89[01:24:26] <H-var> because snitching privately is different from talking publicly duh
90[01:25:46] <Tenkawa> you just did
91[01:26:11] <Tenkawa> we all know who maintains it .. it has a record
92[01:26:22] <alex11> anyway, Suggests means not required
93[01:26:28] <alex11> case closed
94[01:26:38] <Tenkawa> indeed
95[01:26:49] <A|an> I've just read about the BleedingTooth vulnerability and the suggestion is, to upgrade the kernel to 5.whatever...I've never upgraded the kernel outside of a normal release...would doing so mess up your system stability?
96[01:27:23] <alex11> no but the current kernel will probably be patched
97[01:27:37] <alex11> just upgrade the 4.19 kernel in buster whenever there's a fix published
98[01:27:45] <H-var> haha bleeding tooth
99[01:28:02] <Tenkawa> alex11: what you guys running officially currently? (I build straight out of kernel.org)
100[01:28:09] <A|an> yeah, how do they come with names like that?
101[01:28:11] <alex11> just the stock debian kernel
102[01:28:37] * Tenkawa needs to look up what its on lately
111[01:30:29] <A|an> I don't evern use bluetooth, I was just wondering, if one were so possessed, is upgrading to a non-Debian blessed kernel a bad idea?
120[01:32:26] <A|an> I'm not doing anything, I was just wondering if the floor would collapse and the ceiling fay away if you installed the latest and greatest.
128[01:35:39] *** Quits: meLon (~meLon@replaced-ip) (Remote host closed the connection)
129[01:35:51] <H-var> CoronaBleed
130[01:36:00] <sney> debian is very good about supporting the stable kernel with new security fixes, so even though it's based on 4.19 from upstream, it's more current irl
141[01:39:32] <Pegasus_RPG> Hello. Any apache httpd gurus here? I have a strange situation where most files in a certain directory (password-protected,) won't display/download when accessed via HTTP, but will via HTTPS. SOME txt files do display/download though and there's no error in the Apache logs and FS permissions are identical on ones that do and don't display. Where the heck else can I look?!
142[01:40:05] <sney> Pegasus_RPG: I try to avoid apache, but I suspect .htaccess
143[01:40:10] <ratrace> Pegasus_RPG: there's a whole legion in #httpd
144[01:40:46] <Pegasus_RPG> ratrace: that channel is dead atm... sney: not present in the directory in question
147[01:41:34] <sney> also make sure you're testing in an uncached browser, incognito mode or whatever
148[01:41:44] <Pegasus_RPG> ratrace: The browser acts like it gets a zero-byte file.
149[01:41:45] <ratrace> you always get something.... so what is it. what's status code? what's access log sayin? error log?
150[01:41:46] <sney> browser caching will always ruin your day when testing stuff if you let it
151[01:42:09] <ratrace> right, so tried with curl or wget to remove the browser cache nonsens from the equation?
152[01:42:31] <Pegasus_RPG> ratrace: access log shows just a GET of the file in question as expected. I did but need to feed it auth creds... *looks up how*
153[01:42:51] <ratrace> Pegasus_RPG: with what status code?
154[01:43:00] <ratrace> (and length corresponding to the file?)
155[01:43:35] <Pegasus_RPG> Ah so "502 Invalid headers in response"
156[01:46:11] <Matrox> if i add my gpg signing key to git with (git config --global user.signingkey MYKEY) is it permanent, or i have to run this command everytime i start a new shell?
157[01:48:20] <alex11> it would actually be really awkward if one day you had to update to a whole new branch to get a kernel security fix because it just isn't possible to patch the older one
158[01:49:37] <sney> debian doesn't typically move fast enough for that, either. it'd be a mess.
182[02:03:21] <ufeindschiff> Hi, has anyone ever ran into the issue before that when selecting either install or graphical install on an installation image to start the Debian installer, it resets the machine instead of booting into the selected option? This happens both in UEFI mode as well as legacy mode
183[02:04:32] <dvs> ufeindschiff: I've only heard about this when the .iso is written incorrectly to the media.
184[02:05:39] *** Quits: Kel (~KelMonsta@replaced-ip) (Remote host closed the connection)
269[03:58:07] <Matrox> my system had a power failure and shutdown abruptly. in that case is it better to just reboot and let the system recover the ext4 journal? or boot from a liveusb and run fsck -f on the ext4 partitions? I loaded the usb and run fsck right after the power failure, fixed some inodes, and on boot never saw the recovering journal message.
270[03:58:35] <sney> linux is very good at recovering from power failures in general
271[03:59:01] <Matrox> does the journal recovering do anything more than fsck?
272[03:59:05] <sney> using a rescue usb can't hurt, exactly, but it's overkill. when I lose power I just turn the machine back on
273[03:59:24] <Matrox> when i do that it usually says recovering journal for a bit
274[03:59:30] <Matrox> and boots
275[03:59:31] <sney> the 'recovering journal' messages that you see are part of e2fsck
287[04:00:54] <dpkg> Posting the same question in several places at the same time (IRC channels, news groups, mailing lists, forums) is impolite; your time is NOT more valuable than everyone else's. Your question might be answered elsewhere, meanwhile we are wasting our time doing research for a problem you've already solved. Cross-posting can also make you look like a spammer and get you k:lined. See also <multiple ask> <hurry>.
288[04:01:04] <alex11> what happens if you lose power on a package upgrade/dist upgrade?
296[04:03:43] <sney> you might need to boot in single-user and run an 'dpkg --configure -a' before proceeding normally, if things seem hairy. but this isn't windows and power failure recovery is considered normal use
317[04:11:24] <alex11> e2fsck in multi user mode, then dpkg --configure -a in single user, then continue the upgrades in multi user?
318[04:11:30] <alex11> does it matter which mode?
319[04:12:05] *** Quits: dionysus69 (~Thunderbi@replaced-ip) (Remote host closed the connection)
320[04:12:09] <mtlsw> alex11, just ignore him. he's giving bad advicce. If you lose power, you should always run e2fsck from a single-level -- that is if you can remount "/" to read-only -- or use a Live-CD/USB to do the e2fsck.
321[04:12:35] <mtlsw> alex11, after just reboot into "normal" boot and perform your "apt-get -f install"
322[04:14:29] <alex11> i'm sure the reference manual or something has an entry on this, or the wiki, i should've checked those
323[04:14:55] <mtlsw> alex11, ++ any other commands -- it should trigger something as "dpkg --configure -a"
324[04:15:07] <mtlsw> alex11, it's for any operating system.
325[04:15:20] <mtlsw> alex11, especially for journalled filesystems.
357[05:09:37] <rk4> i've been abusing debian for quite a few years now, and come to think about the only time i've had notable problems from immediate loss of power, was with JFS2
359[05:11:24] <rk4> there's a sea of 'appliances' running linux that don't really have a soft power off button, but that we just yank the power to. one supposes in this adverse environment things have become fairly robust
386[05:29:20] <rk4> ok 'best practice', 'bad practice', are opinions
387[05:29:37] <rk4> maybe the opinion applies to most people, maybe it doesn't
388[05:29:45] <rk4> it is fine to switch to root through su.
389[05:30:53] <leoni> rk4: Alright. The only reason I've been using su instead of sudo is because I believe it's just the traditional way of using root. This is purely biased.
413[05:49:54] <PamiPetteri> alex11: what are you trying to do? what is the app and what are the commandline arguments? I mean the whole command with its parameters.
414[05:50:16] *** randomgry is now known as gry
415[05:50:22] <alex11> i have no current problem, i'm just wondering hypothetically what you do in case of a power loss on package upgrades/dist upgrades
416[05:52:19] <PamiPetteri> alex11: sudo apt -f install or apt -f install as root user, if you enter the root user mode via su or the login mechanism, either virtual console or (not recommended) via the graphical mode (display manager).
417[05:52:52] <alex11> right, someone was suggesting do a e2fsck first
418[05:54:29] <PamiPetteri> alex11: another command is "dpkg --configure -a" as root, you can login as root via the shell, the display manager or with "sudo dpkg --configure -a". I think "sudo apt-get -f install" or "sudo apt -f install" work better - it all depends on what is needed and incomplete.
419[05:54:48] <alex11> i know how to become root but thanks :P
420[06:00:05] <eb0t> hey i have us keyboard and i need uk.
421[06:00:16] <eb0t> i have looked everywehre and cant get it done
422[06:00:55] <eb0t> i have even installed console-common then install-keymap gb ..whichfails
472[06:41:39] <eb0t> i think i did that PamiPetteri ...but ill recheck to see if for some reason it didnt take into account xorg
473[06:41:53] *** Quits: burrows (burrows@replaced-ip) (Remote host closed the connection)
474[06:41:55] <PamiPetteri> eb0t: it should be in the same place as the locale, possibly. you could also try other options. often in the commandline "sudo dpkg-reconfigure keyboard-settings" works.
511[07:36:27] <hl521> Hey, for some reason I don't think usermod was installed, what package do I install to get usermod?
512[07:37:57] <Gerowen> A quick check in my Synaptic shows that accountsservice provides those commands.
513[07:38:10] <hl521> thanks
514[07:38:32] <hl521> strange, that still doesn't provide it
515[07:38:33] <leoni> hl521: How do you know they are not installed?
516[07:38:48] <leoni> I mean, the program is not installed.
517[07:38:59] <leoni> Do you have manpage for it?
518[07:39:30] <leoni> Perhaps you have to run it with permissions.
519[07:39:42] <hl521> leoni: strange, yes i do
520[07:39:52] <hl521> maybe usermod isn't in the $PATH or something
521[07:39:56] <jmcnaught> hl521: did you use "su" to become root? /sbin/ might not be in your $PATH
522[07:40:03] <hl521> yeah, I did
523[07:40:04] <leoni> Try "su --login root", then try using it.
524[07:40:08] <hl521> stby
525[07:40:12] <jmcnaught> !buster su
526[07:40:12] <dpkg> In buster, su no longer overrides PATH by default, requiring that you use "su -" or "su -l" for login shells (which is not really a new thing at all...). To approximate the previous behaviour, put "ALWAYS_SET_PATH yes" in /etc/default/su (create it). See replaced-url
527[07:40:28] <leoni> I was going to mention that.
528[07:40:42] <hl521> yeah that works
529[07:40:42] <leoni> You gotta run su with --login.
541[08:05:54] *** Quits: bn_work (uid268505@replaced-ip) (Quit: Connection closed for inactivity)
542[08:07:03] <Pegasus_RPG> Hello. I have a system whose kernel I just updated that uses linux bonding. When it boots now, I get "networking[<pid>]: error: Another instance of this program is already running." What the heck does that mean in this context? Which program??
581[09:34:11] <Pegasus_RPG> Fixed my issue...apparently ifupdown2 isn't ready for prime time yet. I just replaced it with ifenslave and ifupdown and all is well.
609[10:10:08] <hata> just curious, why does this output appear when running apt install? Never seen it before, not sure if it's something I did or if it's part of something updating replaced-url
610[10:12:38] <Montresor> You have needsrestart installed.
611[10:13:17] <hata> ah, perhaps it got installed with docker
764[12:41:45] <powerhouse> I just loaded up debianlive10 no problem but my mouse and keyboard didn't work =( any solutions? I need to boot up using debianlive to use lvresize to resize a partition
774[12:47:08] <ratrace> so that's cinnamon then. I asked if that was gnome, in case the live gnome also defaults to using wayland, in which case you could try the xorg session. no idea about cinnamon.
776[12:48:45] <powerhouse> okay great, how do I try the xorg session? is that a different iso?
777[12:49:01] *** debhelper sets mode: +l 1147
778[12:50:17] <dka> Hello, I have a laptop from a third person who uninstalled the linux kernel, now he can't start up the desktop. I have the username, he forgot the password from root and his self
782[12:51:19] <ratrace> powerhouse: I don't even think cinnamon uses wayland, but if it did, xorg would be present anyway for compatibility reasons. with gnome, you select xorg session on login by clicking the cog icon. no idae about cinnamon.
783[12:51:51] <ratrace> dka: if there's no kernel....... how do you imagine to boot into it?
784[12:51:59] <dka> I don't imagine
785[12:52:04] *** Quits: frgo (~frgo@replaced-ip) (Remote host closed the connection)
786[12:52:13] <nkuttler> dka: use a live cd..
787[12:52:23] <dka> now he says he is on ubuntu xD
788[12:52:25] <ratrace> dka: then you'll have to chroot from a live env and reinstall the kernel
789[12:52:27] <nkuttler> dka: chroot into filesystem, install kernel, run passwd?
792[12:52:45] <dka> nkuttler, I can't login into system to reinstall kernel
793[12:52:58] <nkuttler> dka: please re-read what i wrote
794[12:53:16] <dka> Yes he as the usb key used to install the OS
795[12:53:21] <ratrace> I think dka doesn't know what "chroot" means
796[12:53:22] <dka> since he is on #ubuntu, should I ask support here?
797[12:53:24] <powerhouse> ratrace, okay will try gnome
798[12:54:09] <nkuttler> dka: fwiw, the fastest thing to do in such scenarios is usually to reinstall
799[12:54:15] <ratrace> powerhouse: well that's not what I suggested, I only said what _could_ be the reason on gnome under wayland. cinnamon probably isn't using wayland at all, I don't know which compositor that would be even.
800[12:54:36] <ratrace> dka: no, you should ask in #ubuntu
801[12:54:53] <dka> ... the uSB key used shows the Debvian GNU/Linux instgaller boot menu, so he is on ubuntu
805[12:56:01] <ratrace> dka: you'll need to bind /dev, /proc and /sys into the chroot in order for grub and initramfs to install properly. watch out if /boot is separate, you'll have to mount it into the chroot too
806[12:56:10] <absinthe> i'd go with nkuttler's suggestion... backup his personal files and reinstall.
807[12:57:10] <dka> He says the copy fail probably because file name being too long
810[12:58:32] <ratrace> I'm gonna side with absinthe and nkuttler on this one. Backup personal files, reinstall. Or else this is blind leading the blind....
811[12:58:37] <absinthe> ah, so there's a boot loader now ?
812[12:58:45] <dka> instead of having grub, he is having a splash screen
813[12:58:52] <Brigo> is it possible to uninstall the kernel in debian? should'n that trigget a wave of uninstalled packages, basically the whole system?
814[12:59:10] <absinthe> no i'd just uninstall the kernel, but with a big warning like
815[12:59:12] <ratrace> Brigo: nope, the kernel has very little deps, just initramfs and iirc grub
816[12:59:20] <absinthe> are you sure you want to deinstall the kernel youre running ?
817[12:59:21] <absinthe> orsl
818[12:59:25] <ratrace> which makes it installable nicely into contianers
819[12:59:30] <absinthe> oslt*
820[13:00:05] <ratrace> Brigo: nope = no wave of installed packages. yes = it's possible to uninstall the kernel
829[13:03:12] <Brigo> if the system were Debian Stable i would try to chroot into it with the netinstall usb and try to install the kernel and fix grub, i think it worth a try
856[13:14:18] <ratrace> orcacomputers: I'm at loss what that means, sorry. Isn't the debian gnome live environment just that? the regular gnome env? but anyway ..... you're looking for _wrong_ things. if there's no cog, no way to select xorg session, then it probably already IS xorg session and not a wayland compositor
857[13:14:55] <ratrace> orcacomputers: which then means another thing ... your kbd and mouse work on the login screen?
865[13:18:05] <orcacomputers> ratrace, yeah when I boot up it goes to gnome patition editor which is fine, idc just want my keyboard and mouse to work. when i boot into regular gnome environment i cant do anything because of no key board and mouse. okay. yes keyboard mouse does work at gnome partition editor screen, i have to make a selection to go into debianlive
900[13:45:43] <Brigo> dka, the password is encrypted, you can't access it from anywhere. You have to get root privileges from other means, like chroot.
902[13:48:47] <carandraug> hi! Is there a method to find a package given a file md5sum? I have a file on my system that dpkg-query does not know where it came from, I believe it might have been an install script that moved it. But if there's a way to search by md5sum that would help
906[13:51:31] <ratrace> that's a very weird filename... sure it's not an .a or a.so?
907[13:52:01] <carandraug> ratrace: I'm replying to bug #972703 where I'm asked for the md5sum of that file. I guess they plan to find the package given the md5sum so I guess there must be a way
911[13:53:42] <ratrace> If there's a way, I don't know it. was just curious which file it is
912[13:53:56] <carandraug> ratrace: the goal is trying to find what packages were installed that led me into this situation. I can just give them the md5sum, they find the package, then I try to reconstruct the history. I was hoping to save them one step
913[13:54:36] <ratrace> I'm thinking if there's a md5sum in a db, then apt-file should find it by path ... but I suppose yes it's possible the file was moved during an upgrade so current databases don't have the path
1002[14:59:58] <unixbsd> I do always install on part1 netbsd with grub2 because it is a non automatic grub, and then, I can have several partitions. I copy initrd and vmlinuz and modules, from live dvd, copy the dvd content onto /cdrom and then, all works like a charm. No installer needed, just debootstrap.
1003[15:01:23] <unixbsd> even fdisk on debian modifies my mbr without asking
1019[15:09:05] <epsilonKNOT> theres like two people sebastian and mo who are managing all blas lapack things
1020[15:09:15] <epsilonKNOT> mo is awol afaik
1021[15:09:21] <epsilonKNOT> need to send messages to sebastian i guess
1022[15:10:06] <jelly> just send mail to the team address
1023[15:10:50] <jelly> ,maintainer src:lapack
1024[15:10:51] <judd> Sorry, there is no record of 'src:lapack'.
1025[15:11:27] <jelly> ,maintainer liblapack-doc
1026[15:11:28] <judd> Package lapack version 3.4.2-1~exp3 was uploaded by Sylvestre Ledru on 2012-09-29, last changed by Sylvestre Ledru and maintained by Debian Science Team.
1027[15:11:42] <jelly> great
1028[15:11:58] <jelly> Maintainer: Debian Science Team <debian-science-maintainers at lists.alioth.debian.org> # that thing
1029[15:12:14] <epsilonKNOT> noice thanks!!
1030[15:12:29] <epsilonKNOT> i'm not familiar with debian interfacing and looking up packages
1031[15:12:39] <jelly> epsilonKNOT: also, most debian devs are on irc.OFTC.net now, not freenode
1032[15:12:45] <epsilonKNOT> O:
1033[15:12:49] <epsilonKNOT> not cool
1034[15:12:58] * epsilonKNOT throws a radiation banana grenade at devs
1035[15:13:00] <jelly> irc.debian.org = irc.oftc.net for about 15 years now
1036[15:13:21] <epsilonKNOT> that email address is longer than my thesis
1037[15:13:50] <jelly> and there's a #debian-science channel over there
1051[15:22:09] <genr8_> E: The repository 'file:/mnt/Debian10.3.0amd64-1 buster Release' is not signed. N: Updating from such a repository can't be done securely, and is therefore disabled by default.
1068[15:28:44] <iamjfk11> genr8_ Debian does not currently sign the Release files on its CDs; apt can be configured to always trust packages from CDs so this is not a large problem.
1069[15:28:51] <iamjfk11> that's refer to apt secure
1070[15:29:01] *** debhelper sets mode: +l 1153
1071[15:29:02] <jelly> epsilonKNOT: but it's still weekend, and who knows how many people there are just lurkers like me
1073[15:29:19] *** Quits: amartin (~willmart@replaced-ip) (Remote host closed the connection)
1074[15:29:33] <epsilonKNOT> i'm chill with waiting
1075[15:29:41] <epsilonKNOT> infact, i'm so chill i need a second bra
1076[15:30:03] <epsilonKNOT> but yea
1077[15:30:17] <epsilonKNOT> that mkl package is scary and a HUGE PITA
1078[15:30:42] <genr8_> iamjfk11, i realize that, but i was commenting on the person who said "sure just add file:/cdrom" job done... its not so simple
1096[15:37:04] <iamjfk11> According to the official documentation at this time, rebuilding a Droplet to a different base OS distribution can cause issues and is not recommended. Instead, you should create a new Droplet using your chosen distribution and migrate to it.
1097[15:37:04] <iamjfk11> But from the error that I’m seeing I think that you might have to update your repository list to match the Debian repositories. You could try to check the contents of /etc/apt/sources.lit on your rebuilt droplet and the new Debian droplet and make sure that they match.
1100[15:38:25] <genr8_> apt-secure mentions you can use the "deb [allow-insecure=yes] " option in sources.list. then it STILL gives those warnings, but doesnt disable the repo
1137[16:32:04] <SanchoPensa> since a week back, or so, I cannot seem to copy-paste by highlighting and middle mouse anymore, I am on Xfce. How do I fix this?
1138[16:32:10] <wyoung> and the other 118 permutations of gender these days
1139[16:33:07] <SanchoPensa> wyoung: is this a Linux help channel, or a debate on gender non-sense channel?
1140[16:33:36] <ratrace> well truth is, not only "guys" are here so....
1141[16:34:13] <nkuttler> hey guys is not a gendered greeting to most English speakers
1142[16:34:17] <SanchoPensa> ratrace: ya, well, I have been watching about 1.000 80ies and 90ies sitcoms, where women adress each other with hi guys, so...
1143[16:34:36] <wyoung> nkuttler: Yeah it is though, use folks, group, channel, even gang
1144[16:34:59] <wyoung> SanchoPensa: You still need to be respectful of peple
1145[16:35:02] <wyoung> people*
1146[16:35:03] <SanchoPensa> and other than that, would I appreciate it, I you keep your OPINIONS to your selves, and focus on the topic.
1147[16:35:04] <nkuttler> wyoung: please try to be more accepting of people who speak a different English variety than you do. you just show intolerance
1148[16:35:06] <SanchoPensa> well I am.
1149[16:35:39] <SanchoPensa> I consider it not so much respectful, when people believe, they have to boss other people around with their OPINIONS.
1150[16:35:46] <wyoung> SanchoPensa: Read the channel guidlines, there's a big section about distrimation
1151[16:36:05] <SanchoPensa> wyoung: I am not discriminating agains anybody.
1152[16:36:05] <nkuttler> wyoung: what you are doing is linguistically motivated discrimination
1153[16:36:16] <wyoung> nkuttler: rofl
1154[16:36:20] <SanchoPensa> exactly.
1155[16:36:56] <SanchoPensa> other than that, I am here, because I have a Linux question, not to be lectured on your OPINIONS.
1156[16:36:59] <wyoung> nkuttler: I believe speaking english in this channel is also in the CoC
1166[16:37:48] <nkuttler> wyoung: which CoC and which channel?
1167[16:37:57] <wyoung> nkuttler: This one
1168[16:38:09] <nkuttler> wyoung: care to show me a link to a CoC?
1169[16:38:13] <SanchoPensa> and the guidelines are set by the community. I have been part of this community for 25 years, I have contributed plenty, and I do not agree with your linguistic fascism
1170[16:38:16] <wyoung> nkuttler: It's in the topic
1171[16:38:34] <nkuttler> wyoung: there's no link in the topic
1173[16:38:37] <ratrace> one doesn't need a CoC to know that greeting everyone, not just men, is polite. ANYHOOO that's been rehashed now, parties notified , so pls, let's move on
1195[16:42:35] <derpadmin> I did not quite follow because I am sleep deprived, but imo guys encompass girls
1196[16:42:43] <derpadmin> or group
1197[16:42:44] <SanchoPensa> I have been entering this channel vor 25 years now with the same greeting. hey, guys. Wich I personally consider to be gender neutral.
1216[16:46:06] <SanchoPensa> wyoung: it is not an error in my perception. And YES, you are being intolerant, since you keep lecturing me OFFTOPIC in an Linux TECH HELP CHANNEL on your OPINION. The one, that I do not share.
1217[16:46:24] <SanchoPensa> wyoung: It is not for you to jufdge, what kind of person I am.
1218[16:46:29] <wyoung> Don't worry, everyone makes mistakes, just learn from them.
1219[16:46:37] <derpadmin> I think I will have to side with SanchoPensa, nothing disrespectful was said here and guys emcompass pretty much all genders imo
1224[16:47:21] <wyoung> derpadmin: Genders with penises, yes
1225[16:47:28] <nkuttler> another: it's not like facts matter to some people
1226[16:47:39] <SanchoPensa> wyoung: since I am not on the same page as you are, and since I find absolutely nothing wrong about it, YES, it is an opinion. A liberal mainstream OPINION.
1227[16:47:42] <derpadmin> well, girls at work say guys...
1243[16:49:32] <another> wyoung: merriam-webster considers it unisex. if you disagree with that common perception... well, that's just like, you're opinion, man
1244[16:49:46] <wyoung> Webster isn't English
1245[16:49:55] <wyoung> Oxford iss
1246[16:49:58] <wyoung> is*
1247[16:50:02] <nkuttler> jmcnaught: i would not classify that as overreaction. a person's language is a part of their identity, it's at their very core. being intolerant of language that's widely in use is quite hostile
1248[16:50:15] <SanchoPensa> things have the meaning that you ASCRIBE to them, and wyoung I do ascribe a different meaning to the word guys, than do you.
1249[16:50:15] <SanchoPensa> Yet still I am not lecturing you.
1250[16:51:04] <SanchoPensa> so.
1251[16:51:13] <ratrace> SanchoPensa: see now you're being pushy and bully about this. could've dropped it long time ago but no.... now YOU need to lecture about who is or isn't "libdem".
1252[16:51:29] <SanchoPensa> can we now pleas talk about it, why my middle mouse button does not copy paste any longer? THANK YOU!
1253[16:52:08] <SanchoPensa> ratrace: Look. I have been entering this channel for the last 25 years with the same greeting.
1254[16:52:24] <SanchoPensa> this is the first time in 25 years, that somebody feels an urge to lecture me about it.
1255[16:52:46] <SanchoPensa> I want to set this straight, since I do not want to get lectured on it again the next time I do.
1256[16:53:07] <ratrace> SanchoPensa: well fine, you said so, your conscience is clear, so let's drop the subject
1257[16:53:12] <another> 'nuff said. don't feed trolls. another out.
1258[16:53:24] <SanchoPensa> I am not being pushy, I am making it clear, that I am part of the community, and as a part of the community, I feel, like I have a right to state, that I feel otherwise about this topic.
1259[16:53:56] <SanchoPensa> since I have been lectured. It wasn't me who started it. Right?
1272[16:58:27] <SanchoPensa> appears to me like the "key-binding" of the middle mous button somehow got lost
1273[16:58:35] <absinthe> no
1274[16:58:37] <derpadmin> did you check the mouse mapping in the configs? something change? I know, sometimes, some programs bork my copy paste too (looking at you kmail)
1275[16:58:50] <absinthe> well, could be, but not imho
1276[16:59:20] <SanchoPensa> point is, I can't even find a setting within Xfce to change the button-mappings of my mouse...
1278[16:59:57] <absinthe> sorry, archwiki is kindo of my bible, even on other distros
1279[17:00:04] <absinthe> kind of*
1280[17:01:00] <SanchoPensa> absinthe: that's ok. I do, however, not quite know, what to do with it?
1281[17:03:44] <Voltum> does debian 10 buster, have nftables support in the kernel?
1282[17:04:04] <Voltum> do i just install nftables package, setup its config and i am done?
1283[17:04:16] <Voltum> (or is ther any service to enable)
1284[17:04:40] <absinthe> im not sure but it seems "clipman" is the copy/paste thing used on xfce, perhaps there's an option to make it use the primary or secondary clipboard
1290[17:06:21] <SanchoPensa> absinthe: maybe I'm a step further through that hint; when I higlight something, it does also not show up in my clipman xfce panle-plugin...
1291[17:06:35] <SanchoPensa> only, if I explicitly copy it...
1292[17:06:50] <SanchoPensa> this is really weird...
1293[17:07:13] <another> that sounds like standard behaviour
1357[17:45:47] <rudi_s> Voltum: iptables has an iptables-apply command which performs a rollback if you don't acknowledge it after a timeout. Not sure if nft provides something similar.
1358[17:45:49] *** Quits: linzst (~linzst@replaced-ip) (Remote host closed the connection)
1359[17:46:04] <ratrace> Voltum: with iptables, the trick I use to not lock myself is this: iptables-restore /etc/iptables/rules.v4 ; sleep 30 ; iptables -F ; iptables -P INPUT ACCEPT
1360[17:46:09] <rudi_s> (Btw. you can do that with iptables to, just pass the same ruleset to iptables-restore and ip6tableas-restore, using -4 and -6 restricts it only to ipv4 and ipv6.)
1361[17:46:36] <ratrace> Voltum: the config file is for iptables-persistent; but the idea is to set it, sleep and flush unless I can ctrl-c because I didn't disconnect myself
1377[17:51:46] <ratrace> Voltum: right, packet filter config is not persistent across reboots withotu something to reapply it
1378[17:52:48] <rudi_s> ratrace: Sure, but DROP is more consistent and also easier to read (as you can easily follow the rules and don't have to look at the top for the default policy). However, I almost never use iptables directly anyway (as it isn't atomic) but instead use ip{,6}tables-restore to apply the ruleset.
1389[17:55:35] <rudi_s> Btw. in Debian there's iptables-apply which does basically what you said with iptables -F/-P (except it restores the last ruleset).
1391[17:55:59] <ratrace> personally i'm still using the iptables interface because I don't wanna bother learning yet another firewall language that MIGHT very well be replaced soon with ..... yet another, BPF based
1392[17:56:12] <rudi_s> But I just use iptables-restore (and hope I don't lock myself out).
1393[17:56:25] <rudi_s> (Otherwise I'd have to use the console/serial to get access again.)
1394[17:56:34] <rudi_s> Me too, I don't see much advantage in nftables.
1395[17:56:49] <ratrace> hence the restore; sleep; flush trio in case you DO lock yourself up
1396[17:56:57] <jhutchins> Voltum: What you need to do is set up a script that will load the new rules, then if you don't connect and shut the script down within a certain time, it reverts the changes.
1397[17:57:04] *** Quits: dez (uid92154@replaced-ip) (Quit: Connection closed for inactivity)
1398[17:57:10] <ratrace> obvs, must be done through tmux or screen
1399[17:57:49] <ratrace> but these days I have a test VM where I test the ansible scripts which includes iptables rules .... no longer modify it with vim in production :)
1415[18:26:33] <bla> Can someone share a "debianic way" of doing container orchestration? :) on multiple tutorials I die a little each time pipeing to bash is suggested or downloading go binaries from somewhere.
1419[18:27:48] <bla> There might be no better way currently. Debian has swarm which turned out to be problematic. It has docker, but running k8s in docker is not production-way to do it. It seems.
1420[18:28:28] <bla> Mesos might be easy to handle... but maybe you just have a nice solution?
1421[18:28:50] <ratrace> k8s is the defacto way for mass container deployment these days
1422[18:29:28] <enocoffee> what scale are we talking about?
1423[18:29:46] <bla> Is there some resource of doing it with debian... "cleanly" or just grab rancher amd use it?
1435[18:34:14] <enocoffee> last time i've build a k8s cluster i did it using kubeadm. but rancher might be the right way to go so i won't give a definitive answer here.
1459[18:41:06] <ratrace> k8s scales yes but my question is, is the complexity overhad really needed for a project that small. k8s benefits are in vast deployments where you need to constantly change number and calass of containers. starting, stopping, failover'ing, etc...
1460[18:41:20] <ratrace> 6-12 sounds like something you set up once and it stays like that for good
1461[18:41:28] <bla> Currently it's a Debian on physical machines, in containers, everywhere.
1462[18:42:02] <bla> ratrace: physical servers are static, but containers, environments, etc. change a lot.
1463[18:42:17] <ratrace> we have few dozen dedicated servers and zero containers (other than what you "containerize" with systemd service units) ... so I'm trying to understand other people's use cases
1464[18:43:20] <bla> ratrace: well. We can click "deploy" on jenkins and have the app on the test environment automatically. It's compatible with what is being run on the dev's laptops. How do you do it with systemd alone?
1465[18:43:32] <bla> Deploying on prod isn't much further from that too.
1471[18:44:51] <ratrace> I see zero need for containers like docker.... so I'm trying to understand why others are jumping to containers for such small deployments
1472[18:45:09] <bla> Ok. We're happy with running containers and not VMs on laptops though. Nor having to carry a golden image, or update and distribute a large image.
1473[18:45:27] <bla> Well, you should probably try for something.
1474[18:45:54] <bla> Having "docker-compose up" set up 10 services on dev's laptop, in a compatible way across all devs laptops,
1475[18:46:09] <bla> and be able to change dependencies on the go and just rebuild what needs to be rebuilt is really handy
1476[18:46:27] <bla> I'm all for docker + docker-compose. I start to have problems with doing a bit bigger, "medium" orchestration.
1477[18:46:52] <bla> Because on-cloud k8s seems to be a standard, fine; I'd like something wayy simpler - but achieve high-availability in case of app/containers/servers failures.
1478[18:47:01] <ratrace> but aside from testing, why you need containers in prod, if they're just regular debian things?
1483[18:48:37] <bla> You can of course build a .deb file and install it. That's kinda similar, but docker is rather simpler.
1484[18:48:53] <ratrace> we achieve that with ansible playbook and homogenous debian installations, ie. all is debian 10, in prod and testing
1485[18:48:55] <bla> Unsure how it impacts development.
1486[18:49:15] <bla> I've got devs with ubuntu on laptops, bullseye or buster on laptops. They'd have to use VMs
1487[18:49:20] <ratrace> so apt install services and deps, set up configs and it works 100% reliably on any such installation
1488[18:49:33] <enocoffee> it's an efficient workflow and the automated recovery is a nice feature. treating everything as disposable, hosts, storage, containers.. .simplifies recovery and administration.
1489[18:49:41] <bla> Also - docker is idempotent and "static" once built. You don't get system cruft which can fail your testing/build or anything
1490[18:49:42] <ratrace> no, you have one VM where you pull stuff from git, for example, deploy, find what needs adjustment, and then redeploy that to prod
1491[18:50:00] <ratrace> but alright, I see how a docker would be tempting to lazy it ou
1494[18:50:26] <bla> "find what needs adjustment" is something we don't have to do really.
1495[18:50:37] <ratrace> I'd never in a million years use it as dockerizing promotes bad habits, unpatched code, unstandard practices, heterogenous ten thousand different versions and cases and subconfigurations and blah...
1496[18:50:45] <bla> Well - we do it during development. Later it's just there.
1499[18:51:23] <bla> You can probably say same about packaging stuff in .deb. It makes you lazy. ;p In some ways.
1500[18:52:05] <ratrace> no because, eg. openssl update is an `apt update` trigger away. with docker we'd have to rebuild and redeploy massively more amount of stuff just for single lib update.
1501[18:52:06] <bla> If required I can move my app out of debian servers too. But hopefully I won't have to.
1527[19:09:22] <ratrace> doesn't. you have to autoremove (or there's a package that does that iirc), and grub always lists all the kernels it finds in /boot
1535[19:14:54] <hmuller> i guess that's the question, how does the system know to keep the latest and the previous
1536[19:15:14] <verblendet> hi, ok, quuestionm, installed latest debian from netinstall, 10.6, choosed sddm to be my windowsthingy, i created a user, but its not on the list and there is NO option to enter username. where is sddm.conf? nothing in /etc, thank you
1537[19:15:33] <ratrace> hmuller: it doesn't. if you forgot to autoremove and accumulated 10 kernels, all 10 would be found and listed by grub
1539[19:15:53] <verblendet> i created a user with uid ~500, its a config thingy i guess, so where to configure it and how to get a simple username enter field like in good old days?
1540[19:15:54] <ratrace> so the "rotation" (and I put that in quotes because it only looks like it) comes from the function of autoremoving packages
1551[19:18:10] <verblendet> and how to configure sddm to list users below userid 1000
1552[19:18:14] <verblendet> thank you
1553[19:18:28] <ratrace> /etc/init/sddm.conf ?
1554[19:18:46] <hmuller> ratrace: yes, I get that. I am just curious what/where the logic is that keeps the current and previous kernel present even after autoremove
1567[19:21:04] <Onyx47_> there's a startup sctipt in /etc/init.d/sddm but I don't see it either, might be that it doesn't get created unless you change the defaults?
1568[19:21:06] <verblendet> there is no /etc/init/sddm.conf, and if debian netboot doesnt install debian....
1569[19:21:30] <verblendet> i will remove and reinstall completely
1570[19:21:31] <ratrace> verblendet: did you actually install the sddm package?
1572[19:21:34] <Onyx47_> the .conf file I mean, it's not there for me either
1573[19:21:52] <verblendet> yes i do and if i do dpkg-reconfigure it let me choose sddm
1574[19:22:09] <ratrace> and did you try creating the file?
1575[19:22:46] <verblendet> no i didnt create the file as i dont know where deboan expects it as default will be /etc/sddm.conmf but you mentioned /etc/init too, thats weird
1592[19:28:57] *** Quits: Grldfrdom (uid391113@replaced-ip) (Quit: Connection closed for inactivity)
1593[19:29:02] <jaggz> how do I get the actual name of a source for apt install -t
1594[19:29:23] <ratrace> weaselhtml: default on debian gnome is to use the wayland protocol, but xorg is there, you can select the session on login
1595[19:29:39] <verblendet> ok, created a minimum sddm.conf
1596[19:29:50] <ratrace> note: it's just protocol selection. mutter and kwin are both wayland and xorg compositors.
1597[19:29:57] <weaselhtml> ratrace: I sure hope they don't start adding in netplan
1598[19:30:06] <weaselhtml> If that happens, I'm going to another distro
1599[19:30:16] *** orblivion is now known as orb
1600[19:30:16] <ratrace> what has that got to do with wayland?
1601[19:30:28] <jaggz> (I look at its apt/sources.../foo.list and have its URL.. and its hostname.. but that error)
1602[19:30:30] <weaselhtml> Well, I really don't want massive changes to happen to my debian systems
1603[19:30:31] <verblendet> but wth? userids from 500 on are occupied by nobody and systemd crore something? go away devil!!!! ouch, that hurts and breaks good old userlists i used for decades
1604[19:30:32] <jaggz> apt install cuda -t developer.download.nvidia.com
1605[19:30:44] <jaggz> The value 'developer.download.nvidia.com' is invalid for APT::Default-Release as such a release is not available in the sources
1606[19:30:46] <ratrace> wayland is next gen rendering thing and replacement for xorg. netplan is some ubuntu-only NIH nobody else, and many ubuntu users either, cares about.
1607[19:31:04] *** Joins: Prints (~333@replaced-ip)
1608[19:31:17] <jaggz> verblendet, seriously? like hard-coded assigned and given away?
1609[19:31:20] <ratrace> verblendet: erm.... y u blame systemd? uid<1000 are system and reserved. your fault if you create a conflict
1610[19:31:37] <jaggz> I didn't know <1000 were system/reserved
1611[19:31:41] <ratrace> now you do
1612[19:31:54] <verblendet> grrrml jaggz, systemd-coredump is 999, and nbody abviously 65534, i set maximum userid to 99999 :D oopsi
1613[19:32:04] <jaggz> I do not truly know yet, except my deep faith in you
1614[19:32:20] <verblendet> ratrace, i blame systemd YES, as uid from 500 to 100 was reserved for ME? k?
1615[19:32:39] <verblendet> 1000 i mean ;)
1616[19:33:29] <ratrace> well, you also get to keep the pieces when you break it.
1617[19:34:32] <jaggz> so how do I install cuda? replaced-url
1618[19:34:38] <verblendet> where is systemd.conf? to configure uids?
1619[19:34:50] <jaggz> cuda : Depends: cuda-11-1 (>= 11.1.0) but it is not going to be installed
1620[19:35:31] <verblendet> jaggz, using nvidia cards? there are properitary packages i remember, and they are more up to date and stable
1621[19:35:37] <jaggz> "but it is not going to be installed" could be a bit more verbose
1625[19:37:19] <verblendet> ratrace, and i guess this is done somewheer during initialization and hardcoded - so it is documented somewhere ;) otherwise this rule does not exist
1626[19:37:32] <jaggz> verblendet, actually, thanks for mentioning that. I was using the recent updated official packages from backports, but was trying the toolkit from nvidia, but I just found there are some other new packages available in official which might be what I'm looking for
1627[19:37:34] <ratrace> verblendet: it's defined and documented by login.defs and its manpage
1630[19:38:20] <ratrace> min and max uids are very much arbitrary decision. "system" ones are such because of convention and packaging decisions. sure, as oyu can see, you can designate and UID to your user. question is only what are you going to break by going against the convention
1646[19:40:02] <jaggz> verblendet, kdeconnect lets your media playback stuff be accessed by your mobile
1647[19:40:11] <jaggz> so if you have something playing it can be paused on your mobile
1648[19:40:28] <verblendet> jaggz, i remember, but my kdeconnect is broken since months
1649[19:40:35] <jaggz> sorrow :(
1650[19:40:48] <jaggz> I was using nvidia's installation guide but .. that's where I ran into problems
1651[19:40:59] <jaggz> I believe from a conflict between debian official stuff and the nvidia repo
1652[19:41:18] <jaggz> (and debian knows why "it will not be installed", but for years they've decided not to say WHY it's made that decision not to install it)
1653[19:41:23] <verblendet> still on debian 8 for daily live... i really like to upgrade now, new shony fast computer, but new kde ui kills me
1654[19:41:29] <jaggz> it's just.. y'know.. not going to.
1655[19:41:51] <jaggz> i'm at stable
1656[19:42:14] <jaggz> and it appears they fixed some of the stuff in the nvidia packages during this last update, making it more complete.
1658[19:43:09] <verblendet> i remember setting up a cuda workstation for gpu computing using debian 10 amd official nvidia without problems last year - all workded fine, but it was sitting in a serverroom and only doing remote stuff and computing, not used as a desktop machine too
1730[20:31:12] <jhutchins> Hmmm... bet it wouldn't be too hard to set up something that would open a certain set of web sites first thing in the morning, so you get to your computer and it's all there, you don't forget to check the spam filter for three days or miss a rocket launch.
1731[20:31:53] <jhutchins> Speaking of which, Rocket Lab Electron on the 28th.
1732[20:32:16] <ratrace> jhutchins: isn't that default on browsers? continue where you left?
1750[20:42:35] <debhack> Hi guys! I've been fighting for quite some time now with this stupid problem and I can't really figure out what's going on. I'm experiencing screen tearing (both in my laptop and external monitor, so not hardware related) when I open/close windows. It's weird because it only happens when I use dwm, I tried with XFCE and no problem. I have an integrated intel graphics card (i915 chipset) which is perfectly well supported by the xf86-video-intel module. I
1751[20:42:35] <debhack> have tried with the "TearFree" option, uxa acceleration method and it made no difference. So far I'm surviving by just setting a solid dark wallpaper which matches the background of my terminal (this way the tearing/flickering is not that evident) but I can still see it when I open/close light themed gtk apps. Ant hints on where to look next? Thanks!
1752[20:43:15] <sney> how new is it? some newer intel chips require firmware
1753[20:43:18] <sney> !i915 firmware
1754[20:43:18] <dpkg> Some Intel UHD GPUs made after 2015 require firmware from userspace for all features to be enabled. This includes Skylake, Kabylake, Broxton, Cannonlake and possibly others. Ask me about <non-free sources> and install firmware-misc-nonfree to provide.
1755[20:44:39] <ratrace> debhack: I think it's about compositor (xfce) vs no compositor (dwm)
1756[20:45:18] <sney> I was wondering about that too but I had to look up whether dwm is compositing
1789[21:11:33] <debhack> I installed the firmware-misc-nonfree and it didn't solve the issue... I also tried installing picom (I did not configure it though) but same thing...
1790[21:12:18] <debhack> how did you solve it in your i3 config, ratrace?
1799[21:21:39] <debhack> I'm messing around with vsync in picom and no difference either... I'm afraid I'm gonna have to stick to the "dark wallpaper" method :S
1820[21:36:45] <ratrace> debhack: where'd you get it?
1821[21:37:49] <debhack> picom from the repos (I'm using debian testing), the solution by reading the man and tinkering with options that could be related...
1897[23:01:57] *** Quits: auk (auk@replaced-ip) (Remote host closed the connection)
1898[23:02:52] <leoni> Is there a similar command to aptitude why, that is not aptitude?
1899[23:03:50] <leoni> I have this web browser installed, called hv3, it's extremely outdated and it's not currently in development, yet I think it came with the default installation of the desktop enviroment. Weird?
1921[23:12:21] <jmcnaught> "aptitude why …" does not necessarily tell you why a package was installed in the first place, it just shows current dependency chains.
1966[23:40:16] <c-c> diginet: well I looked at my system and the super fast nvm ssd seems to get ~100% use with only system services running. Not sure how to interpret that. Also tested installing iotop and $ iotop -oPa - similar result, but can tell the written size is small. Keep digging.
1967[23:40:52] <diginet> it's been stuck at the same position for 10 min now
1972[23:41:53] <nkuttler> diginet: monitor disk usage
1973[23:42:11] <sney> if you're having bad i/o performance on a disk and you can't find a software explanation, it's likely that the disk itself is failing
1974[23:42:29] <diginet> I see :(
1975[23:42:32] <sney> SMART data is often not readable via USB so this could have been a long time coming
1976[23:42:53] <sney> (or if it's new, a factory defect)
1977[23:43:05] <diginet> maybe I need to buy a new external hdd
1986[23:53:37] <sney> most of the ones I've seen are just a regular laptop disk in a plastic box. but spinning disks seem to survive a little better when they're not constantly being moved around.
1987[23:53:54] <diginet> is there a way to convert a fat partition to ext4 without deleting data?
1999[23:59:07] <genr8_> probably 3+ factors going on. majority 2.5" (laptop), whole lotta physical movement while still powered on, somethin with the USB interface makes it not hold up (possibly active power management)