel modules
361[04:13:57] <sney> the firmware blobs from kernel.org are
technically closed source, different distros have different
approaches to this. but anyway the kernel loads the module and the
module tries to load the firmware, if listed. (you can see them in
the output of e.g. modinfo amdgpu)
491[07:25:55] <mason> I did a debootstrap install, and I ran
into a problem - in the debootstrap install, usrmerge seems not to
be applied. /bin, /lib, etc are not symlinks into /usr. Is this
normally just done by the installer, or is there a package that
ensures that the links exist?
492[07:26:57] <mason> It seems that on a standard install, these
links exist without the benefit of the usrmerge package, and
I'm curious as to what does it.
493[07:28:53] <mason> (The specific issue I hit is that most
things seem to install to /lib, and in a non-usrmerge set-up,
systemd units installing into /usr/lib/systemd/system don't
start automatically, even if enabled.)
496[07:30:16] <Unit193> debootstrap has an option to enable or
disable it, newer versions have it on by default. The package is one
way to do it after the fact.
498[07:31:29] <mason> Unit193: Any thoughts as to why the split
world doesn't work? For this package, if I simply moved the
units to the /lib hierarchy, they started working on boot.
They'd work with manual systemctl invocation while they still
lived in /usr.
499[07:31:52] <Unit193> I avoid "merged" systems as it
breaks stuff.
500[07:33:18] <mason> Unit193: Interestingly, this appears to be
the opposite, where unmerged broke. I don't like the idea of
usrmerge generally, but I'm curious about this. My suspicion is
that the package assumes other things are in /usr that aren't,
but I'm early enough on that I'm still confused about the
package working when invoked manually with systemctl, but not
automatically.
510[07:36:10] <mason> Unit193: Nah, the deboostrap flag for
merged usr was it. I can track down *why* it broke on my own, but
now I know why systems default to /lib being a symlink now.
513[07:39:21] <mason> (I also hit a bug today where migrating a
VM from an Ubuntu host to a Debian host saw it get a different
"predictable" interface name. That's good fun too. I
need to start defaulting to net.ifnames=0)
533[08:01:43] <mason> ratrace: Yeah, the man page for this guy
notes unmerged as the default.
534[08:02:13] <ratrace> says here --merged-usr is default
535[08:02:21] <ratrace> so ubuntu's isn't?
536[08:02:33] <mason> Correct. At least the one shipped in
18.04.
537[08:02:48] <ratrace> k, good to know if I ever have to
deboostrap from 'buntus
538[08:03:37] <mason> My main curiosity now is understanding why
on an unmerged system, my unit files installed to
/usr/lib/systemd/system didn't come up on boot.
zfs-mount.service for instance - I could drive it via systemctl, but
it didn't happen on its own. I was able to enable it.
542[08:04:03] <mason> Ah, I was wondering. Yeah, still doing it
on LUKS here too, as I don't like the ZFS native encryption.
543[08:04:27] <mason> If nothing else, it presents a pain when
you try to send stuff to another system that's not on native
encryption.
544[08:04:55] <mason> ...whereas LUKS is well-understood by
everything and works nicely.
545[08:05:08] <ratrace> I don't like it either because it
doesn't encrypt everything. dataset names for example, which is
a huge deal in my book
546[08:05:22] <ratrace> LUKS ftw
547[08:05:26] <mason> Yep.
548[08:05:45] <mason> Plus, multiple keys. LUKS and GELI have
them. ZOL native encryption doesn't.
549[08:06:46] <ratrace> as for that zfs-mount thingy, of course
you checked the journal? usually it's due to non-empty
mountpoints, which in case of systemd happens if you have datasets
on /var or /tmp ; zfs-mount is alien to linux native way of
mounting, so systemd has no idea it's a mountpoint,
prepopulates with files _before_ mount get the chance
552[08:07:19] <ratrace> ZFS is really a very alien filesystem on
Linux. In many things, it's fighting with Linux
553[08:07:38] <mason> ratrace: Yeah, that seemed not to matter.
Clear mountpoints, wouldn't mount. 'systemctl restart
zfs-mount' and they'd be mounted.
554[08:08:05] <mason> So it's something about the unit file
living in /usr/ and not /lib/ on an otherwise "merged"
system that seems to matter.
555[08:08:10] <ratrace> but then the reason for failure on boot
should be in the journal
556[08:08:50] <mason> I'll move 'em back to /usr and
let it fail, and see. I didn't see anything notable in
journalctl, but I'll look again.
557[08:09:10] <mason> Note that it didn't fail on boot, it
didn't *try* on boot.
558[08:09:43] <ratrace> iirc it was always /usr/lib, even on
strech (unmerged)
559[08:10:26] <ratrace> systemd doesn't even look
elsewhere:
replaced-url
560[08:10:36] <mason> ratrace: So, the difference between
"look, it all works" and "nothing happens" was
whether the units lived in the /lib hierarchy or the /usr hierarchy.
I'm betting on a systemd bug, but I'll have to track it
down.
561[08:10:53] <ratrace> elsewhere == locations like /lib and
such, outside of the path search series it does
572[08:13:07] <mason> /usr is a dataset but it's
canmount=off so that anything in /usr is in root, but anything in
sub-datasets is properly in a sub-dataset
573[08:13:34] <mason> Yes, enabled. Again, the difference
between working and not is which hierarchy the unit lives in.
574[08:13:35] <ratrace> mason: legacy mounted?
575[08:13:56] <ratrace> so uh.... that's catch22? systemd
needs /usr mounted to run units in it which ..... mount /usr ? :))
576[08:14:07] <mason> No, only /tmp and /var/log are legacy, as
they get clobbered in initramfs otherwise
583[08:16:38] *** Quits: luna (Lunateris@replaced-ip) (Ping timeout: 256 seconds)
584[08:17:11] <ratrace> mason: there's a package, usrmerge,
that converts unmerged to merge'd system. for example when you
upgrade from stretch, and I used it in such situation. check out it
and see if it messes with systemd config or binary
588[08:18:09] <mason> ratrace: Yeah, I know about the package,
and now I know why the links exist on various installs without it
thanks to Unit193. The remaining bit is why where the units live
matters.
589[08:18:11] <mason> Yes.
590[08:18:18] <ratrace> so /usr IS a mountpoint
591[08:18:27] <mason> BRB, going to reboot with the units back
in /usr and see if the journal has anything.
592[08:18:30] <mason> No.
593[08:18:31] <mason> sigh
594[08:18:31] <ratrace> and /usr/lib is on that dataset, NOT on
rootfs
595[08:18:37] <mason> Both incorrect.
596[08:18:45] <mason> brb anyway
597[08:18:47] <ratrace> oh right you said, canmount=off
606[08:25:58] <mason> Dunno, now of course with the units moved
back to /usr it's coming up. Maybe I'm just over-tired, or
maybe I didn't have something enabled that I thought I had.
607[08:26:44] <mason> Still, I've learned a bunch of
interesting stuff, so it's not a loss.
608[08:27:26] <ryouma> there is also weird stuff, like "#
-*- coding: utf-8 -*-" which sets a varible whose identity is
not obvious from the syntax
609[08:27:31] <ryouma> name*
610[08:27:44] <ryouma> mischan
611[08:28:03] <mason> Good, I was trying to relate that.
612[08:28:06] <mason> Failed.
613[08:28:19] <ryouma> :)
614[08:28:58] <ratrace> can't rmeember if that was python2
or vim :)
615[08:29:09] *** Quits: Sierra (~sierrakom@replaced-ip) (Remote host closed the connection)
618[08:29:33] <ryouma> (this is like the second mischan i have
ever done in this channel. first time i was rudely told off for
offtopicness even though it was obviously a mischan :/)
625[08:32:55] <Unit193> mason: TBH, it's sort of sounding
like you're in a state of half-merged... Either installing the
merge package or trying to correct things with debsums seems perhaps
the way to go?
631[08:35:04] *** Quits: teej (uid154177@replaced-ip) (Quit: Connection closed for inactivity)
632[08:35:32] <mason> Unit193: There are no symlinks in / and
the interesting bit to me was that my ZFS packages were the only
things installing units in /usr/lib/systemd/system. I'd have
thought that would be the default location, given the move towards
usrmerge.
633[08:35:48] <ryouma> so the idea is that dist-upgrade does not
merge /usr/bin with /bin by default but nwe installations do?
635[08:35:57] <mason> I mean, after the merge, everything's
effectively in /usr/lib, but things all seem to install to /lib.
636[08:36:25] <mason> ryouma: From the wiki usrmerge page,
it's still something optional and in the future.
637[08:36:59] <ryouma> oh, huh, so you're just trying stuff
out?
638[08:37:55] <mason> ryouma: Nah, I'm moving my
workstation from Ubuntu to Debian while I've got a week of
vacation to buffer me from needing it for real work, and I whapped
into an oddity that I now suspect was my not enabling a unit, or a
target rather.
644[08:39:03] <mason> Unit193: Yeah. I might redo the packages
to default to that, as they feel forlorn and lone in there by
themselves, even if they're working.
666[08:45:05] <mason> Unit193: I based it on the upstream
"custom package" instructions, but I integrated some of
the patches from the backports ZFS DKMS.
667[08:45:35] <blodkorv2> Are there a list of guides and
tutorials about securing a debian webserver that is well known to be
good and have correct information?
668[08:45:40] <mason> ...and tossed in a metapackage to pull it
all together, kind of like what Debian's team does with
zfsutils-linux
783[10:24:20] <hmht> I made a .deb with a Depends, but it turns
out I also need to run update-alternatives for the dependency to get
it to work. Can I declare this in my .deb?
784[10:24:42] <lrvick> Hey all. I have been trying to use debian
snapshots for deterministic builds, and for some reason over and
over and over my install fails to install openjdk-11 from recent
snapshots: E: Failed to fetch
replaced-url
785[10:24:59] <sney> hmht: ask in #debian-mentors on OFTC
786[10:25:19] <Unit193> Sounds like a personal package, though.
820[10:54:56] <ThoMe> is it only this line in my sources list
deb
replaced-url
821[10:57:07] <sney> dpkg: stretch-lts
822[10:57:07] <dpkg> Security support for Debian 9
"stretch" from the Debian Security Team ended July 6 2020.
The <LTS> team will continue to provide limited security
support for some architectures and a subset of packages until June
30 2022 (total 5 year life). See
replaced-url
906[12:08:56] <dar8> Hi, I know this may sound like a weird
question (And perhaps not the right place to ask) but does anyone
knowif there is a difference beetween the ejection of a device (USB,
external Hard drive) beetween Debian (GNU/Linux) and Windows ?
907[12:10:01] *** Quits: dselect (~dselect@replaced-ip) (Quit: ouch... that hurt)
916[12:13:22] <dar8> Ok thanks. Weirdly i have three external
drive My Passport that can be ejected on both OS, but only Debian
will shut it down ... On Windows i have to remove the USB cable to
make the drive stop ...
917[12:13:29] <ueb> I'm trying to make a bi-directional SSH
tunnel
918[12:13:49] *** Quits: mxco86 (~mxco86@replaced-ip) (Quit: Ta ta)
947[12:29:13] <sney> anything you're connecting to on
localhost should work as long as the daemon in question is listening
on localhost. regardless of any routers on the other end of the
cable
948[12:31:16] <ueb> sney ssh -R 33061:localhost:3306
remote.server.com , should link remote 33061 to local 3306,
isn't it?
950[12:32:17] <ueb> and on remote server I try to reverse
connect with: ssh -p 33061 localhost
951[12:32:40] <sney> ssh: connect to host localhost port 33061:
Connection refused
952[12:32:46] *** Quits: tolecnal (tolecnal@replaced-ip) (Quit: server maintenance)
953[12:32:49] *** Quits: dar8 (d4c6ca0a@replaced-ip) (Remote host closed the connection)
954[12:33:14] <sney> I can't help you troubleshoot every
part of this right now, but just take it piece by piece until it
works, and keep reading the manual
963[12:36:40] <ueb> on StackOverflow some people had the same
problem
964[12:37:14] <ueb> they figured out the solution to the
ssh_exchange_id: Connection closed by remote host error. Execute
both chmod 700 ~/.ssh and chmod 600 ~/.ssh/authorized_keys
965[12:37:47] <ueb> I tried but ~/.ssh/authorized_keys does not
exist
997[12:55:15] <pwnd_nsfw> Goooood morning everyone! I'm
having issues with love2d/love. I have an rtx 2060, up to date
buster, nvidia-driver installed. I receive: libGL error: No matching
fbConfigs or visuals found, libGL error: failed to load driver:
swrast. When I run love2d, with or without a file as argument, a
window will pop up, but won't render anything inside of it.
1001[12:55:50] <pwnd_nsfw> I've been up and down google,
seeing that I should remove libgl.so.1 or so, but that's the
only version of that I have of libgl.so, if I recall it's name
correctly
1002[12:56:50] <pwnd_nsfw> The program completely hangs after
that. I have to completely exit the console for the program to exit,
I can't click the x, nor ctrl+c in console
1003[12:57:31] <pwnd_nsfw> I have love2d installed using snap,
and I think I eventually ended up removing love that I believe is
installed by default
1004[12:57:42] <ratrace> "libGL error: failed to load
driver: swrast" sounds like it wants an opengl feature not
available in the current version of opengl you have
1005[12:58:04] <pwnd_nsfw> That's what I guessed when google
results had suggested deleting one file over the other
1006[12:58:05] <ratrace> 2060 is really fairly recent card, might
need latest kernel, latest nvidia drivers, latest mesa and/or xorg
1007[12:58:32] <pwnd_nsfw> This install is probably not older
than a month
1044[13:10:49] *** Quits: noosanon (~user@replaced-ip) (Remote host closed the connection)
1045[13:10:52] <pwnd_nsfw> I have backports in my repo, I just
don't know how to reference the driver lol
1046[13:11:01] <Haohmaru> for me it's like this: crapdows is
only getting "worse" on many levels, there's no hope,
so i'm "stuck" with linux.. luckily, debian is
getting bett0r and bett0r
1064[13:14:39] <pwnd_nsfw> I didn't know how to reference
the package
1065[13:14:55] <Haohmaru> ratrace if the package has
"surrounding" packages that also get installed, when you
add -t - would apt also install those surrounding packages from
backports?
1068[13:15:23] *** jeweet_- is now known as jeweet_
1069[13:15:36] <ratrace> with -t you enable the repo for that
transaction, so it'll pull in deps from there too
1070[13:15:44] <Haohmaru> hm
1071[13:15:53] <ratrace> which means..... DON'T apt upgrade
-t unless you mean it
1072[13:15:59] <pwnd_nsfw> lol
1073[13:16:10] <Haohmaru> in synaptic you gotta click each of
them and Force Version
1074[13:16:13] <Haohmaru> >:/
1075[13:16:14] <pwnd_nsfw> will it ignore upgrading this action?
1076[13:16:19] <Enissay> I want to extract some info from the
output of a given command using regex. Input has multiple lines and
I want to extract different parts from different lines and print
them in a formatted one liner. What would you advise to use for the
extraction (with perl regex support) ?
1077[13:16:29] <Haohmaru> ratrace nah, i use synaptic
1078[13:16:34] * pwnd_nsfw braces for impact
1079[13:16:36] <pwnd_nsfw> It's installed
1080[13:16:54] <ratrace> pwnd_nsfw: you don't need -t for
upgrade unless you want to bump all packages to their upgrades from
backports
1081[13:17:09] <ueb> lsof -i -P -n | grep LISTEN
1082[13:17:11] <ratrace> once you install the driver, it'll
be tracked for upgrades from that repo with regular apt upgrade
1083[13:17:17] <enoq> debian on the desktop is probably a good
idea once flatpak gains more traction
1084[13:17:18] <pwnd_nsfw> If I don't put -t, it'll
ignore this is what I was asking
1085[13:17:28] <ueb> this should tell me listining ports
1086[13:17:32] <pwnd_nsfw> understood
1087[13:17:38] <ratrace> enoq: debian is just fine for desktop.
not all of us want !sns
1088[13:17:42] <pwnd_nsfw> brb
1089[13:17:50] *** Quits: pwnd_nsfw (~UncleCid@replaced-ip) (Remote host closed the connection)
1091[13:18:04] <enoq> ratrace: yeah, there are some pieces of
software though that sometimes cause you trouble
1092[13:18:17] <Haohmaru> yeah, it's fine for firefoxing,
CAD'ing, coding, etc..
1093[13:18:23] <enoq> depends on what you run though
1094[13:18:40] <ratrace> sns is often available through
backports, and there's !ssb if it isn't
1095[13:18:48] <enoq> everything related to chromium for instance
but I don't know how debian handles that
1096[13:19:07] <ueb> I see that 3306 is not listening, but it
should listen after "ssh -N -R 33061:localhost:3306
remote-host", isn't it?
1097[13:19:09] <ratrace> snapd is also available for those very
rare cases when upstream doesn't package its software properly,
so snap is the only way
1098[13:19:17] <ratrace> dunno how good flatpak is on debian,
haven't tried it yet
1099[13:19:35] <enoq> does debian package chromium or is it a
separate repo?
1100[13:19:42] <ratrace> enoq: debian tracks latest chromium with
a small delay due to testing
1101[13:19:54] <enoq> I see, so electron should be fine as well I
guess
1110[13:21:12] <ratrace> desktop linux suxx in general.
"crapdows" is way better there, occasional ragefit
notwithstanding
1111[13:21:24] <Uncle_Cid> :p
1112[13:21:37] <Uncle_Cid> I do enjoy what I can do, but the fact
that I have to reset everything on reboot is pretty ass
1113[13:21:41] <ratrace> I stopped recommending linux for desktop
to people who don't know what linux is. Windows works for them
just fine.
1114[13:21:50] <Uncle_Cid> For sure
1115[13:22:18] <ratrace> and personally I run only linux,
everywhere, and i3-wm on desktop because I can't stand the
bloated behemoth restaurant-sink of gnome and other big DEs
1136[13:36:03] <dpkg> Shiny New Shit Syndrome is a serious
disorder, which usually breaks out into an epidemic every time
something new is released. If you have SNS, ask me about
<backports> and <ssb>; these are better options than
upgrading to <testing> because it is a <moving target>.
1137[13:36:42] <genr8_> !ssb
1138[13:36:43] <dpkg> First, check for a backport on
<debian-backports>. If unavailable: 1) Add a deb-src line for
sid (not a deb line!); ask me about <deb-src sid> 2) enable
debian-backports (see <bdo>) 3) apt update; apt install
build-essential; apt build-dep packagename 4) apt -b source
packagename 5) dpkg -i packagename-ver.deb To change compilation
options, see <package recompile>; for versions newer than sid
see <uupdate>.
1139[13:36:56] <genr8_> shinyshitbackports ?
1140[13:37:31] <ratrace> genr8_: Simple Sid Backport, but eh....
close enough :)
1145[13:39:26] <genr8_> i just had 47 days uptime with 0
problems, lost power due to a power company outage, came back up,
and 6 hours later i hard locked, everything except for mouse cursor.
1146[13:39:49] <genr8_> now im questioning what the frick that
was from
1150[13:43:00] <genr8_> 06:59:10 HELLOGOODBYE
systemd-logind[924]: Power key pressed <-- it detected that much.
I pressed it a few times to see if the logs would show it, and it
did.
1152[13:43:07] *** Quits: rgr (~rgr@replaced-ip) (Remote host closed the connection)
1153[13:44:13] <ratrace> it's unlikely it logged that and
didn't log whatever caused the lockup. lockups without logs are
indicative of hardware issues and system going belly up at the
kernel level, not allowing userland to commit logs
1157[13:45:15] <genr8_> yes you're right. i see now in
syslog: Jul 23 06:34:48 HELLOGOODBYE kernel: [22309.147868] NVRM:
Xid (PCI:0000:0a:00): 79, GPU has fallen off the bus.
1158[13:45:42] <ratrace> oh the GPU picked up drinking again? :)
figures.
1160[13:47:20] <genr8_> it did seem like the GPU crashed. but the
mouse was working... nothing else was though. I wonder if it was
recoverable. I forgot to push SysRQ or whatever the key to do weird
stuff is
1161[13:47:34] <genr8_> what would you have done ?
1162[13:48:00] <genr8_> changing TTYs did not work
1188[13:58:33] <genr8_> the diverted ones are not even active,
and the right .so's are mapped from the debian packages with
the "alternatives" symlinks
1229[14:11:26] <Uncle_Cid> No one will know why "just
restart it" works pretty well too
1230[14:11:36] <Uncle_Cid> that's a lie, but it fits our
story now
1231[14:12:23] <ratrace> especially Debian users are very
confused. Debian wants often contradicting things, and then we all
suffer, instead of a (vocal) minority suffering and crying
themselves into devuan beds.
1232[14:12:27] <Uncle_Cid> I've realized I'm always in
here with an nvidia related issue
1233[14:12:37] <Uncle_Cid> rofl
1234[14:13:18] <Uncle_Cid> Welp, with suffering comes growth and
wisdom I suppose
1235[14:13:27] *** Quits: Orcs53 (~Orcs53@replaced-ip) (Quit: I am leaving.)
1236[14:13:35] <ratrace> at the price of how many tables flipped
1252[14:26:42] <sdisavona> I want a minimal system where
I've core software (like browsers/email clients/etc) I keep
updated, while I don't really follow all the tens of packages I
install and just use rarely.
1253[14:26:48] <sdisavona> I'd also like having the
stability of debian stable with the bleeding edge from sid.
1254[14:28:34] <ratrace> how does that work.
1255[14:28:37] <sdisavona> To achieve this I thought about
installing stable with the bare minimum I need all the
time/everyday. Everything else would be installed/obtained as
flatpak/appimage + GUIX/Nix.
1256[14:29:45] <sdisavona> So basically I'd just use either
GUIX or Nix to install everything else, which may be more up to date
than the stable equivalent.
1257[14:30:05] <ratrace> !frankendebian
1258[14:30:05] <dpkg> When you get random packages from random
repositories, mix multiple releases of Debian, or mix Debian and
derived distributions, you have a mess. There's no way anyone
can support this "distribution of Frankenstein" and
#debian certainly doesn't want to even try. Ask me about
<reinstall>
1259[14:30:19] <sdisavona> I couldn't find a lot of
experience and reports about this. Any thoughts?
1260[14:30:28] <ratrace> just gave you one :)
1261[14:31:20] <sdisavona> @ratrace I agree with the
frankendebian thing, but I also think it's slightly different
from what I'm thinking about.
1270[14:35:50] <ratrace> I mean, if that works for you, go ahead
of course. having some part of the system "stable" and
another "unstable" is better than having it all
"unstable".
1271[14:36:36] <ratrace> but nix is kinda contradicting that
"I don't really follow all the tens of packages",
you're very much your own system integrator with that.
1278[14:39:46] *** Quits: iliv (~iliv@replaced-ip) (Quit: "<paniq> you know when i walk out the door,
there is plenty of stupid people. i open irc, there is plenty of
intelligent people. so the choice comes easy.")
1279[14:40:06] <ratrace> sdisavona: I don't have any really
1308[15:01:39] <nift4> Hm. I can't get into #debian-next as
reffered by the topic.
1309[15:01:53] <greycat> *sigh*
1310[15:01:59] <greycat> !debian-next
1311[15:01:59] <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
1396[16:02:57] <hiya> I am running Debian 10 stable, suddenly
without me doing anything special my laptop's speaker is giving
me hissing noises as of it is broke
1397[16:03:06] <hiya> Is there something I can do to test things
1398[16:03:22] <hiya> It happened immediately after I was trying
noVNC software in Firefox
1399[16:04:25] <shtrb> See if you raise or lower the volume if
that solve the issue (had that with flatpak )
1400[16:04:51] *** Quits: bashcu (~Thunderbi@replaced-ip) (Remote host closed the connection)
1460[16:55:37] <dpkg> Nagios (formerly NetSaint) is a popular
open source computer system and network monitoring application. It
watches hosts and services that you specify, alerting you when
things go bad and again when they get better. Packaged for Debian as
nagios3.
replaced-url
1470[16:57:27] <dpkg> Nagios (formerly NetSaint) is a popular
open source computer system and network monitoring application. It
watches hosts and services that you specify, alerting you when
things go bad and again when they get better. Packaged for Debian as
nagios4 or nagios3.
replaced-url
1471[16:57:47] <annadane> well, if you're still on jessie
then we can't help you
1472[16:57:51] <annadane> might as well just say nagios4
1503[17:22:29] <fuxxy> I want to install a single package (and
it's deps, if needed) from the 'nextstable' branch,
but keep the rest of the system in the 'stable' branch. Is
what I'm attempting to accomplish a 'backport'? Just
curious what keywords to google.
1504[17:22:54] <greycat> !buster-backports
1505[17:22:54] <dpkg> Some packages intended for Bullseye (Debian
11) but recompiled for use with Buster (Debian 10) can be found in
the buster-backports repository. See
replaced-url
1506[17:23:45] <greycat> You cannot use compiled programs from
future releases on your stable release. That way lies disaster if
you try. Backports are recompiled to work on the stable release.
1510[17:26:42] <fuxxy> so packages.debian.org lists the package
version I want in the 'bullseye (testing) branch. Is there a
way to verify it exists in the 'buster-backports' repo
before I add the repo to sources.list?
1511[17:26:58] <greycat> What is the NAME of the package?
1517[17:27:47] <greycat> It hasn't been backported. Judd
would have said if that were the case.
1518[17:27:51] <greycat> ,checkbackport deluged
1519[17:27:53] <judd> Backporting package deluged in
sid→buster/amd64: all build-dependencies satisfied using
buster.
1520[17:28:02] <greycat> Judd believes that you can build your
own backport of it, easily.
1521[17:28:05] <greycat> !simple sid backport
1522[17:28:05] <dpkg> First, check for a backport on
<debian-backports>. If unavailable: 1) Add a deb-src line for
sid (not a deb line!); ask me about <deb-src sid> 2) enable
debian-backports (see <bdo>) 3) apt update; apt install
build-essential; apt build-dep packagename 4) apt -b source
packagename 5) dpkg -i packagename-ver.deb To change compilation
options, see <package recompile>; for versions newer than sid
see <uupdate>.
1557[18:03:25] <vincent-> Hello. Just a simple question. Other
package managers like "yum" or "dnf" are able to
install package providing a file. For instance, you can do this:
"dnf install /usr/bin/python3.7" and it will install the
python3.7 package. Is "apt" able to do the same?
1558[18:03:40] <greycat> yes
1559[18:03:54] <greycat> only you don't give a pathname. you
give a package name.
1560[18:04:11] <vincent-> greycat, that's the point, I want
to give the path :-)
1561[18:04:14] <greycat> (reads the question again) I suspect you
are looking for the apt-file package.
1562[18:04:24] <greycat> ,info apt-file
1563[18:04:27] <judd> Package apt-file (admin, optional) in
buster/amd64: search for files within Debian packages (command-line
interface). Version: 3.2.2; Size: 37.6k; Installed: 90k; Screenshot:
replaced-url
1564[18:04:59] <vincent-> That's for searching. I don't
mean that. I mean, you do "apt install /usr/bin/whatever",
and it will install the package which provides that file.
1565[18:05:47] <towo`> open a whishlist bug on apt, provide your
code for apt as patch
1566[18:06:19] <EdePopede> vincent-: find out which package
provides the file you want and install it then
1567[18:06:26] <greycat> how the flip do you know a full freaking
pathname but not a PACKAGE name
1568[18:07:00] <greycat> "is it /usr/sbin/foo? no, is it
/usr/bin/foo? no, is it /bin/foo?"
1570[18:07:43] <vincent-> So, the real use case is not that
simple. I have an Ansible configuration which is installing
whichever package provides the "/usr/bin/unbuffer" tool.
It works for "dnf" but not for "apt".
1571[18:07:47] <EdePopede> soon there will only /usr/bin/foo be
left
1572[18:08:00] <petn-randall> vincent-: The problem with that
approach is that in Debian at least several packages could provide
/usr/bin/foo, so it's not quite as simple as you might think it
is.
1573[18:08:42] <vincent-> petn-randall, I see. Well, that's
good to know.
1574[18:08:50] *** Quits: cHi-ses- (~Username@replaced-ip) (Remote host closed the connection)
1619[18:25:09] <greycat> If you're not root, then how the
hell do you think you're going to make use of an IPv6 address?
1620[18:25:28] <ComputerTech> irc network
1621[18:25:42] <Dagger> you do. line 17
1622[18:25:52] <annadane> !installit
1623[18:25:52] <dpkg> Your box does not come with every
application, tool and utility known to debiankind installed already.
If you find that the program you've been told to use isn't
there, install it. Also ask me about <search>. If someone
suggests an application to you, it's highly likely that
it's available via apt-get or aptitude.
1624[18:25:52] <petn-randall> ComputerTech:
2605:e000:100b:8036:a1f:71ff:fe03:d33d seems to be your adress. The
other starting with fe80:: are link-local addresses that are not
unique on the internet.
1625[18:26:04] <ComputerTech> ok thank you petn-randall :)
1626[18:26:06] <ComputerTech> appreciate it
1627[18:26:11] <Dagger> ...Linux allows regular users to open
sockets too, y'know
1666[18:54:27] <somiaj> What are some useful ways to
install/connect to vm's using libvirt (kvm-qemu) on a headless
(well at least xorgless) server? Here I'm wondering about the
standard vnc/spice connection to do the install?
1667[18:55:19] <somiaj> Or would it be better to create a base
image (say minimial buster install + sshd) and just copy that into
place, configure it with virsh, and connect to it with some key
pair?
1668[18:55:30] *** Quits: gryffus (~gryffus@replaced-ip) (Quit: This computer has gone to sleep)
1669[18:56:05] <somiaj> I know vnc can be done remotely, but with
virsh how hard is this to setup? I have personally just use
virt-manager up to this point (and done a little bit with virsh, but
always had xorg so I could use virt-viewer)
1674[19:04:54] <petn-randall> somiaj: The problem with
pre-installed images is that things like SSH host keys then
don't change. And probably other details that could be harmful
if shared between VMs.
1676[19:05:32] <petn-randall> I used virt-clone and virt-sysprep
in the past, although I'm considering bootstrapping images
cleanly when I revamp my setup in the future.
1691[19:13:37] <petn-randall> somiaj: So virt-clone just makes a
copy of the VM, and virt-sysprep mounts the image, and wipes things
like log files and SSH host keys from the image so they get
recreated on boot.
1692[19:14:16] <somiaj> cool, reading up on that now. Can you
virt-clone to different size image files?
1693[19:14:34] <somiaj> just thinking it might be easier to only
install a min system once and just use that as a base if possible
1700[19:20:38] <somiaj> and while I'm asking questions,
share files between server and host, virsh have tools for that, or
should nfs or somethign simlar be setup?
1703[19:22:53] <petn-randall> somiaj: If you use .qcow2 images
those are quite compact. You can also run some command to compact
them even more (don't remember which, one of the virt-*
commands).
1706[19:23:32] <petn-randall> somiaj: At work I back up full VMs
daily, but for my private stuff I only back up the data, as I can
recreate and configure the VMs easily with my ansible playbooks.
1707[19:23:53] <somiaj> thanks again
1708[19:25:27] *** Quits: tuxmania (~tuxmania@replaced-ip) (Remote host closed the connection)
1710[19:26:17] <petn-randall> somiaj: Really depends on how much
back space you have, and how much data you're willing to loose
if things break down, and how long you're willing to need to
restore that data.
1741[19:41:40] <tesstet> i'm using kde plasma right now but
my system keep freeze sometimes, i'm pretty sure it's
related to my gpu but i don't know what is actually causing the
crashes
1742[19:41:52] <tesstet> annadane why people should stick to one?
1743[19:41:53] <annadane> do you have an nvidia card?
1744[19:42:20] <tesstet> yet and i'm using nouveau drivers
1745[19:42:33] <annadane> try installing the proprietary driver,
i had crashes with nouveau + plasma before
1746[19:42:42] <annadane> nouveau is crippled because they get no
support from nvidia
1747[19:42:51] <tesstet> i see but i'm not willing to
install these ahaha
1759[19:44:36] <petn-randall> tesstet: You might want to try a
newer kernel from backports if this is a relatively new GPU. Most
more subtle bugs get fixed in newer kernels.
1760[19:44:50] <petn-randall> For new hardware, that is.
1804[20:34:18] <genr8_> tesstet, run 'sudo tasksel' to
install additional DE's from a dropdown menu, (like when you
chose your original one in the installer)
1805[20:34:52] <genr8_> or just figure out the right package name
from here
replaced-url
1812[20:39:00] *** Quits: fflori (~fflori@replaced-ip) (Remote host closed the connection)
1813[20:40:02] <tesstet> nice thanks
1814[20:40:40] <nlpqda> Anyone has an idea how to install debian
armhf on raspberry pi? directly imaging the iso file to sdcard is
neither bootable nor tells which dirve to be used as target. I tried
to boot my rpi3 from usb but it didn't work
1815[20:41:56] <sney> there's some info here
replaced-url
1836[20:55:08] <nkuttler> rander2: dpkg -S rt2800 ? looks like
it's part of the kernel. missing firmware?
1837[20:55:52] <sney> !ralink firmware
1838[20:55:52] <dpkg> Firmware from userspace is required by the
<rt61pci>, <rt73usb>, <rt2860sta>,
<rt2870sta>, <rt2800pci> and <rt2800usb> drivers.
Ask me about <non-free sources>, then install the
firmware-ralink package to provide.
1839[20:55:54] <rander2> nkuttler, lot of stuff , but it's
not present in lsmod
1848[21:01:18] <petn-randall> rander2: No worries, just a
heads-up. For the future, it's ok to ask in one channel, and if
you haven't gotten a response in 1-2 hours, ask in the other.
1879[21:35:10] <AndyAndyBoBandy> I'm finding that installing
and other operations done with 'apt -yqq ...' are noisy
and include progress bars. Am I misusing the quiet flag?
1880[21:35:52] *** Quits: cdown (~cdown@replaced-ip) (Remote host closed the connection)
1894[21:46:56] <aiRness> Hello, is the gcc-9/libgcc-9-dev still
have unmet dependecies on sid ? Any ETA on this ? It's been
almost ~2 months
1895[21:47:04] <greycat> !debian-next
1896[21:47:05] <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
1923[22:03:55] <n_1-c_k> apt has got in a mess here, could
someone repeat about what information helps diagnose? apt policy or
something?
1924[22:04:26] <greycat> !bat
1925[22:04:27] <dpkg> [Basic Apt* Troubleshooting]. To diagnose
your problem, we need ALL OF THE FOLLOWING information: 1. complete
output of your apt-get/apt/aptitude run (including the command used)
2. output from "apt-cache policy pkg1 pkg2..." for ALL
packages mentioned ANYWHERE in the problem, and 3. "apt-cache
policy". Use
replaced-url
1940[22:17:04] <KOLANICH> The another patch I have already sent
is a patch for reprepro. In fact I have sent its maintainer a email
with the link to my git repo pretty long ago. No answer. Some days
ago I have sent a bug and a patch to the bug tracker. Still no
answer. Is the maintainer alive? Does Debian accept PRs at all?
1941[22:17:04] <n_1-c_k> my apt trouble,
replaced-url
1942[22:17:05] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
1943[22:21:03] <greycat> n_1-c_k: are you doing this without a
terminal? Or is the TERM variable somehow mangled/unset?
1944[22:21:34] <n_1-c_k> In an emacs shell buffer. TERM is
'dumb'.
1945[22:21:48] <greycat> Well, that explains the errors about the
terminal.
1956[22:26:15] <greycat> Edit the
/var/lib/dpkg/info/texlive-base.postinst script, putting "set
-x" somewhere near the top of it. Then try again and see what
happens. The goal is to figure out which command is failing, and
ideally *why*.
1968[22:32:54] <greycat> Apparently "ucf" is failing. I
don't know what that is, so you'll have to figure out
why... maybe that filename that it was told to purge is already
missing or something? I have no idea.
1986[22:40:58] <dpkg> To provide command output in English
instead of your native language, set your locale to an English one
(e.g. C) prior to running the command, e.g. "LC_ALL=C apt-get
-f install".
1987[22:41:22] <greycat> Other than that, I would guess that your
sources.list is not complete, or you haven't run "apt-get
update" since you last edited it.
1988[22:41:25] <greycat> !buster sources.list
1989[22:41:26] <dpkg> A suitable /etc/apt/sources.list for
"Buster" has the lines: "deb
replaced-url
1990[22:43:06] <n_1-c_k> I changed that 'purge' line to
read 'if [ -f $file ]; then ucf --purge $file; fi' and I
think it's installed muttprint now, which is hopeful...
1991[22:43:47] <n_1-c_k> I got 'needrestart is being skipped
since dpkg has failed' but maybe that's not so bad.
2020[23:13:43] <rocketmagnet> i'm on debian 10 and want to
install wxwidgets, why can't i just install the libgtk-3-dev
package ?? i get a few dependency problems
2021[23:14:02] <rocketmagnet> i need it desperately
2022[23:14:27] <tarzeau> rocketmagnet: which version of
wxwidgets?
2077[23:28:18] <greycat> rocketmagnet: keep following the chain
until you figure out what's CAUSING the errors. Try installing
the next one.
2078[23:28:54] <greycat> Or, if you wish, you can try the generic
shotgun approaches: run "apt-get -f install" or "dpkg
--configure -a" and see if one of them coughs up a useful error
message.
2103[23:40:15] <dpkg> In buster, su no longer overrides PATH by
default, requiring that you use "su -" or "su
-l" for login shells (which is not really a new thing at
all...). To approximate the previous behaviour, put
"ALWAYS_SET_PATH yes" in /etc/login.defs. See
replaced-url
2126[23:55:15] <greycat> Well in any case, it clearly came from
outside of Debian, and it's clearly clogging up your package
management. Figure it out.