8[00:01:41] <d1g1ta1> lwp: The new USB image got me further (asks if I want to do a graphical (full or VESA) or install, but when I choose any of the three options I just get a black screen.
9[00:01:59] <Gryllida> trek00: maybe you know another software for this? i'd like to blur background of webcam, or replace it with another background.
14[00:04:25] <Gryllida> trek00: okay, thank you for clearing up the situation with obs. Is there a way to somehow make apt prompt about this issue before proceeding with the install?
26[00:07:06] <[E]sc> i'm getting systematic noise from the top left side of my laptop when using 5.5.0 kernel and gnome-shell 3.36.1. There's no noise in BIOS or tty or LXDE or kernel 5.3.0. I turned off gnome-extensions and the sound dies down eventually, but starts back up again when i open an application. i suspect the fan is being overused, but can't be sure. how can i fix this?
27[00:10:58] <sponix> trek00: the jellyfin issue solved itself for now _again_ ... Going to try it with remote access enabled one more time, and see if it continues the odd behavior
34[00:15:20] <[E]sc> trek00: the noise turns on the moment i log onto gdm3. after shutting down services, the sound still persists. when i turn off the gnome-extensions, it eventually dies down. however, if i click on say thunar, it'll start the clicking/grinding noise again. i had htop running but it didn't show me anything that stuck out
35[00:15:55] <trek00> [E]sc: which GPU do you have?
66[00:28:27] <sponix> trek00: I seem to have solved the jellyfin by disabling remote access, uPnP, and DLNA. And the xfwm4 is a matter of setting my composition type to "none"
67[00:28:49] <sponix> trek00: and for me the firefox fix was just using ESR instead of regular firefox ;)
104[01:00:27] *** Quits: M1zuki (~Vella@replaced-ip) (Remote host closed the connection)
105[01:01:08] <warsoul> trek00 just finished almost die
106[01:01:09] <warsoul> lol
107[01:01:19] <warsoul> i think i put something back wrong
108[01:01:26] <warsoul> pc dindt was turning on
109[01:01:38] <warsoul> i had to unplug gpu and memories again
110[01:01:40] <warsoul> then it worked
111[01:01:56] <warsoul> hopefully doenst freeze again
112[01:02:20] <sponix> annadane: I found the issue, "jellyfin" is trying to do Microsoft networking calls ... No wonder it is going nuts. Amazing how Microsoft can corrupt anything
113[01:02:43] <annadane> oh dear
114[01:03:26] <sponix> System.Net.WebSockets.WebSocketException (7564911): The remote party closed the WebSocket connection without completing the close handshake. ---> System.Net.Sockets.SocketException (110): Connection timed out
115[01:03:26] <sponix> at Microsoft.AspNetCore.Server.Kestrel.Transport.Sockets.Internal.SocketAwaitableEventArgs.<GetResult>g__ThrowSocketException|7_0(SocketError e)
124[01:13:53] <saptech> I found and followed this link for the Printer Utility, replaced-url
125[01:14:55] <saptech> It seems to want lsb but it's been deprecated in Buster. Is there anything else I could try to get the Printer Utility to work?
150[01:39:44] <brutser> when i connect my laptop to a monitor, i wan the monitor sort of "take over" from the laptop screen, i can do this easily on boot, check if hdmi is connected, if so, then turn off eDP (laptop monitor) with xrandr - but if i now unplug the hdmi cable, i got no screen anymore obviously, so my question is, can you automate this a little more, so when
151[01:39:44] <brutser> you connect/disconnect the hdmi cable, it works smarter?
153[01:40:38] *** Quits: mibo (~mibo@replaced-ip) (Remote host closed the connection)
154[01:42:57] <brutser> solutions i find are that i need to keep running things in the background every X seconds, to check if a monitor has been attached/detached
175[01:55:01] <AndreasLutro> brutser: just make sure you use `udevadm monitor` to find the appropriate event that happens when you plug/unplug your monitor and you should be good
200[02:05:55] <emv> the funny thing about librtlsdr-dev is that another popular distro derived from debian also has a version 0.6.0-3 but it has different patches applied.
202[02:07:47] <brutser> AndreasLutro: /devices/pci0000:00/0000:00:01.0/drm/card0 < according to that output, would my line look: KERNEL=="card0", SUBSYSTEM=="drm", ENV{DISPLAY}=":0" < where can i read this part?
229[02:40:26] <iiwwii> hi all. ive moved a debian installation to a new machine with a different network hard. i have a keyboard and mouse plugged in too. it doesnt get an ip address. starting dhcp i see "failed to load rtl_nic"
230[02:41:09] <iiwwii> i think it's missing realtek drivers what are availible as "firmware-realtek" from some apt repo
231[02:41:33] <iiwwii> what#s the easiest way to put these on a usb stick or something to install them offline?
246[03:03:12] <lwp> d1g1ta1, did you modify the .iso image at all before you booted it ?
247[03:03:58] <iiwwii> it works :) thank you. one strange thing. before when i booted it would get an address automatically. now it only works after typing "dhclient"
248[03:04:10] <iiwwii> is there a systemd unit or something i can enable?
262[03:17:14] <lwp> d1g1ta1, if you are stuck, in the graphical installer, you can switch to a rescue shell with ctrl-alt-F2, and try to see what is the problem.
265[03:19:16] <iiwwii> sney yes that's it. the name of the interface changed - thanks :)
266[03:19:28] <sney> np
267[03:20:04] <Tjowers> Screw X -- Wayland does all the work with a better architecture. So run wayland server, no need for GLFW, install wayland-client, write a client that opens windows, then from there we hit it with vulkan -- anyone down? We need to get a serious GUI to linux and start getting the general public switched over
268[03:20:45] <sney> I think that is the general idea
269[03:21:40] <Tjowers> We need a GUI, a gaming engine, a gaming engine IDE, a better terminal, a better shell (lower priority), and a better file explorer (drag and drop important)
270[03:22:11] <Tjowers> And libreoffice needs to either step it up or get dropped and make sure users can find microsoft office in the app store
272[03:22:39] <sney> I'm sure most of those things are on the respective development roadmap(s)
273[03:23:06] <sney> but this is a support channel, so soapboxing is probably not going to accomplish much. you could always join one or more of those projects and contribute, though.
274[03:23:35] *** Quits: iiwwii (562c26f3@replaced-ip) (Remote host closed the connection)
278[03:25:11] <Tjowers> No I'll set up a repo for all the things that need to be made with a full description of what needs to be done, then I'll start working on it on the side, other people will probably join, I want a fresh compilation of linux as high as it can get with the latest gcc, binutils, and glibc -- cloned and building the master branches As-We-Speak
279[03:25:54] <Tjowers> Idk if I should use debian as the base or go with arch
280[03:26:03] <lwp> Tjowers, maybe you could sponsor some student programmers to do some of the actual work
281[03:26:34] <znf> If you need all those things, I don't think linux is the right OS for you
282[03:26:51] <annadane> get the gsoc people to do it when they inevitably come in here looking for projects :P
283[03:27:21] <Tjowers> 100 -- I'll write the header files then get some eager "interns" to start working on it, give them the build environment so they don't have to worry about how to get their dependencies and fix their build errors -- the libraries for this stuff is pretty straight-foreword
285[03:27:49] <lwp> Tjowers, if you are thinking of starting an entire new distro, you probably need to fine-tune you sales pitch to be more tempting than the above rant
286[03:27:58] <sney> people have been making derivative distros with lofty goals for decades. as it turns out, "this stuff is pretty straight-foreword"(sic) is little more than famous last words, in most cases
287[03:28:00] <Tjowers> Wayland server and client replaced-url
288[03:28:06] <sney> the most successful example was probably ubuntu
289[03:28:14] <znf> I mean, LibreOffice alone is a mammoth. And as much as I like Open Source software, it will never actually reach the capabilities of MS Office.
291[03:29:03] <znf> Libre is "good enough" if you just want to open up a document, or just write something simple, but when it comes to authoring tools, it pales in comparision to MS Office :-/
292[03:29:06] <znf> Which is quite sad
293[03:29:08] <sney> your best bet for the most progress is to contribute work to projects that already exists, rather than making your own more different OS with blackjack and hookers
294[03:29:54] <Tjowers> No libre is disgusting if you haven't been traumatized by the world of engineering
295[03:30:45] <znf> I'll repeat: LibreOffice is /good enough/ if you want to open up a document. That's it.
296[03:30:59] <Tjowers> No they aren't moving fast enough and aren't projecting anything exciting enough to get people motivated
297[03:31:00] <znf> Anything else more complex, nope.
298[03:31:24] <sney> LO will probably never escape its staroffice roots. someone needs to build an office suite from the ground up. will they? idk, I doubt it, everything is moving into the browser
299[03:31:38] <znf> Heck, even Google Docs pales in comparision to MS Office
300[03:31:40] <Tjowers> Yeah, if you want to actually use Linux you need stuff like that to look sexy
301[03:32:00] <Tjowers> Cover your mistakes with spinning bars
302[03:32:17] <znf> I've had to use Google Spreadsheets for some JSON data a few times, and the experience was horrible
303[03:32:43] <Tjowers> Yeah, look, either it gets stepped up or it gets dropped entirely -- or change the name to Excel Viewer
304[03:33:06] <znf> what's stopping you from creating something from scratch? :P
305[03:33:34] <sney> welp, not like anyone could convince me not to join an indie distro project in 2005, so I hope whatever project you actually get off the ground succeeds
306[03:33:36] <znf> I look forward to it!
307[03:33:45] <Tjowers> Need an IDE for developers to easily make apps like they do for iPhone and OSx it needs drag and drop and colors and hide everything super magic, like disney world level
308[03:34:05] <znf> You mean, like.... VS Code, which, last time I checked, runs on linux?
309[03:35:08] <Tjowers> Yeah I'll start writing the projects out in repos over the next few weeks, I'm working mostly on making connections with people here in san fran -- get something rolling, then I can talk people into using it for free -- in-OS-purchases will work well enough
310[03:35:39] <znf> Can I invest $1 in your project, so when you eventually grow I can reap $100.000 ?
314[03:36:46] <avu> they must already have millions for such a project to have a chance, given the many man-years of work it'll need, I doubt your $1 is wanted ;p
315[03:36:51] <Tjowers> I want to start a business for real-estate property-shares trading online -- the Estate Exchange -- I know a few people in SF that work at the high-level for commercial real-estate
394[04:38:36] <annadane> how do i stop MATE from blanking my monitor after a period of inactivity? i have the mate power manager installed, i installed mate via mate-desktop-environment-core but there doesn't seem to be anything in power management settings
420[04:52:40] <somiaj> Yea, I'm surprised the option is not there (or maybe it is hidden somewhere). I don't use any DE, but do know about the default console blanking of xorg.
421[04:54:27] <somiaj> sometimes things are orginized in such a way that doesn't make sense. For instance this could be under 'display settings' and not 'power management'
446[05:07:26] <annadane> isn't it nice when you answer your own questions?
447[05:07:39] <sponix> all that node stuff is a pain in the ass. I highly recommend messing with it in a VM, or some container type, and not letting it foul up your Host OS
487[05:41:19] <somiaj> annadane: similar to pip, if you just start using npm to install stuff you can create a mess that is a bit hard to handel and have stuff installed outside of .deb (dpkg/apt)
523[06:04:28] <[E]sc> i'm getting systematic noise from the top left side of my laptop when using 5.5.0 kernel and gnome-shell 3.36.1. There's no noise in BIOS or tty or LXDE or kernel 5.3.0. I turned off gnome-extensions and the sound dies down eventually, but starts back up again when i open an application. the gpu temperature is around 36C, the cpu is at 41 C and the cpu fan at 2900 rpm. how can i fix this?
538[06:08:54] <somiaj> Exaeta: also explan what it is exactly you have done that you expect to work.
539[06:09:20] <Exaeta> somiaj: get this program working. the program is supposed to listen on a port, and accept http requests. strangely, it works on my desktop, but not the server
540[06:09:22] <genhaoqi> [E]sc: use top to find CPU intensive processes, or maybe the driver can not contral the fan well,
541[06:09:22] <Exaeta> code is the same
542[06:09:29] <somiaj> genhaoqi: I could see hardware going back creating different noises, like a capicator, hd's use to often make noise, though less common with SSDs
543[06:09:30] *** Quits: donofrio (~donofrio@replaced-ip) (Remote host closed the connection)
556[06:12:43] <somiaj> does this pogram produce any logs you can look at?
557[06:13:28] <somiaj> but various tests you can see, is before you run the program check to see if anything is listing on that port with netstat, after you run the program, check netstat again and see if it is indeed listing on the prot.
558[06:13:44] <[E]sc> somiaj: it's like a systematic scratching that's like a ticking noise.
563[06:15:41] <[E]sc> genhaoqi: when i switch to tty it stops or when i use LXDE. right now Xorg is using 1% of the CPU then it's gnome-shell. so i'm lost and i'm on SSD.
566[06:16:23] <somiaj> wonder if there is some hardawre issue in which the higher gpu ussage (it is close to the monintor that this noise is coming from) is causing an actual audiable noise.
567[06:18:25] <genhaoqi> [E]sc: sometimes driver problem will always turn on your Graphics card
568[06:19:22] <somiaj> though it is stange it is somehow tied to the kernel, but in general unless you have a fan/speaker/some moving part, your computer shoudl not be making noise.
570[06:20:40] *** Quits: nsk_nyc (~nsk_nyc@replaced-ip) (Quit: My computer has gone to sleep.)
571[06:21:10] <[E]sc> somiaj: speakers are on the front of the laptop, but sound is coming from top left. i don't know if it's considered coil whine or whatever.
615[07:17:07] <k3tan> i have 5 drives, all are mounted in /mnt - 2 drives show up in the sidebar on nautilus, and 3 are shown in 'other locations' - why is this?
1019[10:12:22] <Wndows10> quickly isnt possible lol
1020[10:12:30] <Haohmaru> why?
1021[10:12:50] <Wndows10> reason i use ubuntu or mx linux etc is becasue it can be used out of the box. i tried debian before.. its a nightmare for casual user to install
1022[10:12:56] * Haohmaru gives Wndows10 a few RedBulls
1048[10:19:25] <Haohmaru> Wndows10 wut r ya saing, there's like 5 or more different desktop environments to choose from, each with different looks and feng-shui
1049[10:19:40] <Wndows10> i cant do the feng shui
1050[10:19:44] <Haohmaru> i'd recommend LXDE for simplicity
1051[10:19:45] <Wndows10> i like gui and sofware centers
1052[10:19:54] <Haohmaru> software centers?
1053[10:19:58] <Wndows10> but i need a fast one that runs off a usb
1054[10:20:04] <Wndows10> lubuntu maybe
1055[10:20:14] <Haohmaru> lubuntu is crapbuntu with LXDE
1056[10:20:25] <Haohmaru> just get debian and select LXDE from the installer
1127[10:46:59] <oznt> hi guys, I have a debian machine which must have ntp installed. The problem is that dhcp renewals are expected to be done with chrony. Is there anyone here who knows of an alternative to autorenewing the dhcp address ?
1269[12:30:10] <ksk> Hi, I wanted to check out a kernel > 5.5 because of the improved ryzen 3000 support (thermal sensors etc), is there a "debian way" of doing that?
1319[12:55:43] *** DarkiJah is now known as PopeIsAntiChrist
1320[12:56:53] <Gryllida> what package provides logging of cpu temperature to file every minute, together with time stamp of this temp? a laptop here powers off randomly and I would like to check whether this is cpu temp or not
1321[12:58:25] <ksk> mhhm, seems I am just building 5.6 from upstream using "make deb-pkg"
1322[12:59:14] <ksk> Gryllida: I could use such a tool, too - and get rid of my "watch sensors" :)
1352[13:26:33] <ChrisWi> hi guys ... just installed stretch and asking myself how do you guys stop network ? systemctl restart networking is not doing ... ifdown enp2s0 is giving error 'ifdown: interface enp2s0 not configured' but ip addr is showing interface enp2so ... so I am a bit confused now
1353[13:27:44] <ChrisWi> even systemctl stop networking.service is not doing anything ... ip addr is still showing interfaces
1365[13:33:09] <Haohmaru> as for networking, i don't think i've wanted to stop it, but i usually deal with it via the network manger (it's in the system tray)
1366[13:33:11] <ChrisWi> Haohmaru: sorry for that typo ... it is enp2s0
1368[13:34:47] <ChrisWi> Haohmaru: it is a server ... no X and hence no 'system tray' ... only console. What is the usual way to configure networking and giving the ability to start/stop/restart it ?
1412[13:56:26] <cyveris> And I only want you to read the manpage you've been directed to so that you can understand for yourself how to utilize documentation and administer your system.
1413[13:56:43] <f8e4> i have it open, its cryptic for those :D
1414[13:58:03] <cyveris> It can be cryptic, but taking the time to understand how to use that documentation will pay off massively for you.
1431[14:09:03] <ratrace> f8e4: you must disable PasswordAuthentication, because with it an attacker can bypas keys and go straight for trying out passwords
1454[14:27:21] <kristijonas> debian stretch, i try to set my wifi adapter (iwlwifi) on monitor mode. after five to ten seconds it automatically gets back to managed, which I dont want to happen. Can I can debug/fix it?\
1539[15:26:00] *** Quits: Anderson69s (~Anderson6@replaced-ip) (Remote host closed the connection)
1540[15:26:18] <ratrace> shtrb: the pidging AppArmor policy is missing mmap allow on those files in name="" attribute
1541[15:26:38] <ratrace> now whether it should have that, or not, I don't know. do you see any broken behavior in pidgin?
1542[15:27:03] <shtrb> not yet , the browser had opened the link as expected (for now)
1543[15:27:39] <ratrace> shtrb: btw it looks like those are all due to nvidia acceleration in the browser.
1544[15:27:39] <shtrb> those seem like some kind of temporary files
1545[15:27:52] <shtrb> Wow , but how ?
1546[15:28:03] <ratrace> what do you mean how?
1547[15:28:26] <shtrb> how did you know I have nvidia , I don't see anything in the traces mentioning that
1548[15:28:41] <ratrace> shtrb: because I've written many AppArmor profiles and I recognize the paths
1549[15:29:01] *** debhelper sets mode: +l 1266
1550[15:29:05] <shtrb> oh !
1551[15:29:51] <shtrb> so I need to search for a solution with pidgin or nvidia packages ?
1552[15:30:15] <ratrace> ideally, pidgin would invoke the browser with a Px profile transition, but.... default AppArmor policies are usually incomplete or outright broken.
1558[15:32:58] <ratrace> shtrb: problem is broken nvidia abstraction in the default apparmor package. it's missing some paths, like the ones you mentioned. so if you want, you can fix pidgin's profile and allow those specific paths, but I don't know how the pidgin's profile is actually dealing with the browser invocation, I don't have the profile at hand now
1559[15:33:45] <shtrb> These had not been random paths ? I thought it was just random temporary values .
1562[15:35:05] <ratrace> shtrb: they're semi-random, the patterns are really: /tmp/.gl* , @{HOME}/.nv/.gl* and my personal "favorite" /tmp/#[0-9][0-9][0-9][0-9]*
1585[15:50:24] <brutser> when i install debian with openbox on a laptop, user directories are created, such as desktop, documents, downloads etc. that is part of xdg that creates them right? - when i do the exact same install on a virtual machine, the user directories are NOT created. what can be the difference and what would trigger the installation of them? i know i can
1586[15:50:24] <brutser> manually installed xdg-user-dirs and use that to create the directories but i like to know why and what creates them on the laptop install, but not on the vm :s
1678[17:09:33] <brutser> i dont understand this, if i install same installation on laptop (debian+openbox), user directories are created, such as documents, downloads etc. - but, neither xdg-utils or xdg-user-dirs is installed. -- if i do the same install on kvm guest, NO user dirs are created and the only way to force this is to install xdg-user-dirs package, at least,
1679[17:09:34] <brutser> that's my only way to force it -- so what is creating the user dirs on the laptop and not on the virtual machine?
1681[17:11:07] <PsynoKhi0> I've mounted /home over NFS and now a couple users realised they forgot to copy over a a bunch of files... any way I can access the old "home" without unmounting the NFS share?
1718[17:31:31] <somiaj> ladderff: does /run/screen exist at boot, or only after you run screen?
1719[17:31:45] <ladderff> hmm, would have to reboot to check
1720[17:32:10] <themill> ladderff: that's an unusual mode for /usr/bin/screen
1721[17:32:21] <somiaj> since you said you chagne it after each boot, I was wondering if some service is running inside a screen, causing /run/screen to be created at boot with different permissions. Though themill might have spotted the issue.
1722[17:32:35] <ladderff> yeah, I agree about it being unusual...
1723[17:33:04] <themill> if you change /usr/bin/screen, then you have to change the code that is creating /run/screen (/etc/init.d/screen-cleanup or /usr/lib/tmpfiles.d/screen-cleanup.conf)
1724[17:33:20] <themill> see /usr/share/doc/screen/README.Debian.gz
1725[17:33:30] <ladderff> I couldn't say how /usr/bin/screen ended up being 4755
1754[17:51:47] <brutser> i dont understand this, if i install same installation on laptop (debian+openbox), user directories are created, such as documents, downloads etc. - but, neither xdg-utils or xdg-user-dirs is installed. -- if i do the same install on kvm guest, NO user dirs are created and the only way to force this is to install xdg-user-dirs package, at least,
1755[17:51:48] <brutser> that's my only way to force it -- so what is creating the user dirs on the laptop and not on the virtual machine?
1790[18:09:20] *** Quits: RadoS (~cheater@replaced-ip) (Remote host closed the connection)
1791[18:11:02] <glguy> (In Debian 10.3) How do I configure wlan0 not to be configured. I'm reading that I should change something in /etc/network/interfaces, but that only points to /etc/network/interfaces.d, and that directory is empty. Where is the configuration that would cause wlan0 to come up?
1802[18:17:33] <somiaj> glguy: depends on how you manage your network cards, /etc/network/interfaces is only one way to manage them, if you have nothing about your wlan0 in the interfaces file (or interfaces.d directory) then something else is managing your network as diogenes_ poitns out (wicd is used by some DE's too)
1804[18:17:38] <silverballz> i'm having trouble understanding if wlan0 is up or not up and if you are trying to disable it or enable it
1805[18:17:44] <somiaj> diogenes_: what desktop enviorment did you use?
1806[18:17:51] <glguy> I don't want wireless enabled
1807[18:17:55] <glguy> it turns on, I don't want it on
1808[18:18:20] <somiaj> glguy: define turns on? Is it connecting to the network, or does the interface just show up in ip a with no address assoicated with it?
1809[18:19:11] <diogenes_> somiaj, you confused me with glguy :)
1810[18:19:24] *** kreyren_ is now known as kreyren
1812[18:19:58] <somiaj> diogenes_: haha, I was gonna metnion that as diogenes_ pointed out, check nm, but then got confused
1813[18:20:05] <glguy> somiaj: It was connecting to a network. I commented out the SSID and WPA password in /etc/wpa_supplicant/wpa_supplicant.conf but I'm still trying to understand how the interface comes up
1814[18:20:26] <glguy> or how it was coming up
1815[18:21:14] <somiaj> glguy: there are various ways to configure the network, in general you should choose one and stick with it. Did you install/use a desktop enviorment, if so which one?
1856[18:43:28] <Kobaz> without having to apt-policy query each individual package
1857[18:43:39] *** Quits: leorat (~leorat@replaced-ip) (Remote host closed the connection)
1858[18:44:02] <sney> blackop: full-upgrade has nothing to do with your problem, which is that the config file mentioned by libkmod has bad syntax. you can probably safely delete that file.
1860[18:44:41] <sney> diogenes_: running apt full-upgrade will do exactly the same thing as running apt upgrade in 99% of cases on stable. it's not necessary but it's unlikely to cause any problems
1865[18:46:08] <sney> edge cases undefined by a standard debian stable system
1866[18:47:27] <diogenes_> so maybe blackop hit that 1%.
1867[18:47:58] <sney> obviously not, did you even read the paste?
1868[18:49:06] <sney> there's some wonky modprobe.d config in there. it's probably gunk from a previous release, or something he got from stackexchange that's not compatible with current kernels, etc. the error messages are triggered by the kernel dpkg-reconfigure
1913[18:59:30] <randompleb> What happened to radare2? It was removed because upstream didn't want to maintain older versions shipped with debian and they were openly hostile towards Debian.
1914[19:02:25] <f8e4> who has hardened their sshd before, what suggestions?
1924[19:10:52] <Kobaz> what's a quick way to tell if there's any installed packages not from my current release, ie older distro, or new/backports ? without having to apt-cache policy on each individual package?
1957[19:22:02] <sney> and your security issues have already been reported and is being worked on, I guess since you came here first that means you didn't check: replaced-url
1962[19:23:01] <zapotah> sney: its just that considering that theres a new variant of it that is moving to not just doing cryptomining but actively infiltrating infrastructure, four days is too slow
1963[19:23:18] <zapotah> and autoremoval wont help
1964[19:23:19] <joepublic> zapotah, your full refund is on the way.
1965[19:23:27] <zapotah> joepublic: its not about that
1966[19:23:58] <zapotah> but some things do warrant some serious urgency
1967[19:24:12] <zapotah> and it _is_ in that case distro maintainers security teams responsibility
1968[19:24:21] <joepublic> I see evidence of developers working on it, and you complaining about it; guess which one of those has value?
1969[19:24:46] <zapotah> joepublic: the upstream has already published fixed packages for buster even
1970[19:24:54] <zapotah> and all the way to debian 7
1971[19:24:54] <sney> I think you may be under the impression that #debian is full of developers who can fix these things on demand. a) this channel is mostly other users who help out in their spare time and 2) the bug is being worked on! what more do you want!?
1972[19:25:00] <zapotah> sney: im not :P
1973[19:25:02] <joepublic> fixed packages for buster != fixed buster packages.
1974[19:25:27] <sney> the most recent email in that bug thread is from today. oh no, too slow
1975[19:25:45] <zapotah> sney: have you followed this thing today and yesterday?
1980[19:27:01] <sney> no, I only learned about it from you about 5 minutes ago, but it seems to me that the debian maintainers are already on it, so again: what do you want
1992[19:28:20] <joepublic> then your rhetorical skills have been failing you.
1993[19:28:27] <zapotah> propably
1994[19:28:35] <sney> I guess I can see the panic now. this affects you personally, I take it?
1995[19:28:46] <zapotah> it actually does not
1996[19:29:22] <Kobaz> hmm
1997[19:29:43] <sney> I was only familiar with ansible/chef/puppet but I guess this is another one of those, and the security implications are obvious with a RCE.
1998[19:29:49] <zapotah> but since its easily exploitable in so many ways, i can see any webhoster, shared platform or any public facing service that has any remote code execution thing in it being able to attack that frontends salt-master, its a pretty serious issue
1999[19:29:52] <sney> popcon numbers are not super high replaced-url
2000[19:29:53] <Kobaz> having problems with an aptitude regex... aptitude search '~A ^(?!testing).* ~i' getting : Unknown term type: "".. this is a valid regex though
2011[19:34:08] <sney> I take it this is the kind of vuln where it's not a problem if you do your orchestration through a dedicated management vlan or vpn, but a lot of devops types have never even heard of that concept
2012[19:34:31] <greycat> the zdnet article gives me two CVEs, which can be checked up on, at replaced-url
2014[19:35:25] *** Quits: kristijonas (~kristijon@replaced-ip) (Remote host closed the connection)
2015[19:36:44] <zapotah> sney: its the type of vuln where if theres even a single point of code execution from the minion towards the master, youve full access
2016[19:36:59] <zapotah> sney: ie. if you manage public facing webhosts with it
2017[19:37:17] <zapotah> and that page has any vuln that allows for any kind of code execution
2018[19:37:48] <zapotah> since the minions, by design, have to have access to the master
2020[19:37:54] <greycat> There's nothing anyone in this channel can do about it. The issues are known, and presumably the security team is working on official updated packages.
2021[19:38:08] <zapotah> tds: thank you, was trying to refind that link
2024[19:39:02] <Kobaz> having a weird issue with aptitude search... trying to select packages that are installed from distro buster: aptitude search "~A buster ~i" and this is including packages from 'testing'
2025[19:39:59] <Kobaz> for example, in the returned list is showing remmina, but apt-cache policy is showing the current remmina package is from testing... why the inconsistency ?
2026[19:40:14] <sney> Kobaz I think you can have multiples and negatives, so something like ~Abuster ~A!testing etc maybe?
2084[20:20:47] <greycat> Jesus fucking christ, NO IT ISN'T, because most people don't USE SALT, and what the fuck would the topic SAY? There is no action that people can take yet.
2085[20:21:06] <zapotah> greycat: but there are actions
2086[20:21:11] <greycat> Name one!\
2087[20:21:12] <Kobaz> actions required: unplug
2088[20:21:21] <brutser> how come with the same installation, on physical laptop xdg-user-dirs is installed and generated, but on virtual machine, it is not installed - it's exact same installation of debian 10 and openbox (scripts) - what determines the need of xdg-user-dirs??
2089[20:21:21] <Kobaz> problem solved
2090[20:21:23] <zapotah> greycat: upstream has packages for debian of all versions
2091[20:21:43] <zapotah> greycat: the issue is with debian upstream packaging building against python 2.7
2092[20:21:52] <zapotah> which is eol
2093[20:22:00] <joepublic> as I mentioned earlier, "packages for debian" != "debian packages"
2094[20:22:08] <greycat> "/topic #debian Some random internet person named zapotah says you should abandon debian's salt package and use upstream instead of waiting a couple days for the security team to fix it"
2112[20:25:19] <Thedarkb-Desktop> Is it possible to install packages from sid on buster without wrecking the system?
2113[20:25:35] <cyveris> !ssb
2114[20:25:35] <dpkg> First, check for a backport on <debian-backports>. If unavailable: 1) Add a deb-src line for sid (not a deb line!); ask me about <deb-src sid> 2) enable debian-backports (see <bdo>) 3) apt update; apt install build-essential; apt build-dep packagename 4) apt -b source packagename 5) dpkg -i packagename-ver.deb To change compilation options, see <package recompile>; for versions newer than sid see <uupdate>.
2115[20:25:35] <xnaas> I fucked up and should've made my /var partition much larger.
2116[20:25:35] <oxek> joepublic: oh wait, *this* is not an official channel?
2117[20:25:36] <annadane> that's more of a support channel, if you really want a response *maybe* try #debian-devel on the same network
2118[20:25:37] <greycat> Only VERY SPECIFIC packages that have no dependencies.
2119[20:25:54] <greycat> Like if it's a shell script or something. If it's compiled, your answer is HELL NO.
2120[20:26:06] <joepublic> debian's official irc network is oftc
2154[20:31:09] <greycat> 14:31 Ignoring ALL from zapotah
2155[20:31:17] <zapotah> why do i bother...
2156[20:31:20] <cyveris> greycat: Ignore him, honestly. He also sits in #pfsense and bitches endlessly about software he doesn't pay for.
2157[20:31:28] <zapotah> cyveris: i dont
2158[20:31:32] <zapotah> but whatever
2159[20:31:35] <cyveris> Yes you do, I watch you constantly.
2160[20:31:41] * greycat figures out that when zap said "I don't have it enabled", he meant "popcon", whereas I meant "salt"
2161[20:32:03] <greycat> Ain't nobody usin' salt. You're just all worked up for nothin'.
2162[20:32:06] <zapotah> cyveris: then you misunderstand me gravely
2163[20:32:08] <joepublic> popcon is a survey to be extrapolated.
2164[20:32:21] <joepublic> yeah, it seems your rhetorical skills have been failing you more places than just here.
2165[20:32:32] <greycat> Why do you think I divided the number of salt-common installs by the number of libc6 installs?
2166[20:33:03] <annadane> yeah i'm not sure why people are using popcon as a metric; in any case some people clearly do use salt so the more info that can be provided to the debian project the better, if a bug report doesn't exist, file one or bring it up as i mentioned in #debian-devel on oftc
2167[20:33:42] <greycat> There is a bug report. I could /lastlog to find it. There are CVE entries. Including in the security-tracker. I posted their URLs earlier.
2168[20:33:59] <joepublic> the bug is active as of today, aiui
2169[20:34:21] <annadane> there you go, then
2170[20:34:23] <joepublic> serious stuff, worm-trojan-rce-cryptojacking all in one
2171[20:34:33] <annadane> it will be fixed when it's fixed
2172[20:34:45] <greycat> If anyone here actually *uses* salt and has a suggestion for how to mitigate it ("run xxx stop and don't restart it until the patch is out"), I'd be glad to hear it.
2174[20:35:00] <annadane> oh i did find a potential mitigation i can link, i don't know whether it is one
2175[20:35:03] *** Quits: Newami (~Newami@replaced-ip) (Remote host closed the connection)
2176[20:35:16] <joepublic> mitigation is addressed at the replaced-url
2177[20:35:32] <zapotah> since the general discussion has shifted towards focusing on my views on how things should be handled from a sane software engineering perspective from actually discussing this seriously, i guess its time to give up on finding some non-biased ground to discuss this on
2178[20:35:33] <Kobaz> some saltsploiting
2179[20:35:50] <joepublic> #debian has an official channel over on oftc if you want to try your luck there
2193[20:39:04] <zapotah> for the record, i did not bring anyones personal views on anything to this, i simply want to not understate the gravity of the exploit and apparently some people want to downplay it just due to animosity towards my approach to the issue
2195[20:40:08] <greycat> So I guess the saltexploit web site is only written for actual salt users who will know how to "turn off your Salt Master" and "create a new salt-master" and so on. Of course, that second part is contingent on "Install all updates for your distro", and those don't exist for Debian yet.
2196[20:40:09] <oxek> just so I learn something, is the package 'libc' something that exits on every debian install?
2200[20:41:08] <zapotah> anyway, for you who did read, take the problem seriously, tell whoever administers systems to take it seriously and tackle this immediately and not wait for the distro to do something "several days later"
2201[20:41:09] <annadane> i hope i didn't come across as hostile though i suspect it's not addressed to me
2202[20:41:14] <greycat> other mandatory packages are base-files, dpkg, etc.
2206[20:43:00] <brutser> guys, i got this issue that i cannot solve, i install a basic debian and openbox from install scripts, when installed on a laptop, xdg-user-dirs are installed and generated, when installed on a vm, no such package nor directories. i install pcmanfm, i only install python-xdg but no other xdg tools, when i do apt-cache rdepends xdg-user-dirs, i only
2207[20:43:00] <brutser> find libglib2.0-0 that is installed on the system, but on both the laptop and the vm - so why this difference??
2208[20:43:25] <annadane> oh god there's 5 versions of the malware
2209[20:43:34] <zapotah> cyveris: one last thing, complaining about free software when theres security issues or serious shortcomings to it, its warranted, if everyone is a yes man, nothing goes forward
2210[20:43:38] <annadane> it's spectre meltdown all over again
2211[20:43:48] <oxek> brutser: probably the vm pulls different dependencies than the laptop?
2212[20:43:55] <oxek> check out what the installed tasks were
2215[20:44:44] <oxek> annadane: have spectre/meltdown had any reported in the wild exploits being used?
2216[20:44:53] <annadane> idk
2217[20:44:56] <oxek> not downplaing them, just asking for update
2218[20:45:16] <oxek> whereas this thing looks like it is being actively exploited
2219[20:45:18] <zapotah> oxek: most required root access
2220[20:45:22] <annadane> spectre was always considered the harder one to exploit, and meltdown got (mostly?) patched, i stopped paying attention to that months ago
2221[20:45:30] <brutser> oxek: ok, looking dpkg.log i don't find, what is the place to look?
2222[20:46:38] <oxek> brutser: there was an install log somewhere, let me find it
2224[20:47:38] <oxek> I think it is the files in /var/log/installer
2225[20:48:03] <oxek> you might find some answer there, or try diffing the files for differences between laptop and vm
2226[20:48:09] <annadane> i do think we shouldn't yell at people who are only trying to help, but whatever
2227[20:48:53] <oxek> I doubt we're yelling at anyone unless they have some fancy text-to-speech on their end :)
2228[20:49:22] <zapotah> the unfortunate thing is that theres so much bad engineering around that challenging it is seemed hostile
2229[20:49:28] <annadane> well, there were all caps involved
2230[20:49:59] <annadane> i agree with 99% of the admonishments in this channel but i feel a major exploit is something that should at least be listened to
2237[20:51:05] <zapotah> but i guess im on ignore now so...
2238[20:51:09] <oxek> zapotah: please don't pursue that line of argumentation further
2239[20:51:14] <annadane> especially when said package affected by the exploit appears to be packaged in debian
2240[20:51:19] <greycat> The only actual *advice* I've seen so far is «read replaced-url
2241[20:51:21] <annadane> (as most things are, it being debian)
2242[20:51:42] <annadane> which program* affected by
2243[20:52:05] <joepublic> well, shut it down and nuke it from orbit need not involve any commands at all, necessarily
2244[20:52:16] <[E]sc> my laptop makes systematic scratching noises, as if it has something stuck in the fan when i'm using conky on my DE. is there a way to fix that? i'm on nvidia geforce mx150 440.82 kernel 5.5.0 using gnome 3.36.1
2245[20:52:18] <zapotah> oxek: fine, the arguments against me simply seemed nonsensical and based on something else than facts
2246[20:52:33] <zapotah> but i guess not everyone can think rationally
2247[20:52:49] <zapotah> we cant have nice things because of it
2262[20:55:26] <oxek> Does anyone actually read the topic?
2263[20:55:29] <greycat> No.
2264[20:55:43] <greycat> Changing the topic is a massive pain in the ass, and will not help anyone.
2265[20:55:51] <zapotah> i guess people have degenerated that far from irc...
2266[20:55:56] <zapotah> oh well
2267[20:55:59] <brutser> oxek: yes i find installation of xdg-user-dirs in /var/log/installer/syslog
2268[20:56:26] <brutser> it's hard though to read that file to determine what package caused the installation to occur
2269[20:56:33] <greycat> If you can actually *construct* advice that would be helpful to people, you can put it in a bot factoid with a useful name, and it can be triggered when people come here asking about the Salt bugs.
2270[20:56:59] <oxek> brutser: you could try `aptitude why xdg-user-dirs`
2271[20:57:01] *** Quits: ecbrown (~user@replaced-ip) (Remote host closed the connection)
2272[20:57:02] <greycat> But without even a bot factoid to point to, there's nothing we could even *fit* into the topic.
2278[20:58:15] <greycat> or simply the execution of the program, if the execting process has sufficient privs
2279[20:58:35] <zapotah> but i will leave as i see the consensus is pointing the opposite direction, hopefully someone will find the solutions elsewhere
2281[20:58:44] <shibboleth> does debian (org) use salt?
2282[20:59:19] <zapotah> at the very least i will have left the burning hatred of me peronally carve some synaptic memory of this issue to the apparent regulars
2283[20:59:24] <quotemstr> greycat: Is there a way to rerun postinst on arbitrary packages?
2311[21:08:29] <sawgood> Hi: Debian people: I put a file in /etc/init.d/ called mystartup.sh (and) I hav it set with 0755 ...... but when I restart the box: mystartup.sh (does not run)
2312[21:08:45] <sawgood> I know it works because I can run the file manually after reboot: and it works
2313[21:09:19] <greycat> sawgood: you should seriously consider learning systemd. But if you *must* write new init.d scripts, the piece you're missing is the symlink from /etc/rc2.d/ that begins with "S"
2314[21:10:02] <greycat> in all honesty, you would be better off just running the script directly from /etc/rc.local instead of trying to write and manage an init.d script
2316[21:12:54] <sawgood> greycat: perfect that is what I'll do
2317[21:13:02] <greycat> !rc.local
2318[21:13:03] <dpkg> /etc/rc.local may be used to run simple commands at boot time. It exists by default in jessie or older; in stretch or newer you need to create it. Don't forget the <shebang> and be sure to chmod 755 it. rc.local is considered a hack, a stopgap, or a temporary band-aid; see <systemd>
2319[21:13:10] <sawgood> I just want a few route commands to run after each reboot
2335[21:17:51] <greycat> dpkg, literal CMD: salt (.*?)
2336[21:17:51] <dpkg> "cmd: salt (.*?)" is "($1): <action>pours salt onto $1's slug-like corpse and watches it dissolve into a quivering pile of protoplasm."
2342[21:21:16] <annadane> dpkg, saltexploit is <reply> Some mitigation measures for the Salt exploit can be found at replaced-url
2343[21:21:16] <dpkg> annadane: okay
2344[21:21:34] <greycat> dpkg, salt is <reply>Salt is a tool used for remote mass system management. A vulnerability has recently become public (March to May 2020) and is being actively exploited. See <replaced-url
2345[21:21:34] * dpkg pours salt onto is <reply>Salt is a tool used for remote mass system management. A vulnerability has recently become public (March to May 2020) and is being actively exploited. See <replaced-url
2348[21:21:45] <dpkg> i forgot saltexploit, annadane
2349[21:21:45] <sney> lol
2350[21:21:58] <greycat> dpkg, forget CMD: salt (.*?)
2351[21:21:58] <dpkg> i forgot cmd: salt (.*?), greycat
2352[21:22:03] <greycat> dpkg, salt is <reply>Salt is a tool used for remote mass system management. A vulnerability has recently become public (March to May 2020) and is being actively exploited. See <replaced-url
2353[21:22:04] <dpkg> greycat: i already had it that way
2354[21:22:14] <sney> I just fixed it in a query
2355[21:22:27] <sney> you have to start with "no, " to correct an existing factoid
2356[21:23:01] <greycat> It's not correcting an existing factoid. The other factoid is a parametrized trigger.
2357[21:23:10] <greycat> It had to be completely removed.
2367[21:25:34] <dpkg> from memory, slug is the Sydney Linux User Group. Meetings are normally held on the last Friday of each month, at the University of Technology. replaced-url
2377[21:28:40] <brutser> ok, so i now know that libglib2.0-0 is recommending xdg-user-dirs and firefox-esr and viewnior (both packages i have installed), depends libglib2.0-0 - so i know why libglib2.0-0 is installed, but this is installed on both the laptop install as the vm installation, so the question remains, why xdg-user-dirs will create the user directories only on
2378[21:28:40] <brutser> the laptop installation and not on the vm installation
2379[21:28:50] <sawgood> greycat: ON the fast and cheap: couldn't I just make a new file called: /etc/rc.local (and) put my commands to run post-boot in that file?
2381[21:29:17] <greycat> Yes. Make sure it's got a shebang and +x perms. But you won't know *when* it runs, necessarily.
2382[21:29:42] <greycat> Network stuff is generally time sensitive (has to be run *after* interfaces are up), so rc.local may be chronologically impaired.
2383[21:30:58] <brutser> actually i must refine that , because even though libglib2.0-0 recommends xdg-user-dirs, it is NOT installed on the vm installation, but it is installed on the laptop
2392[21:36:49] <sawgood> greycat: testing /etc/rc.local now .. (new question) if I go with the /etc/network/interface (process) and use the (UP and DOWN) commands: can those commands use (route) vs (ip route)
2393[21:37:05] <greycat> it can be any command you want
2394[21:37:32] <sawgood> greycat: I think you've helped me before long long ago .. thanks
2395[21:38:52] <sawgood> When I rebooted: it did not run/load /etc/rc.local (but) I can run /etc/rc.local manually and they work: so I guess I'll try the up/down commands in /etc/network/interface (for best results)
2397[21:41:58] *** Quits: Jerrynicki (~niklas@replaced-ip) (Remote host closed the connection)
2398[21:42:41] *** Quits: HelloShitty (~narayan@replaced-ip) (Disconnected by services)
2399[21:43:01] <BadPractice> hi. I have a i7-8650U which has UHD 620 graphics. I would like to find out if i have the correct driver installed (and if not how to install it most maintainable on debian)
2400[21:44:15] <sney> hi BadPractice, intel graphics are supported out of the box most of the time. do you have working video?
2406[21:46:46] <sney> BadPractice: ok, sounds like it's working. intel contributes drivers directly to the linux kernel so there is no extra work to do in almost all cases
2411[21:48:12] * sney figures they were expecting a nouveau/nvidia situation
2412[21:48:30] <jhutchins> BadPractice: It's not usually a matter of poor performance caused by the "wrong" driver.
2413[21:49:01] *** debhelper sets mode: +l 1294
2414[21:49:06] <BadPractice> sney, yea exactly. I just wanted to make sure i use the fast one and not the open one. Seems like on intel its fast and open B)
2421[21:51:25] <sney> yeah, the hardware isn't amazing, but that driver fully supports that hardware is all
2422[21:51:50] <BadPractice> yea... i guess my perfomance issue is not a shitty driver as much as a shitty gpu. Well... at least my laptop is super light :D
2462[22:09:30] <oxek> I think the closest thing I found is replaced-url
2463[22:09:47] <oxek> but it is not in debian repos, and I can't tell if it is any decent
2464[22:10:20] <_DeLa_> jhutchins: Thanks!
2465[22:10:39] <jhutchins> oxek: I think it's more of a process than a program. You have to configure multiple other items.
2466[22:11:17] <jhutchins> oxek: As far as I know there isn't a package that does it for you.
2467[22:11:31] <jhutchins> oxek: PAM is not a plugin type of system.
2468[22:11:57] <greycat> The P literally stands for Pluggable.
2469[22:12:17] <sney> and the M stands for modules!
2470[22:12:18] <greycat> It's designed for you to write your own auth plugins (in C).
2471[22:12:26] <oxek> I am not using it to login to my system, I am using it to generate codes I input into e.g. Amazon AWS when I want to login
2472[22:12:41] <jhutchins> greycat: WEll, yes, but you don't just drop the module in a folder and it automatically takes effect, you have to reconfigure the stack.
2473[22:13:50] <jhutchins> oxek: Password generators are a dime a dozen, the trick is getting AWS to recognize it.
2479[22:15:25] <sney> jhutchins: remember those rsa fob things people had for dialup 2fa in the late 90s? they do that, more or less
2480[22:15:31] <oxek> and it generates it based on a 'secret' that you have, which proves possession of this secret to the webservice without revealing the secret
2481[22:15:49] <oxek> it's used for logging into google as well
2482[22:15:57] <jhutchins> sney: Yeah, we had RSA fobs for a couple of functions at my last job.
2483[22:16:02] <oxek> or github, or gitlab, or reddit, etc. it's everywhere
2484[22:16:10] <greycat> I still have one of the fobs. It's dead though.
2502[22:26:14] <dpkg> i heard dbgsym is Packages that end in '*-dbgsym' contain the symbols required for debugging executables and libraries. The dbgsym packages are automatically generated packages that are in a separate archive; add a line like "deb replaced-url
2503[22:26:20] <somiaj> It seems they are on snapshot.debian.org
2510[22:27:35] *** SunWuKung1 is now known as SunWuKung
2511[22:27:37] <greycat> Before stretch, there weren't any -dbgsym packages, and they weren't in a separate repository. There were simply -dbg packages.
2527[22:33:22] <f8e4> is there postgrest for debian? replaced-url
2528[22:33:24] <Kobaz> W: Download is performed unsandboxed as root as file '/var/cache/apt/archives/partial/perl-doc_5.28.1-6_all.deb' couldn't be accessed by user '_apt'. - pkgAcquire::Run (13: Permission denied)
2531[22:34:40] <greycat> I'm not even bother going to search for that for you. Do it yourself.
2532[22:34:58] <sney> !warning
2533[22:34:59] <dpkg> Yerp, lots of software outputs warnings. Kernel module warnings on boot, mplayer warnings, GTK warnings, X11 warnings, build warnings, gpg warnings. Don't be scared - informative output is a GOOD THING. Consider yourself warned, and if the program works as expected, be happy.
2534[22:34:59] <greycat> also I'm not even going to bother proofreading, apparently
2568[22:51:40] <sney> everyone knows not to add deb lines from random branches and derivatives so you don't create a frankendebian full of incompatible binaries. but what you *can* do if you're willing to tinker with it and solve problems as you go, is add a deb-src line and use the debian toolchain to automatically build your own packages on the debian version that you're using. this often works painlessly especially with standalone applications that don't depend on a
2571[22:52:35] <somiaj> _DeLa_: I don't see any bug reports either, so it appears that the users of debian don't have any issues with that package that they have reported to debian.
2572[22:53:33] <ndegruchy> _DeLa_: I imagine it has a lot to do with what you consider a threat vector. The link talks about a low number of encryption rounds, is upstream unwilling to increase this amount, or switch to a stronger (by default) method with more rounds?
2573[22:53:51] <_DeLa_> somiaj: TBH I just assumed that the github issue would still be relevant as it is from summer of 2018 and the buster package is just a few months older (January 2019)
2574[22:55:11] <_DeLa_> ndegruchy: right....well ... I decided to add a cryptomater folder into my seafile cloud for the more important stuff ...
2575[22:56:12] <ndegruchy> _DeLa_: I don't see anything on the upstream github regarding this issue, so it's either unreported or it's not considered an issue
2576[22:56:41] <_DeLa_> ndegruchy: Thanks for checking
2577[22:57:19] <somiaj> _DeLa_: I'm just pointing out that it isn't reported in debian. If there is a security issue / flaw with the pakage it should be reported to debian, though as ndegruchy maybe it isn't consider an issue.
2578[22:57:57] <somiaj> I don't know the software enough to give any statement on it is an issue, just poitning out that no one has reported it to debian
2582[22:59:56] <_DeLa_> Hoping not to offend anybody with the following thought: Seafile is open-source but it still is from communist china *cough* ...could this be an issue?
2583[23:00:44] <sawgood> Hi: I tried adding (up / down) statements in my /etc/network/intefaces (but) upon reboot: none of those commands see to work (here is an example)
2594[23:02:34] <sney> sawgood: it's also considered a best practice to use full paths, so 'up /sbin/route blabla'
2595[23:02:37] <somiaj> _DeLa_: in general if it is well audited no, but if there isnt' anyone looking for security issues, that could be a problem.
2596[23:03:02] <greycat> well, maybe it has to be run *later*, then ... maybe something else adds that route, after a few milliseconds, and your "route del" fired too quickly. who knows.
2597[23:03:22] <NieDzejkob> Hi there! I'm trying to report a bug regarding the rustc package. The `reportbug' tool says there's a new version in experimental, so I want to make sure the bug still reproduces in the new package. I added an 'deb replaced-url
2598[23:03:22] <NieDzejkob> DebianExperimental)
2599[23:03:35] <_DeLa_> ndegruchy: ok, version 7+ is in bullseye right now
2600[23:03:44] <sawgood> sney: I have (3) lines (1) add and (2) deletes
2601[23:04:02] <sawgood> it does the (1) add of a default GW , but not the (2) deletes
2602[23:04:06] <_DeLa_> somiaj: good, because people are obviously keeping an open eye....
2603[23:04:12] <NieDzejkob> (additional context: the rest of my sources.list is sid)
2628[23:13:11] *** nsegkos is now known as nksegos
2629[23:14:03] <sawgood> This box sits behind an ISP modem: which uses a routed /29 public IP subnet and still runs a RFC1918 192.168.1.x (LAN) ...
2630[23:14:58] <sawgood> only Debian: still sees the 192.168.1.x LAN (other) Linux distros never even know its there ... they don't see the 192.168.1.x LAN ... only Debian does ..
2631[23:14:59] <trek00> sawgood: what if you remove all the up lines?
2632[23:15:47] <sawgood> trek00: If I remove them and reboot: I'll have a default GW of the 192.168.1.x LAN upon reboot
2634[23:16:14] *** Quits: jthomas (~joseph_th@replaced-ip) (Remote host closed the connection)
2635[23:16:18] <trek00> sawgood: well something else configures your card, may be network-manager? networkd?
2636[23:16:26] <sawgood> 1st up line gets processeed, but not the next two lines (to remove things)
2637[23:17:11] <sawgood> trek00: I think so becausae my /etc/resolv.conf (also) gets auto populated for the 2 DNS servers on the 192.168.1.x LAN (and) it does not use the static entries I've put in many times
2638[23:17:21] <trek00> sawgood: dhcp?
2639[23:17:24] <greycat> perhaps your up lines are processed BEFORE the gateway line
2640[23:17:37] <sawgood> I will systemctl mask NetworkManager
2641[23:17:43] <greycat> having a gateway line that you don't actually want is quite silly
2642[23:17:51] <greycat> also, network-manager ignores interfaces in /e/n/i
2643[23:17:53] <sawgood> never thought of that: I thought only CentOS had NetworkManager (my bad)
2650[23:22:03] <sawgood> the ISP modem: issues DHCP out for a 192.168.1.x LAN (in addition) to having a routed /29 subnet ... (its not in full bridge mode), but only Debian boxes seem to see the 192.168.1.x LAN .... no other OS ever even know about it when they have a /29 IP assigned ...
2651[23:22:07] <znf> So, I'm trying to do a PXE install of Debian10, which works fine, but I'm also trying to pass a preseed file. I've added the preseed/url=replaced-url
2684[23:50:33] <Thedarkb-Desktop> Why is /sbin/ not in the path on default installs?
2685[23:51:19] <sney> /sbin has never been in path for a normal user
2686[23:51:25] <sney> you may be looking for this though
2687[23:51:26] <sney> !buster su
2688[23:51:26] <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/login.defs. See replaced-url
2689[23:51:29] <Maizum> hi room
2690[23:51:52] <Thedarkb-Desktop> Yeah, that's what I was after.
2691[23:55:18] *** Quits: dvs (~hibbard@replaced-ip) (Remote host closed the connection)
2692[23:55:48] <znf> 'su' overwrote the PATH? When the hell? I've been doing 'su -` for 10+ years at least
2693[23:56:38] *** Quits: Maizum (~ttt@replaced-ip) (Quit: Lost terminal)
2694[23:56:55] <sney> I switched to 'sudo -i' sometime in 2006 or so and never looked back, but yeah, apparently debian's su was doing nonstandard stuff the whole time