this is #debianan IRC-Channel at freenode
(freenode IRC service closed
2021-06-01)
0[00:00:01] <kreyren> How do i get ssh working on debian? I
spent around 4 hours with it now and the only thing i figured out is
that if i smash keyboard with my hands then it's going to break
in half and that it looks like new with super glue
1[00:00:05] <kreyren> sudo find /var/log -type f -exec grep
ssh {} \; | ix
54[00:34:40] <TomTheDragon> I'm trying to build some third
party software with a debian source package... Is it possible that a
package is cross-compile only?
71[00:41:49] <judd> Search for nvidia/gv100/acr/ucode_load.bin
in buster-backports/amd64: firmware-misc-nonfree:
lib/firmware/nvidia/gv100/acr/ucode_load.bin
97[00:53:33] <SanchoPensa> themill: since you are here, ole
chiefrocker, can I please adress you with a very pesky problem? I
was in the process of upgrading to buster, when I ran out of space
in the /boot partition, hence the upgrade routine aborted. now I am
stuck in a condition, where I do not get any error messages, but
autoremove suggests to remove half the desktop. By installing gnome
and KDE, I have eliminated about 20 of 50 lines with packages, but
there a
98[00:53:33] <SanchoPensa> re still some 30 lines of packages
left, that autoremove suggests to remove. obviously I am missing
some metapackage. have you by any means any idea, how I could find
out, what meta package is missing?
99[00:53:59] <SanchoPensa> themill: here is what it looks like:
replaced-url
100[00:55:33] *** Quits: dbristow (~dbristow@replaced-ip) (Read error: No route to host)
214[01:30:58] <karlpinc> SanchoPensa: The buster release notes
contains a section for what to do when things go wrong with a buster
upgrade. I don't know if that will cover your problem but it
might.
215[01:31:09] *** Quits: Immanuel (~Manu@replaced-ip) (Ping timeout: 246 seconds)
216[01:31:55] <SanchoPensa> karlpinc: thank you! I have actually
flown over them, but nothing jumpend my eye, maybe I need to read
more thoroughly...
217[01:32:01] <wskyx> eh mate-volume-controll-applet constantly
crashing is driving me nuts
218[01:32:15] <SanchoPensa> rpifan: do you start pulseaudio
manually?
240[01:36:56] *** Thargoid is now known as stwalkerster
241[01:37:09] <tds> that's definitely what i see on this
buster machine though, eg
242[01:37:11] <karlpinc> SanchoPensa: People always want to see
everything the <bat> factoid asks for to fix your kind of
problem, if you want to present another paste.
246[01:38:47] <karlpinc> SanchoPensa: IIRC the buster release
notes also said to do an "apt-get upgrade" before the
"apt-get full-upgrade". I don't know if you've
tried that.
247[01:38:55] <SanchoPensa> tds: fascinating! in my case too!
but ps aux | grep pulse returns 1305 0.7 0.0 979168 15036 ? S<l
Oct28 2:33 /usr/bin/pulseaudio --start --log-target=syslog
248[01:39:01] *** debhelper sets mode: +l 1507
249[01:39:18] <SanchoPensa> karlpinc: yes of course! standard
routine!
251[01:42:22] <SanchoPensa> i have another very odd issue... i
get a warning, that the firmare for the g100 nvidia graphics adapter
is missing. so, it appears, I need backports.
253[01:43:42] <SanchoPensa> so, I install backports in the
sources.list, I call the package form apt-get -t buster-backports
install firmware-misc-nonfree, yet still I get the package from
buster, instead of packports, tells me is the newest version, but
that one is from january, and backports is from july
254[01:43:47] <SanchoPensa> why is that?
255[01:45:00] <SanchoPensa> firmware-misc-nonfree is already the
newest version (20190114-2).
318[02:18:06] <karlpinc> SanchoPensa: You might also want to
read the release notes section on cleaning up after an upgrade, if
you've not.
319[02:19:13] <SanchoPensa> karlpinc: I am right now in the
process of discovering some highly interesting pieces of software as
deborphan and debfoster. but so far they did not help me top
determine, what I am missing.
320[02:20:18] <karlpinc> SanchoPensa: It's only missing for
me if I want it and it's not there, in which case I install it.
321[02:21:06] <SanchoPensa> karlpinc: the problem is: my upgrade
routine to buster was interrupted, because my /boot partition ran
out of disc space
326[02:22:33] <SanchoPensa> I was able to reduce the number of
packets, that it suggests by about 40% by installing gnome and KDE.
or in other words, I am missing some damned meta-package, but I dont
know which one.
327[02:23:19] <karlpinc> SanchoPensa: I don't know.
Upgrading to buster is going to involve removing a lot of old stuff.
You could try using tasksel. That's the big-bucket-o-packages
installer.
328[02:23:24] <SanchoPensa> so, repeating apt-get full-upgrade
does not do the trick
329[02:23:45] <SanchoPensa> karlpinc: yeah but not texlive and
stuff
349[02:27:36] <dpkg> In order to troubleshoot your problem with
apt-get, apt or aptitude 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
350[02:27:37] <SanchoPensa> oh, dpkg has the answer, got iot
351[02:28:13] <SanchoPensa> hehehe
352[02:28:27] <SanchoPensa> karlpinc: are you kidding me? for
all 37 lines of packages? :D
355[02:29:43] <karlpinc> SanchoPensa: I don't solve these
sorts of problems. (Usually.) But yes. It is asked for for a reason.
Should be simple enough to cut and paste. All the packages are on
one very long line in your paste. So: apt-cache policy ... For one
long command.
363[02:31:56] <themill> yagnii: see "dpkg-buildflags";
many packages will do "DEB_BUILD_MAINT_OPTIONS=hardening=+all
dpkg-buildflags". Globally enabling things like -pie tends to
break things in ways that are only spotted at runtime not at
buildtime, so there's not a lot of desire to go there
378[02:37:39] <SanchoPensa> themill: I was in the middle of
upgrading to buster, when I ran out of diskspace on /boot. I fixed
that, but I seem to have lost some meta-package
379[02:38:13] <themill> most of those library packages re only
in stretch and not in buster; there are newer versions installed.
380[02:38:27] <SanchoPensa> themill: sudo apt-get full-upgrade
returns no errors, but suggests to autoremove those packages in the
pastebin link
381[02:38:37] <themill> also, you seem to still have stretch in
your sources.list which is likely unhelpful for apt
382[02:38:55] <SanchoPensa> themill, but there are packages like
texlive...?
399[02:41:56] <dpkg> SanchoPensa: Please pastebin the contents
of your /etc/apt/sources.list and /etc/apt/sources.list.d/*.list.
The easiest way to do this is to pastebin the output of: head -v -n
-0 /etc/apt/sources.list{,.d/*}
400[02:41:57] <SanchoPensa> geez, yes there was one in
/etc/apt/sources.list.d/base.list
406[02:42:52] <alexandros_c> Good day, I am running debian 10
with plasma 5.14 and it will boot into sddm. When I go the console
and check on it status there are problems and it say that it
connected successful but all I get is a black screen. Can anyone
help me with this?
407[02:43:52] *** Quits: year_of_love (~yell@replaced-ip) (Remote host closed the connection)
408[02:44:42] <SanchoPensa> themill: ok, one error source
less... but. you are saying, that removing all those packages might
be safe after all?
415[02:46:39] <allizom> what are you doing, what's the
issue. please rephrase
416[02:46:44] <alexandros_c> Good day, I am running debian 10
with plasma 5.14 and it will not boot into sddm. When I go the
console and check on it status there are problems and it say that it
connected successful but all I get is a black screen. Can anyone
help me with this?
478[03:18:45] <crestfallen> hi how do I get into the debian (10)
sound system. I have a bluetooth device that is recognized and
connected but I'm getting no sound
490[03:22:08] <crestfallen> put differently, I'm getting
sound out of the system, so I should say what could be the issue if
I'm connected to a synched device and the device sound checks
itself?
493[03:23:33] <alexandros_c> allizom, the strange this I have
been rebooting that for a day and have no success getting into sddm.
but after asking for help all of a sudden when I rebooted it started
sddm go figure
494[03:24:23] *** Quits: czesmir (~stefan@replaced-ip) (Quit: Lost terminal)
495[03:25:09] <allizom> all's well that ends well, to put
it with a quote
496[03:25:13] <SanchoPensa> themill: man, you really never seize
to amaze me! in all those years, you haven't once, not a single
time said anything, that was wrong. So, I made that leap of
confidence, and autoremoved those 37 lines of packages. You what
happened? NOTHING. :D You were right. apparently they were all
stretch packets after all.
534[03:51:40] <mathieu> I've been running sid since forever
and it seems like I hit one of those weird issues that happen once /
5 years. My sound card device has disappeared. snd-intel-hda is
loaded and the same issue happens with the 3 kernel packages I have
installed (4.19, 5.2, 5.3). Is there a known issue going around?
884[09:13:23] <darxmurf> Which owner/group/perms should I set on
a shared folder on a SMB server linked to ActiveDirectory where
users are managing access rights on shared ressources with windows
ACL?
977[10:06:17] <Janni> After having updated my firefox-esr to
68.2.0esr on my debian buster all my extensions no longer seem to
work. They are installed and marked as active but they have no
effect whatsoever.
978[10:06:21] <wskyx> r/2
979[10:06:23] <StyXman> Janni: o/
980[10:06:33] <Janni> Deactivating and reactivating doesn't
help, neither does deinstalling and reinstalling.
981[10:06:49] <StyXman> Janni: I have a friend who had a similar
issue, lemme find her BR
982[10:06:53] <Janni> I haven't tried with a fresh profile
yet. I was hoping this was a known issue.
1099[10:46:12] <acebrianjuan> thanks for your honesty guys!
1100[10:46:26] <acebrianjuan> that's all I needed to know
1101[10:46:27] <kalpu> Hi, trying to build a debian package: I
have a postinst and a preinst file, both have a shell function which
is the same. Can I somehow create a third file and postinst and
preinst "include" that one?
1104[10:47:15] <at0m> all depends what the machine's task
is. for example, I know a light & sound guy that runs his
lightshow on a Win98 machine. it just works. it's not for his
email or surfing, it's a dedi machine.
1118[10:50:28] <teclo> morning, on a clean install of Debian
10.1, the server is connected to the Internet but does not route
packets from other machines. I did echo 1 >
/proc/sys/net/ipv4/ip_forward but the server still does not route.
Am I missing something ?
1119[10:50:51] <kalpu> wrksx: Thanks, but the included file is
not installed on the system. I just don't want to write the
same function in preinst and postinst. Do you have any idea?
1120[10:50:52] *** Quits: andy__ (~AndikaSag@replaced-ip) (Remote host closed the connection)
1126[10:52:06] <wrksx> kalpu: I never packaged anything myself so
take that into account... But I guess you could add an arbirary file
in your package, let's say "functions_def.sh" and
include that in your pre/post install scripts
1148[10:55:18] <vlt> Sometimes I'm the light & sound
guy. There a projects from 2003 where I use an old Debian *image*
and write it byte by byte to the laptop before the show. I
don't even bother with a virtual machine.
1155[10:56:35] <ratrace> teclo: most likely you need just SNAT
for LAN machines to access the WAN, but if you need the reverse, aka
"port forwarding", that's DNAT then too
1156[10:56:46] <ratrace> (and those are your keywords for google
if you need examples ;) )
1157[10:57:13] <teclo> ratrace: but I took the start_firewall.sh
script from 8.x and put it on 10.1, but when I run it, I got no
networking anymore
1158[10:57:27] <teclo> but I guess the start_firewall.sh script
needs to be adapted
1226[11:43:43] <teclo> morning, I'm having something
annoying with my ppp0 connection. In /etc/resolv.conf, I got
nameserver 192.168.1.1 so the machine points to itself because
it's running dnsmasq, and then in /etc/resolv.conf.dnsmasq,
there I got the DNS servers of my ISP. But everytime I do pon or
poff, it erases /etc/resolv.conf with the ISP's nameservers.
How do I stop the pppoe tool from doing this ?
1227[11:43:43] *** Quits: anothertorusr (~anotherto@replaced-ip) (Remote host closed the connection)
1228[11:43:43] <Tyson_911> how to connect a backdoor without a
netcat listnere
1229[11:43:43] <teclo> I already commented out defaultroute and
replacedetfaultroute
1230[11:43:43] <teclo> in /etc/ppp/peers/dsl-provider
1231[11:43:43] <teclo> perhaps should I comment out usepeerdns
1232[11:43:44] <teclo> pppoe gurus, please raise their hands ! :)
1233[11:43:44] <Tyson_911> how to connect a backdoor without a
netcat listnere
1234[11:43:44] <Tyson_911> how to connect a backdoor without a
netcat listnere
1235[11:43:44] <Tyson_911> how to connect a backdoor without a
netcat listnere
1236[11:43:44] <Tyson_911> how to connect a backdoor without a
netcat listnere
1237[11:43:44] *** Quits: Tyson_911 (~Tyson91@replaced-ip) (Killed (Sigyn (Spam is off topic on freenode.)))
1264[11:57:52] <jomofcw> I use FTP (via VSFTPd) and SFTP (via
openssh_server) but when my user connect then can see the full tree
of files (then can edit only what they have in home). I would like
to create an access FTP or SFTP to a temporary user and I want him
to see (and be able to write) only the /patch/to/directory I choose.
How can i achieve this, please ? (using Debian 8 Jessy)
1275[12:02:45] <ratrace> jomofcw: no, SSHd. Look up
"ChrootDirectory" in the sshd_config manpage, and read it
_carefully_ especially about directory permissions.
1321[12:35:01] <NeedHelpThanks> yes I know, but I don't
understand. Isn't the release of the kernel I'm using also
the version of the kernel I'm using? should be?
1329[12:36:23] <ratrace> technically, release is really 4.19.x
because Debian tracks the whole LTS branch and stays with it for the
duration of Stable, but that .x would break the versioning, so
it's .0
1330[12:36:32] <NeedHelpThanks> so if I've understood:
release is X.XX.X, like.. "Official kernel full version",
and version is like small patches added up to that version
1342[12:39:34] <ratrace> NeedHelpThanks: the "release"
in this case reflects the branch used and debian specific revision.
Debian tracks LTS kernel branches, in this case that's the
4.19.x branch
1350[12:41:57] <ratrace> I mean it's way more complex than
that, but in this case, it's basically that. Here's more
on versioning in debian:
replaced-url
1353[12:43:28] <ratrace> NeedHelpThanks: in this case, the
package is NAMED linux-image-4.19.0-6-amd64 and that's also its
release reported by uname -r . However the actual package version is
4.19.67-2+deb10u1. It's a bit more complex when you consider
the linux-image-amd64 metapackage that currently sits at
4.19+105+deb10u1 which represents 105th revision of the 4.19 branch
with deb patches
1354[12:43:36] <NeedHelpThanks> ratrace my version prints a
different revision, x.y.z(.m.n.....)-(REVISION) than uname -r,
version ends in -2 and uname -r ends in -3. I just want to
understand the concept, the essence of the difference between -r and
-v, so would you say release is equal to Windows 10 and version is
equal to Home/Pro/Enterprise (like different versions of the
1355[12:43:37] <NeedHelpThanks> same release, like a few
different changes over the "full package")?
1362[12:47:19] <ratrace> eg. if it were only
"linux-image-amd64", then you couldn't have multiple
versions of it. This way, you have
"linux-image-4.19.0-6-amd64" package name, versioned
4.19.67-2+deb10u1, and before that there was "fra$ dpkg -l |
grep linux-image
1363[12:47:20] <Lope> ratrace, hey bro, I'm busy testing
multiple LUKS unlock with the for loop idea. It's not working.
The dropbear ended and the system started booting, but got stuck and
wanted me to type the password for the 2nd LUKS volume. After I did,
it finished booting.
1367[12:47:48] <ratrace> NeedHelpThanks: anyway you get the point
I hope
1368[12:48:26] <NeedHelpThanks> ratrace my version (Windows NT
5.1) is 4.18.20-2, and my release (Windows XP SP 2) is 4.18.0-3, so
how come the release (something that comes AFTER the version) is
LOWER than the version?
1369[12:48:27] *** Quits: Juzzika (~Juzzika@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1370[12:48:34] <Lope> ratrace, ssh root@testhost -p 1234 -i $key
'for i in {1..2}; do echo -n "superSecReT" >
/lib/cryptsetup/passfifo; done'
1374[12:49:04] <Lope> ratrace, it unlocks the first luks only? :/
1375[12:49:17] <Lope> and then the dropbear ends. Weird huh?
1376[12:49:29] <Lope> I'm going to try login and do it
manually just for curiosity sake.
1377[12:49:38] <ratrace> NeedHelpThanks: well it makes sense when
you think about it. It's version X.Y of a X.0 release, that
started with .0 and is now .Y as it reflects the X(.0) BRANCH
you're in
1379[12:51:25] <NeedHelpThanks> ratrace it would make sense if
the version is X.Y and the release is X.Y+1, because the release is
released after the version, the release builds on top of the
version, so the release shouldn't be X.Y-20
1380[12:51:35] <ratrace> Lope: I used to use this with passfifo
to remotely unlock:
replaced-url
1381[12:51:43] <Lope> this is super weird. I echo'd the
password to fifo manually inside a SSH session to dropbear. Nothing
appeared to change. I waited 5 seconds and just repeated the
command. Then it looked like it was working but now dropbear has
ended, it continued booting but is asking for me to type in the LUKS
pass for the 2nd LUKS :/
1382[12:51:52] <ratrace> NeedHelpThanks: it should because
that's how revisions work in debian versioning
1383[12:51:55] <Lope> (basically same experience as the for loop)
1385[12:52:13] <NeedHelpThanks> so they go backwards instead of
forward
1386[12:52:38] <ratrace> NeedHelpThanks: debian doesn't
alter the upstream version from which it branched of, that's
why there are revisions and modifications appended to the basic
version
1387[12:52:50] <jomofcw> Sorry I was afk
1388[12:53:03] <ratrace> NeedHelpThanks: however in case of the
kernel, like I said, that revision is baked into package name to
allow co-installation of multiple versions.
1389[12:53:20] <ratrace> NeedHelpThanks: and it represents only
the BRANCH that debian is basing on
1390[12:53:26] <Lope> ratrace, haha ghetto forloop buddy.
I'll try adding the sleep 1 like you have.
1391[12:53:55] <jomofcw> ratrace OK thanks, I give it a try
1392[12:53:58] <ratrace> Lope: oh yes now that you mention it, I
think that sleep is important. You asked before and I said I
didn't think it was needed....
1393[12:54:05] <dka> How can I create a ISO of windows from linux
debian?
1394[12:54:13] <Lope> Yes, good memory.
1395[12:54:42] <dka> I am stuck with woe usb it keep saying my
disk is mounted while it's not
1396[12:54:48] <dka> should I format it ?
1397[12:54:50] <wskyx> dka: you mean what, put it on a
thumbdrive?
1398[12:54:58] <dka> Yes USB KEY
1399[12:55:16] <wskyx> easiest way use a windows vm with
winsetuptousb
1400[12:55:27] <NeedHelpThanks> Debian releases a version,
let's say 4.18.20-2, and from there Debian releases releases,
so if my version is 4.18.20-2, I could understand a release called
4.18.20-3 or 4.18.21, but how can I have a release (4.18.0-3) which
is under my version? that makes no sense. Sorry, I know you've
trying to explain but I'm just too stupid to understand,
1401[12:55:28] <NeedHelpThanks> nevermind
1402[12:55:35] <Lope> ratrace, I'm using a mechanical HDD,
not sure if that makes any difference. (like if I need a longer
sleep)
1403[12:55:46] <wskyx> or dump the iso data to the drive and
manually install the bootloader
1410[12:56:18] <Lope> ratrace, I'm testing on an old laptop
that's right infront of me so I can get my stuff right before I
go doing it on a remote server and suffer the pain of a shitty
remote KVM.
1420[12:58:21] <ratrace> Lope: you should be able to test it
manually. ssh in, echo one, observe what happens on the laptop
screen, it should re-prompt for the other, try echo again... maybe
the sleep is too short?
1427[12:59:24] <dka> Is there a tutorial I can follow for Debian
10 and Creating Windows 10 USB Bootable Key ?
1428[13:00:00] <Lope> ratrace, I logged in manually and entered
the first password. I counted (inaccurate) 12 seconds and then it
disconnected my dropbear session and continued booting, then asked
for the 2nd LUKS device's passphrase.
1434[13:00:43] <ratrace> Lope: ah... how many entries you have in
crypttab? you should have one for each device iirc
1435[13:00:55] <Lope> ratrace, I know there's a special
thing you can do with crypttab that allows you to physically type in
a password once that's used on multiple LUKS
1436[13:01:04] <Lope> So maybe that will work...
1437[13:01:16] <Lope> I've got 2 LUKS and 2 entries in
/etc/crypttab :/
1438[13:01:44] <ratrace> Lope: yes there was something about
systemd methinks but I'm not sure that made it into Buster
1440[13:02:35] <ratrace> Lope: frankly, I wouldn't know
then. the passfifo method worked for me, but I stopped using it.
these days I don't use cryptsetup-initramfs at all, and have a
custom initramfs "local-top" script that holds boot until
I provide the key.
1445[13:03:09] <ratrace> Lope: the premise there is simple.
initramfs will execute all those scripts in order, and
"local-top" is exec'd before root devices should be
made detectable
1446[13:03:53] <Lope> ratrace, well your custom initramfs thing
is probably what allows your method to work.
1447[13:04:05] <Lope> ratrace, I wonder if I put my non-root
thing first.
1449[13:04:10] <ratrace> so the script I use simply waits for the
keyfile to become available, then once found (a simple for loop with
sleep testing on path existence) it proceeds to cryptsetup luksOpen
the devices, exits with 0
1450[13:04:19] <Lope> Because once it gets the root LUKS it
continues booting and that's where things go wrong.
1458[13:05:54] <ratrace> Lope: btw it's not a custom
initramfs. I use regular initramfs mechanics for that. You can place
your "scripts" in /etc/initramfs-tools/scripts
1459[13:06:55] <Lope> LOL. I reversed the order of the
/etc/crypttab entries but it still used the first password to
decrypt the root LUKS!? hahaha
1460[13:07:20] <Lope> I'm going to try put my root fs after
the other one in /etc/fstab
1461[13:07:28] <Lope> (just experimenting for curiosity sake)
1462[13:07:32] <ratrace> Lope: this is the currently WIP version
I have (it's bound to be more secure with that temporary
keyfile, and remote fetchign from key server will be added)
replaced-url
1463[13:08:07] <ratrace> Lope: I use literally this script in
/etc/initramfs-tools/scripts/local-top/unlock-root.sh (the filename
is arbitrary) and no cryptsetup-initramfs or crypttab
1464[13:08:07] <Lope> But that actually wouldn't help,
because /etc/fstab is inside the root LUKS.
1465[13:08:17] <ratrace> update-intramfs -u is all that's
needed to get it in
1466[13:08:27] <Lope> so it's clearly decrypting the root
LUKS before it bothers looking at fstab.
1467[13:08:29] <JyZyXEL> Lope: have you tried the
"initramfs" crypttab option?
1468[13:08:42] <Lope> JyZyXEL, can you elaborate please?
1474[13:09:11] <Lope> ratrace, looking at your comments and links
now.
1475[13:09:12] <JyZyXEL> Lope: if you define all entries in
crypttab with the "initramfs"-option
1476[13:09:18] *** Quits: NeedHelpThanks (4f9d19ca@replaced-ip) (Remote host closed the connection)
1477[13:09:27] <ratrace> I think that option is removed now (not
recognized), but there used to be a bug where cryptsetup
wouldn't work in initramfs without it, something something
1487[13:11:07] <Lope> man crypttab says: The initramfs hook
processes the root device, any resume devices and any devices with
the initramfs option set. These
1488[13:11:07] <Lope> devices are processed within the initramfs
stage of boot. As an example, that allows the use of remote
unlocking using
1489[13:11:07] <Lope> dropbear.
1490[13:11:19] <Lope> Sorry about all the unexpected linebreaks.
Sounds like a winner.
1491[13:12:02] <Lope> JyZyXEL, i think this is going to work.
1505[13:14:13] <Lope> okay now I tried with the for loop of 2.
It's stayed at the initramfs stage of the boot and is asking me
to type the 2nd LUKS's password. So the 1 second delay
wasn't enough for this mechanical disk and maybe the CPU etc.
1506[13:14:41] <ratrace> Lope: even with that
"initramfs" option in crypttab?
1510[13:15:19] <JyZyXEL> remember to update-initramfs afterwards
1511[13:15:27] <Lope> ratrace, the initramfs option helped a lot.
It made the boot process stay in initramfs until the 2nd luks was
unlocked. But 1 second probably wasn't enough delay for the
mechanical disk and 2000 series i7 laptop 2/4 CPU
1523[13:18:40] <Lope> okay, I manually entered the 2nd pass via
dropbear interactive ssh session, it booted :)
1524[13:18:58] <Lope> ratrace, what are the benefits?
1525[13:19:15] <ratrace> Lope: problem with that remote unlocking
of root is..... barely documented anywhre. there's a number of
"tutorials" and at least when I was doing it the first
time few years ago, most of them didn't work, and were
different between debian and ubuntu
1527[13:19:34] <Lope> ratrace, yeah, we are pretty fringe here.
1528[13:20:11] <ratrace> very fringe, especially with debian
specific scripts and systemd's own meddling in the affair --
with good intentions, but I don't think that feature is yet in
Debian
1529[13:20:36] <ratrace> so I said to hell with all that, Imma
build my own unlocker, with blackjack and hookers.
1531[13:21:25] <Lope> heheh what? what about duke nukem saying
"shake it baby"
1532[13:21:27] <ratrace> Lope: the premise is VERY simple.
initramfs executes a set of "scripts" (execd during boot,
where "hooks" are exec'd during update-initramfs) in
order with the SOLE purpose to make the rootfs available for
autodetection and pivoting to
1533[13:21:41] <Lope> okay I modified the for loop with a 15
second delay. It wasn't enough lol.
1534[13:22:12] <Lope> I think using the 1 password many luks
thing will be more robust anyway. But keen to try this for curiosity
sake.
1535[13:22:21] <ratrace> so, you create a script (I use the
"local-top" stage) which sits and waits for the key to
appear, then runs cryptsetup unlock. the only drawback of this
method is that you need to have a hook that puts cryptsetup and
needed crypto modules in initramfs -- things that the
cryptsetup-initramfs package would do
1536[13:22:49] <ratrace> unfortunately, can't have
crypsetup-initramfs AND empty crypttab both -- ie. to take advantage
of it preparing the initramfs with all the tools and NOT using
crypttab but your own script
1540[13:24:08] <ratrace> I primarily do it because I want the
ability to look up keys from our central keyserver AND wait for a
local file at the same time, thus: 1. prevent console unlcking, 2.
allow remote key fetch, 3. allow a ssh session to log in and provide
the keyfile.
1546[13:25:19] <ratrace> I dont' know.... this way I have
100% control via VERY simplistic "local-top" stage script,
and a tiny hook that preloadds initramfs with cryptsetup binary and
required crypto modules
1551[13:27:08] <ratrace> these two are all I need and
initramfs-update -u:
replaced-url
1552[13:28:15] <ratrace> and like I said, I'll be extending
that local-top script to fetch the key remotely and try one or the
other until one of them suceeds
1554[13:29:26] <ratrace> it'll be a bit more involved as I
need to add whole openssl, curl and build a local CA (so I
don't need to trust the ca-certificates)
1557[13:30:30] <Lope> ratrace, I tried adding
",keyscript=decrypt_keyctl" to my crypttab, but got error:
entries E: /usr/share/initramfs-tools/hooks/cryptkeyctl failed with
return 1.
1558[13:30:36] <ratrace> that hook is btw the default
cryptsetup's hook, stripped down for all the carp I don't
need
1560[13:31:02] <ratrace> JyZyXEL: maybe you meant that? I
actually used the cryptsetup's initramfs hooks
1561[13:31:10] <Lope> ratrace, oh crap, I just read the line
after the example "Unfortunately, this currently doesn't
work on Debian systems using systemd init due to a bug "\
1568[13:33:15] <ratrace> Lope: and the last sentence of that
paragraph? that's the bug I seemed to half-remember... you need
"initramfs" in crypttab to avoid that bug
1578[13:36:31] <ratrace> Lope: if only passfifo was a shell
script, I could dissect it and find out what it does. I gave up
googling for its source code
1579[13:36:49] <Lope> it worked. it said "using cached
passphrase for (2nd luks)
1590[13:38:46] <Lope> I first knocked on wood 3x, threw some salt
over my shoulder. patted my stuffed penguin and gave it a hug for
good measure.
1591[13:39:07] <ratrace> all the required steps, agreed. and
then? :)
1592[13:39:16] <Lope> With the most important things taken care
of I crossed my fingers and then...
1593[13:39:47] <Lope> I installed the keyutils package, ensured
crypttab entries all had
luks,discard,initramfs,keyscript=decrypt_keyctl then
`update-initramfs -u` and linus is my uncle.
1594[13:39:55] <han-solo> Lope: tell about the error just before
you installed `keyutils` package
1595[13:40:03] <ratrace> holy undocumented gotcha galore
skagsack! :) now quick, immortalize that knowledge into a debian
wiki! patting the penguin included!
1596[13:40:22] <Lope> ratrace, cool, where is this wiki?
1605[13:43:15] <Lope> What I dislike about virtualbox mainly is
the USB oracle drivers, and the UUID's from hell that the disk
images use. They are non-portable in the most ungodly way.
1609[13:43:56] <ratrace> unborn: it's in stretch-backports,
and not at all in buster. I do second using qemu instead
1610[13:44:23] <Lope> the only thing virtualbox has going for it
in my experience is that it has a virtual 3D accelerator which can
be useful, for example if you want to be able to run the blizzard
console in a VM.
1611[13:44:42] <JyZyXEL> does virtualbox still use DKMS or did
they get everything mainlined?
1612[13:44:52] <unborn> qemu uhm my mother would not be able to
use windows with it... its too complicated for her.. and i
don't like it as well.
1613[13:44:55] <ratrace> Lope: don't forget passing usb
devices in
1614[13:45:10] <Lope> ratrace, you mean IOMMU stuff?
1615[13:45:19] <Lope> for the USB controller?
1616[13:45:45] <unborn> Lope: that means I would have to create
new virtual hdd and install windows again - right?
1617[13:45:46] <ratrace> I think that was it yes, passing a USB
device directly into the VM
1618[13:45:51] <Lope> ratrace, oh you mean just connecting a USB
device to the VM? well yes, but you need the oracle USB driver blobs
to get that working.
1620[13:46:13] <ratrace> JyZyXEL: since 4.15 the linux kernel
contains the GUEST modules mainlined, so that part is no longer
needed to be installed separately
1624[13:46:34] *** Quits: enoq (~textual@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1625[13:46:40] <ratrace> no wait
1626[13:46:46] <ratrace> you can convert a VB image into qemu
1627[13:46:51] <Lope> unborn, you can install crap from previous
versions of debian in the current version of debian.
1628[13:47:13] <unborn> nah, I would like to move forward lope
1629[13:47:20] <Lope> you just change your sources to the old
debian, update, then install what you want, then change the sources
back to the current debian.
1630[13:47:43] <unborn> ratrace: the reason why I did never used
qemu was lack of proper and easy documentation
1633[13:48:36] <Lope> unborn, virt-manager is 70% easy to use.
You will need to use qemu-img to manipulate images, but other than
that you can do everything in the GUI.
1635[13:49:01] <ratrace> Lope: dunno, I never had problem finding
stuff on how to use qemu. I do prefer to use qemu-system-x86_64 from
the cli though, no GUI thingies.
1637[13:49:15] <NetTerminalGene> can rx570 gpu reclock itself
high or low frequency depends on workload with free amd driver?
1638[13:49:36] <Lope> unborn, i dunno how you feel about binary
blobs in your system, but when I first moved to linux in 2012/3 I
was still very in my windows comfort zone and I installed VMWare
Workstation in linux and it's truly a masterpiece. It blows the
user experience of virtualbox and virt-manager into non-existence.
1640[13:49:41] <unborn> Lope: its for my mom, she needs to be
able to click and start.. you know, shes 63years old and would not
get any command line stuff in old scull
1643[13:50:49] <darxmurf> Anybody got a working documentation
about how to setup a SMB share using windows AD ACLs?
1644[13:50:57] <unborn> Lope: yap I tried vmware.. got license
somewhere. I did not like it.. my hardware was not powerful.. can
virt-manager be done on machine like this? hold on I pass the specs
now
1645[13:50:59] <ratrace> unborn: if you/she is using GNOME,
there's also GNOME boxes iirc, with very simplified interface
1646[13:50:59] <Lope> unborn, lol, you know I tried to get my dad
to use linux. I still try. I try to remind him every time he has a
problem in windows that windows is less relaible than Linux and that
it's not actually easier to use, it's just slightly more
familiar.
1649[13:51:34] <Lope> unborn given that your mom is 63 she could
probably just use linux software like libreoffice and firefox, what
else could she need?
1651[13:52:02] <eypo> unborn, just in case
replaced-url
1652[13:52:19] * eypo agrees with Lope
1653[13:52:26] <Lope> unborn, if your hardware is too unpowerful
to run a VM then that's even more reason to try get your mother
to just use linux software.
1658[13:53:59] <unborn> eypo: I wont have frankenstiendebian
rather then force her (and my self) to be up to date.. Lope I think
she need to get retired - its some database apps they using in
windows for stock and ordering.. that's why I am pissed off.. I
don't use windows since debian 5 too.. just repairing it for
other people.
1659[13:54:00] <Lope> The huge benefit if you get your parent to
run linux is you can just SSH in and do stuff for them in seconds,
and there's very little maintenance to be done.
1663[13:54:52] <eypo> My father (about 70) was 'forced'
to linux using 'My way, or the highway'. You want my
support, you use linux. You want windows, go ask Microsoft for
support, and don't forget to pay the licenses for windows,
office and so on.
1664[13:55:04] <Lope> legacy. Old business apps tend to work well
on WINE
1665[13:55:13] <unborn> Lope: she knows.. but she have still 7
years to get to retirement.. I will found my way otherwise I will
purchase some cheap amd
1669[13:56:05] <unborn> hell Lope no wine on any linux box never
ever.. wont see that unfinished unsecured crap in any proper linux
box in our house or family (politely saying)
1670[13:56:16] <eypo> unborn, quit being microsoft support for
your relatives for free
1673[13:56:59] <Lope> eypo, i tried that strategy as well.
problem is he uses a few windows programs like lightroom or
whatever. Part of moving OS is the willingness to try different
software. A lot of people hate learning stuff because they hate the
feeling of not knowing what they're doing.
1674[13:57:01] <unborn> lope I will install virt-manager, it
seems to me nice.. docs are also not that bad. I am thankful for
your suggestion
1675[13:57:10] <Lope> It's a really terrible place to be
(for them).
1676[13:57:47] <eypo> Long time ago I realized that relatives
have two types of computer problems :
1681[13:58:32] <eypo> type 2: those that when you say you're
going to install linux, they magically dissapear
1682[13:58:41] <unborn> eypo: I know.. its just her as she
don't have an choice.. and its my mother, you know even I would
rather...I have to try to help her :)
1688[14:00:11] <unborn> eypo: my mom is using linux since 2011
gladly trust me.. but her work place wont even discuss it.. Lope do
you have any suggestion for tips and how to's for virt-manager?
1689[14:00:28] <unborn> I mean more info.. so I can learn my
self..
1695[14:01:14] *** Quits: l30n0x1 (uid304748@replaced-ip) (Quit: Connection closed for inactivity)
1696[14:01:39] <Lope> unborn, don't let virt-manager create
your virtual disks for you. create them yourself with qemu-img. If
you use qcow2 you can use CoW
1713[14:05:09] *** Quits: hello (67d4e394@replaced-ip) (Remote host closed the connection)
1714[14:05:15] <Lope> Unless of course you start the VM and then
she accesses the VM via a VNC client.
1715[14:05:39] <unborn> thanks, let me play with it.. as I said
if she would not be able to get it into her brain I purchase some
cheap windows machine for her..
1716[14:05:41] <Lope> You could have a VNC client connect to the
VM automagically on boot or with a desktop icon etc.
1717[14:06:21] <Lope> nah, she'll be fine, just make it nice
and easy for her.
1720[14:07:28] <Lope> you can even install a VNC server in the VM
(like realvnc) and connect a VNC viewer to that to get cross
clipboard stuff working.
1725[14:08:39] <Lope> You won't get clipboard sharing
working if you use qemu's VNC server. You will if you use a
SPICE virtual display adapter and install windows guest drivers
though.
1726[14:09:01] <Lope> virt-manager is a GUI, that is probably the
package name.
1734[14:12:09] *** Quits: we6jbo (~we6jbo@replaced-ip) (Remote host closed the connection)
1735[14:12:27] <unborn> Lope: this looks fantastic! is that
virt-manager somehow long time project in debian (perhaps you
don't know) ? I mean support..
1736[14:12:32] <Lope> as ratrace mentioned you can convert your
virtualbox disk image to qemu format, but whether windows freaks out
about the hardware changing is another story.
1739[14:13:05] <Lope> BTW did you know you can download windows
10 ISO directly from MS and install it and never activate it. So far
the only downside is they show a watermark.
1743[14:13:31] <Lope> if you haven't activated windows then
hardware change won't be an issue
1744[14:13:47] <unborn> Lope: well if any windows freakouts I
would rather install her new system from scratch.. for conversion
its only 40 gigs with all data she have there.. no hard feelings for
it.
1745[14:13:50] <ratrace> Lope: and they do so in full screen
steam games too! The gall! :(
1746[14:14:06] <Lope> ratrace, oh yes I've heard that. But
it would probably cause license deactivation and a lovely 20~50
minute phonecall with microsoft to re-activate.
1747[14:14:17] <ratrace> Lope: not with win10
1748[14:14:23] <Lope> ratrace, lol, never tried that.
1755[14:15:33] <ratrace> no idea :) I just know a few things
about win10 that were important for my installation of it under a VM
for the purpose of playing Destiny 2 with passthru'd gpu
1756[14:15:46] <Lope> you can buy a possibly dodgy windows
activation key for $50 or less. I've not tried it.
1757[14:16:32] <ratrace> but since Destiny 2 moved to steam and I
eagerly tried it in proton only to find it deliberately refuses to
work under proton, I rm'd the whole thing and vowed never to
touch stuff that actively refuses to run under linux even wine.
1758[14:16:32] <Lope> Before my GPU died I was playing overwatch
on windows 7, much nicer :)
1759[14:16:39] <ratrace> Lope: I got mine for ten bucks on ebay
1763[14:17:04] <unborn> Lope: ps.. if you have windows 7,8 pro
you can activate it to 10 for free.. all the times.. payload is
privacy which is non in 10.. that is why I was a bit upset with this
whole virtualbox thingy.. she have windows 7 up to date and she not
need and wont update to 10 now I have like month to update whole iso
of 7 to sp-whatever-mumber.. hey man that virt-manager looks
promising
1764[14:17:27] <Lope> windows 7 is nice if you only use windows
for games and if you have an intel CPU because of all of intels
security vulnerabilities. Faster than new windows with all the
patches.
1765[14:17:53] <ratrace> isn't Win7 like EOL now?
1766[14:17:57] <unborn> ratrace: as you correctly mentioned: the
win10 license allows you to migrate it wherever whenever - this is
because you don't own os, you renting it
1767[14:18:02] *** Quits: luneff (~yury@replaced-ip) (Quit: This computer has gone to sleep)
1769[14:18:18] <ratrace> unborn: you were also renting it before,
and it was stricter with being unable to move from hw to hw :)
1770[14:18:19] <Lope> ratrace, if it runs the games you want to
play, who cares
1771[14:18:21] <greycat> Yes, most people should be off of
Windows XP by now. Oh wait, you said Windows 7, didn't you.
1772[14:18:34] <ratrace> greycat: yes that's EOL too innit?
1773[14:18:45] <greycat> A bit, yes.
1774[14:18:49] <Lope> ratrace, the only downside of windows 7 is
it doesn't support the OBS feature where it copies directly
from the GPU framebuffer to NvEnc.
1775[14:19:01] *** debhelper sets mode: +l 1542
1776[14:19:07] <Lope> But apparently windows 8 can do it.
1777[14:19:13] <ratrace> Lope: well I care. installing random
third party executable code via steam would NOT be my favorite,
especially on unpatched EOL'd win, would NOT be acceptable to
me :)
1778[14:19:31] <unborn> greycat: great to see you.. I was looking
for virtualbox replacement for old lady in my house.. so we got
discussion about it.. as I need to learn a bit more to move my self
from virtual box
1788[14:20:24] <ratrace> unborn: ordinarily I'd always
recommend qemu, but since you have an existing installation, of win7
no less, perhaps the easiest way is to install VM from their debian
repos and call it a day.
1793[14:21:23] <Lope> ratrace, yeah, win10 is fine. I think you
can disable the security mitigations in win10 anyway.
1794[14:21:25] <unborn> greycat: wrong! qemu or virt-manager? as
I know you for long time, your answer would be starting with q.. but
its for old lady (she needs easy gui)
1795[14:21:35] <ratrace> Lope: I wouldn't want to :)
1796[14:21:45] <greycat> unborn: I don't do virtual
machinery
1797[14:21:46] <ratrace> unborn: there's no OR in that
relation
1798[14:21:52] <Lope> ratrace, well if you're on a games
only machine, it's fine :p
1809[14:24:07] <unborn> Lope: I agree! ratrace steam? pornhub is
better.. or perhaps I am too old for games.....oh wait dedicated
gaming console is best ;)
1810[14:24:08] <ratrace> technically, proton is a nice VALUE
ADDED thing Valve did to WINE which is a gift from linux gods :)
1847[14:30:18] <Piraty> i assume python is a meta package
depending on python2 currently, which will be advanced to depend on
python3 once py2 is dead in debian-stable ?
1865[14:33:23] <BCMM> Piraty: python2's upstream EOL is very
soon, and efforts are underway to remove it from debian, as it will
have no security support
1866[14:33:37] <Piraty> i know that it's EOL, hence my
questions
1867[14:33:59] <BCMM> Piraty: no, it is not going to be
indefinitely included in future debian versions after it is EOL
1877[14:35:44] <BCMM> and if it's not included, the
"python" package obviously can't depend on it
1878[14:35:58] <greycat> 09:32 themill> the intention is that
there will be no package called "python" in the next
release.
1879[14:36:06] <BCMM> Piraty: i'm not sure what you intended
to ask, but i'm trying to answer <Piraty> i.e. it
installed python2 in debian11,12,13 still?
1880[14:36:26] <ratrace> Piraty: this is not (just) Debian
packaging decision. Python mandates what {/usr}/bin/python is meant
to be. Currently with PEP-394.
1881[14:36:50] <ratrace> also don't conflate path to
`python` binary and what it represents, with package name.
1882[14:37:30] <Piraty> ratrace: so if PEP-394 mandates python2
to live behind /usr/bin/python, why not setup that symlink then
1892[14:38:49] <greycat> ratrace: near as I can tell,
they're using a broken Docker image which is missing that
symlink, and they are sitting here arguing about it for an hour
instead of just creating it.
1893[14:39:00] <ratrace> greycat: agreed
1894[14:39:11] <unborn> Lope: i have storage parameter error
40960m requested allocation > 9154m available... got more then
100Gib on disk
1895[14:39:36] <Piraty> i'm trying to help here and report a
possible inconsistency
1896[14:39:48] <Piraty> won't do again, promise
1897[14:40:08] <BCMM> Piraty: you installed a package called
python2 and got a command called python2. if you install the package
called python, you'll get a command called python.
1904[14:41:27] <unborn> Lope: Unable to complete install:
'unsupported configuration: CPU mode 'custom' for
x86_64 kvm domain on x86_64 host is not supported by
hypervisor'
1905[14:41:33] <unborn> whats that?
1906[14:41:35] <BCMM> ratrace: python2 depends directly on
python2-minimal
1907[14:41:41] <ratrace> BCMM: doesn't {/usr}/bin/python
exist even with just python2.7 package installed?
1908[14:42:10] <BCMM> ratrace: python depends on python-minimal
(the package that provides the /usr/bin/python symlink)
1909[14:42:17] <greycat> /usr/bin/python, /usr/bin/python2 and
/usr/bin/python2.7 come from the package named
"python-minimal" in buster.
1935[14:44:06] <ratrace> as in, it won't be installed with
"regular" python2.7 (or -minimal) package, no
metapackages?
1936[14:44:10] <greycat> My eyes are not good enough to
immediately see that there are three *different* things named
python*-minimal in the output results
1937[14:44:22] <ratrace> I can't test myself right now...
1938[14:44:33] <BCMM> ratrace: yup. it's not an alternatives
thing, it's just a regular old symlink owned by the
python-minimal package
1939[14:44:33] <ratrace> and apt-file aint' helpful :)
1952[14:48:10] <ratrace> BTW... I rescind all I said about
PEP-394, it seems it got changed since 2011 and now `python` MAY
refer to pyhton3 ..... *flips table* *goes away yelling and
grumbling about revisions of existing PEPs instead of having new
PEPs that explicitly disable/outdate older ones*
1964[14:52:05] <greycat> "env python3" would be better,
right?
1965[14:52:18] <Habbie> yes
1966[14:52:19] <ratrace> yea that's what I mean
1967[14:52:27] <BCMM> greycat: such a script will fail almost
everywhere else, too. which is why people don't do that.
1968[14:52:50] <ratrace> it's made even more complicated by
the fact there's no `python` restrictions in virtualenvs, or at
least there wasn't before when `python` was supposed to be
strictly just 2.x
1969[14:52:52] <Habbie> i haven't read the entire backlog,
but my position has been for a while that scripts should use
'env python2' or 'env python3' or be 100%
explicit
1970[14:53:05] <BCMM> ratrace: no, `env python` is still
inappropriate for a script that requires python3
1971[14:53:32] <themill> unless you're on arch and you
don't care about what breakage you force onto everyone else
1977[14:55:46] <ratrace> "For scripts that are only expected
to be run in an activated virtual environment, shebang lines can be
written as #!/usr/bin/env python, as this instructs the script to
respect the active virtual environment"
1978[14:56:21] <Habbie> yes, my py3 venv has 'python'
symlinked to 'python3'
1979[14:56:23] <ratrace> which means the version is implied
through creation of the venv and the programmer knows, implicitly,
the version, which is contrary to PEP-20 :)
1980[14:56:24] <Habbie> and that's not something i did
explicitly
1994[15:00:07] <ratrace> the sooner we move to python3 only and
forget 2.x ever existed, the better. but then we'll gleefully
embrace other problems like point release.... because damn, girl, u
wants some asyncs? you MUST require 3.7 or no python for u!
2076[15:22:10] <horribleprogram> Ede|Popede: so I gotta learn how
alsa vs pulseaudio works
2077[15:23:17] <horribleprogram> Ede|Popede: it's a nice
project because I've never really dealt with audio, how it
works, etc
2078[15:23:27] <Ede|Popede> horribleprogram: i'd say more
"how do i tell mpd what to do using HTTP and how do i grab its
data", though for the latter i'm not sure if it
wouldn't embed (optionally at least) things into its stream (i
remember i may have some mp3 here with an embedded cover)
2079[15:23:39] <wrksx> thanks Habbie and greycat
2080[15:24:07] <Ede|Popede> indeed. and using a webbrowser
(espacially those modern days monsters) just to listen to music is a
mega overkill.
2134[15:55:54] <graphicsv> jelly: wow. seriously? that's
quite fun. I was thinking about packaging latest version of Mesa and
Vulkan SDK on Debian buster. What'd say?
2158[16:05:57] <vinod> Hi. I have a canon Image Class M241d
printer. IT's scanner isnt working .. Kindly help.EVen Simple
Sane, xsane has been installed, still not detecting
2159[16:06:06] *** Quits: inerkick (6a4c14b1@replaced-ip) (Remote host closed the connection)
2178[16:08:57] <jhutchins_wk> vinod: less /var/log/messages,
dmesg | less
2179[16:09:15] <BCMM> vinod: this is just plugged in with USB
right? it's not the network version of that device?
2180[16:09:33] <vinod> i have connected with USB
2181[16:10:30] <greycat> Basic guide to reading large log files
with less: Shift-G jumps to the bottom. b goes back a page. Space
goes forward a page. 1 Shift-G jumps to the top. / lets you search
forward (downward) for a string or regex. n repeats a search.
Shift-N repeats the search in the opposite direction. ? lets you
searhc backwards (upwards).
2219[16:18:27] <SaveTheChildren> mquin (the freenode staffer) is
running a massive server farm at 208.67.222.220 and we need to all
DDoS it in the name of justice.
2238[16:30:03] <aZz7eCh> hi #debian - what would be the easiest
way to check sshfs mount is definitely mounted properly before rsync
con job runs? sometimes (rarely) the file server may have been
disconnected - powerout/cable knocked out - who knows what reason,
but it causes loss of connection on my desktop here. I have to rerun
my script to reconnect the sshfs connection. if i dont.. i ened up
with my system trying to back up to /mnt/sshfs_folder* which is
local,
2239[16:30:04] <aZz7eCh> instead of remote obviously.
2240[16:30:34] <greycat> there is a "mountpoint"
command that will tell you whether a directory is a mount point
2258[16:33:47] <aZz7eCh> greycat, so, change my cron jobs to
running a bash script instead that calls rsync, using mountpoint
command as a condition ... k
2277[16:41:33] <jelly> aZz7eCh: I did not know of the mountpoint
command. Nevertheless there are 2-3 different failure modes.
2278[16:43:14] <aZz7eCh> alright thanks for the info. I've
downloaded ya script so if I end up facing troubles with my first
method I will refer back to your way
2281[16:44:23] <aZz7eCh> ((never had an issue with sshfs
connection - by far the fastest way of connecting and always been
reliable. only issues i have encountered are when actual power out
takes the server offline or cable gets knocked out for extended
period of time. obvious reasons.
2282[16:45:07] <greycat> Oh, if the symptom is "it's
mounted but nonresponsive because server is down", then
mountpoint probably won't handle that.
2283[16:45:21] <greycat> In cases like that, NFS is even worse.
2284[16:45:24] <aZz7eCh> greycat, it does i've just tested
it.
2285[16:45:36] <greycat> So it's being unmounted on the
client? Good.
2286[16:46:30] <aZz7eCh> mountpoint shows it as NOT a mountpoint
if the sshfs connection is no longer existant. and if i rerun my
script to mount it - it doesn't return an error like
"already mounted/in use" or anything, it just does it
2304[16:54:25] <Ede|Popede> lxde PLUS xfce? no idea what effects
this may have. one single daemon running with the command to turn of
the screen after 10 minutes is enough.
2306[16:55:16] <Ede|Popede> in xfce's power manager window 3
of 4 tabs have some relevant settings.
2307[16:55:35] <Haohmaru> there is also another little weirdness
that might be related.. when i suspend, it sorta goes into
screenlock mode + turns off the monitor, doesn't actually
suspend
2333[17:06:42] <teclo> hi, back after some time... for my
firewall rules on Debian 10.1, the wiki says I should do
iptables-restore < /etc/iptables.rules ... but where do I put
"echo 1 > /proc/sys/net/ipv4/ip_forward" then ?
2334[17:06:57] <jelly> Lope: you can run /usr/share/mdadm/mkconf
and overwrite the current file with that output
2335[17:06:57] <teclo> I can't put it in /etc/iptables.rules
I think
2336[17:07:40] <jelly> Lope: and that's basically a wrapper
for "mdadm --examine --scan"
2337[17:08:40] <Ede|Popede> Haohmaru: why do you even mix 2 DEs?
do you really need something from both of them? you could switch to
a text terminal and see what's running inside your login
manager tree. or check all the relevant programs from BOTH DEs. one
by one, tab by tab, setting by setting.
2412[17:50:00] <maxmoon> Hi, I am looking for a
"ready-to-use" Linux distribution, which is minimalistic,
privacy friendly and can run on older netbooks/notebooks. I only
want to change the window manager to i3, but the whole system should
be running out of the box, like audio should work, usb should mount
automatically (support for most file systems), etc. Any idea? Thanks
for your help.
2463[18:05:54] <dpkg> Edit /etc/apt/sources.list, ensure that the
two main Debian mirror lines end with "main contrib
non-free" rather than just "main", then
«apt-get update». But bear in mind that you'll be
installing <non-free> software. These may have onerous terms;
check the licenses. See also <sources.list>.
2477[18:09:11] <greycat> If it's a desktop system, onboard
ethernet, motherboard video chipset, about 3-6 years old, there is a
very good chance it'll work out of the box if you started with
the nonfree firmware installer.
2478[18:09:24] <jelly> maxmoon: if you're very very careful
to choose hardware that's 100% supported by the versions of
software in Debian release you're installing, yes
2483[18:09:46] <greycat> If it's a laptop, if it uses USB
wifi, if it has an nvidia card, or if it's got a hybrid
nvidia+intel video combo, you may have to struggle.
2484[18:09:54] <jelly> amd GPUs will need firmware
2485[18:10:09] <greycat> I'm hoping the nonfree firmware
installer has the GPU firmware.
2502[18:16:12] <greycat> it's in sid, and judd thinks you
can backport it
2503[18:16:43] <Miles8of9> i'm using stable... sid is
unstable
2504[18:16:56] <greycat> !simple sid backport
2505[18:16:56] <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>.
2506[18:16:57] <Miles8of9> should i migrate tu unstable?
2507[18:17:11] <Tinerion> Hi ! I have a problem with a package
used by my scan/printer and nammed brscan3. When I want to install
it, it say "The package brscan3 must be reinstall, but
it's impossible to find archive", and when I want to
uninstall it completely, it say "brscan3 is in incoherent
state, you should reinstall it before try to delete it".
2513[18:18:32] <Miles8of9> debian-backports is a repository?
2514[18:18:56] <greycat> Yes, but this package isn't *in*
it. I checked. That's when I checked whether it could *be*
backported by an ordinary person. Judd said yes.
2535[18:26:00] <judd> Backporting package
nvidia-legacy-304xx-driver in sid→buster/amd64: all
build-dependencies satisfied using buster.
2536[18:26:04] <Tinerion> jhutchins_wk with a file downloaded on
the official website of my printer, and which don't seem to
work, I was thinking that the installation has totally failed, but
it's seem to have fail during the installation.
2537[18:26:13] <greycat> And voila, judd says it can BE
BACKPORTED BY AN END USER.
2538[18:26:29] <greycat> Whether that's really truly true,
who the hell knows, but it's what the bot said, so it's
worth a try if you need it.
2548[18:29:19] <Miles8of9> no, i don't understand..
it's not in buster-backports, ok, it's on sid, jessie and
stretch, so if i have buster i can't install it... right?
2549[18:29:27] <greycat> using upstream's video drivers is
generally NOT recommended
2558[18:30:10] <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>.
2559[18:30:14] <greycat> Follow THOSE instructions.
2577[18:37:58] <Ether_Man> Anyone know of anything similar to
OpenMediaVault, but is more, shall we call it native to Debian? As
in, using debian's own repos and updated together with that,
rather than use its own thing completely using OldStable and
OldOldStable?
2578[18:38:26] <jhutchins_wk> Tinerion: Unfortunately, the only
way you're going to be able to get support for that is from the
manufacturer, and they often aren't helpful with linux.
2579[18:38:47] *** Quits: PwnKitteh (~PwnKitteh@replaced-ip) (Remote host closed the connection)
2585[18:40:31] <Tinerion> jhutchins_wk okay, there is no way to
uninstall a package in this situation? It's not important if I
don't install it, I just want to delete it
2586[18:40:32] *** Quits: PwnKitteh (~PwnKitteh@replaced-ip) (Remote host closed the connection)
2590[18:41:28] *** Quits: PwnKitteh (~PwnKitteh@replaced-ip) (Remote host closed the connection)
2591[18:41:30] <karlpinc> Tinerion: That's what I was
thinking. There must be a way to delete it. (I don't suppose
there's a ppd file or something like that hidden inside?)
2610[18:44:16] <Tinerion> karlpinc no ppd file here :/
jhutchins_wk it's a .deb file
2611[18:44:23] *** Quits: oish (~charlie@replaced-ip) (Remote host closed the connection)
2612[18:44:36] <Ether_Man> Mathisen, not quite. It gives a
unified configuration space for multiple services. Like I can as an
example configure both NFS and SMB from a single place.
2613[18:44:36] <karlpinc> cinesc: You asked already. It's
polite to wait longer. I don't see geteltorico anywhere. (But
I'm running stretch.)
2615[18:45:24] <greycat> It's also polite to fix your typos
and to expand upon your initial question to reflect the knowledge
you have gained while waiting.
2616[18:45:30] <jhutchins_wk> Ether_Man: Sure you can. The
command line.
2617[18:45:35] <greycat> I'm guessing you made a typo from
the original "El Torito".
2618[18:45:35] <karlpinc> Tinerion: Take a look at the apt logs.
And look at the package itself. .deb files have a list of stuff
that's installed. I think. Which means it could be un-installed
manually in theory. Your real problem is what it's done to
apt's database. I don't suppose you do backups?
That'd be the easy answer.
2619[18:45:41] <jhutchins_wk> Ether_Man: It's a matter of
editing text files.
2631[18:48:26] <Mathisen> Ether_Man, i see, well i would still
argue you can do that on a regular debian :) i do understand your
point im not trying to be smart ass here.. for example webmin
alterntives like
replaced-url
2645[18:51:17] <karlpinc> Tinerion: Yes, follow
jhutchins_wk's advice. (I thought you'd already tried and
failed to remove the package with apt's tools.)
2646[18:51:31] <jhutchins_wk> Tinerion: There are a lot of
suggestions on the web, and you can check the manpages for those
tools to see if they have a "--force" option.
2650[18:52:03] <wasamasa> cinesc: one is the name of a package,
the other a name of a command
2651[18:52:17] <wasamasa> cinesc: the package doesn't need
to have the same name as the command it provides
2652[18:52:24] <karlpinc> Tinerion: It is always possible that by
opening up the vendor's .deb you might find something inside
that works. (But I'd steer clear of executable code.)
2653[18:52:47] <jhutchins_wk> !force remove
2654[18:52:47] <dpkg> [force remove] 'dpkg -P
package-name'; if no joy, 'dpkg -L package-name' this
will give you a list of files to delete when the next step is done.
Go to /var/lib/dpkg/info and 'rm package-name.postrm'
&& 'apt-get remove --purge package-name'. Remove
listed files (dpkg -L) and report a bug if it is the case.
2655[18:52:52] <cinesc> ok that makes more sense now
2661[18:55:50] <jhutchins_wk> Tinerion: This kind of thing
happens when a package is built for a different distribution
(ubuntu) or for a significantly different release.
2666[18:57:20] <Tinerion> okay, I see! dpkg -L seem seems to be
the good solution, i'll try it. jhutchins_wk so that should be
because it's a old printer and the package is probably not
update since a lot of time, right?
2680[19:00:24] <jhutchins_wk> Yeah, I have a similar problem with
a Brother 2040. 32bit wrapper won't install on 64b, Brother has
no reason to write a new driver when they can just sell you a new
printer.
2731[19:28:01] <wxb> I'm going to read more on this, but
just so I don't waste time, is it possible to dual boot Debian
Linux and to LUKS the Debian part?
2745[19:35:35] <trek00> wxb: i can't speak about windows,
but it should be no difference installing a plain or a luks
encrypted linux in a dual boot configuration
2811[20:00:19] <trek00> wxb: if it's your first time, you
should make a backup of your data before partitioning, or better
make some tests on an empty harddisk
2812[20:00:21] <aaii> p8m, I dunno, I change my script to send
messages by IRC way that's easier.
2813[20:00:23] *** Quits: Immanuel (~Manu@replaced-ip) (Ping timeout: 276 seconds)
2814[20:01:08] *** Joins: Immanuel (~Manu@replaced-ip)
2831[20:06:50] <_DeLa_> errm... from <dmm remove>: is
"dpkg --remove --force-depends $(aptitude search
'?narrow(?version(CURRENT),?origin(Unofficial Multimedia
Packages))' --disable-columns -F%p)" without quotes really
one long terminal command I am supposed to run?
2832[20:07:01] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
2854[20:13:01] <apteryx> yesterday someone mentioned in #emacs
that Debian is using 'package.el' to manage Debian's
Emacs packages initialization in site-start.el. I've looked at
the source packages of Debian, but didn't see site-start.el
being customized thiat way. Could someone confirm/infirm that this
is done?
2859[20:16:02] <karlpinc> apteryx: My site-start.el says to look
at /usr/share/doc/emacsen-common/debian-emacs-policy.gz. But really,
look at sources.debian.org at the patches and see if there's a
patch that does something. That'd be definitive.
2881[20:24:39] <Ankman> Mathisen: i am also admin of a Mint
computer of my neighbor. it has some systemd files but lacks
systemctl and others. since it's debian based that should be
ontopic here
2885[20:25:30] <wxb> Ankman, Mint is based on ubuntu, no....?
2886[20:25:40] <greycat> it's not safe to assume anything
about Debian based on one of its derivatives, *especially* when it
comes to networking or init systems
2887[20:25:44] <Ankman> isn't ubunti based on debian?
2900[20:27:47] <greycat> the popcon data answers the poll, sort
of, but it's a self-selected sample of installed systems, and I
have a funny feeling that the people who choose sysvinit are
strongly correlated with people who would opt out of popcon
2901[20:27:51] <Ankman> well if ubuntu is based on debian and
mint is based on ubuntu (which is based on debian)...
2904[20:28:25] <Ether_Man> Is there any package for playing some
kind of sound over pc speaker when bootup is complete and when
shutdown/reboot is initiated? I can seemingly only find how to do it
for x upon login which is completely different from what I want
2907[20:29:05] <Ether_Man> Have to create my own startup script?
2908[20:29:06] <trek00> greycat: yes, i was asking here because
of this
2909[20:29:12] <greycat> You wouldn't need a
"package". Just write a script to do it, and write a
systemd unit to launch that script at the appropriate times.
2918[20:31:37] <trek00> greycat: i'm wondering if a
conversion script from systemd to *init is still useful to debian or
not
2919[20:31:39] <Ankman> wxb: on the other hand ubuntu went its on
way using upstart instead of service. although service might still
had been supported
2920[20:31:45] <greycat> !stretch sysvinit
2921[20:31:45] <dpkg> To replace systemd with sysvinit in
stretch, simply "apt-get install sysvinit-core" and
reboot. To switch from systemd to runit in buster or sid,
"apt-get install runit-systemd", reboot, "apt-get
install runit-init", and reboot again.
2922[20:31:51] <Ankman> p8m: thanks
2923[20:31:53] <Ether_Man> greycat, that one CAN do something,
doesn't mean it's always preferred. We can keep
reinventing the wheel every time we need one, but it's
inefficient when one can simply use one that already exists
2924[20:32:01] *** Quits: Akuw (~ajsbox@replaced-ip) (Remote host closed the connection)
2933[20:35:50] <Ankman> a decade ago or so i shot a Star NL-10
matrix printer on a flea market just to see if it works under linux,
when the machine still had a parallel port. it worked, but another
complaint came in about the noise. that would not have happened in
the 80s ;-)
2934[20:36:43] <trek00> Ankman: buy an apartment on a busy street
:)
2935[20:36:44] <greycat> A dot matrix printer basically receives
straight ASCII data over the parallel port. No drivers, no
nothin'.
2969[20:50:37] <dpkg> "Does anyone have X or use Y?" is
taking a poll, not asking a good question that IRC helpers can
answer. Don't do it or sussudio's army of militant badgers
will hurt you. Also see <ask> and <bad polls>.
2970[20:51:05] <Ankman> oh
2971[20:51:14] <Ankman> sorry then
2972[20:51:18] <greycat> /msg judd popcon irssi
2973[20:52:33] <p8m> Ankman: weechat is the hotness these days
2981[20:56:53] <kopax> I have a 4G modem in my laptop. I can
send/receive SMS using my card, so it means the modem work, I just
cant browse online. Any idea how to do ?
3009[21:08:26] <greycat> You have two wireless interfaces,
wlp0s20f3 and wwp0s20f0u7. The former is "UP", so
it's apparently the one that's configured and working. The
other one has no addresses assigned, and isn't "UP".
3011[21:09:00] <Xalys> I have libnss-extrausers installed. When
specifying a user in /var/lib/extrausers/{passwd,shadow}, and having
the primary group of that user being a group in /etc/group (not
libnss-extrausers), the user is unknown (id: ‘xalys’: no
such user). When changing the GID to one that is in
/var/lib/extrausers/group, everything is ok. How do I fix this?
3027[21:12:07] <Xalys> karlpinc: yes, nsswitch.conf doesn't
seem to have any options related to this. I've peeked in the
docs but this seems to be undocumented. Hoping someone is in this
channel who knows libnss-extrausers from its source...
3031[21:12:52] <karlpinc> Xalys: Try the readme.debian.
3032[21:12:54] <karlpinc> !readme.debian
3033[21:12:55] <dpkg> README.Debian (or README.Debian.gz) is a
document found in the /usr/share/doc/$packagename/ which explains
any Debian specific details in the package's operation or
configuration.
3034[21:13:22] <Ankman> kopax: does "ping -c 1
192.168.0.1" return a success?
3035[21:13:33] <kopax> of course it is my LAN router
3036[21:13:38] <Xalys> I did. Nothing there
3037[21:13:53] <Ankman> so the route is set.you should have
access to the internet
3039[21:14:32] <karlpinc> kopax: 'traceroute' might
provide more info.
3040[21:14:49] <kopax> Ankman, the route is set but through the
WIFI, if I disable the WIFI I cant speak to you because I have no
internet, later I want to use the 4G not the wifi
3041[21:15:01] <kopax> that wont help
3042[21:15:09] <kopax> karlpinc, the 4g card is not connected to
any network
3043[21:15:14] <Ankman> may be the DNS is messed up. try to load
172.217.20.142 in your browser
3044[21:15:16] <Ankman> oh
3045[21:15:18] <kopax> it just as the GSM network but no internet
network
3046[21:15:19] <wxb> Flash drive is in, install on its way.
3059[21:16:46] <wxb> karlpinc, Ah, no worries. Thanks anyway.
I'll have to find the missing driver after the install.
I'm at the partition step. Should I select 'manual'
and then create space for the Debian install?
3065[21:18:26] <karlpinc> wxb: You can also go to a vt and really
do it manually.
3066[21:18:42] <wxb> karlpinc, Sorry, vt?
3067[21:18:46] <karlpinc> !vt
3068[21:18:47] <dpkg> well, vt is Virtual Terminal, like the
console; change VTs using Alt+Left/Right or Alt+F1, Alt+F2 etc. To
get from X to a VT, use Ctrl+Alt+F1 (and Alt+F7 to get back, most
likely). You can also use "chvt" to switch VT. VT is also
used to mean Intel VT-x or VT-d (virtualization support; the AMD
counterpart is called AMD-V or Pacifica), ask me about <xve>.
3085[21:23:36] <Ankman> then let the debian installer let install
itself. it might ask to encrypt Debian. go for it
3086[21:23:39] <maxrazer> Well, Debian locked up on me for the
first time in a long time. Running unstable. Exited Zandronum Doom
port and XFCE became non-responsive a second after that. No virtual
terminals would respond either.
3102[21:29:00] <wxb> The LVM option says that it will completely
wipe the disk...
3103[21:29:43] <Ankman> oh
3104[21:30:00] <maxrazer> Ankman, I don't understand what
this does and it isn't working. I already rebooted. Can you
point me somewhere and I'll read about it?
3105[21:30:21] <wxb> So, what's the best way to go about
dual booting? This whole partition things is difficult for me...
3106[21:30:33] <maxrazer> wxb, I concluded two drives is the
best.
3112[21:31:06] <wxb> maxrazer, Partitions are a pain, but I have
no choice.
3113[21:31:21] <kopax> Hi again, I am able to connect to my
network by doing :
3114[21:31:23] <maxrazer> wxb, I got two ssd drives when I built
my PC so I could mess around with linux.
3115[21:31:24] <wxb> trek00, No, I don't know how/when to do
it...
3116[21:31:28] <kopax> nmcli c add type gsm ifname cdc-wdm1
con-name dka-free apn free
3117[21:31:43] <kopax> i have a network, dns server configured,
but still cant connect to internet
3118[21:31:44] <wxb> maxrazer, Good idea. This is for someone
else's machine.
3119[21:31:49] <greycat> trek00: I suggested they make one big
partition for files, and one small partition for swap. They demanded
a guide. I got fed up. They declared they'd ignored me.
3122[21:31:59] <wxb> And they want to keep Windows JIC
3123[21:32:08] <karlpinc> Ankman: Better to use LUKS encryption
and do it at install time.
3124[21:32:29] <wxb> karlpinc, I agree.
3125[21:32:32] <maxrazer> wxb, I'd pretty much always keep
windows. I guess you could even think of Linux as a backup if
windows fails.
3126[21:32:35] <Ankman> karlpinc: no idea if the Debian installer
supports this
3127[21:32:37] <wxb> But should I partition before install?
3128[21:32:46] <trek00> wxb: yes
3129[21:33:00] <Ankman> wxb: if you want to keep windows, yes
3130[21:33:04] <trek00> wxb: you need to create a linux partition
at least
3131[21:33:12] <wxb> So I should stop the install process, and
partition with a different tool than the Debian Installer?
3132[21:33:20] <wxb> And then come back and install?
3133[21:33:23] <karlpinc> wxb: And do whatever to make sure that
the windows file system is shrunk and fits in the partition it uses.
3134[21:33:23] <trek00> wxb: to create a new partition you need
free disk space
3135[21:33:24] <maxrazer> wxb, I think some distros have
something to make it easier to dual boot on one drive and can resize
stuff. I don't think debian helps you do that much.
3136[21:33:40] *** Quits: hele (~hele@replaced-ip) (Ping timeout: 248 seconds)
3137[21:33:44] <kopax> This is my ifconfig :
replaced-url
3138[21:33:46] <kopax> any idea why ?
3139[21:33:48] <karlpinc> wxb: I don't know if the installer
will help. But do see the debian wiki.
3140[21:33:52] <wxb> maxrazer, No, some people find partition so
easy. I don't.
3141[21:34:07] <Ankman> wxb: after Debian is fully installed next
to windows you have the opportunity to envrypt the Debian partition
3142[21:34:09] <wxb> trek00, Should I do that during the install,
or before?
3143[21:34:15] <karlpinc> wxb: And see the install guide. It has
info on installing to dual boot.
3144[21:34:22] <trek00> kopax: are you able to ping your router?
3145[21:34:38] *** Quits: alexertech (~xb@replaced-ip) (Quit: Fades into the darkness)
3146[21:34:49] <trek00> wxb: if you have free space on the disk,
you will create a linux partition during install
3214[21:47:47] <dvs> kopax, you need to connect another
laptop/desktop to chat so you can change the setting on the one
you're currently using.
3215[21:47:48] <trek00> kopax: not, but please help me to
understand your network setup: you are on a pc and want to connect
via wifi to this linux and then let this linux to connect to the
internet via another network, right?
3216[21:48:01] <kopax> Ok
3217[21:48:05] <kopax> I am starting up another laptop
3218[21:48:09] <Ankman> ip route add default via "what ever
the IP of your GSM device is"
3219[21:48:38] <wxb> This is an ordeal... Windows (not
Debian's problem) is having issues freeing up space forthe
partition. I'm in the Windows tools and there is free space.
Anyway. Not aksing for help on that here, just waiting for the job
to finish...
3222[21:49:04] *** judgmental is now known as FleaFart
3223[21:51:37] <wxb> I'm confused, the type is x64, and it
says :C is SDD, do I need the amd64 iso?
3224[21:51:45] <wxb> *SSD
3225[21:52:16] <trek00> wxb: yes, if your system is 64-bit, you
need amd64 iso
3226[21:52:33] <Ankman> wxb: haven't done partioning for
decades. back in 1998 i used defrag.exe in windows to optimize the
sapce. then ran fips.exe (holy crap, this is long ago so today you
don't do this) to partion the hard drive. then booted from the
CD with linux on and installed it in new created partition. i would
assume today this is much easier
3227[21:52:35] <wxb> trek00, Thanks. I thought so, but was unsure
what the SSD changed.
3228[21:53:23] <wxb> Ankman, I'm using the Windows tool, but
it's telling me that it can only free up like 666M for me even
thoug like 12G is free. Sigh! lol
3229[21:53:28] <dvs> Ankman, today it's called partimage.
3234[21:55:15] *** Quits: donofrio (~donofrio@replaced-ip) (Remote host closed the connection)
3235[21:55:25] <dka> I have now only 1 default gatewat
3236[21:55:32] <Ankman> wxb: try to boot linux from usb/dvd and
initiate the installer. it should allow you to shrink the windows
partition to make space for Debian
3237[21:55:37] <dka> I have no more internet because I have
switch off the wifi
3251[22:00:17] <trek00> dka: what is your network setup? both
interfaces are connected to the internet?
3252[22:00:32] <v0idpwn> Hey. I'm trying to install debian
on another computer, and debian installer gives me a glitched screen
both for the graphical mode and text mode. What could be happening?
3253[22:00:42] <wxb> wxb, They are like 'we don't help
with dual-boot' lol I'm asking how to use a bloody winsows
tool and they can't help. lol
3254[22:00:53] *** Quits: archios (~archios@replaced-ip) (Remote host closed the connection)
3255[22:01:10] <Ankman> v0idpwn: try to use the non GUI installer
3269[22:03:15] <wxb> trek00, I did defrag. To no avail.
3270[22:03:26] <karlpinc> v0idpwn: Are you using the unoffical
installer with non-free firmware included? That can help with
hardware issues. (I don't know how much it ever matters with
video.)
3271[22:03:35] <v0idpwn> trek00: just after selecting the mode I
want
3272[22:04:02] <v0idpwn> karlpinc: I'm not. Should I? Is an
almost "default" build with a Ryzen 3 2200G
3319[22:15:25] <wxb> maxrazer, karlpinc Thanks for the ideas for
Windows. Yep. Not the subject here for support. I think I might use
Gparted to go in there and do the partition.
3325[22:18:05] <karlpinc> wxb: VDH is MS's virtual drives,
which live (presumably) inside regular partitions. So there'd
be 3 layers, the file system, the virtual drive, and the containing
partition. (And then there's the actual drive.)
3326[22:18:09] <kopax> My question is am I missing anything in
/etc/NetworkManager/system-connections/dka-free
replaced-url
3327[22:18:35] <trek00> karlpinc: it seems debian installer can
shrink ntfs partitions
3328[22:18:47] <wxb> trek00, That would be good news.
3331[22:19:04] <karlpinc> trek00: A partition containing VHD
would not be a NTFS partition.
3332[22:19:34] <trek00> karlpinc: oh sorry, i don't even
know what a vhd is :)
3333[22:19:46] <karlpinc> It would be a partition containing VHD,
which would contain partitions and those would hold NTFS file
systems.
3334[22:20:52] <karlpinc> trek00: Like a partition can hold a lvm
pv, which then contains more stuff. Except that this is MS Windows
so everything in totally opaque.
3335[22:22:00] <trek00> karlpinc: well another
ms-not-invented-here-but-renamed-stuff :)
3336[22:22:13] <karlpinc> wxb: I'm just guessing, having
bumped into this before.
3337[22:22:30] *** Quits: kopax (~kopax@replaced-ip) (Remote host closed the connection)
3348[22:25:11] <Gigglebyte> Is it possible to assign folders
different profiles so that some of them open automatically with
descending order and others open automatically with descending?
3349[22:25:49] <v0idpwn> How do I access this command line?
3356[22:26:51] <karlpinc> v0idpwn: I forget. But it's pretty
much the first thing that happens after you boot from the install
media. Maybe you need to choose expert mode (or otherwise read the
screen). I bet the install guide has instructions.
3357[22:27:00] <karlpinc> v0idpwn: It's like the grub boot
menu.
3377[22:35:36] *** Joins: marek (~marek@replaced-ip)
3378[22:35:39] <trek00> v0idpwn: check this
replaced-url
3379[22:35:41] *** Parts: marek (~marek@replaced-ip) ()
3380[22:36:12] <trek00> v0idpwn: If you wish or need to add any
boot parameters for either the installer or the kernel, press Tab
(BIOS boot), or e then down arrow three times then end (UEFI boot).
This will bring the boot command for the selected menu entry and
allow you to edit it to suit your needs.
3474[23:10:47] <trek00> Bacteria: i don't have nvidia,
sorry, usually i would install latest drivers and try with different
configuration (are you using Xorg or wayland?)
3477[23:11:07] <trek00> kadiro: now paste the contents of
/tmp/output.txt
3478[23:11:18] <kadiro> I had to cat that output to get that link
3479[23:11:30] <kadiro> yes i did that
3480[23:11:52] <kadiro> It doesn't support one line output
3481[23:12:29] <Bacteria> xorg
3482[23:12:32] <trek00> kadiro: it should be ok, but you are
using a vte based terminal, may be you need to close all the
terminals and may be kill the vte daemon before try again
3483[23:12:42] <Bacteria> i doubt its drivers
3484[23:13:08] <Bacteria> actually i have no clue
3485[23:13:11] <kadiro> trek00, can i remove the script inside
the profile.d ?
3530[23:22:11] <trek00> kadiro: well, about the path problem, the
setup of the shell seems ok, but if echo $PATH does not show the
correct, something gone wrong, i would try with a different
terminal, also logging from the console should show the correct PATH
3531[23:22:17] <kadiro> I think I need the original bashrc file,
it could be wrong or missed something
3532[23:22:23] <rpifan> i just need to get bluetooth
3566[23:28:59] <dka> how to execute script as root on Buster
after startuP?
3567[23:29:12] <trek00> kadiro: i read somewhere, vte based
terminals can use a daemon to speed up the startup process, may be
you need to restart that daemon? I would try installing another
terminal to see if it is working fine
3568[23:29:58] <kadiro> trek00, xfce-terminal and xterm have no
bin path but from the tty2 it is there
3569[23:30:34] <kadiro> I think they both read the same mistake
3570[23:30:41] <trek00> kadiro: i don't know how to fix
this, but you .profile/.bashrc are fine, the issue is somewhere out
of here
3571[23:30:52] <kadiro> ah
3572[23:31:19] <trek00> kadiro: well, if even xterm don't
have PATH, it's not a terminal issue
3573[23:31:22] *** Quits: jnavila (~jnavila@replaced-ip) (Remote host closed the connection)
3574[23:31:32] <trek00> kadiro: may be the display manager, are
you using lightdm?
3575[23:31:39] <kadiro> yes
3576[23:32:05] <dka> is there a /etc/host.d for /etc/hosts ? to
override conf by adding file ?
3641[23:48:56] <karlpinc> There are ways of specifying how hosts
are looked up in /etc/nsswitch.conf. But I don't know what all
the options are and can pretty safely say that uninformed messing
about will lead to trouble.
3642[23:49:09] <kadiro> trek00, it seems my preference
doesn't have that option about utmp
3643[23:49:10] <karlpinc> dka: ^
3644[23:49:30] <HelloShitty> they sai it could be offensive to
kids in case kids were on the channel (channel subject was
electronics, :s)
3698[23:56:39] <kadiro> somiaj, i will look into that just
curious why it is not at login by default
3699[23:57:00] <HelloShitty> There are 1500 people here making
this channel "low quality"... Also this is too many years
of "low quality"
3700[23:57:04] <ratrace> kadiro: also worth mentioning,
display/login managers dealing with PATH completely differently
(because they don't source any shell rc on login)
3701[23:57:06] <HelloShitty> Debian FTW
3702[23:57:12] <somiaj> kadiro: personally I use method in
.profile from /etc/skel/, which has the following
replaced-url
3707[23:57:40] <somiaj> kadiro: And then I put everything I want
in .bashrc, this way my login and interactive shells are
effictivally the same when running bash.
3708[23:57:44] *** Quits: tagomago (~tagomago@replaced-ip) (Remote host closed the connection)