46[00:23:01] <dvs> but is it a version that's at least as
new as the kernel.
47[00:23:57] <sig_9> is there anything special I need to do in
order to get a second monitor working?? I plugged the second monitor
in and in kde settings it doesn't show a second monitor
67[00:32:54] <sig_9> it is showing only this monitor
68[00:33:15] <BCMM> is this a laptop?
69[00:33:23] <sig_9> all-in-one
70[00:33:24] <sig_9> dell
71[00:33:43] <samantaz> A tablet?
72[00:33:44] <BCMM> makes sense
73[00:34:03] <BCMM> samantaz: dell makes a few one-piece
desktops
74[00:34:10] <sig_9> inspiron 3464
75[00:34:15] <BCMM> like a mac, desktop PC built in to a
monitor
76[00:34:34] <samantaz> BCMM: oh, I see now
77[00:34:57] <BCMM> samantaz: makes sense; the one output
xrandr is seeing is a built-in monitor (eDP)
78[00:35:06] <samantaz> yeah
79[00:35:30] <samantaz> just forgot that those existed
80[00:35:33] <BCMM> sorry, i meant to highlight sig_9 there
81[00:35:43] <samantaz> np ^^
82[00:35:44] <sig_9> connecting using HDMI btw
83[00:36:01] <BCMM> hmm so why is it not even showing
disconnected outputs...
84[00:36:14] <samantaz> Probably because of HW switch
85[00:37:17] <samantaz> I ran xrandr too, and lol, I have 6 DP
on my 1080 XDDD
86[00:38:12] <BCMM> sig_9: do you have firmware for that i915
driver? i don't know about intel in particular but i think
it's possible for the output switching to depend on proprietary
firmware
87[00:38:47] <sig_9> I can try, I'm using the non
proprietary
88[00:39:18] <jmcnaught> firmware-misc-nonfree would be the
package, you could check dmesg or journalctl output for missing
firmware errors.
89[00:39:32] <sig_9> I have that package installed
90[00:40:11] <BCMM> anybody know if this sort of thing is
different between modesetting and xserver-xorg-video-intel?
91[00:40:17] <BCMM> (i don't really use intel graphics)
92[00:40:21] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
101[00:42:39] <sig_9> I don't think this is correct
102[00:43:02] <BCMM> 1. the port is physically present on all
models but only functional if you ordered a discrete graphics card?
103[00:43:23] <BCMM> 2. how can a machine like that even have a
removable graphics card? i would have assumed it was just on the
mobo like a laptop...
104[00:43:24] <sig_9> yeah that doesn't even sound right
105[00:43:50] <sig_9> I will try the proprietary gpu drivers
106[00:44:02] <jmcnaught> Is the HDMI port an input on this
machine?
107[00:44:55] <BCMM> oh actually that could concievably make
sense. i was going to say that that's so hopelessly niche that
nobody includes it as an on-board thing but then again it is an AOI
108[00:44:56] <sig_9> hmm
109[00:45:00] <samantaz> BCMM: There is some GPU slot connector
that I can't recall the name of
110[00:45:14] <samantaz> And they probably use that to reduce
production costs
111[00:45:15] <sig_9> says: the use of this driver is
discouraged if your hw is new enought 2007 and newer
112[00:45:23] <sig_9> I have that driver installed
113[00:45:29] <BCMM> and not being able to plug in a playstation
or whatever might prevent somebody from picking it over a PC +
monitor
114[00:45:31] <sig_9> let me uninstall it and see if that helps
116[00:46:31] <samantaz> BCMM: Probably the same as the HDMI
port on mobos: it's here, in case the CPU has a iGPU. If it
doesn't, you can't use that port
117[00:46:40] <BCMM> sig_9: is there anything written on the
chassis next to the HDMI port?
158[00:54:14] <BCMM> so you can still adjust the backlight when
you've got a console or whatever plugged in
159[00:54:20] <sig_9> yeah
160[00:54:23] <sig_9> I knew that
161[00:54:41] <BCMM> the monitor and PC seem very separate, for
an all-in-one
162[00:54:51] <sig_9> thanks for your time though BCMM
163[00:55:12] <BCMM> would be kind of cool if they'd
included true HDMI-in so you could have the xbox in a window or
something, but that would be a whole lot more expensive
164[00:55:25] <sig_9> cool thing is, I can use it for my old
13" laptop
173[01:00:28] <freem> ok, got an alt system working with good
old stable kernel. WHich tool would you advise me to compare the
/etc of both working and broken system? I'm still curious to
know what happened with initrd not loading modules...
185[01:12:46] <samantaz> freem: any diff tool will do it. For a
gui tool, I use 'kdiff', and either 'diff' or
'git diff --no-index' for a term based one.
186[01:13:19] <freem> yeah, I use xxdiff, but I wondered if
there would be something more suitable for /etc
188[01:13:47] <freem> I'm in awe reading how much files
remained from the installs & purge I did through the years :)
189[01:14:17] <samantaz> kingsley: run the GNU 'du'
(disk usage) utility, with the -s (summary) and -h (human readable)
command line switches on your DVD's mount point
190[01:14:18] <kingsley> us
191[01:14:36] <samantaz> kingsley: if you get '4.5G',
then the DVD is full.
192[01:14:48] *** Quits: jotaxpe (jotaxpe@replaced-ip) (Remote host closed the connection)
193[01:14:59] <kingsley> (Oops! I accidentally typed
"us" in my IRC client's window. Sorry!)
196[01:15:12] * dpkg explains, ever so gently, that if kingsley
doesn't give the channel more information, they can't help
197[01:15:23] <dvs> gasp!
198[01:15:47] <samantaz> kingsley: Haha, yeah wrong window.
Happens often to me too xD
199[01:16:05] <samantaz> As long as you don't type your
root password in here, that's fine xD
200[01:18:12] <kingsley> samantaz: Thanks for reassuring me. Do
you happen to know a command line utility, maybe like
'du', which returns a non-zero status for full DVDs?
205[01:23:51] <freem> hum... my fresh install lack something, I
have broken terminal char sequences... I'm sure it's just
one command to run, but can't remember which one
219[01:35:37] *** Quits: fax (~quassel@replaced-ip) (Remote host closed the connection)
220[01:36:31] <samantaz> freem: /etc/shadow is the system
"password" file. It contains the encrypted passwords of
each user, and the account expiration date. Sometime you *may* find
a .old version of it. But .org is... suspicious
221[01:37:00] <freem> samantaz: well, that's exactly why I
asked
222[01:37:31] <samantaz> have you tried to 'dpkg -S'
it ?
223[01:37:39] <freem> now, that system was a debian 11 chroot I
wanted to use, that I downgraded to debian 10 because my tools would
not work
224[01:37:40] <jmcnaught> Unless you made it yourself when you
manually edited the file.
225[01:37:47] <freem> what's -S here?
226[01:37:54] * freem rtfm
227[01:38:03] <samantaz> find the package that provides this
file
273[02:34:54] <maxtim> I had asked a question about LVMs
earlier, and I just spun up a vm to practice the methods discussed
earlier. In the past, I've also messed around with mdadm. So
I'm curious about anyone's opinion on LVM vs mdadm
274[02:37:29] <dvs> maxtim: vs? They're not related as far
as I'm concerned.
275[02:37:56] <maxtim> dvs, you're right I thought about
the question after I posted
276[02:38:34] <maxtim> mdadm is a software raid, lvm is logical
volume management
277[02:38:41] <maxtim> totally different things
278[02:38:50] <dvs> maxtim: Once of my drives has LVM but I
don't find it very useful. I'd rather use gparted.
279[02:39:15] <jmcnaught> I use both, I have two HDDs in an
mdadm RAID1 which is a physical volume for my LVM volume group.
280[02:39:15] <maxtim> i tend to stay away from gui stuff
281[02:39:35] <dvs> find then, parted!
282[02:40:45] <jmcnaught> LVM is pretty flexible, there is a lot
of overlap with mdadm. With LVM you can choose per-logical-volume
linear, mirrored, striped, raid5 etc.
293[02:44:48] <jmcnaught> rander2: "apt search
cryptsetup" there is more than one package, read the
descriptions with e.g. "apt show cryptsetup-initramfs"
294[02:45:27] <samantaz> dvs: regarding mirrors, can you please
do a line count (wc -l) on your
"indices/files/arch-amd64.files" file?
295[02:45:50] <samantaz> And also a 'find | wc -l' on
your pool folder
296[02:45:51] <jmcnaught> A couple of months ago I switched SSDs
and after manually dealing with /boot and /boot/efi moving
everything else was handled by pvmove
364[03:36:41] *** Quits: jvwjgames (uid290762@replaced-ip) (Quit: Connection closed for inactivity)
365[03:39:51] <freem> maxtim: I use gparted, too.
366[03:41:31] <freem> but I got an excuse, I'm not a
sysadmin, I'm a dev which still need to learn a lot and is
soooo lazy
367[03:43:04] <sney> samantaz: that package has an old
Vcs-Git/Vcs-Browser entry, tracker is just parsing it. it looks like
the actual source is here
replaced-url
369[03:44:34] <sney> collab-maint was hosted on alioth, which
went offline in 2018. there are probably quite a few packages still
referencing it, particularly ones in a RFH state like this
383[03:49:42] <sney> there's a bit of a time investment for
getting a package set up in the archive and getting used to the
tools/workflows, but once it's going, involvement can be pretty
minimal. if you have a free week or two now, and can commit to like
an hour a month after that, package maintenance might be for you!
384[03:51:20] <freem> sney: why is there "a bit of a time
investment for getting a package set up"?
389[03:54:03] <freem> that's... just what I read through
the internet tbh
390[03:54:09] <sney> in short, because debian has standards. you
can't just run a tarball from upstream through some scripts and
be done with it, there are steps that need to be addresssed manually
by the package maintainer, mostly to do with licenses and software
policies
391[03:54:27] <freem> that, it's okay
392[03:54:38] <freem> I enjoy debian because of that
393[03:54:44] <sney> and since debian has been around a long
time and established most of these procedures 10+ years ago,
there's a learning curve for devs who grew up in the git
ecosystem
394[03:55:44] <freem> why do you mention git, here?
395[03:55:47] *** Quits: catman370 (~catman@replaced-ip) (Quit: See you later..)
404[03:59:08] <freem> I have used debian since long, but am now
at a point it no longer meets my wishes. I would like to propose
alternate packages the same quality
405[03:59:50] <freem> I obviously need to learn, but where to
start?
415[04:01:55] <freem> last time I checked, OFTC refused me to
talk
416[04:02:20] <freem> or at least, #debian there refused it
417[04:02:30] <freem> spam security, ofc
418[04:02:30] <sney> you probably just needed to register. but
it's best to join -mentors after you have done some reading and
started work on a package anyway.
420[04:03:35] <freem> do you know if my runit scripts would be
acceptable?
421[04:04:39] <freem> I mean, I use those myself, since ...
well, few years now, and have noticed there's a runit effort in
debian, so I might just help
422[04:05:10] <freem> nvm will read the doc
423[04:05:47] *** Quits: Tobbi (~Tobbi@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
429[04:10:44] <sney> freem: when I got started maintaining my
first package, I read a bunch of the new maintainer's guide,
then started work, with the NMG open in a tab, and kept referring
back to it as I figured out the process.
432[04:11:19] <sney> some people do it differently, I'm
sure you have your own learning process. no you do not have to
memorize 1000s of pages before helping.
433[04:12:59] <freem> well, some of the readings will be needed
anyway
476[05:01:49] <urk> sney> Thanks, that helps a lot. 10.8.0,
and 10.7.0 stable release installers don't install on the fly,
and there is a lot of tweaking that has to be done. Also, I found
10.8.0 won't install if the first partition is formatted EFI. I
deleted the partition and there was no problem.
477[05:01:51] <sney> which you could find out yourself by
looking at 'man lspci', rather than spamming vvv nnnn and
hoping to magically get the reight result
478[05:02:38] <urk> Someone else on here suggested the -vvvnn
parameter, and I wasn't aware of the capability of the lspci to
know that a man review was needed.
486[05:09:27] <urk> drivers for the network card are definitely
not in the stable releases of 10.8.0, and 10.7.0 I am going to check
backports. Driver seems to be nvme.
492[05:15:35] <urk> dvs> nvme is definitely not in backports.
So far only the testing release will install on this laptop. I am
making a decision now to install testing 10.8.0 in the first gpt
partition I created, delete the 2nd partition, and send a message to
the maintainers of backports to see if I can get them to add some
drivers for the Dell XPS 15 7590. The laptop has been out for two
years already, and support for it is minimal. Both Intel, and
493[05:15:35] <urk> Dell have not posted the drivers on their
website, and I find this frustrating, but i have spent enough time
on the problem so Testing release is the way i will go.
514[05:24:38] <velix> Hmm... When ever I install docker,
I'm getting an application, which informs me, which services
need a reboot after update. Which package is this?
515[05:24:38] <sney> sometimes when a device is too new for the
running kernel, you'll get a nonsense result like that
520[05:25:24] <judd> Package needrestart (admin, optional) in
buster/amd64: check which daemons need to be restarted after library
upgrades. Version: 3.4-5; Size: 56.0k; Installed: 273k; Homepage:
replaced-url
521[05:25:30] <velix> Nice name ;) totally fits.
522[05:25:33] <velix> sney: thanks
523[05:25:36] * dvs restarts sney
524[05:25:37] <sney> np
525[05:25:41] <ectospasm> TIL: needrestart
526[05:25:52] <ectospasm> I wrote my own shell function to
determine that.
527[05:26:11] <urk> ectospams> Earlier you indicated I will
see the string "Kernel driver in use: <module name>.
Where would I see this? After doing lspci -k?
528[05:26:30] <sney> for best non-annoying results, disable the
NEEDRESTART_SESSION part so it doesn't pop up a gui window
yelling at you to log out
558[05:57:41] <jmcnaught> urk: lspci would still see a PCI
device. Try "lspci -kd ::0280"
559[05:59:22] <urk> It says Network Controller: Intel
Corporation Device 2723 (rev 1a) And the next line says
"Subsystem: Bigfoot Networks, Inc. Device 1654. This I already
had.
560[05:59:44] <urk> I don't have the name of the driver
which would indicate it is being used if it was loaded.
561[06:00:50] <sney> if it's intel wifi, the driver is
iwlwifi.
569[06:04:33] <dpkg> The iwlwifi Linux kernel driver supports
several Intel 802.11n (WiFi Link, Wireless-N, Advanced-N,
Ultimate-N) and 802.11ac adapters. Firmware is required, ask me
about <non-free sources> and install the firmware-iwlwifi
package to provide. Supported devices and troubleshooting hints are
listed at
replaced-url
572[06:05:34] <jmcnaught> I would install a kernel,
firmware-iwlwifi, and firmware-misc-nonfree from buster-backports
copied by USB stick from another computer if you have no networking
at all.
573[06:05:40] <sney> the factoid might be slightly outdated,
focus on the rest of the message
576[06:07:16] <urk> jmcnaught> Thanks, I have no networking
at all.
577[06:07:50] <urk> At one point the Testing disk did bring up
the network, but not sure what happened. This was the 10.8.0
non-free testing hybrid.
578[06:08:01] *** debhelper sets mode: +l 1019
579[06:08:08] <sney> there is no such thing as "10.8.0
testing"
580[06:08:34] <sney> but yes, the image with non-free, aka the
firmware image, includes the non-free firmware files needed to
support most wifi drivers.
581[06:10:34] <urk> sney> Thank you for correcting me. I had
the distribution number wrong. I see a file called
firmware-linux-free
replaced-url
582[06:10:53] <sney> your link doesn't even match your
question
583[06:11:09] <jmcnaught> urk: look at the bottom of each of my
links for the download section.
584[06:11:10] <sney> slow down and actually read things
586[06:11:36] <jaggz> what's the debian nvidia channel?
587[06:11:40] <jaggz> (is there one specific to it?)
588[06:11:48] <sney> jaggz: for support questions, here is fine
589[06:13:35] <jaggz> sney, thanks :)
590[06:13:41] <urk> jmcnaught> No fimrware-iwlwifi here
replaced-url
591[06:14:02] <jmcnaught> urk: I gave you three links. Each one
has its own download section.
592[06:14:57] <jmcnaught> The actual link to download the .deb
file is in the download section's table under
"architecture" it will either be amd64 (for the kernel) or
all (for the firmware packages).
593[06:14:57] <jaggz> actually I was getting an error running
nvidia-settings, but when running it a second time it worked, so
nevermind that.
599[06:16:50] <urk> jmcnaught> The link to the left has the
names of the network repositories where the file list is located at,
and not a name of the firmware you elude to.
614[06:28:22] <urk> The save button wouldn't activate. I
finally got it saved to the Downloads folder on an older 32-bit
laptop, and will copy it tomorrow to my USB. Need to sign off. Its
getting late. Thanks for all of your help.
619[06:30:22] <jmcnaught> urk: well they're not, so maybe
copy those three links to a text file so you can take a closer look
at them, and try them again tomorrow.
620[06:30:32] <urk> ok. Thanks.
621[06:31:42] *** Quits: urk (~user@replaced-ip) (Remote host closed the connection)
622[06:32:16] <sney> urk has been trying to negotiate their way
out of "hardware too new for buster" for like a week. the
dedication is kind of admirable, if nothing else
627[06:35:00] <jmcnaught> It forced me to find a way to lspci
just the wifi card without piping to grep, not sure if it would work
with all wifi cards though: "lspci -d ::0280" which
selects by device class.
628[06:36:42] <jmcnaught> According to
replaced-url
651[07:23:34] <Ganonk> just ask : how to add repo of backbox
linux? is possible?
652[07:23:47] <alex11> !frankendebian
653[07:23:47] <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>
654[07:23:52] <alex11> most likely not recommended
658[07:28:53] <jmcnaught> I went to the website for this Back
Box Linux with noscript and umatrix blocking all javascript and
third-party requests, and it had an advertisement for "need
paper written?" I would avoid and use Debian instead.
759[09:44:52] <yibotg> Hi I upgrade the qbittorent in 10.8,and
found the qbittorent is crashed when I add a new torrent or launch
the qbittorent.In Github,I found the same issue,but somebody said
because the qbittorent and qt is out of date.I want to know whether
the following upgrade can solve the problem?how to solve the problem
manually?Thank you!
837[11:22:43] <shtrb> diogenes_, just imaging if there had been
systemd on mars :D
838[11:23:43] <ratrace> it'd probably fare better than if
openrc was on mars! "Sir, the rover stopped responding because
the service manager isn't restarting failed servicdes.
It's gone, sir."
839[11:24:13] <diogenes_> shtrb, it would be the first flamewar
on Mars )
857[11:57:59] <shtrb> ratrace, ufw and hostname should be
properly configured, but systemd still will block (with hostname it
will truly fail , ufw has no limit and will just block your boot
forever)
858[11:58:33] <shtrb> I'm speaking from experience , still
trying to find a way to solve the annoying problem of failure to
bring networking target up
859[12:00:43] *** Quits: Christian75 (~Christian@replaced-ip) (Remote host closed the connection)
863[12:03:21] <ratrace> most of problems that occur with systemd
can be attributed to bad distro defaults or poorly written unit.
864[12:04:23] <ratrace> (however, I was just half-joking about
systemd on Mars. YES, absolutely, software in such dedicated tasks
should be as MINIMAL as possible with least moving parts and most
probability of idempotent reconfiguration and restart on failure)
865[12:05:02] <ratrace> I do like the features of systemd but
no, I'd never trust it in such highly critical environments
where you can't walk over to the rover and reboot it
866[12:05:28] *** Quits: deb (~deb@replaced-ip) (Ping timeout: 245 seconds)
869[12:09:06] <ratrace> (I wouldn't trust even openrc btw,
it's too buggy for my taste; I'd probably write a custom
service manager that's just for hat purpose and doesn't
need to cover all the possible use cases of a generic init/process
manager)
870[12:09:15] <ratrace> *that
871[12:10:02] *** Quits: dselect (~dselect@replaced-ip) (Quit: ouch... that hurt)
912[13:14:06] <ratrace> _what_ would fallback to /bin/sh ?
913[13:14:25] <ratrace> and define "fallback"
914[13:14:30] <shtrb> stop the process of loading units and give
me shell
915[13:15:19] <shtrb> That would allow me to get a shell on a
problamtic boot and not one where I pass an init afrument to grub
916[13:17:10] <ratrace> I don't think systemd has that
functionality and I don't think such a functionality would make
sense or be possible because you can't guarantee shell
availability at any point in the boot process
926[13:18:38] <shtrb> my actual use case is with it failing to
bring the networking target, and I wish to be able to debug it
better when it happen
927[13:19:09] <fourstepper> Hi, does anyone here use Greenbone
Vulnerability Manager? Do you build from source or do you something
like sid and the official packages from the repos?
940[13:25:10] <ratrace> shtrb: trying to figure that out. but
the possibility _is_ there I suppose, even if it doesn't have
the ability to do it today; patches welcome yadda yadda yadda
954[13:29:13] *** Quits: jvwjgames (uid290762@replaced-ip) (Quit: Connection closed for inactivity)
955[13:30:08] <ratrace> yeah I didn't know about that until
now
956[13:31:16] <shtrb> The nasty problem with my problem , is
that ufw issue is not permenant it comes and goes, sometimes will
not hapen for days and sometimes would happen several time in a row
992[14:16:58] <ratrace> my love affair with ufw shattered
minutes after trying to use it when I realized that ufw "all
denied by default" is not ..... all denied by default, but the
default implies you maybe kinda wanted dhcp and samba and whatever
the eff was it else there so the fire "wall" was actually
a fire "collander", holey as eff, swiss cheeze has nothing
on it.
993[14:17:26] <shtrb> :D
994[14:17:33] <shtrb> good point
995[14:17:39] <ratrace> and it wasn't obvious how to patch
those holes up, there was some default rules something something you
had to hunt down and close up
996[14:18:41] <ratrace> a proper "deny all" policy
must be of the "bruh, careful or you're shuttered outside
of your system, nowai in" persuasion
1034[14:52:24] <disgrntld> hey guys, I had an old wheezy install
with separate /home and / (root) partitions. I replaced my root
partition with a fresh debian10 install (intending to reuse my old
/home partition) but I ended up with my new /home dir on my root
partition... halp!
1035[14:53:36] <disgrntld> my new debian 10 install (root and the
new home directory) is on /dev/sda1, the home directory I want is on
/dev/sda3
1039[14:54:23] <disgrntld> reinstalling is not a problem if I
need to
1040[14:58:00] *** Quits: D4rk4ngel2020 (~darkangel@replaced-ip) (Remote host closed the connection)
1041[15:02:04] <MegaCarp> debian 10 xfce xarchiver v0.5.4.14 -
can't properly unpack .rar files: can see their list, can
unpack that list but it opens with pad app - and empty. paste:
replaced-url
1044[15:03:14] <EdePopede> disgrntld: move/copy the content of
/home to sda3 (top level, not inside a /home on that partition),
tell fstab to mount sda3 into /home. then also mount it manually if
you don't want to reboot, which isn't necessary. BUT:
logout as user, do this as root.
1048[15:07:37] <EdePopede> MegaCarp: for some reason i have unrar
installed (both free and non-free), the alternative pointing to
non-free. i think free couldn't handle some archive so i added
non-free also. still on 9 btw.
1120[15:46:42] <ratrace> milkt: for nanosecond precision :)
1121[15:46:56] <milkt> got it
1122[15:48:02] *** debhelper sets mode: +l 1082
1123[15:48:06] <ratrace> systemd-timesyncd used to log
periodically sync status, and in that info one could see the
precision from the re-adjustments back toward ref clock. but that
got removed from timesyncd and frankly I'm not sure how to get
that info any more
1135[15:52:45] <koollman> milkt: there isn't that much to
configure in a debian default ntpd config, really. it's
expected to work unless you need to adjust for network filtering
rules
1136[15:53:01] <ratrace> milkt: if I may recommend openntpd.
it's simler than ntpd (isn't also a server), and
there's the possibility of confining via a https source but
iirc that required libressl so isn't available on debian
without recompiling
1137[15:53:33] <ratrace> milkt: actually, I'd recommend
systemd-timesyncd as it's more than good enough for most use
cases, but if you really want ntpd, try out openntpd
1138[15:54:20] <milkt> thank you all, last time i tried using ntp
was ... eh around when i used linux for first time, should be easier
now
1139[15:54:51] <milkt> i will at least go to learn how to
configure ntpd and openntpd
1140[15:55:02] <koollman> I do not recommend openntpd. they are
very opinionated about things. if you want a non-standard ntpd
implementation, try chronyd. At least the deviations have reasonable
justification
1141[15:55:34] <koollman> but, hey, if you type 'date'
and your within a second of expected time ... whatever you use is
good enough :)
1149[16:01:06] <Kobaz> firewall went completely insane and had
like 500k rules in it
1150[16:01:20] <ratrace> koollman: openbsd are opinionated yes,
and some things like libressl is a joke but openntpd is one of few
tools (like openssh) that's useful
1190[16:11:17] <ratrace> koollman: chrony looks like a feature
packed behemoth like ntp .... I'd take 92kB of "just sync
my time pls as best as you can using ntp spec" any day over
that :)
1191[16:11:37] <koollman> ratrace: "using ntp spec" is
the debattable part :)
1192[16:12:00] <Kobaz> if you want lean and mean just use rdate
1193[16:12:01] <ratrace> "keep my time within few msec of
reference" is the part of teh debate I'm most interested
in ;)
1223[16:26:38] <Lope> you're the best bro, I've been
searching and searching
1224[16:27:51] <Lope> ratrace, "... AMD decided to make this
as unnecessarily difficult as possible and locked newer packages to
fail unless being installed on a supported Ubuntu release, so
you're going to need an older release which you can find in the
release notes for that release..."
1228[16:28:15] <Lope> AMD is supposed to be the *GOOD* Linux
experience.
1229[16:28:26] <Lope> vs nvidia
1230[16:28:38] <ratrace> it used to be even worse. then few years
ago they open sourced amdgpu and everyone was, like, whoa! but
apparently that's just an illusion
1232[16:29:21] <ratrace> nvidia has always JustWorked(tm) unless
you tried to use the Transformers setup, and sans occasional hiccup;
their corrupted business practices aside.
1233[16:29:21] <Lope> ratrace, this article is from 2017? How can
it be so bad?
replaced-url
1238[16:30:17] <ratrace> biggest reason I'm staying away
from it, even though I'd _love_ to give nvidia the middle
finger and spend big bucks elsewhere
1247[16:31:30] <ratrace> Lope: those chans, LTT included, are
REEKING of sponsored MONEH. I'd take what they say with a
wagonful of burlapfuls of salt.
1248[16:31:39] <imMute> shtrb: that doesn't look like an
error...
1252[16:32:43] <imMute> shtrb: "status: SUCCESS".
doesn't sound like an error to me. sounds like shoddy code that
thinks it's an error but it really isnt. (or really is an error
but is not reporting what the error really was)
1253[16:32:44] <ratrace> Lope: also, the hardware might be good
but the software reeks. Drivers and userland control panel things.
1254[16:32:50] <neilthereildeil> hey guys. i need to resize my
root partition. is there a way to do this?
1255[16:32:58] <imMute> shtrb: the count is also zero...
1256[16:32:58] <neilthereildeil> i dont want to love my data
either
1257[16:33:07] <neilthereildeil> lose*
1258[16:33:17] *** adming is now known as zeroed
1259[16:33:18] <ratrace> neilthereildeil: yes but offline. Thou
SHALT make backups, or suffer.
1260[16:33:30] <neilthereildeil> how would i do it (even
offline)?
1262[16:33:36] <ratrace> neilthereildeil: which filesystem?
1263[16:33:40] <shtrb> imMute, the problem is that in smartctl I
get "Error Information Log Entries: 921" and I even get
smart alerts about that value and Ijust can't understand
1264[16:34:10] <shtrb> And "Error Information (NVMe Log
0x01, 16 of 64 entries)"
1265[16:34:20] <neilthereildeil> ratrace: ext4
1266[16:34:24] <ratrace> neilthereildeil: with resize2fs. do you
want to shrink or enlarge it?
1278[16:37:02] <imMute> shtrb: it might be the reporting software
just being junk and nothing is actually wrong. wouldn't be
unheard of. but I'd definitely consider that hw suspect
1290[16:40:23] <Lope> I maen if this *doesn't* work,
I'll need to install another distro, or *gasp* windoze.
1291[16:40:36] <Lope> apparently Arch Linux has it in an AUR
1292[16:40:44] <urk> I trying to install firmware for the stable
release 10.7.0, but not sure how to get it into the system. I have
it on memory stick, and only find a method of uploading the driver
for the network card, and not the firmware.
1293[16:40:49] <Lope> which I believe is some sort of non-free
package thing.
1294[16:40:50] <urk> This is for Buster
1295[16:40:55] <ratrace> Lope: maybe debootstrap'd ubuntu
into a container would suffice?
1335[16:47:08] <ratrace> you even gots a friggin cloak
1336[16:47:15] <Lope> ratrace, it's like black magic. It
does A MOTHERTON of stuff. Unfortunately, every time something goes
wrong, it crashes. (which is better than swallowing the error)
1337[16:47:25] <Lope> but yeah, whatyougonnado.
1338[16:47:46] *** Quits: Karnos (~Karnos@replaced-ip) (Disconnected by services)
1339[16:47:51] <ratrace> Lope: I recommend polishing it with a VM
1340[16:48:00] <ychaouche> I thought 3+1 != 4
1341[16:48:23] <Lope> ratrace, thanks, well I generally install
to an SSD, and boot the SSD with a VM, before unplugging it.
1353[16:53:32] <ychaouche> # su -s /bin/bash
replaced-url
1354[16:53:43] <Lope> ratrace, geez snapshots are indispensable.
I had to do a rollback cos I tried to install it in /tmp which
didn't have enough space
1355[16:53:48] <ychaouche> bash: cannot set terminal process
group (-1): Inappropriate ioctl for device
1356[16:53:54] <Lope> And the script isn't set to crash when
something goes wrong
1357[16:54:06] <Lope> So god knows what it did haha. #rollback
1358[16:54:36] <Lope> This is the first time I've done a
rollback on a running system rootfs
1359[16:54:40] <ychaouche> oh, even the - needs to be before any
username
1360[16:54:42] <Lope> Seems to have survived.
1361[16:54:44] <ratrace> ychaouche: - is also in the options
group
1362[16:54:46] <ychaouche> and - is the same as -l
1363[16:56:04] <ratrace> Lope: I wouldn't do that if I were
you. issues might not be immediately obvious
1364[16:56:45] <Lope> ratrace, rollback and reboot?
1365[16:57:08] <ratrace> Lope: I'd reboot from initramfs
somehow
1366[16:57:17] <ratrace> %s/reboot/rollback/
1367[16:57:45] <ratrace> or .... sync, remount ro, rollback,
reboot
1368[16:57:55] <Lope> ratrace, this AMD GPU on Linux situation is
a disgrace. I ran the script.
1369[16:57:56] <Lope> amdgpu_device_initialize: DRM version is
2.50.0 but this driver is only compatible with 3.x.x.
1370[16:58:10] <Lope> Segmentation fault
1371[16:58:12] <Lope> hahahahahah
1372[16:58:13] <ratrace> bwahaha
1373[16:58:15] <Lope> that's running clinfo
1374[16:58:21] <ratrace> Ubuntu it is!
1375[16:58:42] <Lope> ratrace, you keep saying Ubuntu, but does
ubuntu have a driver that you can "just install" and
"just works" ?
1392[17:04:37] <dpkg> Firmware is software to operate electronic
devices, usually contained in EPROM or flash memory. Some Linux
kernel drivers require firmware to be provided from userspace,
notably for <WiFi> devices. Most firmware files are not part
of a Debian release as they do not conform to the <DFSG>; some
are available via <contrib> and <non-free> packages, ask
me about <search>. See also <installer firmware>.
replaced-url
1393[17:04:44] <shtrb> #@%@#@
1394[17:05:28] <shtrb> urk , you need to get debian installer
with firmware (also called non official cds)
1432[17:32:54] <Lope> ratrace, my bad, the live USB didn't
have tar installed. now it ran without errors, but then clinfo
showed "number of platforms 0"
1519[18:18:42] <hegemoOn> got too much trouble with grub-pc
update
1520[18:18:42] <ratrace> what's the purpose of that? by
default on debian, it keeps last kernel (before update) around as a
fallback in case the new one is broken
1521[18:18:51] <hegemoOn> which were blocking the crontab for
weeks
1522[18:19:04] <ratrace> I would seriously recommend against
unattended upgrades, especially if you have automation
1523[18:19:39] *** Quits: D4rk4ngel2020 (~darkangel@replaced-ip) (Remote host closed the connection)
1525[18:19:59] <ratrace> so anyway, with keeping the last kernel
around being just a fallback, what's the reason for keeping
specific (number of) versions? because then some other solution
might be better, that of building your own .debs for those kernels
you specifically need
1526[18:20:20] <hegemoOn> i was jsut asking if it's possible
1550[18:24:07] <hegemoOn> we are not talking about desktop
1551[18:24:08] <jelly> 200MB might NOT be okay.
1552[18:24:11] <hegemoOn> but servers
1553[18:24:15] <hegemoOn> not easy to reboot
1554[18:24:27] <hegemoOn> i will extend /boot when we will
migrate to debian 11
1555[18:24:28] <jelly> hegemoOn, are these physical machines or
VMs?
1556[18:24:34] <hegemoOn> baremetal
1557[18:24:57] <jelly> no idea then. Be careful.
1558[18:25:00] <ratrace> hegemoOn: why aren't they easy to
reboot?
1559[18:25:01] <hegemoOn> im
1560[18:25:13] <hegemoOn> ratrace: cause my client is not devops
minded :)
1561[18:25:18] <jelly> if they were VMs I'd suggest using
-cloud- flavor kernels
1562[18:25:23] <rajivmars> I m using debian 10. as i m new to
linux, i am reading debian-reference. that manual says that there
are six switchable virtual character consoles and normally the
x-window is running on the virtual console 7. following the manual
when i am trying to go to tty1 by using ctrl+alt+f1 (or f2), its not
working. but i can go to tty3 by using f3 instead of f1 or f2.
thereafter, when i am trying to come to GUI by using alt+f7, i got
nothi
1563[18:25:23] <rajivmars> stead alt+f2 get the GUI back. why is
it happening? is is normal or something goes wrong?
1564[18:25:39] <ratrace> hegemoOn: oh political reasons? :)
1590[18:29:19] <rajivmars> jelly: so, is it normal or not?
1591[18:29:20] <ratrace> hegemoOn: but the yaml structure is more
expressive and easier to handle than ansible's
1592[18:29:30] <jelly> ratrace, surely it's "main"
now not "master"? :-D
1593[18:29:41] <jrgcombr> hey guys, I´m looking for right
place to get in touch with the current maintainer of debian´s
cron pkg. I think I found a bug when cron is being used with
pam_group.
1594[18:29:46] <jelly> rajivmars, it's normal, yes, the docs
are a bit out of date
1595[18:29:48] <hegemoOn> ratrace:and to be also more precise, im
onboarding on a more ambitious project to migrate 10000 servers from
RHEL to debian with also adding 50000cores to the server farm fully
debian
1596[18:29:51] <ratrace> hegemoOn: one of biggest problems I have
with ansible is that the roles model is totally useless, since a
task can't ask "list me the roles you're in" per
host
1597[18:30:01] <rajivmars> jelly: ok, thanx
1598[18:30:16] <ratrace> jelly: no, here it's master.
actually, slave driver damned godlike control master.
1599[18:30:19] <hegemoOn> for that im relying on
"automation" based on MAAS with some ansible workflow
1600[18:30:30] <ratrace> jelly: yeah yeah, I know what you meant
:)
1604[18:31:38] <hegemoOn> so for this project i will rely on
ansible, as far as unix team are going to launch automated job from
ansible tower pointing to my git repo for post-instal
1673[19:21:10] <JamesGatz> Hello. I have a debian instance
running in Gnome Boxes that runs postgresql configured to listen on
the external IP. It seems postgresql starts before the external IP
is up (?) so it does not listen. I have to restart the service after
boot. What's the best way to address this?
1676[19:22:20] <dpkg> Interfaces marked "auto" in
/etc/network/interfaces are waited for by things that wait for the
network to be up (like systemd's network-online.target).
"allow-hotplug" means the interface is removable or not
always needed, so network-online won't wait for it.
1693[19:33:30] *** Quits: Nokaji (~Nokaji@replaced-ip) (Quit: "... when the freedom they wished for most was
freedom from responsibility then Athens ceased to be free and was
never free again.” ~ Edward Gibbon (1737-1794) - Decline and
Fall of the Roman Empire, 1909)
1694[19:34:14] *** Quits: epony (epony@replaced-ip) (Remote host closed the connection)
1721[20:08:05] <drocsid> Hello, I recently upgraded my debian
system and it broke my user login. I rebooted te system and logged
in as root. I'm unable to set the password or use `su user` to
login via the user account. I've tried `sudo pam-auth-update`
and also various steps in
replaced-url
1722[20:08:49] <drocsid> Any debian / linux gurus who can help me
to fix my login
1723[20:08:51] <drocsid> ?
1724[20:09:15] <greycat> passwd username
1725[20:09:28] <drocsid> no, I get the password auth error when I
try that
1745[20:17:35] <drocsid> So after rebooting, on the login prompt
I type my user name. I hit return. I seem to immeadiatly get `login
incorrect` after entering my username
1746[20:17:49] <drocsid> and hitting enter, before I even enter
the password field
1749[20:20:19] <drocsid> The same happens when I try to login as
root. The only way I was able to login as root was using the
maitenance or recovery kernel boot option
1753[20:22:47] <drocsid> shtrb: I do not believe so. I did use
the machine at work once but it's my personal laptop. This
happened after a debian upgrade. Before that, I would just login
locally using username / password
1754[20:23:02] <shtrb> drocsid, have you modified your pam
settings in any way ?
1755[20:23:14] <greycat> Well, it sure sounds like PAM is hosed.
1756[20:23:22] <drocsid> shtrb: I do not believe so. How do we
check or repair?
1757[20:23:34] <greycat> What version of Debian *were* you
running, and what did you "upgrade"?
1758[20:24:23] <shtrb> greycat, there's also the root file
system in readonly mode fallback due to an error (which may prevent
him to login )
1759[20:24:54] <greycat> I don't recall seeing that
1760[20:24:55] *** Quits: koniu (~koniu@replaced-ip) (Remote host closed the connection)
1781[20:33:16] <drocsid> The other terrible thing about this
laptop, it has a damaged usb socket that floods the terminal. I
forget how to disable that...
1782[20:35:16] <jelly> if things haven't changed for
bullseye, something like "pam-auth-update --force" should
overwrite all the /etc/pam.d/common-*
1812[20:52:47] <drocsid> johnfg: that's not a debian
specific question, but I assume that you have DNS records you need
to update to point to the new host. You may need to point the CNAME
to the new IP.
1813[20:54:21] <ratrace> CNAME is not IP but hostname. and
let's not assume CNAMEs are needed here. but yes, bottom line
is, if a server changed IP, make the relevant updates in all the
relevant zones.
1819[20:57:26] <johnfg> ratrace: Thanks very much. I knew it
wasn't specific to here at debian, but thought it was worth a
question. I appreciate your help!
1820[20:58:01] *** debhelper sets mode: +l 1080
1821[21:01:32] <drocsid> So I haven't had any luck after
updating packages, I cheked the cache, the pins looked ok to me. Ran
pam-auth-update --force, no luck
1829[21:06:06] <johnfg> Where does debian store ssl server certs?
1830[21:07:14] <nkuttler> johnfg: what kind of server?
1831[21:07:32] <drocsid> libpam-gnome-keyring/unstable
libpam-modules-bin/unstable libpam-modules/unstable
libpam-runtime/unstable libpam-systemd/unstable libpam0g/unstable
<-- my list of installed pam packages
1832[21:07:32] <johnfg> apache2
1833[21:07:46] <nkuttler> johnfg: check the apache config
1834[21:08:18] <sney> it depends on the cert, if you're
using letsencrypt then it'll be in /etc/letsencrypt. otherwise
you might have manually installed it to /etc/apache2/ssl or
somewhere above your webroot
1835[21:08:18] *** Quits: Mister00X (quassel@replaced-ip) (Remote host closed the connection)
1836[21:09:14] <drocsid> those look like standard pam packages,
nothing unusual or exotic there. Then how should I go about
reinstalling the packages or what next?
1837[21:09:34] <johnfg> Yeah, I did that first, and I have one
for the website for when I hosted here, but I'm not seeing it.
1838[21:10:32] <greycat> I use nginx & dehydrated, and the
certificates appear to be in /var/lib/dehydrated/certs/
1841[21:13:09] <jelly> johnfg, apachectl -S will tell you where
the VirtualHost definitions are, and you can look at those config
files to see how they define SSLCertificateFile and friends
1842[21:13:23] *** Quits: johnfg (~johnfg@replaced-ip) (Quit: Lost terminal)
1860[21:27:02] <drocsid> when I cat /etc/debian_version it gives
bullseye/sid. Then I assume that's the installed version
1861[21:27:07] <greycat> *sigh*
1862[21:27:12] <cyveris> What did you put in your sources.list?
1863[21:27:49] <shtrb> drocsid, also apt policy output would be
nice
1864[21:27:55] <shtrb> !paste
1865[21:27:55] <dpkg> Do not paste more than 2 lines to this
channel. Instead, use for text:
replaced-url
1866[21:28:29] <drocsid> I had better fix the flood in my
terminal cause I can't read anything first...
1867[21:29:46] <drocsid> dmesg -D to the rescue!
1868[21:30:30] <drocsid> I have to type all this out manually
because I'm not able to ssh in for the login. sheesh.
1869[21:30:51] <urk> Not long ago someone on here provided me a
link to a Debian Hybrid installer that offers a standard installer,
net install, as well as non-free firmware. Any idea where this is
at? For some reason the installer seems hard to find.
1876[21:35:09] <greycat> ALL debian installers (except the Live
ones) use the same code. They all offer you a set of metapackages to
choose from. If the ones you choose are on your physical media, then
they're installed from there. If not, they're installed
from the web.
1877[21:35:52] <drocsid> heres my sources.list:
replaced-url
1878[21:37:06] <drocsid> cyveris: greycat: shtrb: ratrace: any
issues with ^^ ?
1879[21:37:15] <greycat> you've got "sid" and
"testing-security". and a third-party one (spotify).
1880[21:38:02] <greycat> also, you misspelled the word
"security" in the URL of the testing-security ones, so
even if testing-security actually had any content (it does not), it
would have failed for that
1881[21:38:03] <drocsid> sid stands for testing. I don't
know why they called it testing-security. I can remove spotify.
1882[21:38:13] <shtrb> drocsid, !frankendebian
1883[21:38:21] <greycat> are you fucking kidding ME?!?
1884[21:38:25] <shtrb> drocsid, sid standands for unstable !
1885[21:38:26] <greycat> "sid stands for testing"?!?!?
1886[21:38:27] <greycat> NOO!
1887[21:38:29] <drocsid> greycat: I had to type all this so mybe
I had a typo
1888[21:38:51] <ratrace> drocsid: sid's not testing but
unstable
1889[21:39:35] <greycat> So your *biggest* problem is you're
running unstable but you have no idea how anything works.
1890[21:39:36] <ratrace> EVEN if it were... never mix
"testing, stable, unstable" with codenames, infact, always
use just codenames so when a codename transmogrifies from testing to
stable, you won't have nasty surprises
1891[21:39:44] <drocsid> oh my bad. yeah, I have been running
unstable with the testing security packages. I will revert to
testing... Can I use testing-security with testing?
1892[21:40:12] <ratrace> you haven't been running unstable
with testing security packages because those do not exist
1893[21:40:25] <greycat> Bullseye is already partly frozen. It
might not be *possible* to downgrade from unstable to frozen
bullseye at this point.
1894[21:41:02] <urk> Mmister00X> Yes
1895[21:41:15] <franzopow[m]> I am in testing and I would like to
stay on it even after the release of the new stable. Should I worry?
:P
1896[21:41:32] <greycat> If you want to run bullseye (which is
currently testing, and will eventually be stable), you can add
bullseye, bullseye-security and bullseye-updates lines.
1897[21:42:09] <greycat> If you want to run perpetual testing
(never becoming stable), then you would use testing lines. And no
-security and no -updates because those do not exist for testing.
1898[21:42:28] <drocsid> My bad. Someone should also probably
remove
replaced-url
1899[21:42:46] <ratrace> (also there's that short moment of
worrying after new stable release when testing thaws for a massload
of updates from sid)
1900[21:43:25] <franzopow[m]> <ratrace "(also
there's that short moment "> Yeah basically I was
worried about this
1901[21:44:10] <franzopow[m]> <greycat "If you want to
run perpetual tes"> I would like to do so, but, I do receive
security updates, right? I am just the last one of the 3 to receive
those
1902[21:44:27] *** Quits: Vizva (~Vizva@replaced-ip) (Remote host closed the connection)
1903[21:44:29] <franzopow[m]> I currently run testing since some
months
1904[21:44:33] <drocsid> So it seems like the suggestion is that
I should not try to move from sid => testing, but then I should
remove all non sid entries.
1905[21:44:50] <ratrace> franzopow[m]: there are no security
updates to testing except by virtue of a testing pacakge upgrading
to newer version that contains the fix
1906[21:45:04] <ratrace> which may happen days, weeks after the
fact
1907[21:45:19] <greycat> You can *attempt* to converge to
bullseye by using bullseye lines, but there is no guarantee it will
work, because there might be things in sid, things which you already
HAVE, which are numerically higher than what will end up in
bullseye. We can't know.
1908[21:45:23] <ratrace> for stable, majority of security fixes
are backported through patches to code
1909[21:45:32] <ratrace> (that does not happen for testing)
1910[21:46:37] <franzopow[m]> <ratrace "franzopow: there
are no security"> I see but I at least receive some of them
1911[21:47:05] <franzopow[m]> Security updates to packages are
also security updates
1912[21:47:23] <drocsid> hmm how do people end up with a
sources.list like mine. Maybe also using
replaced-url
1913[21:47:41] <drocsid> Doesn't seem to call out the distro
but uses testing instead
1914[21:47:50] <ratrace> franzopow[m]: yes, sure, eventually they
are. problem is only the delay. if you're okay with that, then
all is good
1915[21:48:58] <franzopow[m]> Yeah I am ok. I like to have
updated packages
1935[21:56:49] <ratrace> we had decades of stable-updates vs
stable/updates and it only took 2021 to change that into a
reasonably sane stable-security. *golfclap* no wonder we need third
party sites to create config files.
1936[21:57:12] <urk> Mister00X> I already tried the standard
installers for the stable releases 10.7.0, and 10.8.0 and they would
not install the Intel network card driver. I assume because of no
firmware. So it does matter, and I will likely install via network.
I was just hoping to create a disk that had both options for future
use since Debian now has a hybrid option.
1937[21:57:16] <greycat> because taking the file you already have
and changing "buster" to "bullseye" is so hard
1955[21:59:25] <greycat> I KNOW I answered this one already.
1956[21:59:32] <drocsid> To some degree I've had a bit of
luck on arch...
1957[21:59:37] <greycat> ALL debian installers (except the Live
ones) use the same code. They all offer you a set of metapackages to
choose from. If the ones you choose are on your physical media, then
they're installed from there. If not, they're installed
from the web.
1958[22:00:34] <drocsid> only trouble I had with it was with the
hadoop install...
2045[23:01:40] <wanko> Hello, I'd just like to ask, how do I
get Auto resize to VM window for QEMU/KVM debian guest in
virt-manager? I installed spice-vdagent, but starting it
doesn't do anything (and it does not seem to want to run as
boot)
2089[23:42:26] <greycat> If the firmware package exists for your
version of Debian, then you first make sure you've included
contrib and non-free in your sources.list. Then do what imMute said.
2090[23:42:38] <greycat> If the firmware package does NOT exist
for your version of Debian, then it gets interesting.
2091[23:43:24] <Swift110-mobile> Ok greycat thanks but why
wasn't this installed automatically when I did sudo apt update
and apt upgrade?
2092[23:43:39] <greycat> !non-free
2093[23:43:39] <dpkg> [non-free] a component which contains
software that does not comply with the <DFSG>. To add non-free
packages to your packages index, ask me about <non-free
sources>. To see which non-free packages are installed ask me
about <non-free list>.
2094[23:43:42] <Swift110-mobile> thats whats confusing me right
now because this is the first time I have sen this
2095[23:43:54] <Swift110-mobile> im already using the non free
debian
2096[23:44:28] <greycat> If you installed with the unofficial
non-free firmware installer, then I don't know why it would
have skipped a package, unless your hardware has changed since then,
or it was added to stable since then.
2097[23:44:37] *** Joins: Kel (~KelMonsta@replaced-ip)
2098[23:45:00] <Swift110-mobile> hardware is the same as
it's always been a t420
2099[23:45:03] <Swift110-mobile> thinkpad
2100[23:45:13] <Swift110-mobile> but I am now installing those
modules