[00:06] how do i update linux-libc-dev in ubuntu 20.04 - it looks like it has an old Critical vulnerability? (4.19.181-1 - I am on this I think) [00:07] pycurious: which vulnerability? [00:09] pycurious: are you really running a two year old kernel from debian? https://launchpad.net/debian/+source/linux/4.19.181-1 [00:09] pycurious: if so, there's dozens of way more important issues than whatever is in linux-libc-dev you gotta fix [00:31] https://packages.ubuntu.com/bionic-updates/linux-libc-dev === wolfdale8331 is now known as wolfdale833 === chris14_ is now known as chris14 === shokohsc9 is now known as shokohsc === jason1 is now known as JasonRipper === xenial is now known as Guest5109 === xenial is now known as Guest2803 [03:18] how do i set my ipv6 address static within the LAN? [03:19] on gnome, network manager, add a new profile [03:20] or in your router, lease reserved for your mac address [03:20] oh ok [03:47] Hi I have an Android question but I cant join #android please help. some app is using my Camera, how do I find which one did it? [03:47] I restricted my allowed apps with camera acess to just Camera app, but it still happens [03:47] I get the Camera On notification [03:47] every time I turn my phone on [03:48] lol [03:53] Why is that funny [03:53] `/join #android` [03:54] guest2: because this channel has 0% to do with android [03:56] you might need to register with Libera === gnuself1 is now known as gnuself === elastic_dog is now known as Guest6188 [05:24] is anybody here good w/lxc containers? [05:25] Muligan: Maybe. If you ask your question, someone might know the answer and be able to help. [05:25] (I use LXC some but I'm not exactly an expert at it :P) [05:25] it's an oddball setup [05:25] i'm simply trying to figure out this system [05:25] there's very minimal documentation that really dives into it [05:26] it's an 'Antman' system by Antsle [05:26] so far, I've been able to migrate the KVM's off into my vSphere cluster, but having an issue w/these LXC's [05:27] That issue being? [05:27] from what multiple lxc commands that I've run, they've yielded as NOT running [05:27] Also is this system running an official Ubuntu installation? [05:28] however the webui of this unit is running, and the two LXC's that are running, the data/webui's to those LXC's are reachable [05:28] it's a Ubiquiti AirController [05:28] (Official as in, you downloaded an image from an official Ubuntu website and used that to install Ubuntu on the device.) [05:28] along with Ubiquiti Mfi [05:28] arraybolt3, no [05:28] Muligan: To be clear, is this Ubuntu at all? And if so, is this an Ubuntu installation that came with the device? [05:29] unfortunately, not [05:29] You might have better luck in #lxc then. [05:29] (This channel only supports official Ubuntu operating systems.) [05:29] coo [05:29] ty [05:29] However the #lxc channel supports LXC in general. [05:29] Glad to help! [05:31] arraybolt3, sad thing is, this box is way old [05:31] outdated, etc. [05:32] some of the converted KVM's I'll end up rebuilding altogether [05:32] Muligan: Hmm... re-reading your questions, you're trying to migrate these containers off of where they are onto an Ubuntu system? [05:32] eventually [05:32] Ah, I get it. [05:32] i'm simply trying to access the LXC environment in order to get in to reset the DB passwd [05:33] Can you use "lxc shell" to get into the containers? [05:33] Possibly using "lxc start" to get it running? [05:33] I couldn't even get into the console of the hosted KVMs, I wound up shutting them down, converting the qcow2 vdisks on over to vmdk, and rebuilt them using the vmdk's on my vsphere environment [05:34] the basic command 'lxc' doesn't exist [05:34] Oh yikes. [05:34] lxc-ls lists 'alpha, beta, and gamma' [05:34] then when I go to get info of each one of those, it shows them as not running [05:34] That's probably something we can't help with then. We don't really know how odd devices like that work. [05:34] Though if "lxc-ls" exists, "lxc-shell" might exist? [05:35] I understand, just looking for some direction [05:35] 1st time really dealing with lxc's at all [05:35] And maybe "lxc-start", too. [05:35] lxc-shell doesn't even exist [05:35] (There actually is an "ls" subcommand in normal "lxc", so that's why I'm thinking... oh, tar, ok nevermind) [05:36] lxc-start might though. [05:36] Muligan: Try "lxc-console" then. [05:36] https://pasteboard.co/btet3JDlH2r8.png [05:36] Looks like these "lxc-*" commands are part of the lxc-utils package, and there's an "lxc-console" command in there. [05:37] Looks like it's on your system too. [05:39] https://pasteboard.co/HmTl0jU5Sg8W.png [05:40] Honestly, I'm lead to believe that this system is just jacked [05:41] i mean, it's running 0.4.0b of Antman, and there clear up to 4.0 now for the release [05:41] Muligan: Try using lxc-start then? [05:41] But yeah, with something this old, it may be easier (possibly *far* easier) to start from scratch. [05:41] https://pasteboard.co/ubtXgvjqrAGI.png [05:42] I will start from scratch, but I'm trying to get in and set some commands to default/reset the webui passwd for this AirControl program by Ubiquiti [05:42] Now maybe it will let you use lxc-console? [05:45] i'm not sure which option of lxc-control to issue [05:47] lxc-ls --active -1 spits out nothing [05:48] Yeah then I'm not sure either. [05:49] hell, even lxc-ls --fancy isn't even supported in this version === bad_1 is now known as badbodh [05:50] use mate it rocks. [05:50] huh? [05:50] !ot | propus [05:50] propus: #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! [05:52] ubottu: okey [05:53] Muligan: I'd still try #lxc. Yeah, your equipment might be ancient, old, and EOL, but they'll have a better chance at knowing what to do than we will, I would guess. [05:54] either way, thanks for the attempt [06:53] good morning === realivanjx7 is now known as realivanjx === realivanjx4 is now known as realivanjx [07:48] Greetings everyone. Is there any way to grab messages from 'dmesg' which are shown in red colour? [07:55] hid3: The color is applied to the message from a text element (ie 'ERROR' in message) in the first place. To get that out, you would do "dmesg | grep ERROR" [07:55] If it's not specifically ERROR: Find the common text element in the message and grep for that text element. === realivanjx0 is now known as realivanjx [10:32] can i create an user with the HOME in a different location? /var/www/HOME_NAME ? [10:42] !home | pydev [10:42] pydev: Your home directory is where all of your personal files are usually kept. For moving your home directory to a separate partition, please see: https://help.ubuntu.com/community/Partitioning/Home/Moving [10:43] pydev: sudo useradd [10:43] -d, --home-dir HOME_DIR home directory of the new account === catties is now known as commies === commies is now known as catties [11:10] hi === blackest_mamba_ is now known as blackest_mamba [11:36] anyone here who considers that faster bugfixes in non-LTS releases are better than staying with an LTS for two years? Is that too optimistic toward non-LTS releases? === realivanjx6 is now known as realivanjx [11:42] clear [11:42] hsiktas[m]: LTS is strictly to put off updating. Not putting off updating is always better if time allows [11:43] sound reasonable... [11:43] soudns [11:43] sounds :) [11:44] sadly plenty of organizations are so buried in nonsense, five years and it can still be hard to find time to update [11:44] everything depends on your system being maintained, though, so, gotta [11:45] hsiktas[m]: depends on what tools you use. LTS is fine for main and restricted, but most programs come from universe and are not or less supported. its a very individual task to check if lts suits you. [11:49] it would be nice if universe would be somehow decoupled from the other repos. Like provide the usual support duration for main/restricted but all supported releases point to a common universe repo that gets regular imports from Debian. (basically similar to how each supported FreeBSD release points to the same package versions in their ports tree) [11:51] that won't be possible as libraries change and some tools have to be recompiled. But thats what Snap is for. Same versions across all ubuntus. [11:53] same nonsense version =P [11:53] BUT same version :D [11:53] as far as you know [11:55] yep. I just assume it as I do not use snap or ubuntu at all ;) [11:57] I don't either, but there's nothing good about snap [12:00] Depends. Allowing decentralized servers (or stores if you want) would be great. Sandboxing itself is a good approach considering users being allergic to read stuff about users, groups, permissions,… [12:01] sandboxing is critical given that ever snap is a block box from upstream [12:01] every* [12:05] sad but true. But everyone has to trust somebody as we cannot read all source codes ourselves. Canonical is probably not the best to trust in, but there are worse. If you look what applications are requested and used, it is a minor issue. [12:05] do you think? [12:05] about this? [12:06] jk [12:06] well, mostly you're trusting debian [12:07] I do trust debian/main. Mostly. But thats far from being an ubuntu installation. === hasley is now known as mahler [12:14] Until a few years ago i didnt know the store was closed source ill admit. Is it not possible yet to set ones own store location yet? [12:19] vlm: You can technically if you use the GPL3-snap-client-stuff and write your own server application. But its easier to take other approaches like flatpak and host a custom repository for it. [12:24] easier to avoid both snaps and flatpak =P [12:26] true. Just use a free/libre distribution, read the basics and you are done. [12:27] koffeinfriedhof: you're the one I like [12:28] Well, I don't. Its awful to search for new hardware nowadays and very frustrating. Intel ME, AMD PSP, graphics and wireless stuff only with blobs… Live could be easier ;) [12:29] f'sho [12:29] meanwhile the world keeps cranking out young adults who don't even know what privacy is, so they can't even miss it or wonder how to get it [12:31] "Privacy? Dunno. Is there a facebook group?" [12:32] Thou shalt not curse. :P [12:39] O:‑) [12:46] koffeinfriedhof: good one, thanks [12:49] Hi all [12:49] Hi BluesKaj [12:49] hi koffeinfriedhof [12:49] hi all [12:49] hi just cool people who are cool [12:50] lmao [13:45] why do i now need a computer reboot for group memberships made with usermod to update? a simple logout/login is not enough. what changed? [13:46] sudo usermod -a -G group username === NatSocSiD` is now known as NatSocSiD [13:48] akik: you don't. A complete log out and back in will suffice [13:49] leftyfb: i just tried it on ubuntu 22.10 and it's not enough [13:51] akik: You're special? [13:51] jhutchins: what the fuck does that mean? [13:51] jhutchins: lets not [13:51] !language | akik [13:51] akik: Please avoid any language that may be considered offensive, including acronyms and obfuscation of such - also see https://wiki.ubuntu.com/IRC/Guidelines || The main channels are English only, for other languages, see https://wiki.ubuntu.com/IRC/ChannelList [13:51] akik: do you have error messages? [13:51] yea keep the guy in chains because he can't behave [13:52] akik: I mean that the behavior is not common to most Linux systems, including Ubuntu. There's something different about how you're doing it. [13:52] leftyfb: no error message, just that the group membership doesn't change until a reboot [13:52] akik: this a desktop or server? [13:52] leftyfb: it's desktop installation [13:52] akik: Is your hardware/storage unusual? [13:52] akik: and did you log out of the desktop or just close your terminal? [13:53] jhutchins: oh you're civil now? [13:53] please stay on topic [13:53] akik: I usually am, although occasionally sarcastic. [13:53] this is just a normal laptop with ubuntu 22.10/gnome [13:54] leftyfb: i logged out and logged back in. please test yourself adding a new group for your account with usermod [13:54] akik: I can't log out of my desktop at the moment, a bit busy [13:54] akik: If you look at the shadow files before you reboot, does the change show there? What GUI are you using? [13:55] jhutchins: i'm now in windows, but i can try again in ubuntu. using plasma 5 [13:55] oops [13:55] using gnome [13:55] jhutchins: shadow files have nothing to do with groups [13:56] akik: did you add a user to a group or did you change the main group the user identifies as? [13:57] leftyfb: you're asking questions i already gave info [13:57] 16:46 < akik> sudo usermod -a -G group username [13:57] leftyfb: The collection of password files that control user and group identification are often collectively referred to as shadow files. I believe there's a document, probably at tldp.org, that explains migrating from a pre-shadow to a shadow system. [13:57] Hello... I just did an unattended installation of Ubuntu on VirtualBox as per the documentation on ubuntu.com.  The user it created seems to not be in the sudoers file. [13:58] Before a kernel update (I'm not sure which one, but I may be able to work it out given time), my bluetooth headphones won't connect automatically when I switch them on without me running bluetoothctl connect "$hwaddr". Does anyone know why? [13:58] *since, not before [13:58] Is there a way to fix it? [13:59] I'm on 20.04 with all updates [13:59] ELFrederich: the unattended installation of Vbox is broken. Do a manual installation. === shokohsc8 is now known as shokohsc [14:00] leftyfb: this same problem happens on arch linux/plasma 5 [14:00] ravage, okay... I should delete this one and start over?  I haven't done anything yet, other than manually adding myself to "sudo" group. [14:00] clarkk: Which kernel? [14:00] ELFrederich: yea [14:01] .... because the root password is set to the same as my user's password.  That also seems like a bug.  Shouldn't most installs not even have a root password at all? [14:01] akik: It sounds strangely like your changes aren't being written to the files. [14:01] akik: try newgrp group [14:01] jhutchins: that isn't the problem [14:01] leftyfb: Great, you solve it then. [14:01] jhutchins: i checked and /etc/group was modified [14:01] leftyfb: i know newgrp command but this is not about it [14:01] akik: Thanks for confirming that. Assumptions often lead to errrors. [14:02] akik: my guess is, even logging out of the desktop isn't enough, maybe try restarting GDM as well. Maybe there's still some process under your username running [14:02] newgrp is problematic that you need to run it then in every shell [14:03] can somebody else test this? [14:03] pick a group from /etc/group that you're not part of (id, groups commands), then sudo usermod -a -G group your_username, logout & login, check id or groups command [14:04] akik: but newgro does work for the terminal right? That means the changes have been made successfully but we're not "logging out" sufficiently [14:04] leftyfb: i didn't test it because it's not a solution i want to use [14:04] akik: please test it [14:04] ok [14:07] leftyfb: yes newgrp works, but it changes the primary group [14:08] leftyfb: and then i'd need to run it in each shell [14:08] should i logout with "kill -9 -1" ? :) [14:09] "If the optional - flag is given, the user's environment will be reinitialized as though the user had logged in, otherwise the current environment, including current working directory, remains unchanged." [14:09] not sure if that'll help the 2nd issue [14:09] leftyfb: if that's for su i know about it but it doesn't apply [14:09] it's not, it's for newgrp [14:09] oh you mean i should test it now? [14:10] su - aki worked and the gamemode group is now visible in that shell [14:10] right, that's a direct login [14:10] leftyfb: newgrp without the - also worked [14:10] as opposed to spinning up a vterm [14:11] so it's working, it's just the DE not being able to do a full log out of the user [14:11] i'll try kill -9 -1 [14:11] that won't work [14:11] sudo systemctl restart gdm [14:12] kill -9 -1 kills all your processes [14:12] that fixed it, even though it's a bit brutal [14:13] now gamemode group is automatically in all shells [14:14] I would try restarting gdm [14:14] leftyfb: don't you feel that's kind of strange thing to do? [14:15] akik: it's troubleshooting [14:15] arch has this same problem [14:15] and i use sddm on it [14:16] and plasma [14:16] akik: you're using ssdm on ubuntu? [14:16] leftyfb: sddm on arch [14:17] leftyfb: i think this is gdm on ubuntu [14:17] akik: restart gdm on ubuntu [14:17] yes [14:17] ok [14:17] Hello everyone [14:17] hi [14:19] I have a problem with gpg IDK why.... , in ubuntu 22, when I do: ""sudo gpg --dearmor oracle_vbox_2016.asc --yes --output /usr/share/keyrings/oracle-virtualbox-2016.gpg" [14:19] akik: btw, I just powered on another laptop I have running Ubuntu 22.04. Adding my only user to a group, logging out of the desktop and back in worked just fine [14:19] leftyfb: sudo systemctl restart gdm didn't fix it, but again kill -9 -1 did [14:19] it's extremmely slow, and show nothing... [14:19] someone have an idea why ? I didn't have this problem in ubuntu 20... [14:19] akik: maybe you have some service running in the background as your user [14:20] leftyfb: maybe something changed in 22.10 [14:20] akik: unlikely [14:20] leftyfb: i don't have anything custom on this laptop [14:20] leftyfb: you used usermod? [14:20] yes [14:20] I did exactly as you did [14:21] leftyfb: and it's shown in id now? [14:21] yes [14:21] after logging out of the desktop and back in [14:21] ok so something is different [14:21] no idea what [14:21] I also did a test right before logging in, at the login page I switched to a different tty, logged in and ran id, it showed the new group appended to the list of my groups [14:28] newgrp works fine after logout/login but the group i added is not by default in id output [14:38] With Ubuntu guest under VirtualBox should I use the Guest Additions CD or should I use open-vm-tools? [14:39] ELFrederich: that is for you to decide based on what each gives you in functionality and what your needs and preferences are. [14:39] I saw this... https://linuxconfig.org/virtualbox-install-guest-additions-on-ubuntu-22-04-lts-jammy-jellyfish [14:40] what is this "multiverse" [14:40] !multiverse | ELFrederich [14:40] ELFrederich: The packages in Ubuntu are divided into several sections. More information at https://help.ubuntu.com/community/Repositories - See https://wiki.ubuntu.com/RecommendedSources for the recommended way to set up your repositories. [14:44] well that installation of guest additions didn't seem to do anything.  I see it under lsmod but I can't resize my window [14:47] thank you so i will add --home-dir to change the default home of users [14:47] those are users that i use for websites [14:47] so i will point their home to /var/www/home1 /var/www/home2 /var/www/home3 [14:47] can i do that? [14:47] i think that /var/www has been created by "www-data" [14:48] shmm i do not thing i can create nested homes :D [14:48] pydev: you can, but I wouldn't. Just set the virtualhost Directory for each site to /home//WWW [14:49] ok [14:49] yes better [14:49] thank you [14:49] lat thing i need those users with their home but i do not need to login with those users into the system [14:49] should i add /sbin/nologin ? [14:50] yes [15:06] jhutchins, sorry for the delay. It broke some time in the last 3 kernels. I'm currently on 5.15.0-69-generic #76~20.04.1-Ubuntu SMP. I'd have to do some investigation to find out exactly which one. What would be your course of action, specifically? [15:14] Since a kernel update, my bluetooth headphones won't connect automatically when I switch them on, without me running bluetoothctl connect "$hwaddr". I'm currently on 5.15.0-69-generic #76~20.04.1-Ubuntu SMP. Kernel 5.15.0-67 also had the issue. I have two others installed - 58 and 60, but haven't yet isolated which one it worked in. Can anyone offer any suggestions, please? What steps should I take to either resolve it, or [15:14] where should I raise the issue? [15:18] Hola. === PeGaSuS is now known as irc_m === irc_m is now known as PeGaSuS === enigma is now known as enigma9o7 [15:56] Hola folks. Wondering if anybody is familiar with this. Trying to access data from a weirdo disk image (Android-based image).. So far it's not mounting under any standard tool. All I have managed to identify is that the fs is f2fs. I've used f2fs-tools then, but also no dice. The offset is basically unknown, and I've even looked at trying to just try them all (1-10,000) which also didn't work. Any ideas? [15:56] I should additionally note that testdisk doesn't appear to have found anything useful either :( [15:57] TahcRahc: maybe try #linux or #android [15:59] hello! i'm on ubuntu 20.04 on my laptop and i dont have the posibility to choose HDMI output for the sound i'm only getting built in auido. I'm using the driver for amd kabini r5 driver that ollows with ubuntu? [15:59] it's the same on 22.04 and 22.10 [16:01] how should I add a search domain to /etc/resolv.conf? [16:02] ELFrederich: you do that on your router DHCP settings [16:03] leftyfb I'm set to Automatic [16:03] ELFrederich: on your router [16:05] leftyfb, #android isn't as active, but I'll check, thnx [16:05] okay... what happened is this.  I have two Ubuntu VMs that I wanted to talk to each other on VirtualBox.  So I created a new NAT network.  When I put both of them on the same network they can communicate.  When they were on the default NAT they had a "search foo.acme.com" but when I put them on the newly created NAT Network they get "search [16:05] bar.acme.com" [16:06] any pulseaudio gurus in here? [16:06] having sound problems with a macbook 12" retina 2017 [16:07] ELFrederich: that would be a setting in virtualbox for the network you created. [16:08] ELFrederich: in this case, virtualbox is your "router" [16:08] leftyfb, yeah... no real setting there.  Don't even see an advanced option [16:08] I just choose the CIDR, that's it [16:09] ELFrederich: maybe hit up #virtualbox [16:09] leftyfb yeah... I'm asking there. [16:09] leftyfb but is there not a mechanism for me to add a line to /etc/resolv.conf in addition to what is coming from dhcp? [16:15] ELFrederich, there should be a place to put additional search domains in the network settings (or netplan if these are Ubuntu Server VMs). Don't try to edit resolv.conf. [16:16] ELFrederich: https://askubuntu.com/a/1211705 [16:18] hi [16:18] im new [16:19] idk what to do [16:19] :) [16:19] hfhzieofghg [16:19] g* [16:19] bn [16:19] y [16:23] uh [16:23] so what is this thing [16:23] or idk what [16:23] what is libera chat [16:24] effumate: this is an Ubuntu support channel. Feel free to /join #libera for help with Libera or other IRC questions [16:24] oh ok === xenial is now known as Guest7981 === halvors1 is now known as halvors [17:11] hello again [17:11] i am reading the useradd doc [17:11] i see -r = create a system account [17:11] what is the difference between a normal user and a system user ? [17:15] pydev: they're defined in /etc/login.defs [17:15] pydev: https://www.cyberithub.com/system-users-and-human-users-in-linux-explained-with-examples/ [17:15] pydev: it's just a id range [17:16] pydev: search for "System accounts" [17:17] oh i read that system users have no home [17:17] i need the home directory [17:17] so it is not system user [17:19] pydev: note that ubuntu is a debian derivate, and uses the same uid ranges as debian. to make sure you apply debian-like standards, i recommend working with the frontends (debian and) ubuntu provides for this purpose: adduser and addgroup - and to refer to their manual pages, too. [17:19] tomreyn, ok [17:20] but basically i need users to run my web projects but they do not need to login [17:20] so i wonder if i need to create normal users or system users [17:22] actually, useradd also follows login.defs, which defines the proper ranges, so the choice between useradd and adduser is just a matter of convenience / preference, i guess. sorry. [17:26] tomreyn, i have a doubt i have created an user with [17:26] useradd -M -N -r -s /sbin/nologin -g www-data myuser [17:27] myuser:x:997:33::/home/myuser:/sbin/nologin [17:27] why /home/myuser ?? [17:28] i'd say: use system users to run services or for any technical purpose which is not tied to an existing person (or group of persons, who just need a single login). but the line can be drawn differently by some tools and for some purposes and workloads. === PeGaSuS is now known as irc_m [17:28] what did you expect the home directory to be? [17:29] pydev: use -d option for that [17:29] tomreyn, i am using -M that says do not create the user's home directory [17:29] is the home mandatory ? why has been set to /home/myuser [17:30] pydev: if it doesn't have a home dir, it probably starts in / i think [17:31] akik, in /etc/passwd i see : myuser:x:997:33::/home/myuser:/sbin/nologin === irc_m is now known as PeGaSuS [17:31] pydev: i don't know about -M [17:31] -M, --no-create-home do not create the user's home directory [17:32] pydev: maybe it's for a case where that dir would already exist [17:32] akik, the funny thing is that if i go do /home [17:32] there is no /myuser directory :D [17:33] so i just see it in /etc/passwd [17:33] pydev: what if you now run "sudo su - myuser" ? [17:34] Is there a place to post screenshots to this channel? [17:34] ELFrederich: Usually we just post them to something like Imgur and then send links here. [17:34] su: warning: cannot change directory to [17:34] so the directory does not exist [17:35] ELFrederich imgbb i guess [17:35] !screenshot [17:35] Screenshots can be made with the [PrtScr] button. Want to show us a screenshot of your problem? Upload an image to http://imgur.com/ and link the created page here. [17:35] k, if nothing particularly recommended I'll use the first thing Google gave me;-)  ... https://snipboard.io/KGuU74.jpg [17:36] pydev: did it start in / ? [17:36] That's on a fresh install of Ubuntu 22.04 on a fresh install of VirtualBox on a fresh install of Windows 11.  My old 20.04 VM which I copied over launches Gimp just fine.  All the text is messed up on 22.04 [17:36] pydev: it did what you told it to do. It created the user and didn't create the home directory [17:36] pydev: useradd -M is decribed to do exactly do what you see - (assign but) not create the users' home directory. [17:37] pydev: what exactly are these users for? [17:38] these users are for running gunicorn it is basically a webserver to run django webapp [17:38] behing nginx [17:38] so basically each website has its own user [17:38] pydev: sounds like a job for docker honestly [17:38] BUT the same group (www-data) [17:39] ELFrederich: this may be an incompatibility of virtualbox and wayland (which 22.04 uses by default) [17:39] tomreyn, exactly it does not create the home but for some reason it assign one home [17:39] effortDee: there's a channel #vbox on the irc.oftc.net irc network, and they have a forum, too [17:39] pydev: it will do that. You can specify a different home if you like [17:39] tomreyn, i see that /var/www is the home for www-data so i suppose i can create /var/www/website1 as the home directory of a specific user [17:39] pydev: because it does so by default, yes [17:40] pydev: i don't see why not [17:40] tomreyn, what? can i create "nested" home directory ? [17:41] hmm... this sucks.  https://askubuntu.com/questions/1447077/display-problem-with-ubuntu-22-04-on-virtualbox [17:41] i mean...inside /etc/passwd i see /var/www with www-data user [17:41] it's "solved" by disabling 3d acceleration.  Everything seems so slow when it's disabled. [17:41] so if i create an user with an home inside v [17:41] /var/www/website1 [17:42] is this possible? [17:42] pydev: you can create directories all over your file system, and set homr directories for users to point wherever, whether that will do what you intend or be good idea security-wise is an entirely different question. [17:42] tomreyn why just gimp is messed up though? [17:43] ELFrederich: i wouldn't know that. it seems the window decoration is, too, so it's not just the application [17:43] tomreyn, yes but in my case i will create an home inside another home [17:43] pydev: as I suggested before, create the users as normal, create the /home, set the " for each web site/server to each users /home/user1/WWW/ [17:44] pydev: you can do so. it is probably a bad idea. [17:46] leftyfb, yes but as we said those are system users so i should not create the home dir no? [17:46] i read in that link you posted before [17:47] pydev: don't set them as system users [17:47] there's not really a benefit here [17:47] is this for separating users in a virtual hosting setup? [17:48] no tomreyn no virtualhosts [17:48] ok i will create normal users [17:48] so what is it? [17:48] it is a web server where i run my rojects but there is no virtualizations [17:49] why? [17:49] why do i have to virtualize it ? [17:49] nginx does virtualhosts [17:49] leftyfb, yes [17:49] virtual hosts...yes i meant docker [17:49] so use docker [17:49] so i set the "virtualhost" to /var/www/website1 /var/www/website2 [17:49] etc [17:50] don't [17:50] set it to /home/user/website1/WWW [17:50] leftyfb, why? is there a particular reason for that? [17:50] ah [17:50] yes yes [17:50] ok [17:50] leftyfb, you mean /home/USER/www [17:50] right [17:50] ? [17:50] up to you [17:50] "virtual hosts" refers to having multiple domain names point to a single (web)server. it is not related to virtualization or containers. [17:51] ok tomreyn so i have virtualhosts [17:52] so why do you need to run a separate web server for each user? [17:52] pointing to a single (web)server and having the (web) server respond differently based on the domain / hostname requested [17:52] ^ more precisely [17:52] tomreyn, yes exactly [17:52] however ok got it [17:52] i will create N users as per the number of websites i need to run [17:52] pydev: so use virtualhosts, not separate web servers [17:52] then create /www inside the home [17:53] leftyfb, it is a bit more complicated because as i said [17:53] i have nginx....yes....but i also have gunicorn that is the web server that is running the django appliations [17:54] then again, you should be running each user and this setup in separate docker containers [17:55] tomreyn disabling Wayland worked.  sudo nano -w /etc/gdm3/custom.conf ... and uncomment the line which disables Wayland [17:56] ELFrederich: congratulations for successfully working around a bug that's not ubuntu's [17:56] leftyfb, never used docker idk ...i wonder what is the benefit of running many containers.... N containers for GUNICORN N containers for NGINX [17:57] pydev: if you're using nginx to support multiple webservers talking to gunicorn, you can do without that if you use docker. Each docker will have it's own gunicorn setup [17:58] tomreyn until the bug is fixed perhaps we don't know whose bug it is. [17:59] ELFrederich: well, it doesn't occur with other virtualizations, that at least makes it unlikely. [18:00] leftyfb, yes exactly but in that case i need to run N different containers for gunicorn other N container for NGINX with different configuration [18:00] seems hard to mantain [18:00] pydev: no, you put all that into each container [18:01] per user [18:01] assume you only have 1 user. Put everything it needs into 1 container [18:01] then duplicate that for each user [18:01] i heard that each container should run one service [18:01] one container per service [18:01] the host OS doesn't need anything installed on it other than docker [18:01] nope [18:03] leftyfb, i think that nginx should be just one...because i only can have one container listening on port 80 [18:03] so just one container for nginx and then N container for N projects [18:07] damn, looks like "heal selection" isn't there in Gimp on Ubuntu 22.04?  ... even after installing gimp-plugin-registry [18:12] ELFrederich: https://askubuntu.com/a/1302593 [18:12] ELFrederich: see also https://www.gimp.org/registry/ and bugs 939660 1393237 1793270 [18:12] -ubottu:#ubuntu- Bug 1793270 in gimp-plugin-registry (Ubuntu) "After upgrading to cosmic, GIMP python plugins no longer work" [Undecided, Confirmed] https://launchpad.net/bugs/1793270 [18:12] -ubottu:#ubuntu- Bug 1393237 in gimp-plugin-registry (Ubuntu) "Many of these plugins don't work" [Undecided, New] https://launchpad.net/bugs/1393237 [18:12] -ubottu:#ubuntu- Bug 939660 in gimp-plugin-registry (Ubuntu) "globally-installed GIMP python plugins are not accessible" [Undecided, New] https://launchpad.net/bugs/939660 [18:15] leftyfb, it's not there.  That link they installed it from a ppa.  I installed mine through apt [18:15] that shouldn't matter [18:18] I had a lengthy workaround for getting the "heal selection" and "liquid rescale" to work on 20.04... I was hoping it would just work in 22.04 [18:19] This is what I had in my notes.... https://bpa.st/IJKJA [18:22] ... grouping/ungrouping the toolbox didn't make it show up [19:09] !WN722N [19:29] Since a kernel update, my bluetooth headphones won't connect automatically when I switch them on, without me running bluetoothctl connect "$hwaddr". I'm currently on 5.15.0-69-generic #76~20.04.1-Ubuntu SMP. Kernel 5.15.0-67 also had the issue. I have two others installed - 58 and 60, but haven't yet isolated which one it worked in. Can anyone offer any suggestions, please? What steps should I take to either resolve it, or [19:29] where should I raise the issue? === Furna is now known as Furna_away [19:39] is there a nice clean example/guide of using gparted (or other) to shrink an encrypted lvm in ubuntu 20.04 so I can give my /boot partition more space and upgrade to 22.04? Do I need a separate boot/live cd (assuming I do)? [19:47] hello guys, i reported some couple of months ago that i got screen flickering with ubuntu 22.04 with latest kernel, but when I connected my external display, was working normally, now i see again the problem with the kernel 5.19.0-40-generic , my pc is a laptop lenovo x1 nano gen2 [19:48] does a brand new/clean Ubuntu 22.04 LTS support installation of php 8.2 via apt? If not, what steps must I take to install php 8.2 via apt? [19:48] i dont see anything weird in Xorg log or dmesg [19:48] and it's weird, when i dont move the mouse it works good enough, but when I use the mouse, the flickering starts. [19:51] bluejaypop: flickering screen, in my experience, is usually due to a cable not being plugged in properly. you have a laptop, though. i wonder if it might be a display driver issue? sadly, i cannot offer any real assistance in this matter. [19:52] sneakyimp, could be, since its working normally in Win10 [19:55] bluejaypop: i remember when i was trying to get blender running optimally on my ubuntu workstation that I downloaded a linux-compatible driver from nvidia. perhaps you could investigate what kind of display chip/card you have in your lenovo, and see if they offer linux-compatible drivers. otherwise, you could look into other generic drivers [19:56] bluejaypop: i've no idea if it will be helpful, but there's an old thread here, might have some useful commands to get video driver info: https://askubuntu.com/questions/1331925/possible-to-install-distinct-drivers-for-two-dissimilar-nvidia-gpus-on-ubuntu-20 [19:57] thank you [19:57] let me find out [19:59] https://i.imgur.com/aawFOTI.png [20:01] ubuntu-drivers devices <- gives me nothing in the output, let me find out in lenovo [20:01] that looks like an intel graphics device...i wonder if intel offers a linux-compatible driver? [20:02] bluejaypop: perhaps here? https://www.intel.com/content/www/us/en/products/platforms/details/alder-lake-p/downloads.html [20:02] i added the i915.enable_psr=0 and it seems a little better [20:02] bluejaypop: or here? https://www.intel.com/content/www/us/en/download/747008/intel-arc-graphics-driver-ubuntu.html [20:02] yes, checking [20:03] that latter link has an 'Intel Arc Graphics Ubuntu Linux Installation Guide.pdf' [20:04] Am I correct in understanding that Ubuntu 22.04 LTS doesn't have an apt package to install php 8.2? === A_Dragon is now known as A_Dragon` [20:07] no, i think is 8.1 [20:08] i use docker for php dev === A_Dragon` is now known as A_Dragon [20:11] bluejaypop: i'm trying to set up a production server. no docker [20:11] PHP 8.1 will no longer enjoy active support in a mere 7 months [20:11] PHP 8.2 has been available for 4 months [20:25] sneakyimp: Distributed versions of PHP have beeing "going out of support" ever since the project picked back up afte PHP 5. You live with it. [20:25] sneakyimp: Nobody who relies on distributionpackages acctually needs bleeding edge PHP. [20:26] sneakyimp: If you do, you get it directly from the PHP project. [20:26] bluejaypop: check https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1958191 [20:26] -ubottu:#ubuntu- Launchpad bug 1958191 in linux-hwe-5.19 (Ubuntu) "[i915] Screen flickering in Ubuntu 22.04 (until i915.enable_dc=0 intel_idle.max_cstate=2 are added) (fixed in 5.17.7 and later)" [High, Fix Released] [20:26] Production Server != latest packages. [20:27] try the kernel boot parameters that helped some users there [20:27] You want well tested software. [20:27] jhutchins: respectfully disagree in broad terms. today's tools like composer et. al really hate older versions of PHP [20:29] jhutchins: and software that's been out for 4 months isn't exactly 'bleeding edge.' they're up to php 8.2.5 now. [20:34] if there is a crash or a bug in one of the software like nouveau, i915 or gnome-shell, etc. Why should we file a bug on the Launchpad? why not directly on the bug track of that software? [20:35] jhutchins: I would also point out that PHP 7.4 was EOL in November of 2022, and no longer receives security fixes. I think the problem is sluggish deployment of apt packages [21:03] PHP 8.2 transition will happen in the next release cycle, meaning Ubuntu will ship it starting in 23.10. Ubuntu does keep supporting the versions of PHP we ship in our supported versions. For 22.04, there is no "sluggish deployment". It is about api/abi stability. [21:04] Now, we could indeed discuss why the transition did not happen yet, even though 8.2 was released back in late November [21:07] The reason is the short gap between PHP and Ubuntu release cycles) === B1773rm4n0 is now known as B1773rm4n [21:33] ravage, thanks ill check [21:33] sorry, i got busy [21:36] sneakyimp: If you actually need that, don't rely on the Ubuntu packages, go directly to PHP.org: https://www.php.net/downloads.php [21:58] sneakyimp: ubuntu patches software in LTS releases. This includes php, regardless if upstream supports it or not [22:01] sneakyimp: That's how most of the distros handle PHP security. [22:02] sneakyimp: Mind you, they don't usually add new features. [22:04] does anyone here rely on the Google Drive implementation in Gnome (Online Accounts)? How does it compare to e.g. rclone? [22:05] hsiktas[m]: rclone is the only thing I've tried that actually worked on the first try. [22:06] for example: opening a pdf, leaving it open for hours or even days, and then making an annotation and saving it. Would that sync back? [22:08] hsiktas[m]: Bad idea, cosmic rays will corrupt your file. [22:09] or the file in the local cache gets orphaned and my changes land in limbo :-/ [22:09] hsiktas[m]: Seriously, there are too many variables to be able to rely on a "permanent" connection to google storage. [22:11] well, it is no issue with the official Drive client on macOS/Windows. I can make a change offline and it will sync it days later [22:11] hsiktas[m]: and I'm sure if Google made a linux client for google drive, it would be the same way. [22:11] hi. how can I find out what versions of pyaudio (package python3-pyaudio) are available for my installed version of ubuntu? [22:12] jim: https://packages.ubuntu.com [22:12] !jammy python3-pyaudio [22:13] jim: apt-cache policy python3-pyaudio [22:13] jim: or: apt list --installed python3-pyaudio [22:14] !info python3-pyaudio jammy [22:14] python3-pyaudio (0.2.11-1.3ubuntu1, jammy): Python3 bindings for PortAudio v19. In component universe, is optional. Built by python-pyaudio. Size 26 kB / 114 kB [22:20] ok, that was a short journey: https://gitlab.gnome.org/GNOME/evince/-/issues/1781 [22:20] -ubottu:#ubuntu- Issue 1781 in GNOME/evince "Unable to open PDF file from Google Drive virtual drive" [Opened] [22:31] I'm told by the folks in #python, that the version of pyaudio (package name python3-pyaudio) is broken, and they're recommending people install pyaudio from pip. in the middle of that conversation, they commented that ubuntu devs are tracking the issue [22:33] jim: do they have a bug report? [22:33] the issue may have something to do with a change in python itself, that stems from version 3.10; I don't know what that issue is, and it might be a non-problem [22:33] I do not know [22:34] but they are confirming that if you get the version of pyaudio from pip, you should be OK with 3.10 of python [22:35] let me try to find out about a bug report === xenial is now known as Guest9059 [22:49] LP #1970912 strikes me as being it btw. [22:49] -ubottu:#ubuntu- Launchpad bug 1970912 in python-pyaudio (Ubuntu) "python3-pyaudio fails with: PY_SSIZE_T_CLEAN macro must be defined for '#' formats" [Undecided, Confirmed] https://launchpad.net/bugs/1970912 [23:16] In v22, is there some way to run fsck at boot?  This didn't work https://askubuntu.com/questions/1352774/how-to-force-fsck-on-reboot-for-ubuntu-20-04 [23:18] Guest65: if the fsck.mode=force kernel command line parameter doesn't work, that's worth filing a bug report [23:23] Since a kernel update, my bluetooth headphones won't connect automatically when I switch them on, without me running bluetoothctl connect "$hwaddr". I'm currently on 5.15.0-69-generic #76~20.04.1-Ubuntu SMP. Kernel 5.15.0-67 also had the issue. I have two others installed - 58 and 60, but haven't yet isolated which one it worked in. Can anyone offer any suggestions, please? What steps should I take to either resolve it, or [23:23] where should I raise the issue? [23:23] I've asked 3 times today. Any chance of some help? [23:24] clarkk, can you try with blueman? [23:25] what shall I do with it reinvo49 ? [23:26] clark, run blueman, in the UI, click search, and associate your device trhough it (cause it remember the device then), then disconnect your headphones, and try to power them up again, and look if blueman connect them back again automaticly. [23:27] clarkk: if blueman does not help, i'd run the 'reportbug' utility and choose a linux-kernel package (output of `dpkg -l *linux*image*`) as the package to report the bug against. if that's wrong then someone will adjust the bug report accordingly [23:28] clarkk, is it a packaged kernel you installed with apt? [23:28] reinvo49, yes [23:28] gry, what is reportbug? It's not installed on my system [23:29] ubuntu-bug on ubuntu :) reportbug is for debian [23:30] clarkk: what sarnold wrote (thanks, sarnold) [23:31] ok, thanks chaps :)