=== jancoow4 is now known as jancoow === Xlusive_ is now known as Xlusive [00:57] sigh, display still flickers even with this newer kernel [00:57] @ 165hz === arraybolt3_ is now known as arraybolt3 === chris15 is now known as chris14 === elastic_dog is now known as Guest5832 === chris15 is now known as chris14 === chris14- is now known as chris14 [02:46] nvidia-smi shows cuda 12 installed. dpkg — list - shows cuda 11.8 - how do i get nvidia-smi to show cuda 11.8 instead of 12? [02:46] pycurious: Do you happen to have multiple installations of cuda? [02:47] (You can possibly have stuff like that if you're using Python venvs with CUDA installed with pip.) [02:48] arraybolt3: looks like it - not sure how to find out if the cuda is coming from venv or the system [02:48] pycurious: Try running nvidia-smi in a new terminal, maybe? [02:48] Also make sure your LD_LIBRARY_PATH variable is cleared (if you can do so safely). [02:50] hey [02:50] arraybolt3: they all show 12 [02:50] pycurious: Hmm... that's weird. You didn't use the CUDA .run script from NVIDIA by any chance, did you? [02:50] so i'm on ubuntu studio 22.10 and i have this weird issue [02:51] periodically plasma will freeze and i can't right click on the desktop or open the start menu but apps that are still running still work just fine [02:51] nope - just apt update/upgrade. I do use https://lambdalabs.com/lambda-stack-deep-learning-software [02:51] and i have to ctrl alt backspace to restart plasma [02:51] any potential fixes? [02:52] pycurious: That installs CUDA using a script that is given to "sh". It's almost certainly installing CUDA 12. [02:52] "It provides up-to-date versions of PyTorch, TensorFlow, CUDA, CuDNN, NVIDIA Drivers, and everything you need to be productive for AI. [02:52] pinkergloop: The plasma that comes with Ubuntu Studio 22.10 is 5.25, which was a very buggy release. Using the instructions in the release notes, add the Kubuntu Backports PPA which should get you to a newer, less-buggy Plasma. [02:52] Notice "up-to-date". [02:53] Eickmeyer: release ntoes for Kubuntu or Ubuntu Studio? [02:53] pycurious: What's more than likely is that you installed CUDA using apt, then installed it using Lambda Stack, and now Lambda Stack's version had overwritten apt's version without telling it. [02:53] pinkergloop: You're running Ubuntu Studio, so obviously, Ubuntu Studio's release notes. [02:53] pycurious: This is why you should almost never install software on Ubuntu without using apt, Snap, or some other real package manager. [02:54] arraybolt3: dpkg --list | grep CUDA - what is surprising is that this only shows 11.8 [02:54] It's also possible that they've introduced weird apt packages with wrong version numbers. [02:54] pinkergloop: Terribly sorry, it's actually in the release announcement. [02:54] pycurious: I see now that they say that using "sudo apt update && sudo apt dist-upgrade" will update your AI stack. [02:54] pinkergloop: https://ubuntustudio.org/2022/10/ubuntu-studio-22-10-released/ [02:54] pycurious: So they probably botched a number. What's the *full* version number of the CUDA package? [02:55] pycurious: I'm guessing it's something like 11.8+really12.0 or something similar. [02:55] (Though that would be an awfully weird way for them to do things.) [02:55] arraybolt3: ii nvidia-cuda-dev:amd64 11.8.0-0lambda0.20.04.3 [02:56] ii nvidia-cuda-toolkit 11.8.0-0lambda0.20.04.3 [02:56] pycurious: Yeah. Probably they've botched a version number. As this is the package from Lambda Labs, Ubuntu has no control over it, and it's unsupported here. You may have to ask them why their package number is different than the shipped software version number. [02:56] Eickmeyer: when i add the repository, i get this message [02:56] This PPA will NOT currently receive Plasma 5.26 for Jammy 22.04, as this would break subsequent upgrades to Kinetic 22.10. [02:56] would this get in the way? [02:57] pinkergloop: Are you on 22.10? [02:57] Eickmeyer: yeah i am [02:57] pinkergloop: Then a message intended for users of 22.04 doesn't apply to you. [02:57] Eickmeyer: okay cool lol [02:58] pycurious: (It's possible they have a reason for the version number being the way it is, so maybe they didn't botch it, but certainly something is wrong or at least weird with their software and packaging.) [02:59] Eickmeyer: tysm [02:59] pinkergloop: yw [03:01] arraybolt3: Trying to find where that code is at their end. [03:05] arraybolt3: thanks for the help. Will look [03:48] hi all [03:48] ubuntu [03:48] :) [03:48] i am here [03:48] what do you want to ask [03:53] heh [04:15] Hello, does anyone know how large the entire old-releases.ubuntu.com is? [04:15] lflare, hard to find out, why? [04:17] even this *one directory* on the thing is going to be brutal http://old-releases.ubuntu.com/ubuntu/pool/main/l/linux/ [04:19] for the rest, recent topic https://askubuntu.com/questions/1448681/size-of-mirror-repositories [04:19] Working on mirroring the entire thing, hoping it's not over a 100 TiBs [04:20] I've already mirrored ubuntu and ubuntu-ports [04:20] useless to do old-releases [04:20] Datacenter I work with, still has certain VMs running very old releases [04:21] oke, then i take that datacenter not seriously [04:21] it'll never repay the costs of the download [04:21] good luck! [04:21] maybe setting up a squid-deb-proxy or similar would give more bang for the buck [04:22] I've thought before it'd be nice to have it all locally, but just can't possibly justify it :) [04:22] I mean, it's always annoying that a customer never wants to update their stuff [04:23] If it's not too ginormous, it's somewhat justifiable with the amount of free space we have lying around [04:23] the whole data preservation thing's also a nice touch I reckon [04:25] *nod* [04:25] I'm sure there's stuff on there I'd like to relive [05:13] hi there, i am looking at a ubuntu mirror like this one, http://babylon.cs.uh.edu/mirror-sites/ubuntu/ inside which directory can i find the installer ISO files ? [05:13] i got the list of mirrors from here, https://launchpad.net/ubuntu/+archivemirrors [05:15] tahts mirrors for the repos [05:15] the installations isos are not ther e? [05:16] okay found another link, https://launchpad.net/ubuntu/+cdmirrors [05:16] https://ubuntu-releases.mirror.net.in/ubuntu/releases/ [05:57] `foo | bar && echo OK` will print no matter what foo's exit code is. It only cares about `bar`. How can I make it care about foo? [06:00] thomas25: I can think of a really complex way to do that using background processes, FIFOs, and variables, but I can't help but think there's an easier way. [06:00] Nevertheless, if you want the crazy way... [06:00] I setfacl -R -m u:seedbox:rwx /home/rtorrent/Downloads but ftp user seedbox doesnt' have write permsisions without 775 [06:01] `{ foo && echo OK } | { bar && echo OK }`? [06:01] Yeah, that's a way easier way. [06:01] Though the "echo" might get in the way... [06:02] I think the "OK" will end up as part of the output to standard out that `bar` then accepts. [06:02] If there was some way to "echo" to standard error, that would work. [06:03] if you're using bash, $PIPESTATUS [06:04] I want to setup NFS shared drive - preferably encrypted - on multiple Ubuntu 20.04 boxes - can anyone point me to a working recipe for this? [06:06] lflare, thomas25: Maybe $({ foo && $(echo OK 1>&2) }) | bar && echo OK ? Haven't tested it, wow that looks messy. [06:06] pycurious: nfsv4 supports authentication with kerberos etc, there are online tutorials for this, but I wanted to mention you can also do plain old unencrypted nfsv3 and then protect/encrypt it with wireguard [06:07] alkisg: nfsv4 - how do i get it setup on 3 Ubuntu boxes and mount it? any pointers? [06:07] https://help.ubuntu.com/community/NFSv4Howto - is this the right link? [06:07] thomas25: Actually, I don't think that will work. Hrm... [06:07] pycurious: you can find tutorials for that, I haven't read any, I just wanted to mention the alternative [06:07] arraybolt3: https://unix.stackexchange.com/questions/14270/get-exit-status-of-process-thats-piped-to-another [06:08] thomas25: ^ [06:08] alkisg: That's better yet still. I was about to come up with this symbol-laden monstrosity :P [06:09] :D === xenial is now known as Guest3250 === five61846 is now known as five6184 === NeoFAT32 is now known as d4rk_b0y === tass5189 is now known as tass518 [08:38] A school kept firefox.snap instead of firefox.deb by mistake; now I did `snap remove firefox` and it's taking hours to "save data of snap firefox in automatic snapshot set #1" [08:38] I see that it creates a multi-GB file in /var/lib/snapd/snapshots; I guess it puts the snap firefox data of hundrends of student accounts in there? [08:39] Any way to remove snaps without that data saving? [09:04] Morning all ❤️❣️😘 [09:13] After waiting for 30 minutes, I decided to ctrl+c it (it crash-dumped), rm /home/*/snap and retry... let's see [09:31] :D [09:34] :-D [09:43] Hi all. Anybody know of an IRC channel for keydb? (They're a project created as alternative to Redis with threading) [09:43] Actually, any chat, not just irc [09:54] Fravialis: see /list if anything is useful [09:54] !alis | Fravialis [09:54] Fravialis: Alis is an IRC service to help you find channels. For help on using it, see «/msg Alis help list» or ask in #libera - Example usage: «/msg Alis list http» [09:55] also the project's homepage lists all ways of comms right on the front site [09:55] so... [09:56] Fravialis: https://lmddgtfy.net/?q=keydb [09:56] ok well, https://docs.keydb.dev/docs/ [09:57] Hei - I'm looking for a nice end-device mangement system with ubuntu, especially for notebooks. Apart from unattended setup (imho, this can mostly be done with PXE / unattended-setup / ansible), this also concerns some way of inventory management (who uses which notebook) and deploying user-based profiles (thunderbird, openvpn, desktop-links), etc. [09:57] I'm not looking for the all-incl. mdm solution for pushing software, etc.... I'm more interested in a less intrusive, more privacy approach, mostly focussing on initial setups. Any ideas? [09:58] you could bake your own image to install from [10:15] Piraty: ... I know that link and I'm not clicking it [10:15] I also know where the documentation is, and I know how to list a server. [10:15] Sometimes channels have unexpected or unorthodox names, and you can't find them this way. [10:16] Moreover, they could be on different networks. Now, I could start using irc server indexers and do a federated search over the entire irc space, guessing as I go along, or... [10:16] I could just ask. [10:31] since 22.04 until 22.10, whenever a new update exists, it is kept back, every time! so if there are conflicts, why does they release it at all? === DrPepperKid is now known as MacSlow [10:45] @ice9 it is probably staged releases. [10:45] look at the release notes for 22.04. [10:47] MekApelsin, what about 22.10? [10:47] "Phased updates in APT" [10:47] i haven't looked at the release notes off 22.10, i haven't used it. === pong_ is now known as beaver === halvors1 is now known as halvors === Rhvs is now known as Rhys [12:59] Hi, I started with mariadb-server and wonder that things are different from OpenSuse. Following https://wiki.ubuntuusers.de/MariaDB/, update mysql.user set password=password('root123') where user='root'; ERROR 1449 (HY000): The user specified as a definer ('mariadb.sys'@'localhost') does not exist [13:00] It's Ubuntu 20.04 [13:05] Hi all [13:06] is there a plan to rewrite ubuntu in rust? [13:08] alip Why on earth would anything in ubuntu be rewritten in rust? And to rewrite apt, dpkg, GNOME, and all that in rust is insanity. [13:08] sarnold: if you are interested i got the installer with my duplicate WWN disks by building my own with a patched curtin lib, as per https://github.com/eest/curtin/commit/1052baa37b81cd348e17d17a2ef545c93ce26505 [13:08] -ubottu:#ubuntu- Commit 1052baa in eest/curtin "Skip looking for wwn for testing" [13:08] *i got the installer to work [13:10] i am thinking maybe it should check for duplicates and skip to the next method if it exist [13:10] Though speaking of APT, I think nala is written in rust and it seems to be an alright alternative to APT. [13:12] because if ubuntu is rewritten in rust, it will be blazing-fast and secure. [13:13] Ubuntu is just a distro. What components of Ubuntu should be rewritten in rust? dpkg, apt, snapd? [13:13] all of them. a system is as secure as its weakest link. [13:14] That is completely crazy talk alip. There's nothing to gain from rewriting tried and true components like apt. [13:15] Besides, the only real component that can really use rust is the linux kernel itself. All apt does is fetch packages, dpkg installs the debs, GNOME is already huge and would be impossible to rewrite in rust. Xorg is on its way out, wayland is just a series of procols. I think you get the point. [13:20] alip, 98% of the source Ubuntu is built from are not under Ubuntus control but come from various upstrams via debian ... you'd have to ask these upstreams to switch ... [13:21] step by step all must be written in rust so they're blazing-fast and secure. [13:21] Is this trolling? I can't really tell. [13:21] well, your packages do not contain source but binaries ... performance is usually more a compiler thing [13:22] anyway, feel free to talk to the various upstreams ... this is rather offtopic for this channel ... [13:41] My touch screen is not working on Ubuntu Unity 22.10. It previously worked on 14.xx. Dmesg output is here: https://pastebin.com/WpY3wr88 === realivanjx3 is now known as realivanjx [13:54] how do you enable ipv6 dhcp on an ubuntu deployed through cloud-init? i added dhcp6: true in /etc/netplan/50-cloud-init.yaml, but it's not working [13:54] 'dhclient -6' does that fine [13:59] mei: you do realize that file is overwritten each reboot? [13:59] it doesn't here [13:59] Or rather, shutdown and new start. [13:59] and btw what's inside it [13:59] it's not anywhere else [13:59] mei: you merely have to read it [13:59] it's not into the cloud [13:59] What [14:00] I know how it works, I wrote an entire Python script to load and modify it if necessary at boot [14:00] It's explained to you in the header of the actual file [14:00] it's probably an hacked version [14:00] ??? [14:00] where i should look for instead? [14:00] You don't have many options [14:01] you could disable cloudinit's network configuration [14:01] it's not default configuration, so the data is coming from somewhere [14:02] I don't understand what you mean [14:02] All data comes from somewhere [14:02] i need to change that data [14:02] and you're saying it get wiped [14:02] but it come again the same [14:02] so where should i modify that data? [14:02] You can't. [14:02] what [14:02] Well you can, but it would be overwritten, read the file [14:02] You have read it, right? [14:03] You have read the comments at the top? [14:03] where is the datasource? [14:03] how do you configure how many linux kernel packages are kept? for example after the default install of lts [14:03] Cloud Init inserts this from its ISO data source which is read only [14:03] with unattended upgrades enabled [14:03] i mean, that data come there every reboot right [14:03] mncheckm: I'm not aware of such a setting, but it could exist [14:03] so when there is a reboot, where this data come from? [14:04] mncheckm: you'd look under /etc/apt/ of course [14:04] mei: Cloud Init inserts this from its ISO data source which is read only [14:04] Not a reboot necessarily, but a shutdown followed by a power on [14:04] Fravialis, thank you [14:04] mei: it's just a package. You can remove it with: sudo apt remove --purge cloud-init [14:04] well i see, but this is some hacked version [14:04] Fravialis, I have 500MiB for /boot [14:04] i'm pretty sure the data doesn't change every reboot [14:05] mncheckm: /etc/apt/apt.conf.d/01autoremove [14:05] mei: cloud-init configures your system at every boot [14:05] leftyfb: yeah but it uses this file that is not wiped [14:05] mei: if you don't want it doing that, remove it [14:05] mncheckm: I think that might be your best bet [14:05] i already changed the access point password and it worked [14:05] but dhcp6: true instead is not working [14:06] Do you understand what cloud init actually is? Do you even want to use it? [14:06] it's already there [14:06] You keep saying it's "hacked"... [14:06] well, after you explained me [14:06] that it's supposed to wipe [14:06] It's not. You're using an Ubuntu cloudinit version [14:06] Or you or somebody else installed it [14:06] So like leftyfb said, if you don't want it, remove it [14:08] If you do want it, accept that it does not work the way you want it to, and with respect to how that works, I've explained that cloudinit, at boot, reads its configured data from its own readonly "cloudinit drive" which is an ISO9660 file system [14:08] is there any configuration file that specific this iso thing? [14:08] so i can check [14:08] That can be regenerated when you change cloud init settings, but you'd have to generate your own custom ISO to make anything permanent [14:08] mei: do you want to use CloudInit? [14:08] mei: is your issue that setting dhcp6: true in your netplan file doesn't work at all or that the setting is being removed by cloud-init on every boot? [14:09] Yeah, that's another relevant question [14:09] adding dhcp6: true to netplan and reboot doesn't give me a dhcpv6 ip [14:09] the line stay there in the file [14:09] mei: what happens when you add it and run: sudo netplan apply ? [14:09] let me try [14:10] mei: it sounds like this might be unrelated to cloud-init [14:10] leftyfb: if he's rebooted his DHCP server, if any, might not even support ipv6 [14:10] dhcp6 is false by default [14:10] i already stated that dhclient -6 [14:10] give me an ip [14:10] I see [14:11] mei: ok, when you decide to focus and answer questions during troubleshooting, I'll be happy to help you [14:12] how did i not answer the questions? [14:12] 1. adding dhcp6: true to netplan and reboot doesn't give me a dhcpv6 ip 2. the line stay there in the file [14:12] mei: so far, you add "dhcp6: true" to your netplan file. You run "sudo netplan apply" what hapens? [14:12] i answered "let me try" [14:13] it hangs for a bit (im in through ssh) [14:13] everything go back working but no dhcpv6 ip [14:13] no output [14:13] mei: what version of ubuntu is this and what sort of device is this? [14:13] it's ubuntu server on raspberrypi [14:14] 22.04 [14:14] mei: ( uname -a ; cat /etc/os-relase ) | nc termbin.com 9999 [14:16] https://pastebin.com/raw/dB9Pa7L4 [14:16] Hi, trying to install ubuntu (server) from a live cd. Something is odd with my partition table so the installer crashes (python error using sfdisk). Q: Can I skip the partitioning/formatting step in the installer somehow? I managed to setup partitions manually with fdisk [14:16] mei: let me know when you run the command I gave you and please paste the outputting URL Here [14:17] well whatever [14:17] wasting time [14:20] Fravialis, thank you it looks interesting [14:21] it fails here, if that matters: https://github.com/canonical/curtin/blob/master/curtin/commands/block_meta.py#L801-L808 === dionysus70 is now known as dionysus69 [14:48] greetings. DOes anyone know a gnome app that shows cpu usage on the top bar? [14:48] DynamiteDan: https://extensions.gnome.org/ search for an extension you like here https://extensions.gnome.org/ [14:49] thanks leftyfb ! [14:58] I have a 49" screen on ubuntu 22.04. I've been searching a good window tiling manager to split the screen into 3 sectors, but I didn't find any good one. Does anyone know any good window tiling manager for ubuntu 22.04? [15:02] Angs: I;ve heard good things about the i3-wm ...never used it myself though [15:04] I used i3, but it worked terrible. Those tools change the Desktop management. I like Ubuntu 22.04's default desktop style and convenient to use. I am looking for a tool that uses the same Desktop UI that I have now [15:05] hello. without apparen reason, when I executed the command `python3 -m pip install --upgrade git+https://github.com/ProgVal/Limnoria.git@testing` I started getting some "X11 connection rejected because of wrong authentication.". any ideas? [15:06] for example, in windows, samsung has "easy setting box" windows tiling manager. I have the same Desktop UI and it just splits the screen virtually. I am looking for a similar solution if possible [15:06] Angs: check the package manager, there are several options listed there [15:08] I don't remember aditing anything on this particular machine regarding X11 or whatever [15:08] s/aditing/editing/ [15:22] eventually i got an ipv6 after a while; journalctl -u systemd-networkd and netplan --debug helped. i don't know it requires some time and my idea is that it's not forcing dhcpv6 but when the router announce that dhcpv6 is available it get honored. while dhclient -6 is instant since it's forcing [15:25] Angs, https://extensions.gnome.org/ ... search for tiling ... === tscott_ is now known as tscott [15:52] leftyfb: i think the trick was to 'netplan apply' before reboot. i expected that a reboot would reload that conf instead you need to commit it first somehow. and it will eventually work after reboot. [15:52] the last chapter there confirm that: https://ubuntu.com/tutorials/how-to-install-ubuntu-on-your-raspberry-pi#4-boot-ubuntu-server [15:53] mei: a reboot will do everything that netplan apply does. netplan apply is not required before rebooting and doing so will serve no purpose [15:54] so that page is wrong? [15:54] okay let me test that from scratch just for the sake of discovery [15:54] that page does not say running netplan apply before rebooting is required. It does mention doing so, but does not clarify that it is required. It's not [15:55] From the man page: netplan-apply - apply configuration from netplan YAML files to a running system [15:55] maybe it could be related to the fact thta cloud-init seems used in a hacky way [15:55] that* === beaver is now known as Guest01101 [15:56] mei: you seem to have a misunderstanding and dislike of cloud-init. I highly suggest you remove the package [15:56] why would i remove the package that bring up my wifi card [15:57] mei: see my first point [15:57] mei, that's not usually cloud-init's job [15:57] i don't like or dislike cloud-init. it's just there [15:58] mei: cloud-init does not "bring up" your wifi card. cloud-init is a configuration management tool. It will overwrite network configs but it's absence will not remove your configs [15:59] mei: I would highly recommend you remove cloud-init to keep your overall system and it's management simple [15:59] all the window tiling managers are actually window manager and each of them have their own desktop environment. Isn't it possible to just install an auto window tiling tool that works on the background and compatible with Ubuntu 22.04's default window manager? [16:00] by hacky way i mean that it doesn't overwrite netplan file [16:00] as i was pointed out previously so many times [16:00] because you didn't tell it to do so [16:01] what? [16:01] mei: did you configure cloud-init with your own custom network config? [16:01] i customized the netplan file [16:01] that's not cloud-init [16:01] i never touched cloud-init [16:02] mei: sudo apt remove --purge cloud-init [16:02] well i figure out how to configure netplan by cloud-init documentation... [16:02] that's why i mentioned it [16:02] mei: to maybe make you feel better, I remove cloud-init on all of my machines by default [16:02] when I right click and select "open in terminal", nautilus starts gnome-shell. Does anyone know what do I need to change in order nautilus to start terminator instead of gnome-terminal? [16:03] anyway, i did everything from scratch. and a reboot was enough. i don't know why the first time it was not working [16:03] Angs: open settings -> default applications [16:05] leffyfb, ubuntu 22.04 only has Web, Mail, Calendar, Music, Video, Photos. It doesn't have default for terminal [16:05] how is cloud-init affecting me anyway? we even discovered it's not doing anything to netplan [16:05] mei: a misunderstanding of how your system runs and configs [16:06] Angs: sorry: sudo update-alternatives --config x-terminal-emulator [16:07] lefftyfb, I already did this, but it still starts gnome-terminal [16:07] Angs: what comes up when you hit "CTRL+ALT+T" ? [16:07] terminator [16:07] Angs: you might have to log out and back in for nautilus to use the new default [16:07] I think something needs to be changed in nautilus config, I couldn't what to change [16:08] Angs: log out and back in or reboot [16:08] I did sudo update-alternatives --config x-terminal-emulator weeks ago === Rhys is now known as Guest5166 [16:08] ok, not sure then [16:08] pretty sure nautilus doesn't have a config for this sort of thing [16:09] Hi all, i wonder if there is a process monitoring tool for ubuntu which evaluates the data in /proc/ My M2 (NGFF) drive is showing a temperature of "50" but I think it's supposed to be reported as milliC , not C (like all the other thermal devices). Is this a kernel issue or something with the drive? https://dpaste.org/EHnS0 [16:11] Angs: https://unix.stackexchange.com/a/547749 [16:11] how can I find the pidfile for smartd? [16:11] FKAShinobi: pidof smartd [16:11] sweet! thanks leftfb! [16:24] leftyfb, I get E: Unable to locate package filemanager-actions-nautilus-extension I guess that package is no longer available on ubuntu 22.04 I tried some many other forum posts, no options were available on ubuntu 22.04 [16:42] btw, my msteams/linux problem (on both an lubuntu box and a fedora 35 box) in which no input devices (microphones) were recognized by msteams is apparently a known bug. the fix is to 1) exit msteams, 2) rm -fR ~/.config/Microsoft [16:49] Angs, i'd try this : https://www.omgubuntu.co.uk/2020/07/open-folder-in-terminal-ubuntu-plugin [16:50] Angs, just change python-nautilus to python3-nautilus [17:03] ioria, perfect! thank you :) [17:03] ok [17:09] Angs, finally, remove nautilus-extension-gnome-terminal [17:10] ioria, now even better. Thanks. [17:10] ok [17:20] hi! I have an issue with kvm in ubuntu ,even though I am pretty sure hvm is enabled, virtmgr gives me the following error: ValueError: Host does not support virtualization type 'hvm' for architecture 'x86_64', not sure but I think it was working till I upgraded to ubuntu 22.10 , but that might not be the case, I have not used it in a while, I reinstalled all the packages, and my windows VM does appear to be running ( the [17:20] graph shows it is running. I just can't "open" it. [17:21] is there a known issue with ubutu 22.10? [17:48] transhumanist_: is it turned on in the firmware? [17:49] yes it is , getting these errors: just a second will post [17:49] https://bpa.st/TZHEE [17:49] transhumanist_: what does 'kvm-ok' say? [17:50] exists, can be used (paraphrase) [17:50] its running a vm, just wont launch the display [17:50] that is can't "open" [17:50] weird [17:51] indeed, I smell a bug, or some weird config problem, I was hoping someone might have heard something about it [17:53] ah ... https://github.com/virt-manager/virt-manager/issues/234 [17:53] -ubottu:#ubuntu- Issue 234 in virt-manager/virt-manager "host doesn't support requested feature MSR(48DH).vmx-posted-intr [bit 7]" [Closed] [17:55] problem is if this error is caused by pci-passthrough, in my case I do not have this much wanted capability [17:55] since it seems like not a ubuntu issue I will ask in #kvm instead [17:56] Are n-trig touchscreen drivers still part of ubuntu base install or would they have to be installed seperately? [18:03] Hi all, I have a chicken and egg question: just installed ubuntu 22, want to install some other stuff but I can't "sudo xyz" because "blahblah user is not in the sudoers file", and I cant "adduser blahblah sudo" because "only root may add a user to the system", and I can't "sudo adduser blahblah sudo" because "blahblah is not in the sudoers file". How do I make it possible to sudo with the user blahblah that I created during install? [18:05] cjard: login with the user account to setup when you instaled Ubuntu. That account is definitely in the sudoers file unless you changed it [18:05] !yy.mm | cjard [18:05] cjard: Ubuntu version numbers are: YY.MM (YY=release year,MM=release month). Each year sees two releases, so just specifying YY is imprecise. See also https://www.ubuntu.com/about/release-cycle [18:06] one sec, i'll refresh my memory of what I did to install in virtualbox [18:07] cjard: getent group sudo [18:07] This will tell you the sudo users [18:08] cjard: or, you can run `pkexec bash` and it will pop up a GUI with the users that can do that [18:09] response to getent group sudo is: sudo:x:27: [18:09] response to pkexec is pop a ui dialog asking for password for "adminsitrator" [18:09] I don't know it [18:10] cjard: did you install this VM using an official Ubuntu installer .iso or was it an image you downloaded? [18:11] install process was: make new VirtualBox VM, Choose downloaded iso (ubuntu_22.04.1_desktop_amd64.iso from ubuntu website) and a place for the file to be saved, next screen pick a user name for virtual box (i chose vb/vb as user/pass), then rest of install was unattended i think [18:11] hang on, i'll pull up the link for the iso [18:12] it was downloaded from: https://mirror.freethought-internet.co.uk/ubuntu-releases/22.04.1/ubuntu-22.04.1-desktop-amd64.iso [18:13] cjard: I can't tell you what went wrong, but a new install would probably be the easiest solution [18:13] If you used "vb" as the user, and you got "administrator", something is very wrong, I'd reinstall [18:13] ok, will give it a go! [18:14] (the user i'm logged on with right now is "vb" - it shows on the login screen etc) [18:14] But it's not part of the sudo group, which isn't something the Ubuntu installer would do [18:15] It's possible that the vbox "preseed" stuff is broken [18:15] When you try to reinstall, ignore that vbox autoinstall stuff, leave it empty, and use the ubuntu installer to provide the information [18:15] should i "skip unattended install" and attend it instead, so installer asks me for eg usernames during setup? [18:16] Right [18:16] has it ever happened to you that unpacking linux-hwe-headers stage takes a lot of time [18:16] Because it's possible that this was caused by vbox misusing the preseed stuff [18:16] it does seem this time it took several minutes [18:16] I don't remember it being so slow during previous upgrade [18:16] I don't remember it being so slow during previous upgrades [18:17] pikapika: it can take a while, especially on slower processors [18:17] the computer is the same [18:18] I never noticed it being this slow any previous upgrades [18:18] I understand of course that compression/decompression can be a very cpu heavy process [18:18] but that begs the question, is this file really that huge this upgrade? [18:19] pikapika: which cpu is that? Also check your free ram, and the output of `top` [18:20] its a 6th gen i3 but as I said previous upgrades on the same machine didn't seem to take that long in this step [18:20] I had 80% or so free ram [18:20] I tend to close everything else when upgrading [18:20] OK check the output of top, some background service might be using the cpu [18:21] There's not a big difference in the size of kernels, no [18:21] cpu was also under 20-30% but I can't check anymore since the upgrade is finished [18:21] And top or ps faux will tell you the running processes, to pinpoint which one stalls [18:21] yeah thats what I was thinking [18:21] there shouldn't be much difference in kernel sizes [18:22] so I was wondering what went funny this time [18:22] It was at unpacking, not at update-initramfs, right? [18:23] yeah [18:23] unpacking [18:23] Also check dmesg for disk issues (e.g. hitting a bad sector may cause big delays) [18:23] (or other hardware issues, which will appear in dmesg) [18:25] I have a feeling it might be something funny about the region of disk it touched [18:26] I'll have to see whether thats true [18:30] pikapika, I have certainly seen unpacking a kernel header package take a long time, even on very fast processors. It has to create and remove a lot of small files, which turns out to be a slow operation on many file systems [18:32] oh right [18:32] It would be slow in previous updates on the same system too though... pikapika said that wasn't the case [18:32] if it involves deflating thousands of small .h files then I can understand it taking time [18:32] but yes also [18:32] as I said [18:32] this instance seems to take longer [18:33] I almost thought it hanged there [18:33] I can't be sure how much worse it is than previous since I don't closely watch and measure it unless something is weird [18:34] I don [18:34] but it certainly felt this upgrade took rather long to unpack [18:34] I don't know why it isn't always slow, it must have something to do with the state of the file system (more/less free space fragmentation, maybe?) [18:35] the free space is certainly a bit low this time [18:35] but I did make sure there was comfortable space for the upgrade [18:53] Can xorg be setup on Ubuntu Unity? [18:54] i would expect unity to depend on xorg [18:55]