[00:00] oh /boot/efi on a supposedly bios booting system [00:01] it's because he installed grub when chrooted from an EFI system (/sys/firmware/efi existed) [00:03] that must have been a different time than when grub said "installing for i386-pc platform", though [00:06] yeah i'm confused - i think i just need two references in there - one to /dev/sda1 on /boot and one to /dev/sda2 on / and thats it [00:07] there is an efi directory in /boot now [00:07] i'm guessing ubuntu tried to create one in order to mount what it thought needed to be there? [00:07] your /boot seems to be a directory on /dev/sda2 [00:07] it is [00:08] /dev/sda1 is mounted at /boot which is a directory on /dev/sda2 [00:08] huh? this makes no sense [00:09] findmnt should provides some clarity [00:09] *provide [00:10] i think /etc/fstab was wrng when i booted and systemd in some way modified the file [00:11] something is really wrong..../dev/sda1 isnt mounted anywhere [00:11] i believe that /dev/sda1 is not in use. [00:11] i concur [00:12] grub will be on the MBR of this (or a different) disk. [00:12] there's only the one disk with two partitions [00:12] sda1 is a vfat file system created for uefi booting, it is meant to be an efi system partition. [00:13] both the root ("/") and /boot file systems are on the single ext4 file system that is on sda2 [00:13] well, the root filesystem contains the boot directory, yes [00:13] i.e. you do not ahave (or need, in this setup) a separate /boot file system (or partition) [00:14] *have [00:14] so it looks like i had uefi and then blew it away and then restored it as an mbr [00:14] well, gpt that is [00:15] testing [00:15] at some point you used to have something set up sda1 for uefi booting. whether this system ever uefi booted though, i do not know. [00:15] morgan-u2: please use #tests [00:16] maybe i should just leave it as is then [00:16] oh chrome has frozen again. I am used to it just quittin but this is the old freeze comeing back. SMH. I was just testing if IRC worked. the text editor didnt. [00:16] and just neglect the 500MB primary partition [00:16] tomreyn, sorry if I said it wrong. [00:16] you remember mecomplaining. just quitting is much superior to freexing. [00:18] morgan-u2: i'm kindly asking you to do your tests on a channel where spurious channel activity makes no difference to anyone. [00:18] (in he future) [00:18] gmcastil: what does 'ls /sys/firmware/efi' give? [00:18] gmcastil: you could wipefs sda1 and remove it (and the note on it) off fstab [00:19] that's if the above directory does not exist [00:19] sure tom, no problem [00:19] morgan-u2: thank you. [00:20] Hello, how do I remove snap completely off my system? [00:21] which ubuntu release are you running? are you aware of the consequences? do you also want to prevent it from being re-installed later? [00:21] Short answer is you don't [00:22] Or maybe I'm wrong [00:22] well i tend to agree with you [00:22] tomreyn: I am using Ubuntu 23.04, yes and yes [00:23] so you don't need a web browser? [00:23] I have installed one via flatpak [00:25] https://www.baeldung.com/linux/snap-remove-disable and https://haydenjames.io/remove-snap-ubuntu-22-04-lts/ are two guides which i assume should both work. keep in mind that a standard 23.04 and later ubuntu install will expect that snap is available / can be installed, so you'll be in unsupportable terrain sooner or later. [00:26] ok thanks [00:26] also note that removing snap's can cause data in ~/snap/ - such as browser profiles, to be deleted without prompting. [00:26] got it [00:27] ooops wrong channel [00:29] EriC^^, no efi entry in sys/firmware [00:30] gmcastil: starting with a slash, though, /sys/firmware/ [00:30] yeah [00:30] so you booted in bios mode, not uefi [00:30] apparently so [00:31] and if you will continue to do so, you won't need the file system on sda1, the partition that is sda1, nor the reference to either in /etc/fstab [00:31] yeah i dont have EFI enabled in the VM - i've never done that either [00:32] so i'm unsure as to why that was ever done that way in the first place [00:33] we will not (need to) determine this reliably. [00:33] yeah its not necessary i dont think [00:49] I'm lost [00:50] where do the trolls gather [00:51] i used to have a boyfriend named "ubuntu" [00:51] he would "play the bongos" on my ass [01:09] tomreyn: best to ban *id622320@* [01:11] jeremy31: i'd discuss this with you / ask about it, but not here. === chris14_ is now known as chris14 === five61841 is now known as five6184 [01:42] Cleaning up some old packages I don't use. Colonel something, I think. It'll be fine. === lucerne3 is now known as lucerne === synapt is now known as nate === bluejaypop0 is now known as bluejaypop === adamus1red_ is now known as adamus1red === driib5 is now known as driib === PowaBanga_ is now known as PowaBanga [02:51] how can I force 'dpkg --configure -a' to respect MAKEFLAGS=-j4 instead of using all the CPU cores (which is causing shutdown maybe from overheating)? [02:52] that depends on what dpkg --configure -a is doing [02:52] building a kernel [02:53] building a whole kernel? what kind of crazy pacakge do you have to compiel a kernel [02:53] amdgpu [02:53] thats aa module, not a kernel [02:54] it's not just a module [02:55] repo.radeon.com/amdgpu-install/latest/ubuntu [03:01] though may just be the module causing it (DKMS) [03:32] coming from UNIX, I have a lot to criticize about GNU/Linux, especially Ubuntu, like in this case it won't respect MAKEFLAGS and seems no way to cancel the process by uninstalling the driver and reinstalling with '--no-dkms' [03:43] Hey there, has anyone here ever changed the default number of Snap revisions to keep and then changed back to the default number? [03:49] suggest me some good irc server/channels [03:49] this isn't the yellow pages of irc... [03:49] then where should i go my man [03:50] not here [03:50] rude [03:50] lol [03:50] can u hack me [03:51] totally [03:51] plz do so [03:52] the original: EFnet & IRCnet [03:53] Connection failed. Error: * Verify E: Failed to validate hostname? [03:54] what does it mean #ircnet [03:56] it's a network, not channel [03:59] hi there I launched a game that uses openal\openal-soft and it appears to be installed with ubuntu jammy (lts). is it because of pulseaudio? [03:59] i didn't think it was included out of the box that's all [04:01] huh? [04:01] i didn't think openal support was out of the box and you had to install the package [04:02] is there a problem? [04:05] rbox: i just want to confirm it is before i tell someone else [04:05] confirm what [04:05] stop trolling [04:05] that libopenal is included out of the box [04:06] i dont think i've ever seen antyhing named libopenal [04:06] so probaly not [04:07] wtf [04:07] jesus christ === zofrex2 is now known as zofrex === John_Ivan is now known as DicktaThor === five61846 is now known as five6184 [06:31] HYello, hello, what was that script which tested the gpu chip and adviced the right binary nvidia driver ? [06:34] paul424: sudo ubuntu-drivers list . [06:38] hello [06:41] bashing-om: thanks [06:41] ohh there is also ubuntu-drivers autoinstall [06:42] Would Ubuntu find and install the proper dirvers for 14-years old laptop ? [06:48] hmm the aformentioned command returned empty line.... [06:48] what to do ? [06:48] would noveau be equally fast for such an old card ? === mIk3_09 is now known as mIk3_08 [06:58] should I do this : [06:58] sudo add-apt-repository ppa:graphics-drivers/ppa [07:08] bluetoothd: a2dp-sink profile connect failed: Protocol not available <-- rings a bell to anyone? all solutions I found are 'remove pipewire and install pulse-bt', but that makes the whole ubuntu-desktop house of cards to fall. any alternatives? === five61841 is now known as five6184 [07:35] would noveau be equally fast for such an old card ? [07:38] I'm trying to install a package using apt-get. When I do 'apt-get -y install cuda', I get "some packages could not be installed..." This is the full error: https://hastebin.com/share/kefurukoki.ruby [07:39] Reading this error output, I guess that I have to install something called 'libnvidia-compute-535'. But when I do 'apt-get install libnvidia-compute-535', the prompt returns nearly immediately with a message saying that I already have the latest version installed [07:40] paul424: not likley I found it barly working on most older cards [07:40] toddc, oki thanks [07:41] !latest | bobdobbs [07:41] bobdobbs: Packages in Ubuntu may not be the latest. Ubuntu aims for stability, so "latest" may not be a good idea. Post-release updates are only considered if they are fixes for security vulnerabilities, high impact bug fixes, or unintrusive bug fixes with substantial benefit. See also !backports, !sru, and !ppa. [07:44] toddc: I'm not sure what to do with that information [07:44] cuda is from nvidia and may need files not avaible yet to ubuntu so you may need to locate then elseware or build them [07:45] bobdobbs: I am reading https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#ubuntu [07:46] toddc: yeah, I'm still trying to figure out the cuda/nvida/ubuntu landscape as well [07:47] toddc: I think I *might* have found a workaround: that package that is already installed? I purged it. [07:47] toddc: then I ran 'apt install cuda' again, and it's running now [07:47] bobdobbs: good luck [07:47] toddc: it's all a bit nerve-wracking, cos if something goes wrong then I might not get video after a reboot [07:48] toddc: thanks [07:49] toddc: k, cross your fingers for me. I'm gonna reboot... [07:49] I am lucky I do need latest and can stick to test stable (old) [08:11] toddc: it worked! I have nvidia drivers working again! [08:11] toddc: And I think I have the cuda toolkit installed property, cos I have stable diffusion running again. I think that that needs the cuda toolkit [09:03] hey. anyone ever had a bug where nemo opened all images in folder instead of the image tha was double clicked for gnome image viewer only? eg. eog img0 imgN instead of eog img0. the curious thing about this is that its only for eog, eg. for shotwell it works correctly. [09:05] ciau [09:06] i apologize : hello ... :) [09:22] is there, in irc, a ubuntu offtopic channel ? [09:24] !ot [09:24] #ubuntu is the Ubuntu support channel, for all Ubuntu-related support questions. Please use #ubuntu-offtopic for other topics (though our !guidelines apply there too). Thanks! [09:26] thanks :) [09:29] https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/2037202 When will the fix for that bug be in the beta tarballs so that i can test? [09:29] -ubottu:#ubuntu- Launchpad bug 2037202 in initramfs-tools (Ubuntu) "Mantic/23.10: PXE boot tries to initialize DHCP before network link is up" [Undecided, Fix Released] === EriC^^_ is now known as EriC^^ === TomTom_ is now known as TomTom [09:57] greetings, Is there any downside to adding a gui to my server? [09:58] From what i heard, it's usually best to use the ubuntu desktop iso [09:58] iomari891: more attack surface, but it depends what exactly you want to do with it [09:59] my home server has a desktop because it's also a media center, but it's behind 2 NATs [09:59] iomari891: i use xpra to open a terminal from my server, then open other programs from that terminal [10:00] also what akik mentions, you can install GUI tools w/o installing a whole desktop [10:00] Some of my clients ae using suse server and some are using centos. Both have GUIs. FOr me it'snt an issue but some of my clients want a gui frontend to work with. [10:01] but it's more 'unstable' because you need 100% connectivity, otherwise the program just finishes and I think it even crashes (in the sense that it doesn't try to do a graceful quit) [10:02] These clients have physical access to their servers or local connections. [10:14] hey. anyone ever had a bug where nemo opened all images in folder instead of the image tha was double clicked for gnome image viewer only? eg. eog img0 imgN instead of eog img0. the curious thing about this is that its only for eog, eg. for shotwell it works correctly. === pikapika is now known as militantorc [11:47] Hi all [12:32] hi all. we have a ubuntu 20.04 vm running on vmware. it has to kernels installed: linux-5.4.0-65-lowlatency and linux-5.4.0-164-lowlatency. at the moment i do not have access to the vmware console, and cannot use the grub meny to select which kernel we fish to boot, so it boots the older -65 version. ive tried changing the /etc/default/grub to GRUB_DEFAULT="Advanced options for Ubuntu>Ubuntu, with Linux 5.4.0-164-lowlatency" and running update- [12:32] grub, but after reboot uname -r stil shows the older kernel. [12:33] i also tried purging the -164 version with apt, rebooting, installing it again and rebooting, with no success. it seems its stuck on the -65 version for some reason. why what how? about to go mad here :P [12:38] Hey there. Can this command be changed to work on just the Firefox Snap or can it only be used for all Snaps? sudo snap set system refresh.retain=3 [12:41] Any reason for why snap is considered better than flatpak? [12:42] Elliria: it seems to be a system-wide setting. as you noticed, there is #snappy for snap specific questions, too. [12:42] NIH [12:43] Sandboxing is more robust [12:43] Yeah, they didn't answer last night and haven't answered today, so I thought I'd ask in here. Thanks. [12:43] coconut: there are many comparisons online, discussing pro's and con's. a relevant advantage of snap would be how it isolates snaps from the rest of the system. flatpaks do this to a much lesser degree. [12:44] oh ok, thanks tomreyn [12:45] This video does a comparison between Flatpak, Snaps, Appimage, and Packages, coconut: https://www.youtube.com/watch?v=9HuExVD56Bo [12:45] *click* [12:59] Will ubuntu ever have a compatile version with the new macbook pro? [13:01] With the M2 pro and M2 max [13:02] never day never ever [13:02] but its unlikely to happen in the near future [13:02] *say [13:02] There's always ashai linux [13:02] https://ubuntuasahi.org/ [13:03] I ask because i need a new laptop soon [13:03] not supported here btw [13:04] then buy hardware from a company that does not work against open source [13:04] you know a good one ravage? [13:05] https://ubuntu.com/certified/laptops [13:06] Hmmm, i would have liked a renew of my thinkpad, but lenovo does not sell it anymore [13:11] jadesoturi: check your /boot/grub.cfg -which files are referenced? check the /linux symbolic link, which kernel image does it point to? make sure you run update-grub after editing /etc/default/grub [13:14] tomreyn: boot.cfg looks correct. its referencing the correct images for all menu choices. but the /linux symlink - where is it located? [13:15] and i have ran update-grub after making changes. also tried the grub-set-default. I see the default and grubenv files are updated with my changes, but after reboot, still returns -65 kernel in uname -r [13:16] sorry, i meant /vmlinuz [13:17] i guess your virtualization uses pvgrub/pygrub or similar to select the kernel to boot externally [13:17] (overriding any configurations you make on the system) [13:17] yeah except for i have 10 other vms where i changed from -generic to -lowlatency with no problem.. one would think pvgrub would have the same effect there? [13:18] i won't make assumptions on how reliable / reproducible your virtualization is [13:19] its a pretty standard vmware esxi setup from what i know, but im in devops not on the infrastructure team so cant say for sure my self:P [13:19] so it might be a good time to talk to the infrastructure team [13:20] vmlinuz is pointing to the correct image(boot/vmlinuz-5.4.0-164-lowlatency) and the vmlinuz.old is pointing to the -65 version [13:20] yeah. already tried but the one guy there that knows anything about linux is unreacheable at the moment:P so trying other ways :) [13:21] cat /proc/cmdline says what? [13:21] -65 .. but WTH? [13:21] what sets the cmdline ? [13:21] it certainly says more than that [13:22] -65 as in its pointing to the old image :) [13:22] BOOT_IMAGE=/vmlinuz-5.4.0-65-lowlatency root=/dev/mapper... [13:23] but should that not be updated with update-grub or something? [13:23] well, you'll need to reconfigure esxi most likely. i'm not too familiar with how it boots. [13:25] ok. thank you, ill follow up with the infrastructure team in the morning i guess... [13:25] i suspect you're using the VMware ESXi - Guest Bootloader - Direct Kernel Boot option there. [13:26] hmm that would be weird... especially since all these VMs are setup at the same time IIRC and no other VMs have this issue. but thank you for the pointer. ill see what they figure out tomorrow i guess === gnuself1 is now known as gnuself [15:33] On a fossa installation, I want to mirror a LV. The VG har only one PV, so I created a partitio nof the same size as thj ecurrent partition holding th PV. so far so good. No wI try to create a new PV using pvcreate, say pvcreate /dev/sdb3. pvcreate does not comply, BUT pvdisplay tells me, that th e new PV has 0/zero PE. How is that? === Juesto is now known as Juest [16:28] i have a laptop that just started having problems - desktop shows up momentarily and then goes black - what could cause this? ubuntu 23.04 on hp pavilion [16:28] eelstrebor: black, or flickering screen? [16:28] it's dual boot and the windows partition comes up ok [16:29] lotuspsychje, black [16:29] eelstrebor: what kind of graphics does your laptop have? [16:29] i think it's nvidia [16:30] eelstrebor: are you on xorg or wayland? [16:31] wayland [16:32] wayland + nvidia is a work in progress still eelstrebor try to logout and boot into xorg [16:32] how do i do that [16:32] logout [16:32] logout. select your user. click on the gear [16:33] this is an auto login [16:33] choose xorf [16:33] *xorg [16:33] i was faster :D [16:33] :p [16:33] logout works anyway [16:33] it remembers the last session type [16:33] can't logout with a black screen [16:34] use recoverymode or !nomodeset to get in your system eelstrebor [16:34] alright, i'll give that a try [16:35] bbl [16:35] or if you can get to a console you can try "sudo /etc/gdm/custom.conf" and change AutomaticLoginEnable to false [16:35] oops [16:35] sudo nano /etc/gdm/custom.conf [16:35] i can't get to a console either [16:37] then nomodeset is a good chance [16:37] !nomodeset [16:37] Systems with certain graphics chipsets may not boot properly out of the box. "Temporarily Add a Kernel Boot Parameter for Testing" as discussed at https://wiki.ubuntu.com/Kernel/KernelBootParameters to add the "nomodeset" parameter there. [16:53] Hi, is it possible to install xubuntu on an USB pendrive with persistent partition, that way, that it's possible to install software, configure it, without losing it on reboot, or is it only for the home directory and data saved on it? Is it possible with Rufus? [16:54] you can install from one USB drive to another with the normal installer [16:55] i had to edit /etc/gdm3/custom.conf instead of /etc/gdm/custom.conf - got rid of the autologin but when i click on a user i get a black screen and it looks like it flashes every few seconds - i'll try nomodeset and edit /etc/gdm3/custom.conf to disable wayland and see if thinks work on a reboot [17:04] disabling wayland made no difference - btw, the graphics card is a Wani [Radeon R5/R6/R7 Graphics] [17:05] why this has been working OK for almost 2 months and not now is puzzling [17:05] you said nvidia earlier? [17:06] yeah i was wrrong - i just did a hwlist and it came up ati radeon [17:07] eelstrebor: then its another problem eelstrebor [17:07] eelstrebor: when did the black/flickering started, after an update? [17:09] not sure - i think i did an update last night but the laptop booted ok this morning and after about 2 hours it statred screwing up - did a reboot and momentarily got a desktop and then it went blank [17:10] eelstrebor: try booting into a previous kernel as a test [17:13] interesting, booting the previous kernel worked ok [17:14] eelstrebor: ok, so assuming its an amd kernel issue, try to find recent bugs on 'linux' kernel version not booting [17:15] eelstrebor: https://bugs.launchpad.net/ubuntu/+source/linux/+bugs?orderby=importance&start=0 [17:22] eelstrebor: if you cant find one right away, you can also file a !bug yourself against the linux package === nicoz_ is now known as nicoz [17:36] ravage, and it doesn't loses it's settings and new installed packages? === lyn is now known as lynorian2 === lyn is now known as Guest5243 === Guest5243 is now known as lynorian2 [19:06] i'm trying to install a system using debootstrap but it seems to fail with the following in the log file: https://termbin.com/0nsl3 [19:07] the cmd i'm using is: debootstrap --verbose $(lsb_release --short --codename) /mnt it also outputs the following [19:09] ln: failed to create symbolic link '/mnt/lib': File exists [19:10] If i cleanup the target dir and re-run output varies, but almost always it's either '/mnt//var/cache/apt/archives/partial/zlib1g_1%3a1.2.11.dfsg-2ubuntu9_amd64.deb' or /mnt//var/cache/apt/archives/partial/usrmerge_25ubuntu2_all.deb [19:11] googling and duckduckgo and stackoverflow hasn't yeileded anything useful, canonical forums too [19:15] halp [19:16] I wonder what happens if i try to install a prev LTS [19:16] W: Failure while unpacking required packages. This will be attempted up to five times. [19:16] W: See /mnt/debootstrap/debootstrap.log for details (possibly the package /var/cache/apt/archives/usrmerge_25ubuntu2_all.deb is at fault) [19:17] full log for this one https://termbin.com/n7x2 [19:17] I don't have an answer for you but might I suggest you lookup some debootstrap videos by "Animortis Fortress" on YouTube. Note that these are unoffical support videos but proved to be super helpful for my last debootstrap install. === dstein64- is now known as dstein64 [19:18] WaV: most of the videos say run this cmd and don't have a "negative path" so to speak, unless this one does? thank you for a potential avenue of reserach [19:22] Well, i just tried with focal and //archive.ubuntu.com/ubuntu/pool/main/z/zlib/zlib1g_1.2.11.dfsg-2ubuntu1_amd64.deb still the same issues [19:54] hmmm, so it seems the issue is the target i just tried a subfolder in /mnt/ (my original target) and it worked [19:59] so passphrase will be stored in the TPM and then TPM will provide it to LUKS to unlock the partition? [20:01] ice9: I don't this happens by default. You have to configure that explicitly. [20:02] to make luks work with tpm. [20:05] My plan is luks to unlock partition luks, and then the key. I'm currious how to make the tpm and how really trusted is it? [20:05] maybe this might help: https://glentomkowiak.medium.com/luks-with-tpm-in-ubuntu-df867cad9a1 [20:07] effendy: thank you [20:13] I'm trying to setup LTS with encrypted zfs, LUKS, and debootstrap my system [20:13] so that will come in handy [20:14] So i just came across the following https://www.reddit.com/r/homelab/comments/6cw7ge/multiple_pools_debian_jessie_root_on_zfs_xpost/ [20:14] I wonder if I need some diferent attributes too [20:46] how to run pyton script with shortcut? [20:51] I am currently running 22.04 on my machine. If I upgrade to 23.04 will I be able to upgrade to 24.04 next year? [20:53] To upgrade to 23.04 you first will have to 22.10. And to upgrade from 23.04 to 24.04 you'll have to upgrade to 23.10 first. [20:54] xangua so upgrading 24.04 will not be an issue right? I am new to Ubuntu and the versions and the things like interim and LTS is bot confusing when coming from Windows [20:56] xangua: That's false as 22.10 cannot be upgraded to, it's EOL. [20:56] And you can upgrade LTS-to-LTS without using the regular releases. [20:56] That's up to you, you can upgrade thru all the releases if you want or just patiently wait for 24.04 and you will be able to upgrade from 22.04 to 24.04 directly [21:18] Yakov: what do you mean by "shortcut"? [22:21] appName: ERROR: cannot open device /dev/input/event5 [Permission denied] | using dev input with my c++ code, how to set permissions? === hikikomori is now known as Whithnail [22:35] hi. my vlc player plays videos only with sound but without a picture. how can i solve that problem? [22:55] ^ i'm responding in #ubuntu-de [22:57] Yakov: you should not change the device nodes' permission, but temporarily run code with elevated permissions (ideally much restricted / functtionally limited code, which maybe passes what it learnt to code running with less permissions) [22:58] kind of what sudo or suid could do [22:58] but this is more of a #programming question really [22:58] this helped https://askubuntu.com/a/609678 [22:59] I'm running Ubuntu 22.04 in virtualbox. But for some reason I can no longer log in. I get to the login screen, enter my password, it works for a long time(30 seconds-ish?), but then it returns to the login screen. Can I somehow run it in failsafe mode or something? [23:24] Slartibart, hello, [23:25] Slartibart, you could change to tty1 (ctrl+alt+f1; note it is ctrl+alt+f5 or f6 or f7 to get back to your normal login screen). tty1 is a command line only. Once logged in, install some other login screen software different from what you currently have, and set it as default, then reboot. (I think "sudo apt install xdm" could do the job.) [23:25] Slartibart, if that works, at least you'll have a working login where to troubleshoot further.