3[00:01:02] <greycat> Network config in Debian can be done several different ways. The most basic is via the "ifupdown" package, which includes networking.service, and which is configured in /etc/network/interfaces (see interfaces(5)).
47[00:44:53] <dpkg> Please do not ask if anyone can help you, knows 'something' or uses 'some_program'. Instead, ask your real question. (If the real question _was_ "does anyone use 'some_program'?" ask me about <popcon> instead.) See <ask> <ask to ask> <polls> <search> <sicco> <smart questions>.
56[00:57:57] <oxek> just to confirm, on debian I should not be manually adding exectutables to /bin or /usr/bin because apt might overwrite them. I should instead place them in ~/.local/bin . Do I understand it correctly?
72[01:03:31] *** Quits: jthomas (~joseph_th@replaced-ip) (Remote host closed the connection)
73[01:03:40] *** Quits: asymptotically (~asymptoti@replaced-ip) (Quit: Executed for war crimes)
74[01:04:13] <somiaj> oxek: you can use /usr/local/bin as well, $HOME is best if only your user needs, /usr/local provides a full FHS heirchy to install self compiled software to debian.
78[01:05:40] <somiaj> note /usr/local will take default over /usr, and could cause conflicts if you have thw same software isntalled locally in /usr/local and from debian in /usr
79[01:05:43] <sponix> somiaj: how does /opt rate ?
93[01:14:41] <ham5urg> I want to replace exim4 after a standard installation. I installed msmtp-mta and looking at the configuration. The examples I've found are using the root account for sending email. Will this be available to all services? What about some service which is not running under root?
94[01:14:52] <pingfloyd> oxek: also there's man 7 hier
95[01:15:06] <oxek> anyone know the order of precedence if the same named binary exists in ~/bin, ~/.local/bin, /usr/bin, /usr/local/bin, /bin?
96[01:15:23] <oxek> I suspect it goes from left to right according to $PATH
97[01:15:29] <annadane> yep, that
98[01:15:31] <oxek> but the ~ directories are not in $PATH
99[01:15:46] <oxek> or at least not there when I do `echo $PATH`
116[01:19:51] <pingfloyd> oxek: so now you need to find an alternate means that it will be sourced after X is up but before your start a terminal emulator.
117[01:19:52] *** Quits: mibo (~mibo@replaced-ip) (Remote host closed the connection)
121[01:20:28] <pingfloyd> there's a lot of different ways to tackled this whole stupid dilemma. Don't rule out maybe the best solution of all, changing which DM you use.
122[01:22:00] *** eth01 is now known as fosshost
123[01:22:22] <oxek> sounds great... why make things simple, right?
124[01:22:27] <oxek> I'll see what I can do
125[01:22:30] *** fosshost is now known as eth01
126[01:23:24] <miskatonic> I use startx from the console, so the profile is already loaded from the outset
127[01:23:30] <ham5urg> The most howtos are describing a configuration of exim via dpkg-reconfigure. But this lacks the access-rights. How can a (non-root-server-)program send an email if the mta is owned by root?
128[01:23:56] <ham5urg> And if exim4 is replaced by mstmp
129[01:24:53] <miskatonic> does lightdm not read Xsession?
130[01:25:01] <pingfloyd> another way, ugly and hackish is say try editing /etc/xdg/xfce4/xinitrc or ~/.config/xfce4/xinitrc < source ~/.profile
131[01:25:19] <pingfloyd> miskatonic: I can't remember
154[01:31:56] <somiaj> I have functions that check if something is in PATH before adding it, so I dno't have to worry about double enteries if .bashrc gets sourced twice
164[01:37:29] *** Quits: null1337 (~WhoAmI@replaced-ip) (Quit: Oops. My brain just hit a bad sector)
165[01:37:37] <somiaj> my .profile checks if I'm running bash, and if .bashrc exists, then sources that, and I do all my config in .bashrc (since my use case doens't need to distniguish between interactive and login shells)
166[01:37:51] <somiaj> I've never tired to see what happens if you create a loop sourcing back and forth.
167[01:38:33] <somiaj> but that does mean my .bashrc often gets source twiced by the time I open an xterm, hence my functions to check things fist.
176[01:42:42] <oxek> somiaj: I thought I need a display manager to have a working lock screen
177[01:43:23] <annadane> lightdm has had tons of reported bugs, it's kind of funny
178[01:43:36] <somiaj> oxek: not as much anymore, it use to be startx would run on a different tty, so you could switch to the tty your ran startx on, but now startx runs on the same tty, and this isn't the case
222[02:15:48] <oxek> ok I edited my ~/.bashrc to add ~/.local/bin to my PATH and it works
223[02:15:58] <oxek> I can now run anything I placed in ~/.local/bin
224[02:16:47] <oxek> however, I spotted that it added ~/.local/bin to the beggining of my path, which makes me think that any program that can add stuff to ~/.local/bin can create e.g. ~/.local/bin/ls which will override the functionality of ls
225[02:17:00] <dvs> Why have a ~/.local/bin? Why hide it?
226[02:17:03] <oxek> so I guess I should add ~/.local/bin to the end of my $PATH, right?
227[02:17:20] <oxek> dvs: I already have ~/.local so I thought I might use it
228[02:17:26] <oxek> whereas I don't yet have ~/bin
229[02:17:45] <mason> oxek: Traditional is ~/bin. And if you're worried about system binaries being overridden, have it show up at the end of your PATH.
230[02:18:11] <dvs> But ~/bin comes already defined?
231[02:18:24] <oxek> dvs: appaerntly not on debian xfce by default
248[02:23:06] <oxek> somiaj: you're right, I did not spot that behavior
249[02:23:12] <somiaj> on my system it is in /etc/skel/.profile, if [ -d "$HOME/bin" ] ; then
250[02:23:32] <somiaj> so if you have the stock .profile it should work, (note not all xsessions read .profile though)
251[02:23:41] <oxek> all I needed to do was create ~/bin and/or ~/.local/bin and they then get added to my path when I source .bashrc or start new terminal
252[02:24:04] <somiaj> no, ~/.local/bin won't, at least on the debian .profile I se
253[02:24:18] <somiaj> oh wait it is there, I can't keep up with all these changes
254[02:24:54] <dvs> somiaj, I see .local/bin in mine too
255[02:25:01] <dvs> how about that?
256[02:25:11] <somiaj> dvs: yea, I overlooked it, didn't realize both were there.
257[02:25:40] <dvs> oxek, so are you still going to use ~/.local/bin
258[02:26:28] <oxek> dvs: I am willing to learn. I don't see much difference between the two, except that ~/.local already exists and contains ~/.local/share so I might as well add bin there
259[02:26:38] <somiaj> seems it was added in bash (4.4.18-1) to close bug #839155
261[02:27:40] <somiaj> oxek: you can arrange your data in $HOME however you like, ~/bin was the old standard, all putting it in .local/bin is put it in a hidden dir
264[02:28:05] <mason> This is the first I've heard of ~/.local/bin - when did it pop into existence?
265[02:28:25] <oxek> I guess if ~/bin was old, then I am gonna go with ~/.local/bin if that's new
266[02:28:49] <somiaj> mason: more and more things are using things like this, ~/.local, seems I have some local python and ruby packages installed there too
267[02:29:10] <dvs> ugh
268[02:29:58] <mason> oxek: It's of course up to you, but traditional paths mean you won't find yourself on systems where nothing is where you expect it. Plus, you know, replaced-url
269[02:30:06] <oxek> dvs: is there any particular problem with that
270[02:30:19] <mason> oxek: Watch the video and THEN decide.
271[02:30:22] <somiaj> freedesktop.org keeps updating their stanadrds.
272[02:30:22] <oxek> mason: oh I fully expect to not find things where I expect them, my memory is not what it used to be :D
273[02:31:02] <dvs> oxek, I don't see why it has to be in a hidden directory.
275[02:32:46] <themill> most people don't seem to want ~/share and ~/lib and so ~/.local makes sense so that you can have the normal hierarchy somewhere within ~/ ; at that point ~/.local/bin becomes quite sensible
276[02:32:52] <oxek> dvs: dunno. Something already created ~/.local/share and is putting whole bunch of stuff in there
277[02:33:03] <dvs> I'd be worried if something wasn't working right because a script is running an old version of a program hidden in a hidden directory.
278[02:33:48] <oxek> dvs: isn't there a saying like "tradition is just peer pressure from dead people"
280[02:34:11] <themill> dvs: forgetting that you put something in ~/.local/bin is no more or less of a problem than forgetting you put something in ~/bin
281[02:34:48] <dvs> but at least I can see bin in my home directory and it could ring a bell
282[02:34:59] *** Quits: cast4dx (~nt80@replaced-ip) (Remote host closed the connection)
283[02:35:38] <somiaj> seems to be the reason for ~/.config, so all configurations are more centrally located than each having their own hidden file/directory, can often tell the age of software by the conventions it decides to use.
284[02:36:16] <somiaj> just hard for me to keep up with all these changes
285[02:38:30] *** Quits: deb (~deb@replaced-ip) (Remote host closed the connection)
286[02:40:15] <oxek> I wonder if ~/.bashrc will move to ~/.config one day
292[02:42:54] <somiaj> also due to backwards compadability, I use fvwm, and it now supports like 8 places to find its config file as the conventions have evolved over time.
294[02:43:38] <somiaj> so even if bash also used ~/.config/bashrc, it would still want to check ~/.bashrc for existance if the new place doesn't exsit to not break existing systems.
325[03:10:28] <DeaDSouL> somiaj: here replaced-url
326[03:10:35] *** Quits: Kralle (~Kralle@replaced-ip) (Remote host closed the connection)
327[03:11:15] <DeaDSouL> somiaj: the docs don't mention anything about xattrs, onwership or ACL
328[03:11:20] <sponix> DeaDSouL: think he is hinting that Debian 10 stable likely has at least a handful of actually supported backup programs available
341[03:15:38] <sponix> DeaDSouL: but to answer your original question. No, I don't know anything about it. If anyone else does, hopefully they will speak up
342[03:15:44] <DeaDSouL> sponix: I just wanted to give it a shot here first..
348[03:16:54] <sponix> I am old school and just use rsync .. It is basic, just like me
349[03:19:02] *** debhelper sets mode: +l 1265
350[03:19:04] <somiaj> in general we only support software that is included with debian.
351[03:19:40] <DeaDSouL> sponix: I use it too for my own workstation and servers.. but the reason I went with urbackup, is it can be used on any OS, and it provides one webui to manage them all,
372[03:30:13] <DeaDSouL> somiaj: it's not just about owner:group, some of them are having special modes, besides I'm backing up /home, /etc, /root, /usr/local and /opt
384[03:44:11] *** Quits: tyzef (~tyzef@replaced-ip) (Remote host closed the connection)
385[03:45:12] <somiaj> I was more poiting out you could backup a lot of your user data on this new place that works for you and other users, but then have a seprate backup for the os. There are tons of backup models.
398[04:01:17] <sponix> tabb: Logitech C920 you can just plug in, and have it start working out of the box well
399[04:03:18] <sney> I have a microsoft one that worked out of the box, too. and it's been years since I've seen someone need extra work for one of the built-in laptop sort. they all seem to be UVC
465[05:09:53] <swivel> hello, quick question: I'm trying to dist-upgrade a small debian container from stretch to buster and it wants to install mailutils and exim4 garbage when this container only runs nginx as a proxy and has no mail anything installed
466[05:10:09] <swivel> how do i prevent this nonsense from getting installed by dist-upgrade
467[05:10:51] <sney> it's probably coming in as a recommend for one of the packages that you already have
468[05:11:19] <velix> Is Exim4 the default MTA of Debian Buster?
580[07:42:50] <Destro> fareast, Install a .deb package and then use apt-get install package (without writing the sudo) and then use dpkg -i package.deb --force-not-root --root=$HOME
595[07:47:56] <luckylinux> Hi everybody. I am having some weird issue for the first time ... I used debootstrap in the past with great success. On this system however, for some reason not all base packages are getting installed, notably base-files and base-passwd, so I'm getting chown: invalid user: 'root:root' and I cannot get much further. Any easy way to solve this ?
597[07:48:45] <luckylinux> a /etc/passwd.lock and /etc/shadow.lock was created. I tried to remove it to se if it could unblock the situation, but it had no effect
621[08:06:36] <luckylinux> themill, I will try reinstalling the required / important packages that seem to be part of the base system with required=$(aptitude search ~prequired -F"%p")
626[08:09:30] <luckylinux> Very weird however. It's the first time it's doing me that. Only thing is that I changed from my local mirror apt-cacher-ng to direct debian server (since I am currently migrating my NAS :S). Maybe debootstrap has a timeout that triggers due to too slow internet connection ?
628[08:11:09] <themill> shouldn't do; it's failure modes are unfortunately very hard to figure out, however
629[08:11:59] <luckylinux> themill, Well this is the only change I did, I cannot figure out why it would do it other. Just changed the debootstrap source to one of the official debian mirrors
630[08:12:17] <luckylinux> after installing those packages it seems to not complain too much anymore
631[08:13:45] <themill> that will have primed your cache, but shouldn't otherwise make any difference. Copying around passwd etc is interesting though. Did you have more than one root account, I wonder?
639[08:17:03] <luckylinux> and on the "host" I just checked. Only 1 root account, I always used this USB stick to setup a ZFS partitioning / zpool, then install using debootstrap and chroot into it. Debootstrap didn't indicate any error except one at the begigging
640[08:17:08] <luckylinux> *beginning
641[08:18:05] <luckylinux> cannot remember the exact error but it was something along the line of "Cannot get Package list / Packages.gz" or something
642[08:18:17] <luckylinux> but then it continued going on without any error or warning
643[08:18:35] <luckylinux> unpacked all files (although the list of unpacked files was quite short to other installs)
829[11:47:01] <ham5urg> At a first-time-ssh-connection I get the remote fingerprint to accept. How can I generate the fingerprint at server side, take him (e.g. via usb-stick) to the client and tell it to ssh before initiating the connection?
877[12:14:13] <ham5urg> My client is getting the SHA256 ECDSA key fingerprint. ssh-keyscan is producing (default) other fingerprint ed25519. How can I get the sha356 fingerprint in particular?
878[12:14:17] *** Quits: edlou (uid413273@replaced-ip) (Quit: Connection closed for inactivity)
895[12:27:01] <oxek> ham5urg: it makes sure that anything that tries to connect is at least recent enough to understand ed25519. It gives me at least the knowledge the connecting program has been updated in the last few years.
896[12:27:45] <oxek> plus it prevents loads of automated malware that only understands rsa or dsa
933[12:49:41] <brutser> on debian host and guest, with vbox and guest additions it works out of the box, with qemu-kvm i cannot get mouse button 4,5 to work (left side of mouse) inside guest vm - it seems not very trivial how to do this, but i cannot imagine it's impossible. can anyone help me out here?
936[12:53:06] <ratrace> brutser: even with full mouse capture (Ctrl-Alt-G I think) ?
937[12:54:09] *** Quits: kreyren (~kreyren@replaced-ip) (Remote host closed the connection)
938[12:55:39] <ham5urg> Has anyone managed to have the private key on some fat32-formatted usb-stick and to login via ssh without complaining "Permissions 0644 for id_rsa are too open."?
939[12:56:26] <ham5urg> Hot to store the key on some fat32 usb stick which does not know about permissions?
940[12:56:48] *** Quits: BenNZ (~Ben__@replaced-ip) (Quit: Everytime I think IQ's must have dropped recently, I remember that this is the internet)
945[13:00:16] <brutser> ratrace: i will check that, but it would be nice if i could get it to work just as the vbox guest additions work, there the mouse buttons all work
946[13:00:24] <brutser> i cannot imagine that cannot be done for qemu-kvm
1006[13:54:42] <ratrace> f8e4: first of all, that include pattern probably won't do what you want because you're missing a full filter with exclusions. second, ~ is invalid . just "steve:" means the default directory upon ssh login, which is usually $HOME . otherwise specify full path: steve:/home/steve/or/wherever/
1007[13:54:59] *** Quits: eliotome3000 (~eliotime@replaced-ip) (Remote host closed the connection)
1009[13:55:38] <ratrace> f8e4: with such search patterns where you don't know the top level dir and want ALL of them that DO have the "dist" as second dir, which DO have "*.whl" inside them, better to use `find` to build a list of files, then use that list for rsync
1013[13:57:08] <ratrace> otherwise you need --filter and you need to include all path elements individually, for all the patterns you want included, followed by -* exclusion. see the manpage, FILTER RULES
1015[13:58:05] *** abdulocracy3 is now known as abdulocracy
1016[13:58:12] <ratrace> f8e4: btw, use -n (--dry-run) optionally with -t as well, to see what would be transferred, until you get the desired pattern correct
1023[14:01:25] <ratrace> well you don't have a simple requirement there. fastest but hackiest way is to use find to build a list, then use rsync --files-from=that_list.txt
1024[14:02:22] <ratrace> find ./ -wholename '*/dist/*.whl' > that_list.txt or whatever you seek.
1054[14:15:52] <ratrace> f8e4: btw, looking at the manpage, it appears you can use --files-from=- to feed the list from stdin, so find ... | rsync ... --files-from=- ./ steve: might actually work. just note that files-from paths are RELATIVE to the src dir in your rsync
1055[14:16:32] *** Quits: Tobbi (~Tobbi@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1086[14:29:17] <ham5urg> I want to write a script to check for apt update and send email via sendmail if something is to be updated. Is there a oneliner to tell 'apt update' to deliver a 0 or 1?
1087[14:29:55] <ratrace> ham5urg: why not use the apticron package
1088[14:29:57] <cyveris> ham5urg: Check out 'unattended-upgrades'.
1089[14:29:58] <petn-randall> ham5urg: Have you heard of the package 'unattended-upgrades' yet? It can do just that for you.
1090[14:29:58] <kreyren> petn-randall, and this is unfrankenstaining -> supported here right :3
1107[14:32:44] <GeneralStupid> Hi, it looks like that the debian installer is not able to create a grub which is capable of booting from a mdarray
1108[14:33:04] <GeneralStupid> Is there an easy way to achieve this or do i have to configure than manually?
1109[14:33:21] <petn-randall> GeneralStupid: Hi, how did you prepare the installation medium?
1110[14:33:30] <cyveris> GeneralStupid: Sure it is. When it asks which device(s) to install grub to, you need to enter one or more of the component devices, NOT the md array.
1111[14:33:41] <ratrace> petn-randall: that must hurt less than managing that ... guixgentooexherbian :)
1112[14:33:54] <GeneralStupid> cyveris: i did grup-install to both devices
1113[14:34:06] <GeneralStupid> petn-randall: prepare? i just bootet from usb
1114[14:34:17] <cyveris> GeneralStupid: If you did grub-install, then you didn't do that during the install process.
1115[14:34:19] <petn-randall> GeneralStupid: Yes, how did you prepare the usb image?
1116[14:34:22] <GeneralStupid> cyveris: i used that same setup with gentoo before, so i did it the same
1117[14:34:39] <joepublic> it's like it's gentoobian day
1118[14:34:41] <GeneralStupid> cyveris: i chosed both partitions
1119[14:34:46] <ratrace> GeneralStupid: you didn't install grub onto md though
1120[14:34:52] <cyveris> ratrace: You shouldn't.
1121[14:34:56] <ratrace> right
1122[14:35:05] <cyveris> GeneralStupid: You don't install grub to a partition. You install it to a disk.
1123[14:35:08] <GeneralStupid> i didnt do that
1124[14:35:18] <GeneralStupid> i installed it two both disks
1125[14:35:26] <joepublic> you can install grub to a partition, useful in certain situations
1126[14:35:35] <GeneralStupid> it didnt work. grub config does not load md stuff
1127[14:35:35] <cyveris> joepublic: You CAN. But in this case, you don't.
1128[14:35:51] *** Quits: icecream (~icecream@replaced-ip) (Remote host closed the connection)
1133[14:36:29] <cyveris> No, it can't. That's why grub needs to be installed to the MBR (or GPT equivalent BIOS boot partition) of one or more of the physical disks.
1134[14:36:59] <ratrace> on wait, with GPT you still install to /dev/sda for example, NOT the bios_boot partition
1135[14:37:09] <ratrace> grub then finds the bios_boot partition and puts its second stage there
1136[14:37:22] <cyveris> Yup.
1137[14:37:23] <GeneralStupid> What i did, i chosed manual partition, selected my physical drives as RAID devices and created a software raid device. md0 was created and mounted. I installed the stuff and then i installed grub via the graphical installer first on sda1 then on sdb1
1138[14:37:36] <cyveris> GeneralStupid: And that's your problem.
1139[14:37:44] <cyveris> You installed grub to a partition, NOT the MBR/BIOS boot partition.
1140[14:37:48] <kreyren> petn-randall, here is the promised image of happy debian ^^ replaced-url
1143[14:37:54] <cyveris> You need to instsall grub to /dev/sda and /dev/sdb.
1144[14:38:04] <cyveris> Otherwise, it won't be located in a place where BIOS knows to look.
1145[14:38:04] <ratrace> GeneralStupid: usually problems with manual raid and debian in my experience is in naming of the md. md0 while you install manually, but it becomes md127 on reboot.
1158[14:40:51] *** Quits: icecream (~icecream@replaced-ip) (Remote host closed the connection)
1159[14:40:52] <petn-randall> :)
1160[14:40:54] <oxek> Anyone know if it is possible to not send "Debian-10+deb10u2" in the client identifier of the ssh client (currently sends ""SSH-2.0-OpenSSH_7.9p1 Debian-10+deb10u2") without recompiling the ssh client? I know I can disable it for sshd server using "DebianBanner no" but I am talking about the client.
1161[14:41:05] *** Quits: Hunterkll (~hunterkll@replaced-ip) (Remote host closed the connection)
1162[14:41:42] <cyveris> My first question is "why?"
1170[14:43:25] <oxek> petn-randall: sshconnect.c:send_client_banner(int connection_out, int minor1)
1171[14:43:41] <ratrace> (and it's just Banner, not DebianBanner)
1172[14:43:42] <oxek> can be observed e.g. when you do a client test on replaced-url
1173[14:44:08] <ratrace> oxek: are you really connecting to untrusted ssh servers?
1174[14:44:09] <GeneralStupid> cyveris: iam pretty sure i installed it on the device and made a mistake in my text. Debian Installer only showed two devices to install grub on...
1183[14:46:27] <petn-randall> GeneralStupid: I also asked about how you prepared the installer because a common mistake it to use things like unetbooting, which mangle the installer, and you end up not being able to write grub to the HDDs.
1184[14:46:27] <oxek> ratrace: Yeah I was wondering if I missed something. I can change that function to omit the debian part, but then I'd have to compile and keep everything up-to-date myself, so I was hoping there was something else.
1185[14:47:04] <GeneralStupid> petn-randall: i used dd to a usb key
1197[14:55:37] <GeneralStupid> oxek: actually it should be really easy. you could download the debian package source, just add one patch and recompile. This could be automated and will only fail if your patch is no longer compatible
1198[14:55:51] <petn-randall> GeneralStupid: I've got a system with mdraid raid1, and I'm using grub2, so I'm certain it does work.
1200[14:56:53] <oxek> GeneralStupid: maybe I will try it. Though I am definitely not yet confident enough in my abilities in patching code and compiling it. The most I've done was backport some package from debian-testing.
1201[14:57:23] *** Quits: DLange (~DLange@replaced-ip) (Remote host closed the connection)
1202[14:57:27] <annadane> thanks ratrace :D
1203[14:57:55] *** Joins: conta (~Thunderbi@replaced-ip)
1204[14:58:15] <ratrace> annadane: could CSS it to open up as a separate tiny window, with more bling! :)
1205[14:58:17] <GeneralStupid> /boot is on your md array?
1251[15:26:29] <petn-randall> Is there a reason you want to speak to them privately?
1252[15:26:34] <stockholm> indian, older then me!
1253[15:26:49] <stockholm> i want to ask him about make
1254[15:27:01] <stockholm> he used to maintain gnu make i think
1255[15:27:05] <petn-randall> Regarding what exactly?
1256[15:27:21] <stockholm> i am looking for an irc channel about gnu make
1257[15:27:27] <stockholm> and google failed me
1258[15:27:42] <annadane> well, #gnu is the channel for general gnu related queries
1259[15:27:44] *** Quits: dastier (~dastier@replaced-ip) (Remote host closed the connection)
1260[15:27:55] <stockholm> yes, that might be a good first step
1261[15:28:23] <annadane> make is also pretty common, if you ask your question i'm positive someone will know the answer
1262[15:29:01] *** debhelper sets mode: +l 1297
1263[15:29:21] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
1264[15:29:23] <stockholm> annadane, ask here?
1265[15:29:39] <annadane> ask in #gnu i meant, though i suppose you can ask here if you're running debian
1266[15:29:52] <annadane> it's up to you, #gnu might be more helpful
1267[15:30:05] <stockholm> i want to use that makefile on both debian and windows
1268[15:30:31] <stockholm> and many of the useful tools are not there on windows
1269[15:30:37] <stockholm> like sed or awk
1270[15:30:42] <annadane> gnu's website also has pretty extensive documentation
1271[15:31:03] <stockholm> oh yes!
1272[15:31:04] <annadane> well... makefiles install things to specific locations and you probably can't do that on windows
1273[15:31:29] <annadane> the *general* way one runs linux stuff on windows is cygwin but i doubt make works like that
1274[15:31:55] <petn-randall> stockholm: You'll likely need to find a way to install all those on Windows, and even then many things will differ, so I don't think you're saving a lot of effort by unifying them.
1275[15:32:34] <stockholm> perhaps.
1276[15:33:38] <annadane> i've heard of xy problems but this is ridiculous :P
1277[15:34:12] <ratrace> annadane: these are xyzzy problems. complete with grues in the night. :)
1286[15:37:56] <annadane> "what is the name of this debian dev" -> "because he used to maintain make" -> "i'm looking for the irc channel for make" -> "because i want to use makefiles on debian and windows"
1287[15:38:02] <annadane> next time, just ask your question :P
1288[15:38:46] <cyveris> We need a bot command for xyzzy that says "nothing happens."
1340[16:09:40] <annadane> unless you mean does less have it natively
1341[16:10:20] <ham5urg> I want to disable all local logins via console. Login via ssh (without password but with id_rsa) is already configured. What's the best approach?
1342[16:10:57] <joepublic> say that again slowly. disable all local logins?
1343[16:11:39] <jolt> f8e4: In "less", press "shift+g" to go to end of file. Bonus tip: "shift+f" to "follow", like "tail -f"
1347[16:12:48] <ham5urg> joepublic, yes. The server are remote and I won't see them again in a long time (hopefully). I want to disable all local logins into the machine
1348[16:13:12] <joepublic> one normally prevents unwanted local logins by using strong passwords.
1349[16:13:25] <joepublic> this will not work for your use case, I take it
1350[16:13:58] <ham5urg> I want to disable the usage of passwords. id_rsa is good for me.
1351[16:14:03] <joepublic> are you also taking precautions against someone booting a live cd?
1352[16:14:16] <invra> ham5urg: local login would be possible with a live ISO anyways... unless you phisically remove USB ports aswell..
1353[16:14:23] <jolt> ham5urg: You can lock out users with /etc/securetty. But BEWARE. Sometimes it's good to be able to login locally, so think about what you are doing.
1354[16:14:28] <invra> ham5urg: you can disable password
1355[16:14:55] <jolt> annadane: There are a bunch more keys, to open in editor etc, it's worth a read.
1356[16:14:57] <ham5urg> If I need to login locally I can boot via usb and change it again.
1357[16:15:03] <cyveris> So can anyone else.
1358[16:15:19] <ham5urg> invra, how? just remove it from passwd?
1359[16:15:27] <jolt> Yeah, you need to set a boot pw, otherwise you can just use init=/bin/<shell>
1384[16:19:17] *** Quits: cnsunyour (~cnsunyour@replaced-ip) (Remote host closed the connection)
1385[16:19:22] <jolt> ham5urg: vipw -s, set their pw-string to just * in /etc/shadow
1386[16:19:29] <joepublic> of course, typing FAepGvl*7MwyzvCbq3v7$SB#SxBMa%ZAIRw*nt!w2$9YKjS5EVsDjm1cjWi#wHUrCG0TB0%9ejiA%ZrcHi@rVKNjqr&XP@04zoE every time you sudo would be something of a pain
1390[16:19:49] <jolt> ham5urg: then their pw will never match, but they are still allowed to login via non-pw measures, like ssh key.
1391[16:20:41] <jolt> (since the nologin suggestion would block their ssh logins as well). You could potentially also make a pam rule for whatever setup you want, but I'm no pam expert, so I can't help you out there.
1392[16:21:38] <jolt> ham5urg: maybe just comment out the contents of the common-password file in /etc/pam.d
1393[16:21:59] <invra> i smell imminent catasthropee
1394[16:22:06] <jolt> :D
1395[16:22:34] <jolt> I have added 2fa for ssh-logins not in my office subnet, also by utilizing pam.d, so it definitely works
1396[16:22:45] <jolt> But yeah, is scary
1397[16:23:14] <ham5urg> jolt, I will try
1398[16:23:39] <invra> i fidn that a strong passwords does the same withot the risk of messing up everything xD
1399[16:23:53] *** Quits: ijash (~ijash@replaced-ip) (Remote host closed the connection)
1416[16:34:25] <ham5urg> invra, I trust it, but I don't want to handle any password anymore.
1417[16:34:43] <GeneralStupid> oh its not an installer issue...
1418[16:35:59] <joepublic> GeneralStupid, do you suppose super grub 2 rescue disc would help you?
1419[16:36:36] <invra> yeah then you can set a whatever crazy password, and forget about it... then just use ssh key authentication... no need to manage it
1420[16:36:58] <jolt> ham5urg: well, setting * instead of the password hash do work.
1421[16:37:08] <invra> ^
1422[16:37:58] *** Quits: Hokedli (~laslie@replaced-ip) (Remote host closed the connection)
1440[16:42:45] <joepublic> I mention super grub 2 rescue because it can boot into just about anything, allowing you to install grub properly from within the booted system
1441[16:43:11] <GeneralStupid> the only difference is single user mode
1442[16:43:19] <GeneralStupid> grub is installed and working
1443[16:43:55] <joepublic> I repent of having so mentioned.
1570[17:41:00] <somiaj> another option is have your own rm script in /usr/local/bin that does trash-put for you
1571[17:41:22] <somiaj> since the /usr/local/bin/rm would take presence over /usr/bin/rm, and you could still use the /usr/bin/rm if ever you wrote out the full path
1572[17:41:35] <somiaj> (though others might be surprised if they use your system and wanted to really rm something)
1573[17:42:04] <somiaj> well othes and maybe scripts, so probably not a good idea in the long run
1574[17:42:39] *** Quits: sti (~ivstojic@replaced-ip) (Quit: leaving)
1575[17:42:47] <wyre> somiaj, well, all of this started because I usually create an alias also for rm, something like 'echo Use 'del' or the full path i.e. '/bin/rm''
1576[17:43:16] <wyre> for both users, root and wyre
1581[17:44:23] *** Quits: GyroW (~GyroW@replaced-ip) (Quit: Someone ate my pie)
1582[17:44:45] <wyre> and I can create those aliases for wyre and root but I'm still able to combine 'sudo rm'
1583[17:44:57] <wyre> and I'm not getting the warning
1584[17:44:58] <somiaj> well same issue with the alias if all root shells get it, you can mess up system scripts who try to remove something and end up not
1585[17:45:06] <miskatonic> whereas tcsh abuses aliases as functions
1586[17:45:28] <somiaj> sudo doesn't use the full enviorment of the user that ran it, so it wouldnt get aliases.
1587[17:45:51] *** Joins: stockholm (~Andreas@replaced-ip)
1588[17:45:59] <somiaj> Though I guess I don't really remove stuff I don't want to, another solution is to have a good backup, so you can just recover stuff from the backup instead of have a trash
1589[17:46:13] <wyre> somiaj, then there is not way to prevent totally a bad rm usage!
1590[17:46:21] <somiaj> the few times I've accidently removed the wrong thing, recovering it from the backup was my solution.
1592[17:47:09] <somiaj> wyre: I think you have to be careful and consider the consequences of how it affects scripts (mostly dpkg scripts that remove stuff on removal or other system scripts)
1600[17:49:44] <wyre> deb, lol, imagine you want delete something in ./ but you forget the .
1601[17:50:21] <somiaj> wyre: personally I would just learn how to use rm correctly, and if you create regular backups using rysnc, you have a backup in case you delete something you shouldn't have.
1602[17:50:28] *** Quits: nickodd (~nickodd@replaced-ip) (Remote host closed the connection)
1603[17:50:48] <wyre> somiaj, and could I not prevent other users to use rm through sudo?
1604[17:50:55] <wyre> just my particular user
1605[17:51:08] <wyre> but ... why not all users in sudo group 🤔
1606[17:51:16] <wyre> ?
1607[17:51:50] <dvs> wyre, then you write a sudo line for your user instead of using the %sudo group
1620[17:55:31] <somiaj> there is already a trash-rm utility, you can install and use, I would just train yourself to get use to that, and keep backups, vs train yourself that rm is safe to use
1632[17:58:58] <wyre> apparently this is not to train xD
1633[17:59:01] *** debhelper sets mode: +l 1307
1634[17:59:02] <somiaj> wyre: trash-cli is a package that contains a set of tools like trash-rm, list-trash, empty-trash, restore-trash, so you don't have to write your own 'del' tool
1635[17:59:15] <somiaj> so you can trash-rm /path/to/file
1636[17:59:21] <wyre> somiaj, ofc, del is just a short alias for trash-put
1641[18:00:09] <somiaj> But I do like that comment, best get use to the command as stated, in case you are on an envirment or system that doens't have your customizations
1664[18:11:40] <dvs> wyre, I'm not saying that it's not working. I'm saying that if you only want one user to be affected, then put that specific user in the sudoers file, not the whole %sudo group
1665[18:11:49] <somiaj> nvidia-installer-cleanup is not required, you could just remove that package, espically since it probably has cleaned up yoru system.
1735[18:39:31] <wyre> what do you mean with "how to use rm correctly"?
1736[18:40:00] <wyre> I know how is rm used, but I don't want a typo causing destruction 😆
1737[18:40:15] <wyre> at least I know the risks of rm usage, I mean
1738[18:41:20] <wyre> and I cannot afford another HDD to do backups ..
1739[18:42:33] <miskatonic> the put it up for anonymous ftp and let the whole world mirror your server ... as suggested by Linus Torvalds over 20 years ago
1742[18:46:24] <somiaj> wyre: as I menteiond from the post, if you train yourself that this custom rm script is common, if you are on a system that doesn't have it, you may find yoruself in a worse case, vs undersatnding rm is permanat and to be careful before using it.
1743[18:46:57] *** Quits: r1ch (~r1ch@replaced-ip) (Quit: My MacBook has gone to sleep. ZZZzzz…)
1753[18:51:11] <somiaj> wyre: I would, as this would transfer to other systems better
1754[18:51:39] <somiaj> if you are in the habbit of always typing trash-rm to remove things, if you are on a system without trash-rm, then you get an error that command is not found, you can either install it or realize oh I have to be careful.
1756[18:52:03] <somiaj> while if you are in the habbit of rm telling you to use trash-rm, you may not fully train yourself and end up deleting something on another system without meaning too
1758[18:52:48] <wyre> it would be trash-put instead trash-rm, but I think I get your point
1759[18:52:52] <somiaj> Your stoping rm from working with sudo and alias rm may work to help train yourself, but if you have a safety net, you may not fully train.
1760[18:53:00] <wyre> trash-rm is to remove thing from inside trash
1762[18:53:23] <somiaj> ahh, thanks, I was just going off a post that said alias rm with trash-rm, so I assumed that was what it did, I don't use such tools
1763[18:54:29] <wyre> I found them more safety when I'm removing personal files from cli
1767[18:56:17] <somiaj> but as for backups, you should take them regardless, 'data that isn't copied to three different mediums can be considered lost', what happens if your hd fails, your machine burns down, etc
1775[19:01:59] <wyre> somiaj, what about an alias for rm as root?
1776[19:02:06] <wyre> that would affect scripts?
1777[19:02:43] <somiaj> wyre: It depends on what shell the root scripts are using. so it isn't a clear cut question, but yes if the alias was present it could have consequences of leaving stuff on your system you told dpkg to remove
1778[19:03:21] <wyre> somiaj, the root shell is fish
1790[19:12:33] <wsky> is there anything special i need to do to keep my stretch under long term support?
1791[19:12:58] <somiaj> you don't have to change anything to get LTS support for stretch
1792[19:13:32] <somiaj> do note that for desktop users, you do loose security support of some software. The webbrowser being the biggest, I know one of chormium/firefox-esr is no longer supported in stretch.
1798[19:21:54] <miskatonic> what is not supported in Stretch?
1799[19:21:59] <jmcnaught> somiaj: as far as I know firefox-esr is still supported on stretch (it's not listed in /usr/share/debian-security-support/security-support-ended.deb9, but chromium is)
1800[19:22:30] <jmcnaught> the package debian-security-support has the files that list support status, and a check-security-status command
1809[19:32:19] <ham5urg> How is apt-listchanges invoked/called from unattended-updates?
1810[19:32:44] <Agiofws> hello
1811[19:33:25] <ham5urg> Man, I should read manual pages more serious: Please note that in the default installation if apt-listchanges is run during upgrades as an APT plugin, it displays NEWS.Debian entries only. This can be changed with the --which option.
1815[19:35:42] <Agiofws> i have just upgraded to buster but i can't install wine i have unmet dependencies winehq-staging : Depends: wine-staging (= 5.7~buster) Unable to correct problems, you have held broken packages.
1816[19:35:51] <ham5urg> Ahh, not in /etc but in /etc/apt. Good
1817[19:36:00] <Agiofws> i need some help
1818[19:37:04] <annadane> Agiofws, winehq has specific instructions for debian users
1828[19:40:04] <Agiofws> annadane, i think installing wine through the link and what i've read through the post ist still a mess
1829[19:41:28] <annadane> i've only installed wine-stable, not staging
1830[19:41:35] <annadane> so i can't vouch
1831[19:41:40] <somiaj> Agiofws: do you really need the newest wine, debian does provide wine and wine-devlopment that works for many things
1832[19:42:06] <jmcnaught> Agiofws: that error looks like a problem with winehq.org's third-party repository, but I noticed on their Debian wiki page that they say to use "--install-recommends"… do you maybe have recommends disabled. Also I agree with ^, why not use the Debian provided packages?
1833[19:42:49] <somiaj> Agiofws: as for your issue, can you paste the output of apt policy (wonder if winehq is pinning its repo like buster-backports)
1871[20:03:10] <annadane> libfaudio mentioned before, has 64 bit and 32 bit versions, which you'll know if you read what i posted earlier, Agiofws, maybe that's also partially the issue
1872[20:03:20] <annadane> somiaj, yeah i know i'm just surprised it came from there
1881[20:06:13] <somiaj> anyways that policy is a mess in my mind. It probably works, but webmin, dmo, lots of third party sources can start to cause issues.
1884[20:06:36] <warsoul> somiaj, you got the pastebin?
1885[20:06:38] <somiaj> that is not the second apt policy command I asked for
1886[20:06:48] <somiaj> warsoul: did you remove nvidia-installer-cleanup package?
1887[20:06:54] <miskatonic> webmin is unsupported since a few years
1888[20:06:57] <somiaj> !webmin
1889[20:06:57] <dpkg> Webmin is a lame web-based interface for unsafe system administration for Unix. Check it out at replaced-url
1890[20:07:20] <somiaj> Agiofws: you are using an old repo from sarge for a piece of software that is known to be terrible and full of security holes. Though I doubt this has issues with your current problem.
1891[20:07:24] <warsoul> somiaj y did upgrade it says to remove it i said YES
1896[20:07:44] <somiaj> warsoul: I might have missed your last output, but really you need to learn how to read and intrupit things here.
1897[20:08:04] <Agiofws> apt remove webmin
1898[20:08:47] <warsoul> there is the pastebin of the error
1899[20:08:48] <somiaj> warsoul: I saw that, and no you did not remove the package. 'apt remove nvidia-installer-cleanup', we should try to get to a poitn I don't have to type out each and every command you need
1900[20:09:29] <somiaj> Agiofws: where is the second apt policy wine-staging I asked for
1901[20:09:44] <warsoul> somiaj, debconf: DbDriver "config": /var/cache/debconf/config.dat is locked by another process: Resource temporarily unavailable
1902[20:09:50] <somiaj> Agiofws: webmin isn't the issue here, but with looking at your sources.list, it apepars you like to mix a lot of things which could cause problems.
1904[20:10:31] <somiaj> warsoul: have you tried to reboot, you have your system in some crazy state due to closing a terminal while running apt the last time, and I don't really have the time to guide you through every step
1905[20:10:40] <somiaj> warsoul: maybe someone else will help you
1907[20:11:24] <somiaj> Agiofws: you can try apt install wine-staging, but you are going to head down a rabit hole and figure out what conflict is preventing this package from installing.
1910[20:12:19] <somiaj> Agiofws: so if wine-staging doens't install, repeate the process of figuring out what packages are causing problems, use apt policy pkgname and then try to install the depends that are causing problems, eventaully you should run into exactly what package is causing problems due to the version on your system or the version it wants to install not matching what the software depends on
1912[20:12:55] <somiaj> again it appears to be a multiarch problem, but now try wine-staging-i386, and keep going until you figure out what package is causing the problem
1913[20:13:16] <somiaj> first use apt policy to see if it is even isntallable
1914[20:13:56] <warsoul> somiaj, you are a really good helper
1917[20:14:20] <somiaj> I'm also a really good procstranator, I should really be doing my actual work
1918[20:14:50] * dvs narcs on somiaj
1919[20:15:01] <somiaj> Agiofws: keep going down the list and pay attention to the output you get. You need to figure out what package is causing the actual conflict
1928[20:17:00] <somiaj> it could be a pacakge on your system, it could be the fact one of these thrid party repos provides a newer version of a package that the older version is needed to satsify some dependency
1929[20:17:20] <somiaj> you need to use a combination of apt policy packagename and go down the rabbit hole of depends to find out where the actual conflict is
1930[20:18:06] <somiaj> it could be you have 'held broken packages', though I find that error not fully correct, it just means apt cannot fix the issue, so you must find it and fix it manually
1934[20:20:13] <somiaj> it is due to conlicts in your sources, probably due to mixing repos. But until you track dwon the actual cause, this is only spectulation.
1935[20:20:16] <Agiofws> Package libunistring0:i386 is not available, but is referred to by another package. This may mean that the package is missing, has been obsoleted, or is only available from another source
1936[20:20:30] <Agiofws> ok i have to go
1937[20:20:30] *** Quits: maxrazer (~siducer88@replaced-ip) (Remote host closed the connection)
1944[20:21:51] <somiaj> in this case you need to figure out why that package is on your system still, 'aptitude why libunistring0' or what package depends on it and is trying to pull it in?
1945[20:21:59] <Agiofws> thank you somiaj and annadane
1946[20:22:07] <Agiofws> maybe tomorrow
1947[20:22:10] <somiaj> anyways, this is why we don't really support mixing here, it can be a pain as you have seen
1982[21:12:34] <boredtobonesbyco> let's say I'm without GUI, with 2 ttys one with irssi and the other with elinks, and I want to copy a string of elinks to irssi and have gpm activated. How do I do it?
1983[21:13:16] <boredtobonesbyco> do gpm works across ttys?
1984[21:13:34] <greycat> Well, if gpm is installed, you should be able to copy and paste with the mouse on the Linux console. Or at least, that's how it worked 20 years ago, which was the last time I actually saw gpm.
1988[21:14:16] <g0zzy> Since (i think) i plugged an additional monitor into this laptop, my mate desktop under Debian got borked. Looks like the monitor settings have perhaps gone queer as the font is boldish and elongated. I tried to reset it with dconf (dconf-cli) to no avail. It doesn't happen when i log in as a different user. What's the best way to restore default?
1989[21:15:49] <g0zzy> An update to that. I just copied the ~/.config/dconf/user from the working to the non-working user and it had no effect on the problem
1990[21:18:50] *** Quits: conta (~Thunderbi@replaced-ip) (Quit: conta)
1991[21:18:54] *** Quits: krabador (~krabador@replaced-ip) (Remote host closed the connection)
1992[21:19:07] <boredtobonesbyco> g0zzy: If I can say, at home, I have a huge HD TV that I replace for the 2 monitors, I put it in the maximum resolution of the GPU and it looks great and better
2023[21:31:35] <inerkick> even i followed this method ikonia replaced-url
2024[21:31:42] <ikonia> right ?
2025[21:31:47] <greycat> If you want to install postgresql in Debian, you type "apt install postgresql" or some variant thereof. And you get the stable version that was tested and released with your stable version of Debian. No "guide" needed.
2067[21:48:22] <cuddylier> I made a LACP bond (802.3ad) of 2 x 1Gb links in Debian 10 which is seeing expected performance of 2Gb but outbound traffic is only going via a single uplink despite traffic going to multiple destination IPs. My interfaces config file is here: replaced-url
2068[21:48:28] <sponix2ipfw> you all are just breaking Debian by attempting to snag software from outside sources that doesn't even match your main Debian version. Should be looking into the Synaptic package manager 1st and using what is already provided, before even starting to search elsewhere
2097[22:13:01] <cuddylier> cyveris: Is it normal though for there to be zero load balancing on the outbound? The inbound load balances perfectly but maybe this is how it's designed?
2098[22:13:25] <cyveris> cuddylier: The default xmit-hash-policy parameter bases solely on MAC, not IP.
2107[22:16:44] <invra> well the package description says it does support rar password extraction.. but when i try it doesnt... its a rar divided in parts... might be cause of that?
2142[22:44:14] *** Quits: crockedgrind59 (~crockedgr@replaced-ip) (Remote host closed the connection)
2143[22:45:13] <LostInTheShell> Hiall, it's me tyzef..... guys, I had tried many things, everythings? my wifi card Atheros AR9285 card don't comes UP..... should I give up? but the netinstall make it work ! why ? how ? no idea, I am just here and there, in between wlp2s0 and the kernel, lost in the shell...
2152[22:49:16] <LostInTheShell> I tried wicd, no way, Bionic Puppy, no way..... this wifi card can connect to my smartphone hotspot from puppy(my only live USB) but can't connect to my kind of box "huawei B612s-25d
2154[22:50:32] <LostInTheShell> 'wicd on buster and puppy scann my box but can't connect..... only the buster netinstall connect to it, only during the install
2160[22:52:27] <dpkg> Puppy Linux is a GNU/Linux mini-distribution. It is not supported in #debian. replaced-url
2161[22:52:33] <LostInTheShell> yes bionic puppy en usb pendrive
2162[22:52:53] <LostInTheShell> yes it was just to test you know
2163[22:53:40] <somiaj> well it sounds like we have tried to track down the issue as best as we could in the past. You don't have some issue with a softblock, have you checked rfkill?
2164[22:54:08] <somiaj> and by doens't come up, do y ou see the card in 'ip addr'?
2165[22:54:10] <LostInTheShell> I don't know how to use rfkill
2166[22:54:23] <LostInTheShell> yes I see it
2167[22:54:28] <somiaj> you iinstall it and run it, it will tell you if you have any soft or hardblocks on the network
2168[22:54:38] <somiaj> so you see the interface in 'ip a' but you don't see it in network-manager?
2169[22:54:56] <LostInTheShell> oops, I am on minimal debian, no gui
2170[22:55:04] *** Quits: xcm (~xcm@replaced-ip) (Remote host closed the connection)
2171[22:55:13] <LostInTheShell> nothing is there on my debian
2172[22:55:34] <somiaj> do you configure the network via /etc/apt/interfaces then (well you said you tried wicd, so thinking you were using a guid)
2173[22:55:48] <somiaj> on common issue is trying to use a gui while having the network configured in /etc/network/interfaces from the installer
2174[22:56:04] <LostInTheShell> ip addr I see my wlp2s0 and it say satate DOWN
2175[22:56:27] <LostInTheShell> i use openbox
2176[22:56:28] <invra> well the package description says it does support rar password extraction.. but when i try it doesnt... its a rar divided in parts... might be cause of that?
2217[23:23:43] <somiaj> longears: what are you looking for? If you have a bug, a bug report vs emailing the matainer directly is prefered
2218[23:24:53] <longears> somiaj: Just saw a line in logs and thought I'd bring a suggestion to avoid it. I'll just email the maintainer later tonight.
2219[23:25:16] <somiaj> tyzef: it could be a kernel difference why it works in puppy but not debian. Usually people have good expereince ath9k, maybe your hardware isn't functioning properly, and due to differences in the kernel that issue manifests in debian.
2220[23:25:25] <somiaj> longears: you should file a wishlist bug vs emailing them directly.
2221[23:25:45] *** Quits: teclo- (42@replaced-ip) (Remote host closed the connection)
2222[23:26:01] <longears> somiaj: Ah, thanks. Sorry, never dealt with mainteiners before :)
2224[23:26:37] <somiaj> longears: in general debian prefers using the bug system to log/keep track of issues with packages, this sounds like a wishlist bug to me, and provide a patch if you have a fix.
2226[23:27:30] <somiaj> though it also might be a minor bug, maybe this is something small issue that could be fixed, vs an option you wish would change, after re-reading your issue, a minor bug might be more reasonable.
2227[23:28:06] <somiaj> bugs.debian.org will describe the process to report bugs and how they are classified
2229[23:28:54] <tyzef> sorry somiaj, I may had not explained properly. debian10+wicd, debian10+wpasupplicant, debian10+networkmanager, puppy..... no one can connect to my box... the only way I got congratulation is during the network configuration in NetInstall Debian10, and it worked only for the install... once debian installed... the end
2245[23:32:45] <dpkg> ath9k is a completely free (no proprietary HAL) Linux kernel driver supporting PCI/PCIe 802.11n wireless LAN devices with Atheros chips. It does not require firmware from userspace. See replaced-url
2246[23:32:58] <somiaj> are you able to relabily get the card to work inside the netinstaller? Or was it just one time. If the netinstaller always gets it, you may need to start digging through dmesg and see if you can spot any differnces, though in debian the netinstaller should be using the same kernel, so I can't think of any differences that could arise.
2247[23:33:43] <somiaj> jhutchins: I don't agree with 'completely free', it just doesn't use userland non-free firmware, but it does have firmware on the chip. Though with ath9k_htc you can aperntally use fully open firmware on that card.
2249[23:34:07] <tyzef> yes but if so my hardware is bad, why debian netinstall can do what others can't? this is...... I don't know
2250[23:34:28] <somiaj> tyzef: does it work in the net install each and every time.
2251[23:34:30] <somiaj> ?
2252[23:35:05] <tyzef> I really would like to see what this netinstall do to connect to my box, it have the secret, yet it's a software, we can't have a coffee together
2253[23:35:29] <tyzef> yes whenever I tried it worked
2255[23:35:45] <jhutchins> tyzef: have you pastebinned your cofig files?
2256[23:36:10] <somiaj> tyzef: the only thing I can think of is the netinstall is a bit more minimial of an enviorment, but it uses the exact same kernel, so this seems odd
2257[23:36:34] <somiaj> I would start by taking a log of dmesg from inside the netinstall after the card is up and working and then compare it to debian install it isn't working and look for differences.
2258[23:37:06] <somiaj> actually it probably uses a slightly older kernel. Wonder if trying to downgrade to the exact kernel the netinstaller used would help
2259[23:37:32] <somiaj> there have been a few security patches since the netinstall kernel was made, and sometimes kernels cause regressions
2260[23:38:03] <tyzef> can I try with debian9, I have an iso
2261[23:38:23] <tyzef> hey, ayyoooo ..... I have a xfce debianb 9 iso
2262[23:38:31] <somiaj> go right ahead, but what I would be intersted in would be the output of dmesg and the kernel the netinstall is using in which it works
2263[23:38:34] <tyzef> maybe I should try it
2264[23:38:37] <somiaj> then compare that to a place it doesn't
2279[23:41:48] <Copenhagen_Bram2> I haven't configured it, just installed it with snapd and it won't finish starting up
2280[23:42:01] <jmspeex> Anyone can help me with my newly installed Debian server that keeps auto-suspending by itself?
2281[23:42:29] <somiaj> Copenhagen_Bram2: sounds like a bug in the snap, which we can't fully support. See if you can configure that software and tell what port to listen on.
2282[23:42:47] <somiaj> jmspeex: that is strange, do you have a DE installed on that?
2283[23:43:00] <Copenhagen_Bram2> When does it auto-suspend?
2284[23:43:12] <tyzef> well somiaj, thank you so much, it's a great progress for me !
2285[23:43:13] <jmspeex> somiaj: yes, I installed gnome to make it easier to configure at first
2297[23:47:48] <inthl> I have a very weird problem with ip6tables - but not with iptables. when using some hostname which ip(6)tables would resolve to all IP addresses, then iptables -I INPUT --fobar -s somehostname.com ..works perfectly fine. but doing this with ip6tables fails with the error message ip6tables v1.6.0: host/network `somehostname.com' not found - this happens to *any* hostname, and also I can resolve the IPv6 addresses there perfeclty fine using
2298[23:47:48] <inthl> `host somehostname.com` or `dig somehostname.com` - any ideas what this is all about? or where/how ip6tables fetches the addresses. the host is perfectly fine with ipv6, also other services like httpd do respond to ipv6 requests, pings from-to and vise versa other hosts work as well
2304[23:50:45] <jmspeex> What I don't understand is how something that obviously wrong can be happening without everyone screeming bloody murder considering that Debian is used on servers a lot. Or am I somehow one of the few affected (why?)
2307[23:51:38] <somiaj> jmspeex: you could change your default target from graphical.target to multi-user.target and not laod the gdm.
2308[23:51:54] <AndreasLutro> jmspeex: because you installed gnome on a server?
2309[23:52:13] <somiaj> jmspeex: sometimes bugs only affect certain combinations of software installed, and most people running servers dont' also run a gui
2310[23:52:32] <somiaj> but just because there is a bug, it might only affect some hardware/configurations
2311[23:53:08] <jmspeex> somiaj: How do I change the target?
2321[23:56:26] <somiaj> but it appears that this is more a gdm issue than debian (seems fedora has a bug too). My guess is gdm is doing what windows does by default, which I agree, probably isn't best
2322[23:57:02] <somiaj> and it could also on affect certain setups, hence why only a few have responded to that bug report