64[01:27:30] <queip> indeed, can you remind me? on any card (e.g. on Intel), can one get even basic Xorg desktop, with only pure main repository of debian stable? or, any plans for it....?? regular VGA/VESA should work with no firmware afaik....
65[01:28:05] <queip> roger21: change to user and run command "startx" "startx xfce4" "xfce4start" and such. Probably you will need to install drivers as above; check also dmesg, /var/log/* (by time)
66[01:28:07] <abrotman> There isn't firmware for intel graphics I odn't think ?
67[01:28:26] <abrotman> roger21: how are you installing X? Did you install a display manager?
173[02:15:07] <Aurora_v_kosmose> How do I make user-slices enforce memory limits?
174[02:15:40] <Aurora_v_kosmose> I had a program with known memory issues limited, and it went vastly over it. Attempting to limit it to 4KB (guaranteed crash) failed, and it allocated significantly more.
175[02:15:47] <dannylee> Debian 9.13 might really be G0D...
250[03:36:08] <maxrazer> I'm getting error that configuration data is incorret, can't select the database, even though it is created and I have the hostname, name, username, password all correct.
391[06:09:08] <ryouma> the cryptsetup man page says this, but it does not tell what the tricks are: "There are a few nice tricks for construct‐ ing a fallback, when suddenly out of the blue, your brain refuses to cooperate. These fallbacks need LUKS, as it's only possible with LUKS to have multiple passphrases."
392[06:09:13] <ryouma> so, what are the tricks?
393[06:12:49] <nkuttler> simply add another passphrase? doesn't sound like a trick though
394[06:13:17] <nkuttler> but then, any "trick" sounds like flawed crypto
402[06:22:19] <ryouma> a quick search suggests it might be possible with scripts placed in soem cryptsetup path, which... i guess... ends up in initramfs or so?
403[06:23:38] <ryouma> sounds kind of mendokusai though whatever it is. brittle or maintenance-heavy or confusing which defeats that purpose. idk.
404[06:25:07] <ryouma> i find debian updating of initramfs and grub to be something i don't want to have to mess with if i don't have to. at least ime.
405[06:26:02] <ryouma> (of course, i have had to...)
415[06:45:53] <ryouma> there was a bug in which crucial drive would blow up and lose data if a trim command was sent. what versions of debian have a kernel that will not allow that to occur unless the firmware has been updated on the drive?
416[06:47:57] *** Quits: Conradish006 (~Conradish@replaced-ip) (Quit: Guess I left :D)
433[07:20:47] <dpkg> Pinning is a method to choose which version of a package to install when multiple versions are available from <sources.list>. Bugs are explained at replaced-url
438[07:25:43] <dpkg> To make buster be preferred over other releases in your sources.list, put APT::Default-Release "buster"; in a file in /etc/apt/apt.conf.d/ (create if needed). Or use "stretch", "stable" or "testing" (make it match the name used in /etc/apt/sources.list). Note that some apt front-ends like Synaptic are reported to ignore Default-Release. Also ask me about <tum>.
460[07:56:10] <Lope> if you `echo $PS1` you'll see something like ${debian_chroot:+($debian_chroot)}\u@\h:\w\$
461[07:56:17] <maxrazer> I thought maybe it could be changed in bash. But using hostnamectl also changed it.
462[07:56:25] <Lope> yeah so that will probably be determined by /etc/hostname
463[07:57:10] <Lope> so you can edit that, then run `hostname -F /etc/hostname`
464[07:57:14] <Lope> then disconnect and reconnect
465[07:57:27] <Lope> or maybe just run `bash`
466[07:57:31] <maxrazer> Oh, it looks like that changed too, but not /etc/hosts which still has 127.0.0.1 set to localhost. I guess that would stop 127.0.0.1 from working.
467[07:57:50] <Lope> /etc/hosts is irrelevant to your $PS1 AFAIK
468[07:58:16] <maxrazer> Oh, actually that just means I can type localhost intead of that ip.
469[07:58:18] <Lope> I don't think you should mess with 127.0.0.1 in your /etc/hosts
470[07:58:36] <maxrazer> Tutorials out there say to change that. That confused me.
472[07:59:06] <Lope> I mean you can do this: "127.0.0.1 localhost bob-pc local.bob.com"
473[07:59:12] <StarOnD> I have a query : How does a device mapper work / take a snapshot ? How does it know when the system is in a "consistent" state, how does it know when to take an atomic snapshot?
474[07:59:12] <Lope> As long as localhost comes first.
475[07:59:25] <Lope> But as I said, pretty sure it's irrelevant to your $PS1
476[08:00:07] <Lope> You should ideally set your /etc/hosts and your /etc/hostname
477[08:01:31] <Lope> Now that I've helped you, I forgot what I was doing... :)
478[08:02:52] *** Quits: rare_energy (~quassel@replaced-ip) (Quit: who knows)
480[08:05:02] <Lope> I created a bunch of partitions with `sgdisk` and now `fdisk -l` is upset. "GPT PMBR size mismatch (14745599 != 7579647) will be corrected by write."
588[09:56:27] <ueb> mysql Ver 14.14 Distrib 5.7.12
589[09:56:48] <aminvakil> abff has asked two great questions which are my questions too
590[09:57:30] <abff> also, backports are a thing
591[09:57:44] <hegemoOn> plop
592[09:57:48] <aminvakil> aha
593[09:57:50] <abff> don't ask me how to use em
594[09:58:06] <aminvakil> ueb: to be sure what's the output of "sudo apt-cache policy mysql-server"
595[09:58:32] <ueb> my distro: Debian GNU/Linux 8.4 (jessie)
596[09:58:40] <abff> dpkg backports
597[09:58:40] <dpkg> A backport is a package from a newer Debian branch, compiled from source for an older branch to avoid dependency and <ABI> complications. replaced-url
667[10:28:52] <aminvakil> read mysql documentation if that is supported though
668[10:29:45] <ueb> aminvakil I need to upgrade to stretch for installing php7.3, but this will upgrade mysql to 5.8... isn't it ?
669[10:29:58] <jelly> ueb, php7.3 from deb.sury.org works, but be careful with their metapackages -- if you install eg. "php" it will switch from pulling in php7.3 to php8.0 when you least expect it
675[10:31:03] <abff> ueb: I'm just giving you bad advice I'll leave you to these smart people
676[10:31:06] <aminvakil> ueb: to buster you mean?
677[10:31:20] <ueb> ok thanks jelly
678[10:31:35] <ueb> abff ok thank you too
679[10:31:45] <aminvakil> aha, you want to use sury repo in stretch to install php7.3
680[10:31:51] <abff> gl
681[10:32:31] <ueb> aminvakil I mean stretch, because it's present on replaced-url
682[10:32:52] <aminvakil> jelly: is partial upgrades supported on debian? abff mentioned apt hold ...? is it supported then or not?
683[10:34:07] <aminvakil> ueb: i don't know about sury, but is there a reason you don't upgrade from jessie to stretch and then to buster and use replaced-url
684[10:34:27] <aminvakil> then you can use php package from official debian repos instead of sury
685[10:34:41] <jelly> aminvakil, I'm not sure what you mean by partial upgrades but the answer is probably no
686[10:34:50] <TheBigK> ueb: afaik u should consider switching to mariadb. Is way easier than switching to mysql 5.8...
687[10:35:04] <aminvakil> jelly: preventing a package to be upgraded when there is an upgrade
688[10:35:10] <aminvakil> then upgrade the other packages
689[10:35:17] <TheBigK> by switching to mysql 5.8 u cant switch to mariadb anymore without needging to do dump and reimport... so quite a consideration...
690[10:35:23] <ueb> aminvakil the problem is that I should not upgrade to mysql 5.8
691[10:35:26] <TheBigK> but we use mysql 5.7 on buster as well... works fine
692[10:35:46] <aminvakil> ueb: the link i've just sent is mysql5.7 from repo.mysql.com on buster
694[10:36:37] <jelly> aminvakil, it's not guaranteed to work well over release upgrades, I'd test first on a copy of the machine with the specific packages that need to be held back.
695[10:36:40] <jelly> !hold
696[10:36:40] <dpkg> hold is a status flag that tells the package manager to not automatically upgrade a package. To hold a package 'echo $package hold|dpkg --set-selections' or 'aptitude hold $package'. Note that prior to <stretch> "aptitude hold" is ignored by other package managers and aptitude won't necessarily use holds set with dpkg; see Debian bug #137771. See also <hold list>, <unhold>.
697[10:36:43] <ueb> aminvakil ok but when upgrading my distro it will upgrade mysql to 5.8 not 5.7... and I should avoid 5.8
698[10:36:46] <TheBigK> mysql 5.7 is also support till 2023 i believe or something? is quite long maintenance
701[10:37:21] <TheBigK> so quite some time to think about to switching to mysql 5.8 or mariadb... we for that matter wanting to switch... we dont like oracle ;)
712[10:38:58] <aminvakil> for sure you should test these on a test machine prior to doing that, installing debian jessie, install mysql-server 5.7 from repos.mysql.com, upgrade to strecth, then to buster, see if anything happens to your mysql-server
714[10:39:11] <abff> aminvakil: thats my fault, I didn't see the repo it came from, in my mind I assumed it was mysql-common from debian repos.
715[10:39:28] <TheBigK> jelly: the way to go in my opinion....but databases are always tricky i feel... especially if u do some trickery like stored procedures or stuff like that
717[10:40:20] <TheBigK> as far as testing went we are very happy with our development environment running on mariadb 10.3 ... we will probably switch production soonish to mariadb
718[10:40:36] <ueb> I've installed mysql-server from repo.mysql.com, so upgrading my distro won't upgrade mysql to 5.8?
719[10:40:47] <TheBigK> check whats done when u do the switch
725[10:42:01] <TheBigK> if u have mysql-5.7 in the repo... ur gonna be fine
726[10:42:57] *** Quits: emOne (~emOne@replaced-ip) (Remote host closed the connection)
727[10:43:05] <TheBigK> u have also good timing for swiching since debian bullseye is around the corner :)
728[10:43:14] <aminvakil> ueb: ^
729[10:45:11] <ueb> I should do these steps: 1) change jessie to stretch in /etc/apt/sources.list 2) apt-get update 3) apt-get upgrade 4) apt-get dist-upgrade 5) install php7.3 from sury...
730[10:45:49] <TheBigK> why the sury php ? im just curious
731[10:46:08] *** Quits: Ericounet (~Eric@replaced-ip) (Quit: Je m'en vais ...)
732[10:46:10] <TheBigK> i always do dist-upgrade only... is that wrong?
733[10:46:37] <TheBigK> afaik upgrade is just security no version upgrades... and dist-upgrade does both?
737[10:47:25] <ueb> TheBigK dist-upgrade will auto upgrade php7.0 to php7.3 ?
738[10:47:53] <aminvakil> man apt: full-upgrade performs the function of upgrade but will remove currently installed packages if this is needed to upgrade the system as a whole.
739[10:48:13] *** bern- is now known as bern
740[10:48:32] <TheBigK> ueb: i dont remember ... is so long ago... but u can check when u do the dist-upgrade ... whats happening
741[10:49:01] <TheBigK> if i remember correclty i did some manual steps to upgrade the php version
742[10:49:15] <ueb> TheBigK yes ok, however php7.0 was installed from repo dotdeb
762[10:55:07] <aminvakil> ueb: what's the problem with buster if you need php7.3 ?
763[10:56:00] <ueb> aminvakil I need to make a double upgrade.. I fear a system crash
764[10:57:23] <aminvakil> you a fear system crash with an official supported procedures which many people have done before you, but you're fine with installing packages from third-party repositories?
765[10:57:25] <TheBigK> yeah... u go for stretch... and if everything runs straight... u go buster... ez pz
766[10:57:48] <TheBigK> aminvakil: full agree :)
767[10:58:19] <aminvakil> worst case, install buster from scratch, configure mysql 5.7 from repos.mysql.com, install php from official repos, restore your database backup
768[10:58:36] <TheBigK> trying to avoid third party repos is my main consideration for almost everything. i always tell the devs... u need a reason to switch... a good one... when u want to use external mirrors
787[11:02:47] <ueb> jelly removing from sources.list ?
788[11:03:26] <TheBigK> usually u configure 3rd party in /etc/apt/sources.list.d/3rdparty.list
789[11:03:37] <TheBigK> and the basica debian things in /etc/apt/sources.list
790[11:04:11] <TheBigK> is easier to handle... u can move the files in /var/backups for example to temporary deactivate those mirrors then
791[11:04:24] <shtrb> ratrace, you will not belive I had enabled the debug service you had suggested, and the systemd shenanigans had ended , that is just crazy when I open debug everything just work as expected :D
792[11:05:23] <jelly> ueb, and uninstalling them.
793[11:05:47] <ueb> I need to mantain mysql5.7 installed from mysql.repo, I can't uninstall it
794[11:06:19] <TheBigK> thats fine... just change to the stretch version then
795[11:06:26] <TheBigK> a minor upgrade will be done then... thats ok
796[11:07:00] <TheBigK> and for php u do the same
797[11:07:09] <TheBigK> u look for the stretch packages.. .and change the mirror accordingly
798[11:07:20] <TheBigK> if u dont everything right... ur good to go
799[11:07:35] <TheBigK> u check with the upgrade which version of php and mysql ull get... and ur good to go
800[11:08:01] *** debhelper sets mode: +l 1068
801[11:08:01] <ueb> TheBigK how to do this check?
802[11:08:04] <TheBigK> apt-cache policy $package will give u exactly which version is going to be installed
803[11:08:14] <TheBigK> before u do the apt update obviously....
843[11:18:47] <ueb> maybe I'm forced to make a new machine...
844[11:18:55] <TheBigK> is it a VM?
845[11:19:12] <ueb> yes it's a VM hosted on cloud
846[11:19:46] <TheBigK> bigger harddrive possible? uprading the filesystem is not a huge deal usually
847[11:20:05] <TheBigK> shrinking can be painful... but growing is usually easy
848[11:20:38] <TheBigK> if u have no seperate /boot ... deleting some kernels can help
849[11:21:13] <ueb> the data drive is upgradable, but the root drive I think not
850[11:21:35] <ratrace> shtrb: but that's not normal :) could be just that systemd isn't failing as hard and loud in debug mode?
851[11:22:03] <ratrace> or... does but doesn't kick you in the shins and leaves you with a semi-working shell you can debug from? :)
852[11:22:23] <shtrb> it could be, their devs probably put a working flow (the oppsite of intentional bug) in debug mode
853[11:22:24] <TheBigK> ueb: do u have an idea whats causing the most diskspace? cause 10 gig can be fine, as long as u just have the system files on it
859[11:25:50] <TheBigK> thats better... but not perfect ... im not sure if 1gb is enough... but since u download the packages to /var/cache/apt/archives u could be fine
894[11:40:34] <jelly> 90MB, what does it have inside, wayland and gui
895[11:41:05] <TheBigK> i have no freaking idea... i dont check regulary ... ubuntu magic
896[11:41:21] <TheBigK> im too lazy to unpack that thing to check
897[11:42:27] <jelly> you can put MODULES=dep instead of MODULES=most in their /etc/initramfs-tools/initramfs.conf to make it behave more like debian defaults
898[11:43:00] <TheBigK> jelly: i might consider that... i have even sometimes problems with 1 gig of boot
899[11:43:15] <TheBigK> but i refuse to raise it... i just refuse :D
938[12:16:37] <ueb> apt-get upgrade is defined as "Minimal system upgrade", is it enough to upgrade Debian to Stretch? Or is it necessary apt-get dist-upgrade ?
939[12:17:04] <ratrace> TheBigK: I intend with upgrade to bullseye
940[12:17:25] <ratrace> but maybe I misunderstood you, xwayland will be needed for programs that don't support wayland protocol
942[12:18:12] <ratrace> ueb: you need dist-upgrade for that
943[12:18:30] <ueb> ratrace ok thanks
944[12:18:36] <jelly> ueb, read the release notes.
945[12:18:37] <TheBigK> ratrace: okay interesting
946[12:18:57] <jelly> ueb, don't blindly do a release upgrade before reading them.
947[12:19:04] <milkt> !tell ueb about jessie->stretch
948[12:19:39] <jelly> or you're going to get into all sorts of avoidable trouble and we're going to laugh at you
949[12:20:03] <TheBigK> im using kubuntu on my desktop and i tried wayland... but copy paste was a problem and then i thought... nah ... i give it with next release a shot :)
950[12:20:10] <jelly> well I am, other people are not asses
951[12:20:38] <ratrace> even the bot will laugh at you
952[12:20:49] <TheBigK> so mean hahahaha
953[12:21:02] <ratrace> Hey, I'm an ass. don't take that away from me.
970[12:23:10] <ratrace> jokes aside, I don't know what Ubuntu's problem is, but I can imagine a lot of NIH code stands in their way, because debian defaulted to waylend for gnome installations and that seems to be not a problem
971[12:23:21] <ratrace> I think fedora has been wayland for some time too now
972[12:23:34] <ratrace> *weyland
973[12:24:05] <TheBigK> i didnt try it for long... may be i could have fixed it. on 21.04 i gonna try again
974[12:24:28] <jelly> my current workstation's running ubuntu as well
975[12:24:47] <ratrace> jelly: already laughed at you for that.
1008[12:40:37] <milkt> probably pasting exact error message can be helpful, but i think installing "debian-keyring" or "debian-archive-keyring" will fix it
1009[12:40:58] <ratrace> ueb: keys for what repo? btw note that upgrading from one major version of debian to another, should (must, even) be done with all third party repos disabled
1010[12:40:59] <docmax> is it possible to prevent apt-get upgrade to re-enable systemd services?
1011[12:41:28] <ueb> ratrace mysql repo
1012[12:42:09] <milkt> docmax: you can mask it i think
1021[12:44:21] <dpkg> 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/*}
1047[13:09:35] <fnstudio> hi i've downloaded one of the (tested) images from here replaced-url
1048[13:09:40] *** cdown_ is now known as cdown
1049[13:10:05] *** Quits: wMw (wMw@replaced-ip) (Remote host closed the connection)
1050[13:10:23] <fnstudio> i know those are not officially supported images, but i was wondering if anyone has any experience with the hdmi audio settings as i don't seem to be able to get any audio out of the pi
1051[13:10:56] <megaTherion> On a default Debian (10) install, what's the default MTA to deliver mails?
1052[13:11:04] <megaTherion> (I mean which one is started and run by default)
1053[13:11:13] <fnstudio> the pi is connected to a tv monitor via hdmi; when i try to play a audio file, the monitor says "unsupported audio signal"
1054[13:13:26] <ratrace> megaTherion: Exim will be pulled in if and when an MTA is needed
1055[13:13:49] <ratrace> unless you first install Postfix, or something like dma
1060[13:20:02] <giorgian> Hello; when I run apt update in a debian docker image (tried :latest, :buster-slim and a few others) I get: "E: The repository 'replaced-url
1083[13:39:28] <giorgian> No, apt -v just outputs the version
1084[13:40:43] <giorgian> TheBigK: it was a local problem, the image itself works fine.
1085[13:40:49] <themill> !apt debug
1086[13:40:49] <dpkg> To debug what is going wrong with apt/apt-get/aptitude fetching from the mirrors, append «-o Debug::Acquire::http=true» to the command line. The output will be lengthy but informative.
1115[14:15:32] <TheBigK> abff: that would be my choice if i consider switching. but i dont know... im not unsatisfied with kubuntu ... usually it just works. there are just little things that are sometimes annoying
1117[14:18:39] <TheBigK> as i started with linux i used to use debian at first... later on i switched to gentoo... when 64bit became a thing... and also because i wanted the newest version of plasma... cause it was buggy... but it was also so shiny new... that i wanted it anyways :))
1118[14:18:55] <TheBigK> and later on decided... i use kubuntu... and since then.. im fine with it
1120[14:19:41] <TheBigK> its so similar to debian... but the packaging is ... good id say... i know i could make it same on debian sid.. pretty much... but its just more work
1188[15:41:47] <ueb> I'm back and I'm trying to fix apt-get issues before upgrading my distro
1189[15:41:53] <ueb> error: There is no public key available for the following key IDs
1190[15:43:17] <n4dir> do something like apt-cache search keyring, perhaps pipe it to | less; compare it with the keys you have installed (dpkg -l | grep keyring). That is what i would do (first of all)
1191[15:44:08] <ueb> I've found that this key is used by: /var/lib/apt/lists/httpredir.debian.org_debian_dists_stretch_Release.gpg
1200[15:47:28] <n4dir> i just meant "apt-cache search keyring", then i go "manually" through the list, and usually quickly find what i search for. ueb else you have to wait for someone more in the subject, my approach really is just the sledgehammer
1202[15:48:15] *** Joins: albert (~albert@replaced-ip)
1203[15:48:29] <miguel_clean> I am trying to understand debootstrap. and it's dependancies. this will run a debian (kind of inside a subdir) staying on the hosts kernel?
1204[15:48:56] <n4dir> miguel_clean: yes. kinda like that
1208[15:49:24] <miguel_clean> how likely is this to suceed on an old stock android kernel?
1209[15:50:05] <n4dir> so it will use some stuff from the "host", like the kernel, and to use it you will have to "chroot" (after you mounted some stuff for/in it, i think /proc /dev and /sys)
1210[15:50:31] <n4dir> miguel_clean: there are apps to create a debootstrap on an android, if i recall correct
1211[15:50:58] <n4dir> playstore, or however it is called these days
1212[15:51:21] <miguel_clean> i wanted to try it myself. this is an old device. and not all this apps are available
1213[15:51:41] <miguel_clean> basically I wonder what debian expects from the kernel?
1214[15:51:41] <n4dir> i think the debian wiki has a page about it too
1216[15:52:07] <miguel_clean> most of the time my debian experience was based on using the kernel shipping with debian
1217[15:53:23] <miguel_clean> are some libs (as libc) not specifically linked against a specific kernel?
1218[15:53:26] <n4dir> it's ages ago i did a debootstrap on an android phone, and i used one of those apps. It was rather easy. More i can't say. good luck
1219[15:53:36] <miguel_clean> heheh thanks :)
1220[15:54:16] <miguel_clean> I am just playing around. don't want to throw away a perfectly good tablet which just became old and lacks support from its creators
1221[15:54:25] <n4dir> yeah, sure.
1222[15:55:18] <n4dir> to me it doesn't seem as if it was a subject you hear of in this channel often. Still ask again or just wait for someone. And, perhaps, try other channels too.
1223[15:55:25] <albert> miguel_clean: have you looked at entirely wiping android and replacing it with debian?
1224[15:56:04] <ueb> apt-get update gives me this error: There is no public key available for the following key IDs 112695A0E562B32A
1225[15:56:07] <miguel_clean> albert: not yet. but I am considering this as well at some point
1226[15:56:44] <miguel_clean> albert: would require me to build the kernel myself I assume. not sure how to get the settings and find out what proprieatary drivers I migth loose in this process
1227[15:57:25] <ueb> is there any way to fix it? I've tried with: gpg --keyserver pgpkeys.mit.edu --recv-key E562B32A
1265[16:14:56] <ueb> E: Sub-process /usr/bin/dpkg returned an error code (1)
1266[16:15:27] <jelly> miguel_clean, you can certainly try to ask Samsung for the exact source codes and build process for the free parts of the code for that device.
1267[16:15:53] <jelly> ueb, grub is always installed onto the whole disk
1268[16:16:15] <jelly> well, if the system is booting in legacy mode.
1269[16:16:29] <jelly> if it's booting in UEFI mode then things are different
1270[16:17:40] <ueb> should I fix anything now about this error?
1271[16:18:20] <jelly> ueb, the actual failure was somewhere above in the logs. After an installation or upgrade or dist-upgrade fails, try to run "dpkg --configure -a" and "apt-get -f install" to make dpkg and apt to back to a consistent state.
1272[16:19:59] <ueb> Failed to start unattended-upgrades.service: Unit unattended-upgrades.service failed to load: Invalid argument. See system logs and 'systemctl status unattended-upgrades.service' for details.
1351[17:32:33] <PaulFertser> fnstudio: for the reference, there should be nothing ARM specific in audio, and as usual for the basic testing you're supposed to be reading dmesg, doing aplay -l, trying to use aplay with some wav file, use alsamixer to adjust the levels etc.
1352[17:32:49] <PaulFertser> fnstudio: no matter if arm or mips or amd64, it's still the same alsa.
1353[17:33:23] <jelly> ueb, where did you get php7.4 from?
1354[17:33:40] <ueb> here is mysql status now that I unmasked it
1407[18:00:33] <hop> one i can use to make xen-pciback grab the device
1408[18:00:38] <hop> jelly: yes, very funny!
1409[18:01:03] <imMute> hop: I see a "Bus 03.Port 1" in my lsusb. The sysfs for that is /sys/bus/usb/devices/3-1 which is a symlink to /sys/devices/pci0000:00/0000:00:1d.0/usb3/3-1 0000:00:1d.0 is the PCI address of the USB controller
1410[18:01:35] *** Quits: genr8 (~genr8_@replaced-ip) (Remote host closed the connection)
1411[18:02:11] <hop> i see bus and device only, no port
1432[18:14:48] <Hobadee> chocolate, most likely the php->mysql connection is attempting over socket and the socket name changed during the upgrade. That would be my guess.
1437[18:16:35] <Hobadee> chocolate, Could be some PHP dependency also. Check version numbers on everything. That "No such file or directory" error is likely because a file was named with it's version, and the upgrade changed that.
1438[18:16:43] <hop> jelly: 17:35 < hop> is there a way to get a usb-device to a xen domu with a stock debian kernel?
1441[18:17:26] <Hobadee> I'm building a Debian VM template - is there a way to have it prompt for IP address settings on first boot? Google is failing me...
1442[18:17:29] <hop> neither qusb nor pvusb appear to work
1443[18:17:32] *** Quits: Souler (~Souler@replaced-ip) (Remote host closed the connection)
1502[19:04:36] <urk> How can i get firmware if there is no firmware prompt? I need the firmware so that my Intel network card drivers may be installed. I would like to explore doing a full install with deboostrap since I have had trouble with the stable release installers on a Dell XPS 15 7590.
1527[19:19:05] <_-Jon-_> I'm hoping someone can assist me with a strange problem. I have a domain, say example.com. If I 'ping <randomness>.example.com' it resolves to some domain parking IP. If I do 'host <randomness>.example.com' it returns nothing
1530[19:20:25] <imMute> _-Jon-_: one might be using a DNS resolver that injects advertising while the other doesnt
1531[19:20:26] <chocolate> I have both php 7.0 and php7.3 installed, but I'm getting this output : This version of ownCloud requires at least PHP 7.2.0 DO I need to set php7.3 as default? what's happening?
1537[19:22:54] <_-Jon-_> imMute: Is there any way to find out? I know it uses /etc/resolv.conf but I'm wondering if there is somewhere else that it might be getting DNS servers from
1542[19:26:21] <greycat> ping does standard hostname lookups which use nsswitch.conf and a normal standard nsswitch will use /etc/hosts first and DNS second
1543[19:26:22] *** mpmc is now known as HorrificShroud
1544[19:26:56] <greycat> note that this is *different* from how host(1) works
1554[19:30:12] <jelly> that said, you can also investigate what ping (really, the glibc resolver library) does, by doing "getent hosts randomstuff.example.com" and "strace -f randomstuff.example.com"
1560[19:31:48] <urk> jelly> Yup, tried both 10.7.0, and 10.8.0, and couldn't get the laptop to work. At one point I was able to get the testing version to work, but now it too won't install, and has the same issue: firmware, drivers. I do have the firmware for Buster, and downloaded it to the laptop I am talking to you from.
1561[19:32:08] <urk> I am going to need to install the firmware in order to do deboostrap.
1562[19:32:19] <Crax> Sorry this is weird to ask but how to mask my ip xD..
1564[19:32:40] <jelly> urk, it's possible the laptop won't work well even with firmware if it's just too new
1565[19:32:58] <urk> There should be a way to reach the firmware from shell that is on a memory stick, but may have to do some manual installation of packages.
1567[19:33:41] <jelly> if you 're doing debootstrap from another live linux, you can just install firmware packages inside the debootstrapped system
1568[19:33:59] <jelly> you need a linux that mostly works to do debootstrap.
1569[19:34:00] <urk> jelly> I researched your argument, and XPS 15 7590 works quite well with Debian stable, but from my own experience there is going to be a fair amount of work getting things up and running. From what I learned, don't create a EFI partition first, and save it for last.
1573[19:34:49] <urk> jelly> I should be able to reach the drivers from shell, but there are other things that will need to be installed in order to add .deb files.
1574[19:34:56] <jelly> urk, "the same issue: firmware, drivers" needs a bit more expansion, what precisely fails with firmware images, and at what point
1575[19:35:22] *** Parts: hop (~hop@replaced-ip) ()
1576[19:35:24] <urk> jelly> Drivers don't initiate because the firmware is missing, and the stable installer doesn't prompt to add firmware.
1577[19:35:27] <urk> That is the problem.
1578[19:35:35] <repulse> What's the correct way to enable debug output for system-config-printer? Just kill the app and run it from command line like "/usr/share/system-config-printer/applet.py --debug"
1579[19:35:39] <urk> I sent a message about this to Debian bugs this morning.
1583[19:36:14] <sney> jelly: background is that urk has been trying to negotiate out of "hardware too new for buster" for over a week. it's a wifi6 iwlwifi nic that isn't supported by 4.19. the bullseye+nonfree installer works fine, but we just keep going around and around...
1584[19:36:17] <jelly> urk, so, again, which specific drivers fail?
1585[19:36:42] <rajivmars> what should be the ideal size for a "/root partition in debian. i have 1 tb hdd and going to make a separate. /home partition
1586[19:36:50] <jelly> urk, so your hardware IS, in fact, too new for stable _kernel_
1587[19:37:10] <urk> jelly> Should be, but not so sure. Only the testing version has worked, and only at one point. My initial thoughts are that the firmware is already there, but something wrong with not being detected.
1588[19:37:12] <imMute> rajivmars: for / or /root?
1590[19:37:15] <ratrace> rajivmars: /root or / (root)?
1591[19:37:51] <jelly> rajivmars, 20GB is enough, use LVM and resize things later. Even if you use LVM, don't let the dumb installer allocate the WHOLE space for /home. Leave some (most) space VG space free.
1592[19:37:53] <rajivmars> ratrace: for /
1593[19:37:54] <urk> jelly> Then how come there are postings on the net that Debian works fine with XPS 15 7590? I looked it up before I bought the laptop, and still believe there are problems with the installer not detecting things.
1594[19:38:08] <ratrace> rajivmars: I wouldn't go below 10GB if that's a desktop. jelly's advice is maybe a sound one.
1595[19:38:09] <imMute> rajivmars: /root IMO shouldn't have it's own partition. as for / with a separate /home, I stick to 4GB to 8GB, but I don't have any GUIs installed, so you probably want to double that.
1596[19:38:12] <jelly> urk, I'm going to trust sney on this over "postings"
1597[19:38:42] <urk> The debian release came out in the fall of 2019, and my laptop came out in the winter. Though newer, it really isn't new at this point. Its two years old.
1598[19:39:15] <urk> jelly> The only thing I am trying to do now at this point is get the firmware into the installation via shell, and nothing more.
1610[19:44:03] <imMute> rajivmars: you probably want to do what jelly suggested and use LVM so you can resize the partition later if it ends up being too small
1611[19:44:06] <ratrace> rajivmars: better question is: why is this even important? terabyte drives are extra cheap today
1623[19:50:04] <jmcnaught> urk: what happened? I thought that you had Debian 10 installed and just needed to install new firmware and kernel packages which you had downloaded. Did you start all over again?
1630[19:58:17] <urk> jmcnaught> I did have testing version installed at one point, and successfully downgraded to stable, but the headers were not correct so I elected to reinstall. Looking back, I should have stayed with testing. And from what I can tell, you definitely want to put Debian in the first partition if doing dual boot. It won't write files systems to the drive if a UEFI partition is created first.
1631[19:58:46] <urk> If I had to do it again, I would have just manually changed the headers.
1632[19:59:24] <urk> The only thing that didn't work was the XFburn, but it wasn't working correctly in testing before the downgrade.
1635[20:01:05] <jmcnaught> urk: if you want to use testing it should probably install just fine unless you are using two dozen partitions or something confusing. If you want stable, it is just a little more complicated. You install without network, then after installation you boot it up and copy the firmware and backports kernel packages by USB and install them with "dpkg -i packagename.deb"
1636[20:02:36] *** Quits: shibboleth (~shibbolet@replaced-ip) (Remote host closed the connection)
1683[20:41:55] *** Cypher1008 is now known as Cypher100
1684[20:43:46] <urk> jmcnaught> Is the firmware installed through a shell? Or can the firmware be installed via shell? I am not getting a prompt to install firmware.
1686[20:44:26] <urk> jmcnaught> Previously I was able to get sid/buster to install, but now it won't do that. I think there is more than one issue going on here.
1699[20:55:37] *** Quits: Vizva (~Vizva@replaced-ip) (Remote host closed the connection)
1700[20:57:02] <chocolate> I need help with this output error in owncloud replaced-url
1701[20:57:42] <tgunr> been search for a hour and can't find how to map my Apple Exended Keyboard which has the key pad and additional F-keys F-13 thru F-19. Seems .Xmodmap is no longer in favor? How do I get Debian 10 with Gnome to recognize all the keys?
1709[21:02:39] <sney> chocolate: is this the system that you upgraded? "failed to get a lock" sounds like there are competing daemons trying to access that file. make sure you only have (1) each of your httpd and php-fpm running.
1714[21:10:21] <jmcnaught> urk: what I am talking about is install Debian 10 *without* firmware *without* networking. Once you boot into the freshly installed Debian system, then you install the firmware and kernel from backports. I am not talking about a step in the installer.
1740[21:38:45] <neilthereildeil> i want to use the remaind of my disk for an LVM VG
1741[21:38:49] <neilthereildeil> how can i do this?
1742[21:39:05] <neilthereildeil> remainder*
1743[21:39:45] <jhutchins> jmcnaught: It depends on whether you have networking after the install. If so then yes, just run apt. If not you'll have to figure out how to download the .deb and copy it to the system, then install it.
1745[21:40:28] <jmcnaught> jhutchins: earlier they were told how to download the packages they need and copy to a USB stick
1746[21:40:45] <jhutchins> neilthereildeil: Follow the instructions for creating a VG. It doesn't hvae to be the whole disk.
1747[21:41:06] <neilthereildeil> jhutchins: would i need to create a partioin in order to create a VG on it?
1748[21:41:34] <jmcnaught> neilthereildeil: 1. create a partition 2. use pvcreate(8) on the partition 3. use vgcreate(8) 4. create logical volumes with lvcreate(8)
1749[21:41:40] <jhutchins> Yeah, the advantages of a stick is that you can get the installer to prompt you for a device containing the firmware.
1752[21:42:54] <sney> in this case, they needed both the firmware *and* the bpo kernel, so no amount of feeding the installer was going to help
1753[21:42:59] <jmcnaught> jhutchins: this is exactly the confusion that they are having. The 4.19 buster kernel simply does not support their wifi card, so it does not prompt for firmware. They either need to install testing/bullseye, or they need to do a basic install of buster without networking and install the loose .debs from a USB stick after the first boot.
1754[21:42:59] <jhutchins> neilthereildeil: Yes, if you don't use the whole device you need to partition.
1755[21:43:47] <sney> I wonder why nobody picked up the unofficial stable+bpo installer project after kmuto stopped doing it. that was always handy.
1756[21:43:50] <jhutchins> jmcnaught: Yup.
1757[21:44:43] <jhutchins> sney: Not enough support probably, or other comittments.
1767[21:47:57] <neilthereildeil> or can i create a partition thats easily extendable later on?
1768[21:49:14] <mentor> neilthereildeil: You can move physical extents in a physical volume around afterwards; you can change the size of a physical volume too. You can also have multiple physical volumes. All of this may eat your children.
1769[21:49:51] <neilthereildeil> you mean this is a bad idea?
1770[21:50:09] <mentor> neilthereildeil: I was speaking to your use of the word "easily"
1771[21:50:20] <greycat> If you use traditional partitions, you can just leave some space unpartitioned. If you use LVM, you'll want to leave a bunch of the VG unallocated, for future use.
1773[21:51:17] <mentor> Yes, you do not need to allocate all the physical extents in your physical volume in one go. Growing things is easier than shrinking them.
1786[21:57:22] <rapier> guys, i got an error when I boot from pendrive. The error is "initramfs unpacking failed decoding failed" how could I please fix that?
1789[22:03:17] <diogenes_> rapier, what app you used?
1790[22:03:34] <jmcnaught> neilthereildeil: no easy way to do this. You could create a VG out of the unallocated 2TB, boot from a live disk to copy the rootfs from partition to LV on VG, chroot into the rootfs while still in live boot to update fstab and update-grub and update-initramfs, and finally pvcreate the old root partition and vgextend it into the VG. Restore from backups as necessary.
1817[22:09:30] <jmcnaught> rapier: Debian ISO images are hybrid images that work for both burning to optical disc or copying to USB storage. Programs like rufus do not take this into account, and attempt to mangle the ISO so it will boot from USB. This messes things up.
1821[22:13:12] <urk> jmcnaught> I am now getting an error "failed to determine code name for the system". Linux is the only partition on the 1 TB drive, and in the first part of the drive. Everything else was deleted. How do I fix this problem?
1822[22:14:02] <jmcnaught> urk: when does this error happen?
1823[22:14:03] <urk> jmcnaught> This error occurred when I attempted to install the base system. I am running 10.8.0
1825[22:15:02] <urk> Actually the complete error is "Debootstrap Error, Failed to determine the codename for the release". And then it asks me to Continue? Or Go back?
1826[22:15:14] <rapier> Is it possible to extract iso to pendrive using winrar and write to on pendrive?
1827[22:15:15] <urk> I am not running Debootstrap.
1835[22:17:45] <mrjpaxton[m]> rapier: You can extract an ISO with `p7zip`. Try this: `7z x <my-file>.iso -o/<directory>`. Of course, replace the <my-file> and <directiory> with your file name and destination.
1836[22:17:52] <urk> jmcnaught> I did not select expert mode. I believe this is 10.8.0 non-free with firmware.
1837[22:18:16] <mrjpaxton[m]> Though, in most cases you can use `dd` to write to thumb drive.
1844[22:21:37] <jmcnaught> urk: is it possible you removed and reinserted the USB stick at some point, or unmounted it while in the installer's disk partitioning step?
1848[22:22:22] <urk> I always unmount disks before removing them. The disk was unmounted on the other, slower, older laptop next to me.
1849[22:22:35] <mrjpaxton[m]> Just a feature request though. I wish you could select "Expert mode" from the actual first GUI/TUI installer window. But at least you can still use Ctrl+Alt+F1 to F8 to use a terminal, any time.
1850[22:23:01] <mrjpaxton[m]> Otherwise, wouldn't you have to reboot to select it?
1851[22:23:19] <urk> There is not enough tools at shell to change the code name. Maybe I can add vim at shell?
1856[22:26:37] <jmcnaught> urk: assuming that the Debian ISO was copied to USB correctly (by win32diskimager if on Windows, or to a device (/dev/sdb) and *not* a partition (/dev/sdb1), this error is likely being caused by 1. bad USB stick or port dropping the connection during the installation or 2. something you are doing during disk partitioning that unmounts the USB
1857[22:28:36] <jmcnaught> urk: if you are still attempting a complicated disk layout with separate partitions or volumes for almost everything, I suggest that just for argument's sake you take the path of least resistance through the installer. Choose all defaults for almost every question, and use the guided partitioning to let it take over the entire disk. It should only take 20-30 minutes, just to see if it works.
1859[22:29:07] <mrjpaxton[m]> Yeah, I'm looking up the error "failed to determine code name for the system", and it looks like a bad write to the USB drive can cause it, or it can be caused by UNetbootin. I would recommend Rufus on Windows or `dd` on Linux to write the ISO if you haven't tried that already.
1860[22:29:25] <jmcnaught> You can always redo it with more complexity immediately after you establish that it works.
1861[22:29:57] <mrjpaxton[m]> I always have at least one or two `dd` commands in my head at all times. :)
1862[22:30:15] <urk> I just ran the onboard integrity checker, and it says "No valid Debian CD-ROM". I suspect this is because of the missing code name.
1863[22:30:21] <mspe> why nobody uses ventoy?
1864[22:31:15] <urk> jmcnaught> There doesn't appear to be an issue with the partitions.
1865[22:31:45] <jmcnaught> urk: how did you copy the Debian ISO to USB?
1866[22:32:02] <nkuttler> mspe: it ate my bootloader
1867[22:32:06] <mrjpaxton[m]> Well, side note, I also just learned that Windows 10 cannot be written with `dd`, but instead has to be done otherwise with something like WoeUSB. So that's always fun.... Those are the only two tools I really use for writing ISOs.
1868[22:32:30] <nkuttler> also they closed my ticket when i said i'd try to reproduce on more machines..
1869[22:33:08] <mrjpaxton[m]> I mean "toy" is in the name, then.
1870[22:33:26] <mspe> what do you mean “it ate my bootloader”?
1881[22:35:32] <mspe> mrjpaxton[m]: it's not my favorite tool, but it's the most convenient for people trying to use one or several isos in a filesystem, instead of overwriting a whole flash drive
1882[22:36:09] *** grindhold is now known as GRINDHOLD
1883[22:36:42] <neilthereildeil> mentor: ohh good idea!
1886[22:37:07] <neilthereildeil> so i have / FS as a 500GB FS in a giant 3.3TB partition
1887[22:37:42] <neilthereildeil> so now im gonna create a VG of the rest of the space in the 3.3TB partition, and migrate the data in / to in, and then delete the current / partition
1895[22:38:32] <greycat> your words, no sense do they make.
1896[22:38:44] <mspe> mrjpaxton[m]: it's a customized grub that partitions your flash drive as 3 partitions: bios, esp and an exfat fs, and you can write any iso in the exfat fs from windows, macos or linux
1897[22:38:52] *** GRINDHOLD is now known as grindhold
1898[22:39:12] <neilthereildeil> so now i have a 500GB EXT4 FS inside a 3.3TB partition
1899[22:39:13] <urk> jmcnaught> Deleted all the partitions as you recommended, but still get the same error "Debootstrap Error, Failed to determine the codename for the release". The problems I am encountering don't appear to be related to my earlier creation of various partitions. The only thing I did different this time is that I deleted the EFI partition which was the first partition, and replaced it with Linux. However, I still get prompted for an
1900[22:39:13] <urk> EFI partition before committing to disk. That error also shouldn't be showing up.
1901[22:39:20] <mspe> and it generates at boot time on the fly a menu with all your iso files in the exfat partition
1902[22:39:28] <greycat> unless you're working in msdos disklabel land, and the 3.3TB partition is an "extended partition", but isn't that too large for msdos disklabel? wouldn't you have to use gpt?
1903[22:39:31] <neilthereildeil> im going to create a VG from the 500GB-3.3TB range in the partition
1904[22:39:42] <jmcnaught> urk: how did you copy the Debian ISO to USB?
1905[22:39:50] <neilthereildeil> and then mvoe the data from the 500GB / FS to the new VG
1906[22:39:52] <neilthereildeil> yes this is GPT
1907[22:40:21] <greycat> gpt doesn't have extended partitions AFAIK, so how on earth do you perceive that there is a "partition inside a partition"?
1908[22:41:00] <neilthereildeil> theres no nexted parititons
1916[22:41:44] <mentor> A volume inside a partition, perhaps
1917[22:41:54] <urk> jmcnaught> with dd
1918[22:41:54] <neilthereildeil> a FS inside a partition
1919[22:41:59] <mrjpaxton[m]> To be fair, I still am using old USB 2.0 flash drives that are 32GB a my rescue drives, since they still work at least. It does take a while to write to them, but I'll consider checking out Ventoy sometime when I get some higher capacity 3.0 drives. Lol.
1920[22:42:03] <neilthereildeil> a 500GB EXT4 FS
1921[22:42:12] <greycat> you keep saying this thing but it MAKES NO SENSE
1922[22:42:16] <neilthereildeil> inside a partition that 3.3TB in size
1923[22:42:22] <neilthereildeil> hmm why doesnt it make sense?
1924[22:42:23] <jmcnaught> urk: and you made sure to copy to the whole device not just a partition?
1925[22:43:00] <urk> jmcnaught> Whole device was copied.
1926[22:43:11] <mspe> mrjpaxton[m]: honestly it's also more reliable to just copy paste isos in a file system than all the writings people attempt with win32diskimager, rufus, etc.
1927[22:43:12] <greycat> Because a file system occupied a WHOLE partition. Not a piece of a partition.
1928[22:43:18] <neilthereildeil> ok
1929[22:43:30] <mentor> Well, that's not really true
1930[22:43:41] <greycat> The only time a file system should be smaller than its partition is temporarily, when you have increased the size of the partition, and have not yet increased the size of the file system to match.
1931[22:43:41] <mentor> Usually a filesystem occupies the space it itself thinks it does.
1932[22:44:43] <jhutchins> LVM can be pretty confusing. Even when I had pretty much wrapped my head around the concept of creating and expanding partitions, I still had a step-by-step cheat sheet that I always referred to.
1933[22:45:00] <mrjpaxton[m]> mspe: I know, but considering the differences between how BIOS and UEFI boot between the DOS partitioning table and GPT, it's difficult to do a "hybrid boot" between everything.
1934[22:45:32] <mspe> mrjpaxton[m]: it features a hybrid boot that also works with secure boot in the latest versions
1935[22:45:45] <mrjpaxton[m]> You'd even need a GPT partition table if you wanted to boot GPT on BIOS, or GPT if you wanted to boot from UEFI at all in the first place.
1936[22:45:46] <jhutchins> mentor: With LVM, you can us a portion of the available space and leave the rest for later expansion.
1937[22:45:54] <mentor> jhutchins: Yes
1938[22:45:57] <mrjpaxton[m]> *GPT BIOS partition I mean
1939[22:46:06] <jmcnaught> urk: "Whole device was copied" does that mean your 'dd' command had an argument like of=/dev/sdb with no number at the end?
1940[22:46:10] <jhutchins> mentor: You can also choose at a later date to shrink the volume and create a second one.
1941[22:46:13] <greycat> A given partition should either contain a file system, or a swap space, or an LVM physical volume.
1942[22:46:19] <mentor> jhutchins: But I filesystem tends not to defer to an upper block layer to define what its limits are.
1943[22:46:26] <mrjpaxton[m]> I'm talking if it was a different ISO besides Debian, though.
1944[22:46:36] <mentor> Or, lower block layer?
1945[22:46:40] <mspe> mrjpaxton[m]: yeah I told you it makes esp AND bios partitions in addition to the exfat one
1946[22:47:11] <greycat> If you have a 4 TB disk and you want to use 500 GB for a root file system and the other 3.5 TB for LVM, that's fine -- you would make two partitions, one for the file system, and one for LVM. Well, and a third for EFI, probably.
1947[22:47:31] <jhutchins> mentor: Generally that's the partition table, LVM has an equivalent. It tells the OS how big the filesystem is, the filesystem does not.
1948[22:48:01] <mentor> jhutchins: No, the filesystem superblock contains the information the filesystem uses to determine its own extent.
1949[22:48:02] *** debhelper sets mode: +l 1075
1950[22:48:12] <mrjpaxton[m]> How can Linux boot in EfFAT though if it's lower than version 5.7 in Debian by default? I don't think GRUB or Initramfs can read ExFAT without their modules, too.
1954[22:48:48] <mspe> mrjpaxton[m]: grub has an exfat module
1955[22:48:51] <jhutchins> mentor: Hm. Superblock... Perhaps so. So we have redundant an potentially conflicting information, the partition table and the superblock.
1965[22:50:38] <mspe> mrjpaxton[m]: then grub with the exfat and iso9660 modules goes fetching vmlinuz and initrd inside iso, load those in ram then boots
1966[22:51:29] <mentor> Well, it all makes sense if you know that a filesystem has its own ideas, and doesn't make sense if you don't. At least, for me.
1967[22:51:43] <mrjpaxton[m]> No, but I really do support the native ExFAT driver. I tried it out jsut as Ubuntu 20.04 came out, and I am really impressed by it. I hope at least Linux 5.7 gets used in Debian 11. I've backported the newer kernel in Debian, and it works great. (I also wanted Wireguard integrated in it, too.) it's so much better than the FUSE one.
1968[22:52:33] <sney> it looks like debian 11 will probably ship with some 5.10.x
1969[22:52:52] <mrjpaxton[m]> Oh wow, that seems too new.
1970[22:52:53] <mspe> mrjpaxton[m]: currently debian testing is linux 5.10
1971[22:53:23] <mrjpaxton[m]> Wasn't Linux 5.4 or something already out when Buster was released?
1972[22:53:47] <mrjpaxton[m]> But it's still using 4.19.
1973[22:53:50] <mspe> no, buster is 4.19
1974[22:53:59] <mspe> oops
1975[22:54:02] <urk> jmcnaught> Name of the file is firmware-10.8.0-amd64-netinst.iso
1976[22:54:52] *** Quits: silverwhitefish (~hidden@replaced-ip) (Quit: One for all, all for One (2 Corinthians 5))
1977[22:55:03] *** Quits: Filohuhum_ (~filohuhum@replaced-ip) (Remote host closed the connection)
1978[22:55:22] <mspe> mrjpaxton[m]: I prefer using my own standalone custom grub image to boot stuff, but ventoy is close to what I wanted to do and already kinda mature despite its name
1980[22:55:59] <mrjpaxton[m]> urk: This might seem weird, but you can also try the live image instead of the net install if you can't get it working. It also has an installer that can update packaged from the Web. I've very recently had success installing 10.8.0 recently with the live+nonfree ISO.
1987[23:01:15] <urk> mrjpaxton[m]> What type and model of computer are you on? I'm on a Dell XPS 15 7590 which came out towards the end of 2019.
1988[23:01:52] <urk> I did not have any luck installing the live image earlier with the stable release. At one point I got testing working, but that was it.
1990[23:02:11] <mrjpaxton[m]> I have a custom PC build. So I'm not sure if that'll help.
1991[23:03:17] <mrjpaxton[m]> I'm just saying the live image can be more versatile, I suppose. So I like it more. The net installer is good for people who want a quick download, and can install specific software to many machines at once, maybe.
1992[23:03:21] *** Quits: Jerrynicki (~niklas@replaced-ip) (Remote host closed the connection)
1993[23:03:55] *** Quits: czesmir (~stefan@replaced-ip) (Quit: Lost terminal)
1995[23:06:16] <urk> I seem to be dealing with multiple problems: laptop is too new and firmware isn't yet supported by stable release, bugs in the stable release. I shouldn't getting prompted for a uefi partition, etc.
2004[23:11:40] <urk> strange. This time the base system is installing. I may want to run diagnostics on the port after the install ever finally gets up and running.
2008[23:14:18] <urk> There is also an error with insertion of the mirror name. After initially selecting a mirror, the name is not carried over to the input field so it helps if the user has a photogenic memory : ) I dont
2009[23:15:52] <dvs> phtogenic? I'll take a picture of your brain!
2010[23:17:11] <urk> I looked up a mirror on the net, and entered it, but the next input field is now asking me to enter the name of the directory in which the mirror of the Debian archive is located at
2023[23:25:03] <urk> dvs> It doesn't like that one, and says "An eerror has been detected while trying to use the specified Debian archive mirror. Possible reasons for the error are: incorrect mirror specified; mirror is not available (possibly due to an unreliable network connection); mirror is broken (for example because an invalid Release file was found); mirror does not support the correct Debian version. Additional details may be available in
2024[23:25:04] <urk> /var/log/syslog or on virtual console 4. Please check the specified mirror or try another one. I assume I can get this error message if there is no access to an outside network, but not sure how to look this up on /var/log/syslog. I am going to drop to shell and take a peak.
2026[23:26:02] <jmcnaught> urk: you cannot access a mirror because you have no network connection because you are using a Debian 10 installer with a kernel too old for your wifi.
2027[23:26:03] <jhutchins> Try dropping the US.
2028[23:26:27] <jhutchins> jmcnaught: Ah, good call.
2029[23:26:36] <dvs> ???
2030[23:26:37] <jmcnaught> urk: the installer should give you an option to continue without a mirror.
2033[23:28:05] <urk> jmcnaught> Incidentally I am in syslog, and the name of the release is buster contrib main non-free, 10.8.0, unofficial amd64, netinst wifi with firmware.
2034[23:29:56] <jmcnaught> urk: with no network connection it would have made sense to use a larger ISO like DVD-1 so you would have fewer packages you need to sneakernet after installation. A netinst however can install a basic Debian system without a network connection, and that is all you need for now.
2036[23:31:50] <urk> jmcnaught> If I can get the firmware on here then I can run debootstrap.
2037[23:32:42] <jmcnaught> urk: I would forget about debootstrap. Install Debian with a normal debian installer image like you are doing now.
2038[23:33:37] <jmcnaught> urk: Once you install even a basic Debian system to the computer, if it boots then you can use a USB stick to copy backports firmware and kernel packages from another computer.
2039[23:33:47] <urk> jmcnaught> It isn't so easy. I had no problem at all with earlier versions of 10. 10.8.0 seems to have some issues with this laptop.
2040[23:35:13] <jmcnaught> urk: have you tried using a different USB stick at any point?
2041[23:35:39] <urk> yes
2042[23:35:48] <urk> That isn't the problem.
2043[23:37:10] <ryouma> (late to the conv but i second the notion that lvm can be pretty confusing. that is why i do not use it. i don't want to have to wrap my head around 3 layers of concepts and edge cases when debugging os. but that's just me and there are many who swear by it.)
2044[23:38:05] <dvs> ryouma: you're not alone
2045[23:38:31] <koollman> if you think lvm is confusing, try explaining to a new linux user how to repartition their storage over irc without causing data losses :)
2064[23:52:10] <urk> <mrjpaxton[m]> Is that a stable release network install?
2065[23:53:18] <mrjpaxton[m]> Oh sorry, those are the live images. But they worked for me. I think you can even install Debian from the live session, if you want to.
2066[23:53:41] <urk> mrjpaxton[m]> I usually use xfce. I'm installing the os on a very fast laptop, perhaps too fast. Later on I am going to experiment with turning down the clock to see if it extends battery life. Battery life and screen brightness were under rated by Dell. I'm getting pretty close to 10 hrs, and 500 nits.
2067[23:53:43] <mrjpaxton[m]> Or they also have the installer if you select it from the boot menu.
2068[23:53:52] <urk> ok thansk
2069[23:54:04] <urk> was this a stable release?
2070[23:54:44] <mrjpaxton[m]> urk: If you pick the live XFCE ISO, then when you install it, it will automatically install XFCE. All those live images are for Debian Stable, yes.
2071[23:54:45] <ryouma> that is extremely luminous