[00:37] HI [00:37] I lost a file. No backups. not in trash can. How can I undelete? [00:38] Intelo: you could try booting to a rescue system and then using extundelete or autopsy or sleuthkit to try to recover it [00:39] testdisk can probably also do something similar [00:40] sarnold thanks but https://imgur.com/9O0qR9b.png https://imgur.com/38SrBVr.png [00:41] Intelo: it probably isn't on /dev/sdb but on /dev/sdb1 or something similar [00:43] https://imgur.com/clzTe8r.png [00:43] sarnold [00:48] Intelo: hmm, is it on that encrypted vgx thing? or is it on one of the partitions from /dev/sda? [00:49] I am giving up. its too much effort [00:49] sarnold thank you either way :)! [00:49] Intelo: bummer :) good luck [00:56] oi === genii is now known as genii-core === M4he is now known as mahe === keypushe- is now known as keypusher [03:14] How can I test a raid array read/write speeds on ubuntu server? [04:23] Is there a way to make "Files" show webp thumbnails as well. I am using ubuntu 18.04 I know it's old can't change it though. [04:24] sure you can, but I accept that you feel like you can't =) [04:25] brandoneliza: https://askubuntu.com/questions/617047/how-to-preview-dds-and-webp-images-on-nautilus [04:26] matsaman, I can't because it's my bosses hardware :) [04:26] but I can compile stuff on it [04:26] sure you can, your boss won't be able to tell the difference, let's be honest [04:26] he can anyway I think I found it https://askubuntu.com/questions/617047/how-to-preview-dds-and-webp-images-on-nautilus [04:27] My favourite is kubuntu but I got hell for it once. [04:27] But I think because of filemanager-actions ubuntu is easier [04:28] ok =P [04:28] for productivity. I got kde configured a bit like gnome after getting inspired by it. [04:28] I hope the above works [05:05] Howdy, i am looking to find the key name of backspace in the format or found in xkb symbol files - how can i find those ? === fcgreg is now known as fcgreg_ === fcgreg_ is now known as fcgreg [05:38] Why does 18.04 LTS have "known not to work" (not for some use cases - known not to work for anything at all) versions of grpc++ in the repositories? [05:38] I get stability and not moving fast and upgrading to every new version, I don't get "known not to work - compile a newer version from source" [05:41] bailsman: where are you seeing that message? in the package description? [05:45] No, if the package maintainer literally put that in the package description I would truly no longer know what to say. It's a random stranger on the internet: https://github.com/etcd-cpp-apiv3/etcd-cpp-apiv3/issues/40#issuecomment-771363306 [05:45] Issue 40 in etcd-cpp-apiv3/etcd-cpp-apiv3 "Floating point exception (core dumped) on running the compiled app" [Closed] [05:46] but it seems indeed to be known not to work - multiple issues reported also upstream. [05:51] Maybe "everyone" knows this but nobody bothered to tell the package maintainer? [05:53] Likely it is due to upstream making breaking changes constantly, whereas distro release cycle is every 6 months [05:54] devops vs engineering [05:54] and with 18.04 that is obviously much longer since it is on a 5 year support cycle [05:56] 18.04 has 1.3.2 so makes sense [06:03] Oh, that makes a lot of sense. Thanks. I've been learning how to do DevOps. I shall try to master the art of making breaking changes constantly. === matsaman_ is now known as matsaman [06:55] I'm probably going about this compiling from source thing all wrong - currently I'm waiting one by one for "undefined reference to" errors and then searching for which -l flag to add - surely there's a tool that automates this and just compiles with whatever libraries I'm using without me having to know and specify that in advance? [06:57] bailsman: there's pkg-config [07:04] What am I supposed to pkg-config on? I did pkg-config --libs grpc++ - which, sure enough prints -lgrpc++ but not any of the other libs grpc itself depends on [07:06] well, packages with names like lib*-dev will typically install a pkg-config file so you can then do e.g. pkg-config --cflags libpng and pkg-config --libs libpng to get the right -I and -l options. So you'll have to do that for each dependancy, yes [07:07] usually the build system takes care of that automatically though. e.g. by running a configure script [07:08] I'm writing a small cpp program that depends on grpc, and the tool I'd like is for it to just transitively add whatever grpc depends on so that I only have to remember that I'm using grpc [07:08] that's what I'm used to from programming in other programming languages [07:08] There's no configure script - presumably I'd have to make one. I'm not sure how - can that be autogenerated too? [07:12] Never heard of gprc, but it does sound odd that you'd have to track its dependencies as well for your own project. Surely it has some documentation to explain how to do it though? [07:12] Why spend 1 minute reading documentation if you can spend an hour manually adding dependencies 1 by 1? [07:21] Looks like grpc installed some cmake files that are adding all those dependencies automatically - so I just needed to reference grpc from a cmake list [07:28] bailsman: apt-get build-dep " [07:37] yes TJ- or even apt-get install! It worked perfectly, in terms of dependency management, except that the version is old [07:38] bailsman: when backporting a newer package I start with build-dep and then figure out what more upstream have added. Usually that is easy by doing a diff of the configure.ac or cmake files === keypushe- is now known as keypusher [08:44] Can't get past 1024X768 on second monitor on Intel HD 4600 (i5-4570) on VGA output port.. [08:44] Primary DisplayPort to HDMI works fine on 1920X1200 on primary monitor. I am on Xubuntu 21.04 [08:50] sima: does "xrandr -q" indicate higher resolutions are available? if not, maybe the EDID is not being received and validated from the monitor [08:55] And if you're using an adapter (e.g. vga to hdmi), it may not pass the edid properly [08:57] alkisg, it's HP 600 G1 computer, it has 2X display port and 1 X VGA out. I just use DisplayPort to HDM cable for first monitor. [08:57] Other monitor is using VGA to VGA cable and is displaying only 1024X768 and not 1920X1080 [08:57] OK, also upload the output of xrandr that TJ- asked for [08:59] Previously I had AMD graphics in it (RX480) with using 2XHDMI, but I pulled it out because I didn't use it that much and RX480 was long enough to interfere with USB 3.0 jack on motherboard .. and I got fed with HP's BIOS constant nagging about pressing 'F1' every time I boot, if USB 3.0 jack is not plugged in.. [09:01] When I run xrandr -q , screen 1 (DP) goes temporarily black.. and it then spits out: https://bpa.st/3BZA [09:02] TJ-, probably, it's connected with VGA. MS Windows didn't make a problem setting up resolution. [09:04] sima: looks like it is using the Xorg default resolutions. You need to look in the Xorg.0.log file to found out what is going wrong. [09:05] sima: if you're using Gnome desktop I can't recall where that is now stored. For other flavours it is usually at /var/log/Xorg.0.log [09:19] TJ-, https://bpa.st/UASA [09:22] sima: is that the SMS24A450 monitor on VGA? [09:24] sima: looks like that is on the HDMI - so no, looks like no EDID being provided === iOllieN_ is now known as iOllieN [09:27] sima: check if there's an edid node in sysfs. "ls -l /sys/class/drm/card?-VGA*/edid" and if so, try reading it with "hexdump -C /sys/class/drm/card?-VGA*/edid" [09:29] I have a strange problem with my bluetooth dongle - it connects with my headphones, the headphones show up as a sound device ( with the respected speakers ) however there is no sound [09:30] thyriaen, the soundserver might need encouragement to change the datasink [09:30] ramblebamble, how can i give the soundserver a jab ? [09:30] e.g. I can have a BS Speaker connected to my Laptop and have one application use the BSSpeaker while another uses the speakers from my laptop [09:31] ramblebamble, ah but in my case when i go into settings i can hit "test speaker" and no sound comes out [09:31] with pulseaudio try pavucontrol, I don't know about pipewire, using the pulseaudio shim for that one [09:31] so it is unlikely on the application side [09:32] well where do you hit testaudio? [09:32] in the gnome sound settings [09:32] Also please do not cross post in other channels [09:32] I see your post on #linux as well, it is considered bad form [09:32] oh okay [09:33] How to set /sys/module/zfs/parameters/zfs_arc_max on 20.04? [09:33] well gnome does not configure the audio-server but uses its configuration only [09:33] the module is loaded in initramfs [09:33] well - in pavucontrol i get little "beeps" in my headphone when i adjust the volume [09:33] yes, so your phones are working, [09:33] you can configure sinks per application [09:34] ramblebamble, but my system sounds for example dont go to my phones [09:34] fling, no guarantee, but have a look at /etc/sysctl.d [09:34] TJ-, no Samsung monitor is on DisplayPort to HDMI cable and is doing allright. [09:36] thyriaen , look at confiugration, output and configuration, when you hear the volumn change in your earphones, it means it is only a matter of configuration because your phones are hooked up [09:36] TJ-, On Philips monitor on VGA, there is ls -l /sys/class/drm/card?-VGA*/edid giving : -r--r--r-- 1 root root 0 Aug 4 10:07 /sys/class/drm/card0-VGA-1/edid , but hexdump not reading anything from it.. === coconut__ is now known as coconut === denningsrogue9 is now known as denningsrogue [10:23] greetings, I tried to update nodejs to version 14 and made a mess. Now I can't remove nodejs nor can I install any other package without having to run fix-broken which gives me the following error: https://pastebin.com/yECJZc4L [10:24] BTW. I'm using kubuntu 21.04 [10:27] iomari892 have you considered using nvm [10:34] ramblebamble: I'll check it. Can that help my current problem? [10:34] yes, nvm, if I am not mistaken, is the node version manager [10:34] it moves the node configuration from a system to a user level [10:35] you can have multiple node-versions running side by side and you update/install/manage each of the versions and their packages seperately [10:45] ramblebamble: sysctl is for /proc/sys/ , not for /sys [10:45] sima2: sorry, was busy. if hexdump cannot read anything then that confirms the monitor/GPU isn't supplying the EDID data to the kernel or the kernel module is failing to find it for some reason [10:53] fling, ok, sry, since we are talking about /sys/module/... it is kernel module, look here https://stackoverflow.com/questions/23425998/can-i-pass-parameter-to-driver-during-insmod-or-modprobe other than that I would also be interested in a solution [10:58] ramblebamble: having trouble running nvm in zsh. Seems to be bash sensitive. [10:58] what is the error? [11:00] no error. there's no nvm command. THe install script updates bashrc which I find strange seeing that my system runs zsh. [11:00] that just means that it is not in your $PATH [11:00] run bash -c "which nvm" [11:01] blank output. Is the executable nvm.sh? [11:02] if so, I'll link it myself. [11:03] well try it out. but looking at https://raw.githubusercontent.com/creationix/nvm/master/install.sh it is aware of zsh, so it might be that you have to restart your shell [11:05] ramblebamble: I restarted my shell a few times [11:06] k, then just go with linking it yourself, it is a problem with finding it in $PATH, not nvm itself. I am sure it will solve your problem [11:06] here a guide https://tecadmin.net/how-to-install-nvm-on-ubuntu-20-04/ [11:06] afk [11:07] OK [11:35] iomari892 problem solved? [11:51] hi there [11:51] buntus [11:53] hi floratalin, if you have a Ubuntu support question just ask. Casual talk takes place in #ubuntu-offtopic. [11:54] i was wondering at how ubuntu is for a business / office setting [11:54] and i see there are different apps like snap and flatpaks [11:54] and which are better [11:55] there's no better, all a matter of personal preferences. Ubuntu ships snap by default. [12:05] mekaneck thanks for response [12:05] i guess the support is best considering the number of users [13:26] Hello! i have a question regarding preseed installation with the 20.04 mini.iso and the alternate legacy server image. Currently provisioning the system in hashicorps packer with this. during the installation i get the error "init: process '/sbin/reopen-console /sbin/debian-installer' (pid 322) exited. Scheduling for restart. in a constant infinite [13:26] loop. [13:28] what would be the package that provides vulkan headers? [13:30] Hi all [13:36] k, libvulkan-dev got it. but it's too old a version for me anyway [13:51] I have the binary download of Second Life...want to learn how to turn it into a debain/ubuntu package or otherwise something that will show in menus === genii-core is now known as genii [14:33] tes === poe[ is now known as sreggin [14:47] hi there! [14:48] waveform , are you here by any chance? [14:50] hjelp, yup, but partially distracted in meetings for next couple of hours [14:51] hjelp, but if it's about the issue you're having with the kernel it might be best to leave extra details in LP: #1937924 [14:51] Launchpad bug 1937924 in linux-raspi (Ubuntu) "kernel upgrade via apt 5.11.0.1015-raspi kills the OS on Raspberry Pi4B" [Undecided, New] https://launchpad.net/bugs/1937924 [14:51] waveform ok, I DM'd you about this already but I looked around and found i.e. problems with the multipathd starting up etc. [14:53] waveform weird that you couldn't replicate it though , because I could replicate it with 2 64-bit Xubuntu's and one 64-bit Ubuntu Server on the RPi4B ... [14:54] I've probed around the problem OS (the other one has no issues) and found a few things in the boot-up that might explain things, for instance 'journalctl -b' shows that i.e. the Device-Mapper Multipath Device controller fails to start, due to it failing to increase buffer size [14:54] well, I'm not surprised I couldn't replicate it with the server image: if a simple upgrade of a server image resulted in busted boot the certification team would almost certainly have picked it up already (that stuff gets tested automatically). But Xubuntu don't have any pi images, so I'm wondering what exactly those are [14:55] waveform Good question :D I might've used Wimpy's desktopify? [14:55] I think I tweaked around with it at one point [14:56] waveform: probably used the desktopify script [14:56] lol... even beat me to it [14:56] Oh yes, I've used that one, I'm not 100% it's on this build though, but definitely at least in the past. Probably in this one too; there was no official Ubuntu Desktop available back then and frankly it's too cumbersome even on some desktops [14:57] I'm not sure that's been kept up to date given that we started releasing "official" desktop images. To my mind the "supported" way of getting a xubuntu desktop on a pi at the moment would to start from the official desktop image, install xubuntu-desktop (with gdm3 login manager), reboot and switch session to xubuntu [14:58] I am on Ubuntu 20.04.2 LTS and screen is tearing horribly, it runs llvmpipe (LLVM 11.0.0, 256 bits) but should use "HD Graphics 530", how can i install the correct drivers? Xorg.0.log: https://paste.ubuntu.com/p/BWRB7tSScw/ [14:58] waveform ; pinpointed the dm-multipath error too, it's not found within the /lib/modules/5.11.0-1016-raspi directory [14:58] desktopify isn't actually supported here, for that you have to turn to Wimpy's discord server for support hjelp [14:59] Mekaneck , but I've gotten valid support from here in the past? :D Alright, where can I find the Discord [15:00] hi, trying to sync 2 hd in 2 ubuntu 18.4 pc's. hd in comp A is fat 32bit /dev/sd1 and in comp B is ext4 /dev/sdb1, the message is: Failed to create folder marker: mkdir /media/hadi57 .... [15:00] hjelp, okay -- but again -1016 is the -proposed kernel and I'm not entirely convinced you've got that installed correctly either (one of the things I tried earlier when trying to reproduce your issue was activating a pinned proposed pocket and upgrading just the kernel to 1016 -- a reboot worked happily afterwards as did iptables) [15:01] hjelp: there's a invite link on the Ubuntu MATE website [15:01] the multipathd is busted, I think that might have something to do with it? [15:01] correction: hi, trying to sync 2 hd in 2 ubuntu 18.4 pc's. hd in comp A is fat 32bit /dev/sdb1 and in comp B is ext4 /dev/sdb1, the message is: Failed to create folder marker: mkdir [15:02] (or is it really required...) [15:02] kinda funny tho that the other OS build works just fine now, I've updated it plenty of times via apt and keeping it upstream, but this one seems to struggle [15:02] (the other Xubuntu for the RPi4B, 64bit) [15:03] hjelp, no, I'm reasonably certain at this point that your kernel install is busted in some obscure way (multipath modules are definitely there in 1016) -- honestly at this point I'd say reflash a fresh image, don't enable proposed on it, and go from there [15:03] is there a nice task/todo software that you can recommend? [15:04] waveform I did try to keep it all synced with rsync -rvai ; I wonder if I could do some sort of a transplant attempt with the Ubuntu Server 21.04 or should I rather flash a desktop image and try to work from that? [15:04] sorry, rsync -rva [15:10] waveform did you mention yesterday that any boot transplanting attempt should be rather made using the 64bit Desktop rather than the Ubuntu Server edition? [15:11] just making sure [15:12] I could however copy the /lib/modules out of the working one, as well as its boot-partition and the /boot/ directory structure .. [15:13] I'll try that one out and see where it leads. I've nonetheless got this backed up so === kyonko000 is now known as kyonko [15:20] can `.deb` packages be installed on Ubuntu? [15:20] I mean older deb packages wrt to ubuntu version [15:21] Helmholtz, deb packages can, old debs?? maybe not, depends on dependencies.. [15:21] sudo dpkg -i [15:21] Helmholtz: "can" but "shouldn't" if there's a chance of breaking dependencies [15:22] oerheks, Is using a Github Action to automatically generate the deb file, discouraged? [15:22] (For installing on Ubuntu) [15:22] with `apt install ./release.deb` [15:22] Trying to install deb packes from outside the specific ones created for that version of Ubuntu you run can quickly become problemmatic as you descend into the Hell of missing or mismatched dependencies [15:22] ..just sayin [15:22] with `apt install ./release.deb` ?? where do you get that command? [15:23] oerheks: apt install has supported .deb's for some time now [15:23] oh oke, it is new to me [15:23] https://github.com/marketplace/actions/build-deb-action [15:26] I 'd say, try it out? [15:30] https://github.com/sharkdp/bat/blob/master/.github/workflows/CICD.yml#L288 [15:30] I wonder why bat guys wrote it themselves [15:31] Helmholtz: what package are you trying to install exactly? [15:32] dunst, but there are lots of packages which I'd benefit if this Action thing works out: polybar and some other utils [15:35] Helmholtz: version 1.5.0-1 of dunst is available in ubuntu 21.04. polybar is at version 3.5.4-1 [15:51] is there a CLI tool to lookup a package version across all(current) ubuntu releases? [15:53] leftyfb, i use 'rmadison ' [15:56] ioria: ah, sweet, thank you [15:57] ok [16:02] when using snap, some of the binaries in /snap/bin are prefixed with the snap name, is there a snap command to install these without the prefix? [16:03] chilversc, i think not, why would you mess with that? [16:03] because they have awkward names, such as google-cloud-sdk.kubectl instead of just kubectl [16:04] chilversc: make an alias? [16:05] can I make a system wide alias? [16:08] chilversc: yes [16:11] actually, a symlink might be better as some scripts might use sh? [16:15] waveform OK I got _everything_ to work by transplanting from the working Xubuntu (copied recursively /boot/firmware , the boot-partition and the /lib/modules), everything _except_ ... analog audio output :D [16:15] ufw works, multipathd works, no startup delay etc [16:16] for some reason though, the analog audio output (3.5mm jack) is not available anymore in the pulseaudio output devices list :< [16:17] I found this thread: https://askubuntu.com/questions/1279046/sound-issues-on-raspberrypi-4 <= the latest reply from late last year adviced to just give up on that with Ubuntu64bit on RPi4B [16:17] but, I have had the analog output working up until last month, so ... [16:17] though back to the original question, by your answers I assume this is actually specified by the snap package and is a choice by the package author rather than something built into snap install to avoid conflicts? [16:19] What do ubuntu shops normally use for directory services? Like for federated user role management? [16:20] I'm so sick of Microsoft AD and our sys admins are about to bring on some Ubuntu server for us, and I want to know how I can replace AD. [16:21] hello [16:23] can anyone help me out with power management in (K)Ubuntu) 21.04? ;) where can I find some howto/wiki about it? my cpu governor doesn't seem to change when I unplug AC... [16:31] ah, found it, its snap aliases, the snap file can define default/auto aliases; https://snapcraft.io/docs/commands-and-aliases [16:31] and I can add a custom alias using; snap alias google-cloud-sdk.kubectl kubectl [16:32] chilversc: good to know [16:33] yup, for simple conflicts between two packages there's "snap prefer" as well, so you don't have to go through and manually change each alias by hand, though that doesn't work in this case as google-cloud-sdk doesn't list an alias for kubectl [16:51] this gist details a method of creating an ubuntu installation in a chroot with debootstrap: https://gist.github.com/niflostancu/03810a8167edc533b1712551d4f90a14. This method shows mounting a boot directory and installing grub. I had thought a chroot was always using my system kernel. How does this /boot partition even have any effect? [16:55] noarb-: it's using the chroot to build a bootable system image, which will require grub and linux-image packages [16:56] waveform OK; I (re)installed all packages that were listed under 5.11.0-1016-raspi via apt (and with the '--reinstall') flag, re-flashed the flash-kernel with '--force', added the 'dtparam=audio=on' to an earlier portion of 'config.txt' and ~everything just about works now! :) Yay! === Booda is now known as tw3ml0w [17:18] thanks for clearing that up, TJ- ... never would've got there :) [17:19] I have a fully patched 18.4 system with an up-to-date tor installed. Yet, when I enable a hidden server it generates me an old style, v2 address, rather than a new v3. What am I doing wrong [17:43] jpmh: I would triple-check that the tor version is really recent and not just the latest package in ubuntu, e.g. with 'tor --version'. [17:44] holgersson: Tor version 0.3.2.10 [17:44] jpmh: for comparision, here I have "Tor version 0.4.6.6." on a Gentoo/Linux [17:45] jpmh: Does 0.3.2.10 support the v3 addresses? [17:45] please help me out with power management on laptop... please :> [17:45] jpmh: Did you set 'HiddenServiceVersion 3' when setting HiddenServiceDir? [17:46] holgersson: apparently not - so now the question is how do I get v 0.4 to my ubuntu 18.4 [17:48] jpmh: https://2019.www.torproject.org/docs/debian.html.en [17:49] HotblackDesiato: TY - heading there now [17:50] holgersson: yep that apt install is EXACTLY how I got tor [17:50] so, that does not help [17:51] Did you happen to see my question above? Otherwise, yes use the Tor apt repository to update. [17:52] jpmh: I guess you didn't add the repo from tor project. A quick look into their webserver shows more recent versions. [17:53] jpmh: but try Unit193's suggestion first [17:53] shuman: Hi, please ask a specific question. Your general question for help is so broad that it's unlikely anyone will help you. [17:54] shuman: E.g. describe what's not working, what did you try to achieve etc. [17:57] Unit193: I'm not seeing your comment. Please repeat [17:59] jpmh: Did you set 'HiddenServiceVersion 3' when setting HiddenServiceDir? [18:00] In earlier versions the default is still v2, though v3 is supported. [18:02] Unit193: I did not - easy enough to add that. Interestingly on my newer machine that is not reqioted. I'll try that [18:02] Unit193: I assume I put that after the directory line, right? [18:02] jpmh: Yeah. [18:05] Unit193: TY SO MUCH - that solved it - really appreciated [18:07] Sure thing! I'm not sure if later versions have fixes you may also desire, I believe in light of the DoS a little while back there were some changes so clients wouldn't be as effectedaffected. [18:09] Unit193: soon enough the server in question will be updated to a newer OS and so the issues will go away. We actually use the .onion address ONLY for our own use as a non-documented way to get to our servers [18:09] holgersson: ty also for your guidance [18:10] jpmh: np, glad that it now works :) [18:12] holgersson: ok :) I'll try to be more specific :D so... first of all I monitor my CPU speeds with intel-undervol measure (yes, I undervolted my cpu using this software) and they never go below 2Ghz (they went down to 800Mhz in arch based systems).. I used cpufreqd and saw that cpu governer is set to performance all the time... so how can I set it up so it goes to powersave when I'm on battery? [18:17] shuman: Sorry, I can't really help you there as I set my governor to ondemand (or schedutil) all the time. [18:17] shuman: Maybe this wiki article from arch linux helps you a bit https://wiki.archlinux.org/title/CPU_frequency_scaling [18:17] I know it's a different distribution, but you already used arch and the tools are the same and the linux kernel is also the same, just newer on arch. [18:17] holgersson: ok, sorry, I lied... auto-cpufreq shows Currently using: performance governor (not cpufreqd) [18:18] ok, I'll try that, thanks [18:24] Hello everyone. On my ubuntu server, when i type "cat /etc/resolv.conf i see this text: # This file is managed by man:systemd-resolved(8). Do not edit. What do I need to uninstall to be able to edit my resolv.conf file directy with no other automatic tools trying to get the work done for me? [18:27] VlanX: I'm hanging out on #ubuntu due to unrelated issues, and I'm really only familiar with devuan, but there was a "pretty much always" works trick that I remember [18:27] VlanX: https://wiki.debian.org/resolv.conf "Making /etc/resolv.conf immutable" 😃 [18:27] I have no idea if that works with systemd, but you'd think it would! [18:27] VlanX: why do you need to manually edit resolv.conf ? [18:27] ubuntu seems very slow to connect to my wifi for several minutes when I resume from sleep [18:27] touch /etc/resolvconf/resolv.conf.d/tail # and add them there? [18:27] in the past I ran a command that I think reset the network stack and reconnected faster [18:28] what's the wifi command that does that again? [18:28] cluelessperson, known sleep issue, reconnect after 60 sec [18:28] why does ubuntu keep changing my default audio device on reboot? [18:29] Batzy: you probably need to tell ALSA which device you want to use explicitly [18:29] hardware enumeration order isn't always guaranteed [18:29] i switch it in the DE UI, where else do i need to change it? [18:30] I'd use one of alsa's core text configs [18:31] ~/.asoundrc might work with pulseaudio still [18:32] leftyfb: I've put dns-nameservers = xxx.xxx.xxx.xxx under /etc/network/interfaces but for some reason it doesnt work [18:32] VlanX: what version of ubuntu? [18:33] VlanX: also, interfaces file doesn't use = [18:33] and it's not plural [18:33] VlanX: dns-nameserver x.x.x.x [18:34] leftyfb: Ubuntu 20.04.2 LTS [18:34] VlanX: 20.04 doesn't use /etc/network/interfaces [18:34] !netplan | VlanX [18:34] VlanX: Netplan is a network configuration abstraction renderer which uses YAML descriptions of a network to work with either a NetworkManager or Systemd-networkd "renderer". More information at https://netplan.io/ [18:35] you got to be kidding me, can it be because of dns-nameservers instead of dns-nameserver? [18:35] there is no = btw, my mistake [18:35] but indeed it was plural [18:35] VlanX: 20.04 doesn't use /etc/network/interfaces. Please look at netplan as linked to abov [18:35] e [18:36] leftyfb: it now does because i forced it to do so. I fkin hate the default method [18:36] Hi i've 2 ubuntu 20.04 machines in my network. 1 wired 1 wireless. Both machines network connection flapping. They can both ping other machines within the network without interruption but pinging the router and outside machines gets interrupted unexpectedly. other machines in the network dont have this problem (centos and freebsd wired wireless [18:37] connections) there is no firewall blocking any machines [18:37] routing tables seems normal [18:38] there are no dropped packages in ifconfig output [18:39] wired machine has static ip configuration wireless machine has dhcp [18:39] any ideas? [18:42] nemo: thanks mate, that resolved it :) [18:43] ubuntu 20.04; I pulled down the latest updates this morning, now it won't boot. [18:45] webchat10: maybe keep an eye on ip ne output on all the machines? I wonder if they've been configured to have the same mac or something? or maybe the router is pretending to have the mac of one of the machines? [18:45] VlanX: to be clear, nemo's suggestion technically breaks how it's supposed to function. netplan is the new method and you should really learn it, it's not that difficult. [18:46] sarnold: already checked it not the case [18:48] VlanX: 😉 [18:50] webchat10: oh dang :( [18:56] hi === keypushe- is now known as keypusher [19:06] Hi All!! [19:06] How to make XAMPP point to library files present in /usr/lib/x86_64-linux-gnu/ [19:06] ? [19:07] Currently it's taking from "/opt/lampp/lib". [19:07] please forget xampp, it is also not supported here, only tryue LAMPP [19:07] Wow!! [19:09] https://phoenixnap.com/kb/how-to-install-lamp-stack-on-ubuntu [19:09] I am following above to install LAMPP [19:09] Is it fine ? [19:10] that is fine, xampp can give weird issues on ubuntu [19:10] !xampp [19:10] We do not support XAMPP installs here. Please use the LAMP stack that is in our repositories; see https://help.ubuntu.com/community/ApacheMySQLPHP for more information. [19:11] https://ubuntu.com/server/docs/lamp-applications is a good start [19:12] Wonderful!! [19:24] ubuntu 20.04; I pulled down the latest updates this morning which included kernel updates, now it won't boot. I'm running off a live disk right now. [19:26] Everything was up to date as of last night. === tinwasp is now known as mavrix [19:29] Guest94: I think I'd approach that by chrooting to the installed system and using apt purge to remove the new kernel packages [19:30] or just booting to the previous kernel [19:30] sarnold Thank you. You're saying to set up an entire chroot environment? [19:31] leftyfb I don't get a menu. [19:31] Guest94: no, just to chroot to your system's installation [19:31] Guest94: if you *can* get that menu that'd be a bit easier ;) I think you have to hold down the left shift when booting to get the menu [19:32] Normally it just boots to the menu [19:32] No buttons required [19:33] so... mount /dev/sdxx /mnt && chroot /mnt ??? [19:35] Guest94: pretty close, yeah; mount /dev/sdxx /mnt ; then a few mount --rbind commands .. then the chroot /mnt bash -- and then do things .. [19:35] so I have to setup an entire chroot environment [19:36] Guest94: I'm not sure what you mean by that.. [19:36] Guest94: the mount commands are in the block after "If needed, you can chroot into your installed environment" on this page: https://openzfs.github.io/openzfs-docs/Getting%20Started/Ubuntu/Ubuntu%2020.04%20Root%20on%20ZFS.html#rescuing-using-a-live-cd [19:36] thank you [19:37] Hmm. Come to think of it.. I don't know what to purge. [19:40] Guest94: you can find logs in /var/log/dpkg.log and /var/log/apt/ to show which packages were installed, when; which packages were removed, etc [19:40] Guest94, i imagine that if you have hwe , you got 5.11 (instead of an 5.8 upgardes), but you need to check your /boot drectory [19:41] sarnold thank you again. I will do these things. I will also hold the left shift key down to see if a menu pops up. [19:41] Guest94: can you define "won't boot"? What do you get on the screen exactly? [19:42] ioria I'm not running hwe, but I think I did see 5.11 come down. It seemed like a large jump in any event. [19:43] leftyfb it's black and every now & then I see the cursor blink. Maybe every 30 seconds or so?? [19:43] yep, so see leftyfb comment above, then mount the root partition and run a 'dpkg -l | grep linux-image and post the output [19:44] I think I need to find a willing usb flash drive to save these commands. Give me a few minutes please. [19:44] Guest94, nope no need for that [19:45] Guest94, mount root ; for i in /dev /dev/pts /proc /sys /run; do sudo mount -B $i /mnt$i; done; sudo chroot /mnt; mount -a [19:49] hello [19:50] I need help with an experiment. If you bring your face close to the monitor and relax your eyes do have double vision? Can you go back and forth between double vision and focused vision? If you go back and forth like 10 times, does it feel weird and uncomfortable, maybe a little dizzying? [19:51] !ot | Samian [19:51] Samian: #ubuntu is the Ubuntu support channel, for all Ubuntu-related support questions. Please register with NickServ (see /msg ubottu !register) and use #ubuntu-offtopic for other topics (though our !guidelines apply there too). Thanks! [19:51] Samian: probably #ubuntu-offtopic is the better place [19:51] thanks [19:54] OK. I now have all the relevant notes saved. However... How & why does the commands from ioria differ from those supplied by sarnold? [19:55] Well not directly by sarnold but as specified by the link sarnold provided. [19:56] Guest94, i just want to know what kernels do you have installed and if they are 'correctly' installed [19:56] Not sure how to proceed with conflicting instructions. [19:56] Guest94: it's not a big deal, either one will work [19:57] Guest94: I like ioria's version more since it's smaller / easier / more compact :) hehe [19:59] well, considering I already did your version, I trust it's OK to reboot... as soon as I figure out which partition I was using at the time. [19:59] Guest94, once you have a working chroot, run dpkg -l | grep linux-image; check your installed kernels, then maybe a run a apt full-upgrade again ; exit and reboot [20:00] ioria 'k. Be a few minutes until I find the partition. [20:00] sy, heading out, gl [20:00] what? [20:01] ugh. [20:01] I didn't do a "full-upgrade". [20:01] I merely did an apt-get update && apt-get upgrade and since it was held back, followed by dist-upgrade. [20:02] But he's already gone. [20:03] all right, I'll continue up to but not including apt full-upgrade [20:08] helllo [20:19] I'm on an ubuntu focal. Just looked at `/dev/` and there are over 60 ttys. Is this normal? I thought linux makes less than 10 by default. [20:22] it's normal, I'm not sure if it's right or not, but they probably only take a few hundred bytes each [20:23] thanks sarnold [20:39] I finally found the partition. Man that took forever. [20:41] https://paste.ubuntu.com/p/gjcNdyfW9f/ [20:41] so it was definitely 5.11 [20:42] sarnold still here? [20:42] leftyfb ^^^ [20:52] So this is /var/log/dpkg.log https://paste.ubuntu.com/p/4NwXy7zHvK/ [20:53] using `apt show ` It shows a section under `Breaks:...` and lists a lot of (mostly kde) packages. What does `Breaks` section mean? [20:54] It means the package you're looking at is not compatible with, and cannot be installed alongside the "Breaks" packages. If you try to install it, it will tell you the "Breaks" list will be removed at the same time. [20:54] if foo breaks bar, then installing foo will force apt to remove bar, they can't be installed at the same time === keypushe- is now known as keypusher [20:55] thanks gdb [20:55] And this is /var/log/apt: https://paste.ubuntu.com/p/z9CwgRjpVW/ [20:55] Certainly! [20:57] So is this the list I should purge? Install: linux-image-5.11.0-25-generic:amd64 (5.11.0-25.27~20.04.1, automatic), linux-modules-extra-5.11.0-25-generic:amd64 (5.11.0-25.27~20.04.1, automatic), linux-modules-5.11.0-25-generic:amd64 (5.11.0-25.27~20.04.1, automatic), linux-headers-5.11.0-25-generic:amd64 (5.11.0-25.27~20.04.1, automatic), [20:57] linux-hwe-5.11-headers-5.11.0-25:amd64 (5.11.0-25.27~20.04.1, automatic) [20:57] Upgrade: linux-headers-generic-hwe-20.04:amd64 (5.8.0.63.71~20.04.45, 5.11.0.25.27~20.04.10), linux-image-generic-hwe-20.04:amd64 (5.8.0.63.71~20.04.45, 5.11.0.25.27~20.04.10), linux-generic-hwe-20.04:amd64 (5.8.0.63.71~20.04.45, 5.11.0.25.27~20.04.10) [20:58] alzgh: It's actuall more nuanced. Check this out: https://debian-handbook.info/browse/stable/sect.package-meta-information.html -- Sections 5.2.1.2 and 5.2.1.3 [20:58] er it's actually more nuanced* [20:59] What I described is "Conflicts" - "Breaks" is also a case of incompatibility but with a possible way out. [21:00] Yes, this is Debian documentation, but for the purposes of "What does this package meta data mean?" this is perfectly applicable. [21:00] thanks [21:00] Sure thing! [21:05] sigh. Something is mounted on /mnt but it won't unmount [21:12] It says target is busy. fuser says it's not [21:17] I guess the channel is asleep [21:18] Guest94: try umount -l [21:18] if it's FUSE, try fusermount -uz [21:18] otherwise use 'lsof' to see what's using it [21:19] anyone tried https://github.com/angelperezleon/who2f2b ? [21:19] looks promising if it can ban entire ranges [21:21] matsaman Yeah, took me a minute to remember lsof but that did the trick. Thanks! [21:23] Well, actually that only worked as far as lsof is concerned... I'm still getting target busy. [21:23] Guest94: by what? [21:23]