[00:36] ah no need to know :D === womp is now known as mofohack === M4he is now known as mahe [01:37] hey I was testing my app and it accidentally rm -r 'd my home directory. I'm on a live usb right now trying to recover the data [01:37] extundelete finds nothing on it though when I run [01:38] Uploaded file: https://uploads.kiwiirc.com/files/79fa269c20d32c1694b8092cbb365c09/pasted.txt === admin is now known as littlepap [02:25] TJ-: It happened again lol [02:25] Oh wait nevermind thats the same person [02:38] it tis [03:51] Hey all, I'm having an issue with fstab that I've been running for years. I suddenly get `cannot access 'data': Transport endpoint is not connected` [03:51] fstab looks like... [03:51] ```#boot WD External 10TB [03:51] UUID=966071E36071CB0F /home/brombomb/data auto defaults 0 0``` [03:51] and blkid looks like... [03:51] ```#/dev/sdb1: LABEL="easystore" UUID="966071E36071CB0F" TYPE="ntfs" PTTYPE="atari" PARTLABEL="easystore" PARTUUID="dc5bd410-f09c-43db-af47-b46943c20b25"``` [03:52] I don't know what to do next to try and solve it. I have unmounted and remounted [03:55] brombomb, you've provided no OS & release details (ie. a clue to your software stack) [03:55] Ubuntu 20.04.2 [03:57] have you tried using a /mnt/ (instead of /home/) ? [03:57] I have not [03:57] * guiverc didn't mean specifically /mnt/ ; but directory inside there; I'd try it [03:58] * guiverc suggestion only; I don't know that it will help [03:58] Nope :( [03:59] same Transport error [03:59] https://askubuntu.com/questions/1090715/fuse-bad-mount-point-mnt-transport-endpoint-is-not-connected maybe (good chance you've seen it though) [03:59] I'm not using sshfs, since it's a usb mount [04:00] I'm open to migrating to fuse or something newer [04:02] yeah it looked that way from what you pasted; it was all I saw & i don't use NTFS [04:08] okay so a few days ago I was trying to clean up some old HDD, and I installed the ubuntu package usbmount to help auto mount USB devices [04:08] well after a reboot that was conflicting with my fstab [04:08] I unsinstall usbmount, rebooted and It's fixed [04:09] :) [04:17] Thanks for rubber ducking for me guiverc === marcustomlinson_ is now known as marcustomlinson === mteo_ is now known as mteo === mybalzit1h is now known as mybalzitch === synapt is now known as nate [05:20] Would anyone here like to shave my back? [05:20] not here Anastasius [05:20] this is ubuntu support [05:23] Okay, I'm on Ubuntu 21.04. Are you willing to support me, lotus? [05:29] ask a question to the channel Anastasius [05:40] Anastasius: shaving your back actually sounds reasonably entertaining. Count me in. === saribro is now known as swinnenb === buzelAzul is now known as buzel [06:46] Croran: You shave my back and I shave your front. Deal? === Milencho_ is now known as Milencho [06:46] Anastasius Croran ok please stop that stuff [06:47] Or what? [06:48] Nobody said anything for hours, so get of your horse, chief. [06:49] Anastasius: this is an ubuntu support channel, so please no other topics here [06:49] off, too. [06:49] !guidelines [06:49] The guidelines for using the Ubuntu channels can be found at https://wiki.ubuntu.com/IRC/Guidelines [06:49] Oh please, stand down. [06:49] he just wants a little support to get his back shaved. [06:50] I know the rules and the guidelines, [06:50] Yeah, mostly that. Shave my back, dude. [06:51] !ops | Anastasius waterytart Croran offtopic rant [06:51] Anastasius waterytart Croran offtopic rant: Help! Channel emergency! (ONLY use this trigger in emergencies) - Pici, Myrtti, jrib, Amaranth, tonyyarusso, Nalioth, lamont, CarlK, elky, mneptok, Tm_T, jpds, ikonia, Flannel, wgrant, stdin, h00k, IdleOne, Jordan_U, popey, Corey, ocean, cprofitt, djones, Madpilot, gnomefreak, lhavelund, phunyguy, chu [06:51] this is hardly an emergency. [06:52] Oh no, they called the cops again. [06:52] Oh dearie me. [06:52] scatter [06:53] https://c.tenor.com/y9zbADwEuzwAAAAS/cheese-it-bender.gif [06:53] there are offtopic channels you can use if you are bored. you don't need to use an official support channel for this nonsense [06:54] Yeah. Been there, done that. I'd rather be here thanks. [06:54] At least for now. [06:57] Mark Shittleworth is my brother in law, I'll have you know. [06:59] Yeah, I'm with you. I don't like him either. [07:03] even I added resolution fo 1920x1080 in xorg.conf, for Intel HD4600 VGA output, is is still limited to 1024X768 instead of selected 1920X1080. [07:04] For some reason 915 driver can't recognize VGA-1 connected monitor's available resolutions === lordrishav is now known as LordRishav [07:29] do vnc servers work with wayland in ubuntu? [07:31] https://wayland.freedesktop.org/faq.html#heading_toc_j_8 [07:32] I saw that. Just wondered if there was any workaround. === tribaal_ is now known as tribaal [08:27] are there any docs describing the difference between separate origins, ${distro_codename}, -security, -updates, -backports, and -proposed? I can't seem to find an official source, most things I find are hearsay or based on the names of the origins [08:27] I'm seeing some oddities with unattended upgrades and I'd like to know which origins I can safely enable [08:28] (and be able to convince my colleagues that I've done my homework about it) [08:31] !sources | bertptrs [08:31] bertptrs: 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. [08:31] !backports | bertptrs [08:31] bertptrs: If new updated Ubuntu packages are built for an application, then they may go into Ubuntu Backports. See https://help.ubuntu.com/community/UbuntuBackports - See also !packaging [08:32] !packaging [08:32] The packaging guide is at http://packaging.ubuntu.com/html/ - See https://wiki.ubuntu.com/UbuntuDevelopment/NewPackages for information on getting a package integrated into Ubuntu - Other developer resources are at https://wiki.ubuntu.com/UbuntuDevelopment - See also !backports and !sponsoring [08:34] thanks, that clarifies -backports and -prososed, although I'm still not entirely clear on the distinction between the others [08:37] !updates [08:37] !security [08:37] Security Updates are dealt with here: https://help.ubuntu.com/community/Security - See also !root, !firewall, !server, and !usn === diskin_ is now known as diskin [09:27] bertptrs: the only distinction is in the name, security are updates from the security team, -updates are everything else === Bardon_ is now known as Bardon === Abrax is now known as _] === _] is now known as Abrax === sharperguy__ is now known as sharperguy === SmokenatorZ1 is now known as SmokenatorZ === SmokenatorZ9 is now known as SmokenatorZ [13:06] Hi all [13:14] hi guys, will the wireguard kernel module come to 20.04 ? [13:15] Lutin: I thought it was already there? [13:16] rbasak I needed to upgrade to 21.04 to have it.. [13:16] it is definitely there in the hwe kernel [13:16] https://packages.ubuntu.com/focal/wireguard, there seems to be a dkms for it [13:16] ... but i also see it in the regular 20.04 5.4 kernel here [13:17] /lib/modules/5.4.0-80-generic/kernel/wireguard/wireguard.ko [13:17] as well as [13:17] /lib/modules/5.8.0-63-generic/kernel/drivers/net/wireguard/wireguard.ko [13:18] mhh [13:18] okay because with docker I wasn't able to use it about 3-4 weeks ago [13:18] so i upgraded [13:19] cbreak yeah installed that... issues or bugs, don't know anymore [13:19] Sounds like you need help to get it working in your particular environment, rather than there being missing support in Ubuntu. [13:21] I have an lxc that runs wireguard in 20.04 [13:21] (but my outer OS is 21.04 so... maybe it just uses that [13:36] Hello people. A friend of mine wants to install Ubuntu GNU+Linux for the first time (first time installing GNU OS) to try it out and he was wondering if he could be able to add another HDD to his laptop and install Ubuntu there. Is it possible? [13:37] Instead of installing two operating systems on one HDD, he wants to install two operating systems on separate HDDs but on single laptop. [13:44] arh, is possible from grub from the first hdd(or ssd if that is the case), as long as the laptop can have another hdd. [13:46] coconut, so it won't make any problem? Another question. Is it possible to install Windows after installing Ubuntu? It's for myself though. My hard disk is encrypted and I don't want to lose my installation. I got my laptop with Ubuntu pre-installed. [13:56] cbreak it uses your 21.04 host one [13:59] arh, if you install windows after ubuntu? Then your bootloader will be the windows version again, so you would need to re-install grub again to make boot option for ubuntu too. Probably duckduckgo.com just finds it well. [14:03] hi [14:03] so I want to list all packages `apt` got from a repository on my system [14:03] I can see that repository during `apt update` [14:03] but I want to find out what packages `apt` actually sees from that reposittory [14:03] any ways? :) [14:04] apt policy [14:08] mikrosis: grep ^Package /var/lib/apt/lists/repository-name*_Packages | awk '{print $2}' | sort -u # replacing repository-name with the repo [14:10] oops, indeed ... my command above only shows where a package was installed from, leftyfb is correct to look at the list file [14:11] leftyfb: thanks [14:11] leftyfb: OK, I ran that command for a particular PPA repo and nothing is listed [14:11] so this means the repo is empty or something went wrong, right? [14:12] mikrosis: look in /var/lib/apt/lists/ for the repo you're looking for and make sure it has a filename that ends in _Packages [14:13] and make sure to have called apt update to make sure the list file is actually up to date [14:13] ls /var/lib/apt/lists | grep ppa [14:13] ppa.launchpad.net_ondrej_php_ubuntu_dists_groovy_InRelease [14:13] apt update indeed lists that repo [14:13] but when I run that grep to list all packages of that repo - it comes back empty [14:14] mikrosis: ppa.launchpad.net_ondrej_php_ubuntu_dists_groovy_InRelease does not end in _Packages [14:14] mikeliuk84: which release of ubuntu are you running? [14:15] leftyfb: ohhh [14:15] leftyfb: but why doesn't it? something went wrong during apt update? [14:15] coconut, thank you. [14:15] what ubuntu release is your machine running [14:16] me? Ubuntu 20.10 [14:16] mikrosis: time to upgrade [14:16] leftyfb: hmm [14:16] mikrosis: especially if it's a server [14:17] launchpad.net/~ondrej/+archive/ubuntu/php only supports bionic, focal and hirsute [14:17] yes [14:17] oh [14:17] that explains it [14:18] mikrosis: typically when a release of ubuntu goes EOL, you start to lose support for it. 20.10 went EOL a little over 3 weeks ago [14:21] leftyfb: how dangerous is a dist upgrade from my version to the latest LTS? [14:22] leftyfb: I took a backup and snapshot [14:22] but I want to avoid downtime. sure, I could spin up a 2nd instance but this is also a cost factor [14:23] leftyfb: hmm, 20.04 seems to be the LTS, not the latest [14:24] if you have a public server you should always stick to the LTS [14:24] mikrosis: unless you are upgrading every 6-9 months, I would not suggest using non-LTS releases. Especially for a server and even more-so for a production environment === southey is now known as foxtrot [14:35] leftyfb: thanks, this was helpful [14:45] Doesn't Ubuntu offer minimal/server ISOs anymore? [14:45] At least I read "legacy" [14:46] mikrosis: https://cdimage.ubuntu.com/ubuntu-legacy-server/releases/20.04.1/release/ first result on google for "ubuntu 20.04 server legacy" [14:48] leftyfb: right, but they call it "legacy" now :/ [14:52] Hm, how can I find out whether my Ubuntu is an Ubuntu server or not? [14:52] lsb_release -a only shows that it is Ubuntu [14:52] mikrosis: did you install a desktop environment? [14:52] Or is Ubuntu server just Ubuntu without the GUI (headless)? Are there any indications? [14:52] no desktop, no [14:53] I don't want any GUI, minimal [14:53] then it's a server [14:53] OK, so there is no other thing than that [14:53] correct === martums7 is now known as martums [15:09] hello everyone I have a question which I don't know what channel it fits best, so I am asking it here. I have an application installed (latest version in repos) on my Ubuntu 20.04 and I am currently working on the translation of that application's upstream version (newer than the one I have installed). Before submitting my translation I want to compile and run the app to see everything is ok. [15:09] My question is, is there a "safe" way to compile, install and run this upstream version without recurring to a virtual machine or uninstall the currently installed version of the app? [15:11] by safe I mean avoiding any possible conflict arising from having two different versions of the same app on the OS at the very same time [15:12] metallic, by upstream version do you mean from a ppa? [15:13] BluesKaj: no, the latest version available in its github repo [15:13] well, I thought upstream version meant "latest developed version", sort of ^^', sorry if I got it wrong [15:14] metallic, then what reason do you have for not using the default version for your OS [15:14] newest isn't always the best [15:16] because the translation I am doing is for the latest version in the repo, which differs slighty from the latest in the Ubuntu repos [15:16] sorry, the translation is for the latest version in the github repo (latest developed version), not for the latest available in Ubuntu's [15:17] to test the translation in action I should compile, install and run the app, but I am unsure if that may cause any trouble knowing I have already installed an older version :/ [15:18] then you should ask in #ubuntu-devel [15:19] I thought of it, but it's channel description clearly states "ubuntu devel (not support)" ^^', so I guess my question is not in order there :( [15:19] metallic: is this a GUI application? [15:20] yes, it does have a GUI [15:20] metallic: then I would pick something to go into a VM [15:21] I wanted to avoid either that or uninstalling the currently installed version hahaha, but I guess there is no way out [15:23] well, I think I will take the VM road. If anyone comes up with another alternative, please let me know :) [15:23] thanks for your time, BluesKaj, leftyfb [15:24] good luck metallic :-) [15:29] anyone can help me what is the problem of configuring the bridge on the VM which has a static IP? https://paste.ubuntu.com/p/z3F8xVsmWG/ [15:31] when I remove the bridge part, VM connect to the internet, But when I add the bridge part to the config file, VM won't have internet [15:44] where do i specify what app opens a filetype? [15:44] right click does not allow this for desktop files and in nautilus it prepends a wrong path hence fails [15:45] for some reason gnome gets less usable with every update [15:46] i want to 1) provide an application path 2) tell gnome to use the path of a file without any modification [15:49] There is no default app/protocol settings page as in Windows? Then they should really add this [15:50] there is, in gnome systemsettings, default applications [15:50] but the 'path' thing i don't understand [15:51] oerheks: gnome systemsettings, default applications [15:51] eh [15:51] https://askubuntu.com/questions/90214/how-do-i-set-the-default-program === Turgenevskaya is now known as TortillaSandwich === TortillaSandwich is now known as TortillaSandwitc === TortillaSandwitc is now known as TortillaSandwich [15:58] huh? where can i define the app and the launch signature (e.g. labview -launch /path/to/file.vi) in there? [15:59] no possibility to even choose anything related to that in the gnome default applications menu [16:00] look i have files ending with .vi and i want to tell gnome to use the program "labview" in the PATH and invoke it like labvies -launch /path/to/clicked.vi [16:03] is that possible somehow? [16:04] de-facto: I am using LXQt right now and I can do what you ask by right-click > Properties > Open with > Customize > Custom command [16:04] yeah i wish gnome had something like that [16:04] they delete everything from their interface [16:04] de-facto: try mimeopen -d file.vi [16:05] de-facto: also, I'm curious what you mean by "in nautilus it prepends a wrong path hence fails" [16:06] de-facto: if you have multiple labview binaries, that might be the issue you want to look into. Using $PATH in something like this isn't recommended. It should be an absolute path [16:06] leftyfb it prepends a prefix to the file path that makes labview fail to find it then [16:07] mimeopen presents me a choice 1) labview 2) other and 1) works [16:07] so how to tell gnome to do it like mimeopen? [16:08] the open with dialogue is pretty useless as it does not allow any modification [16:08] de-facto: so running /path/to/labview -launch /path/to/file.vi fails? [16:08] nope that works i think [16:08] let me check [16:09] "it prepends a prefix to the file path that makes labview fail to find it then" [16:09] jup that works [16:09] yes gnome prefixes the file path [16:11] labview says the file /usr/local/natinst/LabView-2020-64/*/home/user/Desktop/file.vi could not be opened. no application was found to open that type of file [16:11] so the way gnome launches it it fails, but mimeopen and manually opening it from terminal works [16:12] hence my question: how can i modify the way gnome opens files? they removed the option to modify that (i remember it was in the open with dialogue before [16:12] missing a space? /usr/local/natinst/LabView-2020-64/*/home/user/Desktop/file.vi [16:13] anyway, put labview properly in your %path% [16:14] it is in the PATH [16:14] the way gnome launches it is broken [16:14] i need to fix that, hence where can i influence that? [16:17] hm, put it in quotes or backticks maybe? [16:17] how do i tell gnome to put it in quotes? [16:18] i can easily launch it from terminal [16:18] but that line is missing a space too.. [16:18] i cant even see a way to fix things in gnome, where is that setting supposed to be made? [16:18] Any one Where is config file for Resolution list ? [16:20] AmR, there is no resolution list./xorg.conf, it is generated automatic [16:20] is there some simple crontab manager on cli ? [16:25] is there any way in gnome to fix broken file associations? [16:27] de-facto: overwrite the broken config file with a working one? [16:27] which config file? [16:27] Lutin: crontab -e # though you should really be moving over to systemd timers [16:28] mikrosis if you tell me which file is responsible for defining that i can try to fix it, but i dont know which file is that for gnome [16:30] de-facto: Different distro, but this could be the same: https://bbs.archlinux.org/viewtopic.php?pid=638881#p638881 [16:31] normally one clicks open with; select application( if in the path) and be done with it. [16:31] there is no file with custom associations, AFAIK. not human readable [16:31] oerheks: and in this case it doesn't work? [16:31] oerheks: also *nix also uses magic numbers for binary files [16:32] de-facto: as your question is specifically on GNOME's behaviour, you could try to ask it on #gnome . [16:32] well, he made an error with a space in that line he gave us [16:32] and with a * at the end, not a good idea [16:33] can I check that I changed my password successfully without logging out? [16:33] possibly using the `login` command [16:33] ash_worksi: yes, login using a different window/terminal [16:33] (via ssh) [16:33] or use sudo [16:34] will sudo use the new password? [16:34] yes [16:42] hmm by trial an error: cp /usr/share/applications/labview64-2020.desktop ~/.local/share/applications/labview64-2020.desktop ... and then nano /.local/share/applications/labview64-2020.desktop to modify the Exec line to Exec=labview -launch '%F' [16:42] using double quotes there seems to make gnome prepend that broken prefix [16:42] e.g. Exec=labview -launch '%F' works while Exec=labview -launch "%F" is broken [16:46] leftyfb OK... going to read about them... sounds better.never liked crontab :) [17:42] !repos | bertptrs: i think this describes all "pockets" [17:42] bertptrs: i think this describes all "pockets": 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. === mofohack is now known as Lhack === jake1 is now known as sybariten === Lhack is now known as Lago [20:13] well, I figured out that ubuntu consistently fails to suspend if I have no monitors connected to my GPU, or if the connected monitors don't have power [20:17] interesting [20:17] mort: which GPU make? [20:17] nvidia [20:18] mort: does the nvidia driver report problems in the kernel log? [20:19] TJ-: I don't know how I can check really? From what I've read suspend problems are really hard to debug specifically because non-volatile storage gets disconnected [20:19] I can't get it back on without a hard reboot through holding down the power button or cutting the power in other ways [20:20] I know for sure that the GPU is at fault though, I did the pm_trace stuff in the suspend debugging guide on the ubuntu wiki and at the next boot, the only "hash matches" line in the dmesg log was regarding the GPU PCI device === NoResults is now known as NoSQL [21:20] so I have a connection script that uses expect; I have a reset option which resets the password (given a default) [21:21] Hi, I'm wondering if any Thinkpad users can verify this trackpoint issue that I've run into on my new P14s: https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-input-libinput/+bug/1936008 [21:21] Launchpad bug 1936008 in xserver-xorg-input-libinput (Ubuntu) "Trackpad interfering with trackpoint even when disabled" [Undecided, New] [21:23] I'd like to be able to check that the password has been set successfully before moving on... how can I do that? [21:23] ash_worksi: usually it's enough to check the exit status of the command you execute [21:24] sarnold: you wouldn't know how to do that in tcl/expect / [21:24] ?* [21:24] ash_worksi: use expect/sshpass to ssh to yourself or auth with sudo [21:24] leftyfb: I can do that, but I was hoping to automate it [21:25] hmm... are there any packages to get the dev headers for libzfs.so.4? [21:25] ash_worksi: hence expect [21:25] which package does libzfs.so.4 even come from? [21:26] leftyfb: maybe I don't understand [21:26] !info libzfs-dev [21:26] Package libzfs-dev does not exist in hirsute [21:26] cbreak: looks like it's only available in the libzfs4linux in 21.04 [21:26] find libzfs [21:26] ash_worksi: check $? perhaps? [21:26] ash_worksi: (bash variable) [21:27] ash_worksi: if expect executes the thing directly itself, then it should have a way to get the exit status [21:27] ash_worksi: use expect in your script that checks if you can ssh with askpass or use sudo [21:27] !info libzfs2linux-dev | cbreak [21:27] cbreak: Package libzfs2linux-dev does not exist in hirsute [21:27] !info libzfslinux-dev | cbreak [21:27] cbreak: libzfslinux-dev (2.0.2-1ubuntu5.1, hirsute): OpenZFS filesystem development files for Linux. In component main, is extra. Built by zfs-linux. Size 1,782 kB / 10,945 kB. (Only available for linux-any.) [21:27] leftyfb: thanks, that's for the library [21:28] cbreak: yeah, you asked what package the file came from, that's the package it comes from [21:28] cbreak: try installing libzfslinux-dev ? [21:28] indeed [21:28] I wonder why dpkg-query -S didn't find that [21:28] since I did have it installed [21:28] sarnold: yes, but issuing a command and then reading the output to a variable to get n-bytes to check it is a bit convoluted. I was hoping there was a better way [21:28] sarnold: yeah, promissing indeed, thanks [21:29] ash_worksi: the downside of course is that checking the return value assumes the passwd ocmmand will return the right result. it *should*, but you know how that goes.. [21:29] mneil-qc, i am reading it(i have an x1), but i do not understand what "trackpoint" is. [21:30] coconut: bright little red thing in the keyboard [21:30] coconut, the trackpoint is the red pointing stick on the keyboard [21:30] leftyfb: what i mean is, I update my local password manager after running `sudo su -\r ***********\r passwd -u $user\r **********\r ***********\r` [21:30] where '*'s are passwords [21:30] Does anyone know if a driver for these headphones exists or can point me in the right direction?: Bus 001 Device 010: ID 9886:0042 Astro Gaming Astro A20 [21:31] oh, never used it but just worked one second ago i guess [21:31] coconut, mine works but it's not smooth while there is anything in contact with the trackpad [21:31] I just don't want to update the password manager unless I am sure that the `passwd` operation was successful (it's not right now for some reason, but it could be asking me to retype twice because of some arbitrary crap password rules that sudo overrides) [21:31] it becomes erratic and "laggy" if my palm is near the trackpad, even if I've disabled the trackpad [21:32] ash_worksi: why not just use chpasswd ? [21:32] mneil-qc, no i have no issue here on x1 extreme gen2(uefi 1.36 i believe) [21:32] weird, thanks for checking [21:33] yeah, all i can do [21:33] can you see if it's an Elan or Synaptics one? [21:33] it should tell you if you use "xinput list" command [21:33] TJ-: I had problems with that before.... I could probably get that working, but now that I understand how to use `exp_continue` I just loop across `[Cc]urrent password` and `[Nn]ew password` to insert them [21:34] can't I check this with `login` somehow? [21:34] I keep forgetting how to use `login` [21:35] mneil-qc: have you tried to xinput --disable 11 [21:35] mneil-qc, "Synaptics TM3512-010" [21:36] mneil-qc: just one dash, sorry [21:36] tomreyn, I have, but it doesn't do anything apart from disable my trackpad [21:36] even with the trackpad disabled it still interferes [21:36] oh i assumed that's what you wanted [21:36] Ideally it is, but there's still interference [21:37] coconut, thanks. My older thinkpad's trackpoint uses synaptic and it works fine too. This one uses Elan [21:37] okay, i didnt get that [21:37] ash_worksi: with chpasswd, if you do one user at a time, it'll return a non-zero exit code if there's a failure. E.g. echo "me:mypassword" | chpasswd -R /mychroot && echo "Success" [21:38] hmm... it seems /lib is a symlink to /usr/lib [21:38] bad for thinkpads reputation [21:38] mneil-qc: is there something in the kernel log suggesting to use an alternative driver? [21:38] tomreyn, the reason I think it's interference is because I ran "libinput debug-events" and it still registers trackpad touches even when it's disabled [21:38] TJ-: well I did go through the effort of setting the current user to a variable (using whoami) and sudo'ing to root [21:38] so I guess I should be able to do that [21:38] TJ-, no [21:38] and libzfs4linux installs to /lib [21:39] eg `echo "$user:$newpass" | chpasswd` [21:39] what is -R? [21:39] so dpkg-query -S didn't find the /usr/lib true location variant [21:39] ash_worksi: "man chpasswd" [21:40] cbreak: that's part of the fallout of the usrmerge process. with -S only search for the last part of the path, e.g. bin/myexec or lib/libmylib [21:40] mneil-qc: :-/ i'm wondering whether that's more of a firmware or driver issue [21:40] TJ-: interesting, thanks [21:40] TJ-: Apply changes in the CHROOT_DIR directory and use the configuration files from the CHROOT_DIR directory. --- I don't really get what that menas [21:40] means* [21:40] never heard of that process. [21:41] ash_worksi: it means you can test it on a chroot instead of the live system, until it works perfectly [21:41] mneil-qc: my thinkpad (E495) report an Elan device. How do I do your test? [21:41] which is... like a virtual env? [21:42] ash_worksi: yes, usually created by debootstrap [21:42] TJ-, make sure nothing is touching/near your trackpad. Move the trackpoint around and see if the movement is fluid [21:42] okay, so what determines /mychroot in your example? [21:43] TJ-, if my palm is on the trackpad, or if I click the trackpoint buttons and the trackpad detects my fingers, the trackpoint becomes erratic/laggy [21:43] mneil-qc: so, to be clear, we have 2 devices, SynPS/2 Synaptics Touchpad and TPPS/2 Elan Trackpoint [21:43] TJ-, that's right [21:43] Same for me [21:43] mneil-qc: and which one do you disalbe? [21:44] I disable the touchpad (synaptics) [21:44] mneil-qc: mine's disabled.... never noticed this before, but it works!! [21:44] no choppiness/skipping with the pointing stick? [21:45] mneil-qc: you have it totally disabled, or just the "Disable touchpad whilst typing" ? [21:45] it's definitely off [21:45] I tried in the BIOS as well, but it seems the BIOS is a windows-only switch on new thinkpads [21:46] mneil-qc: enable/disable and now touchpad isn't working... resting thumb on touchpad whilst moving trackpoint, smooth [21:46] TJ-, really strange for me then [21:46] mneil-qc: I'm using kernel 5.13 [21:46] I've tried using a new kernel too just in case [21:46] 5.11 [21:47] I've also tried on fedora with a live usb session and it's there too [21:47] more likely to be Xorg drivers though, presumably you're using Xorg not a Wayland compositor? [21:47] I've already tried that [21:48] I'm thinking the issue is libinput still [21:48] TJ-: like does `/mychroot` have to exist? [21:48] ash_worksi: well yes :) [21:49] mneil-qc: I'm on 20.04, Xubuntu [21:49] TJ-: can/should I just make it my home dir? [21:49] TJ-, when I run "sudo libinput debug-events" I can see the trackpad registering things, even if it's disabled and not moving the cursor [21:49] I've tried on i3wm too, so it's not a gnome thing [21:49] ash_worksi: you can make it anywhere, that will have under it ./etc/passwd and ./etc/shadow I think [21:50] mneil-qc: I'll try that here [21:50] TJ-: is it intended to be ephermeral? (after the command executes, does it go away? Should I get rid of it?) [21:50] TJ-, thanks. Basically, the issue makes it so that moving the trackpoint always feels like there's another mouse/pointer trying to move in another direction [21:51] ash_worksi: it's purely a way to test your processes without messing up the system's real /etc/shadow [21:51] mneil-qc: is it possible the scroll region is stuck down? [21:51] TJ-, definitely not [21:52] mneil-qc: which package has libinput ? [21:52] TJ-, xserver-xorg-input-libinput [21:52] found it ! libinput-tools [21:52] ah, right [21:53] TJ-: I mean, I am just kind of confused; I use `docker` and I think that isolates in a similar manner, but that's completely ephemeral [21:54] mneil-qc: I see events from touchpad when it is disabled but they are ignored. They do not affect the pointer being controlled by trackpoint [21:55] TJ-, that's the same for me, except the lag happens whenever the events show up [21:55] ash_worksi: -R simply tells chpasswd to operate on files other than the system files, so it can be tested, or used to configure real chroots as created by debootstrap [21:57] TJ-: so, something like `... | chpasswd -R ... && echo "success"; expect "success": pass insert -ef "$newpass"` [21:59] ash_worksi: no, that would do the pass insert regardless of exit code. You want: if echo "$user:$newpass" | chpasswd -R /my/test/root ; then pass insert -ef "$newpass"; else echo "Error: password change failed"; fi [21:59] ash_worksi: obviously, you'd remove the -R /my/test/root for real changes [22:00] TJ-: `-R` confuses me tremendously, is it affecting the system or not? I *only* want to update the manager if password is *actually* changed [22:00] otherwise I won't be able to login [22:02] ash_worksi: as the man page says, "Apply changes in the CHROOT_DIR directory and use the configuration files from the CHROOT_DIR directory" [22:04] TJ-: I don't know if I'm hopeless or not, but I just don't know what the implication is of that statemetn [22:04] statement* [22:05] ash_worksi, to put it another way it means it will alter a file /my/test/root/etc/shadow given a CHROOT_DIR of /my/test/root [22:06] coryc257: that sounds like if I do that, and say `if ... | chpasswd -R ...; then pass insert ...; fi` then I'll update the manager and not actually update my password (outside of chroot) [22:08] ash_worksi: example: https://paste.ubuntu.com/p/S35NjSP9PS/ [22:09] ash_worksi: precisely! I said, use -R to /TEST/ your scripts do update shadow correctly first, THEN remove the -R path [22:09] Is writing programs for ubuntu on topic, or is there a better channel for that? [22:11] TJ- so `if ... | chpasswd -R ...; then ... | chpasswd; fi` ? [22:11] LadyCailin: 'for Ubuntu' is kind-of limited - usually it'd be "for GNU/Linux" - Ubuntu is a distribution. Unless you mean packaging a program for Debian/Ubuntu [22:11] also, I just realized I can't do `then pass ...` anyway because I need to run that locally from the script [22:12] ash_worksi: but surely you'd put the if...then...else...fi inside your script ? [22:13] No. I guess it is Linux programming, but I'm specifically testing on Ubuntu. But at the stage I'm at, I doubt there will be any specifically Ubuntu things, though maybe, I'm writing very low level code. In any case, if there's a more specific channel, I can go there, otherwise I can just ask here. [22:13] TJ-: well, `chpasswd` needs be run as `send "chpasswd"` where as `pass` needs to be run `exec pass ...` [22:14] ash_worksi: you're embedding this in an expect script? [22:15] TJ-: yes [22:15] TJ-: I mean, it parses flags in bash and then calls expect [22:15] ash_worksi: I think I'm lost as to why this is wrapped with expect [22:15] LadyCailin: well ask away, but you won't mind many programmer/developer types here most of the time [22:15] TJ-: because I am bound by archaic insanity that is requiring both sshkeys AND a password to log into servers [22:16] LadyCailin: there's #linux for more than user support centric linux (here: ubuntu) support, and there's probably a channel for your programming language of choice. [22:16] TJ-: and I am going to great lengths to circumvent this pointless endeavor [22:16] ash_worksi: hmmmm. nasty [22:16] I'll try #linux then, thanks [22:16] LadyCailin: there's also #ubuntu-offtopic [22:17] LadyCailin: if Ubuntu is possibly impacting your program then this is the appropriate place to ask [22:17] TJ-: I started using expect to just login through a jump server (thank you `ssh -J`), but I ended up needing a number of options [22:17] TJ-: such as --screen [22:17] which starts me in a screen session [22:17] or --user john [22:17] which sudo's to user John [22:18] I basically want to interact with my password manager as least as possible [22:18] anyway, the most complicated one is --reset [22:18] which _can_ take a password but has a default (that another admin sets when I lock myself out) [22:19] so the idea is to be able to run `conn some_server -r` and that should connect, run sudo, chpasswd and quit (I seemed to have problems getting back to `interact` from that point) [22:20] Well, basically, I'm writing a "hello world" level program in LLVM IR, which I'm then compiling to an executable, which segfaults. There isn't really "source code", so gdb isn't particularly useful here, and indeed, the program is only one line anyways, objdump -d shows that the whole program is "exor %eax,%eax/retq", yet it still segfaults. I'm a bit at a loss as to how to proceed with... [22:20] ...debugging this, I imagine that there's something more substantially wrong here, such as using the wrong linker options or something like that, but I'm not sure what to check or where to start. [22:20] LadyCailin: sounds like you're not linking in the stdlib entry/exit functions [22:21] I've overridden the entry point, so that got rid of one of the linking errors, so I suppose that works. But I'm not exiting, I'm returning 0, but maybe that still needs the stdlib as if it were calling exit()? [22:23] The linker options are simple right now, it's `ld.lld-12 -o main --entry=main main.o` [22:29] I can see the linker options clang is using with `clang -v` though, so maybe I just have a look through each of these and figure out what each flag does, and try to apply relevant looking flags until it works. === bloodninja123 is now known as bloodninja [22:34] LadyCailin: you might find help in ##programming for something this detailed [22:36] Heh, yeah, ok. I just tried all the options that clang -v shows, and it still segfaults, so there may be something wrong at a previous stage or something. So, maybe I need to go to #llvm or something. Anyways, thanks for the pointers. [22:48] https://xkcd.com/371/ [22:49] LadyCailin: haha :) [22:49] LadyCailin: did you find it? [22:50] Not yet. I've asked in #llvm, I think this might be a problem earlier in the toolchain, despite the lack of errors. [22:50] aw rats, heh [22:51] 174 of the channel members are still on freenode [22:51] LadyCailin: does your program have an entry point? like _start or what ever it is on linux? [22:52] It does, yes. I've overwritten it though, to be main, using --entry=main [22:52] so.. when I asked freenode ubuntu whether anyone was there... no one answered of the 174 people in channel [22:52] stevessss: I understand most of the connections on 'old freenode' are just zombie clients these days [22:52] stevessss, freenode died due to shenanigans [22:53] stevessss: freenode deleted all accounts and channels [22:58] Ok, figured it out. Apparently _start does some necessary stuff, so you need to link in crt1.o (or one of the other .o files clang seems to link in by default) and then don't override the entry point. [22:58] woot [22:59] I'm going to try to figure out what _start does that's so important, because it's strange that it's a linker option if there's no case where doing it results in a working program. [23:00] LadyCailin: well yes; the stack has to be set up in a particular way, as well as registers [23:00] LadyCailin, Book "Introduction to 64 Bit Assembly Programming for Linux and OS X: For Linux and OS X" mentions that and might be useful as a reference if you are building your own compiler? [23:00] LadyCailin: check out the ELF format [23:01] Yes, I am, I'll add that to my reading list, thanks for the suggestion! [23:01] LadyCailin: seen these things yet? :) https://www.muppetlabs.com/~breadbox/software/elfkickers.html http://www.muppetlabs.com/~breadbox/software/tiny/ -- seems like things you'd like [23:03] Interesting. Yes, I may end up using the defaults that crt0 has, but I would definitely like to understand everything that's going in to the program, so I want to start with "what's the smallest possible program" and work up from there. [23:06] Hello [23:07] I am having trouble with Makefile [23:07] its annoying [23:08] sounds like a Makefile alright :) [23:10] make: *** [Makefile:24: all] Error 1 [23:10] trying to compile UnrealIRCd [23:10] look higher up, that's the end of the problems [23:11] What should i do to fix the problem please sarnold ? [23:11] put the full compiler output in a pastebin [23:11] 1 moment [23:12] https://pastebin.com/CGkdiWqE [23:14] Genius: what TLS library do you want it to use? [23:14] missing an ssl dev library [23:15] i just have to install ssl dev library ? [23:15] can be done with apt-get ? [23:15] and then tell ./configure which one to use, yes [23:16] Genius, sudo apt install libssl-dev [23:16] will likely get you where you want to be [23:16] done [23:16] let me check [23:16] recompiling [23:17] Genius: you should probably re-run ./configure [23:17] I already did [23:17] waiting [23:18] Genius, did it get past the first stop? [23:18] its still working [23:18] no errors yet [23:19] fyi, if you do make -j it may significantly speed up compile times [23:19] in the future [23:19] thanks, noted :) [23:20] you can't speed up ./configure though :( [23:20] it's just slow all the time [23:20] Genius, I once tried compiling the kernel on a rhel8 box wi [23:20] th one core, it took me hours [23:20] but its only unrealircd now :) [23:20] compiling done :D [23:20] i love you <3 [23:20] now doing make [23:21] sarnold, just imagine trying to figure out the concurrency of multithreaded ./configure [23:21] I'd just turn the pc off and go outside lol [23:21] coryc257: if only there were a tool that knew how to sort dependencies and execute things concurrently... :) [23:22] seriously though it'd probably hav eto be pretty difficult otherwise someone would have done it by now === Nitrousoxide_ is now known as Nitrousoxide [23:28] Genius, it make it all the way to the end? [23:30] yes i really appreciate guys :D [23:31] awesome, I'd recommend you contain it in a vm, with with selinux/apparmor if you plan to run it on your machine and open it up to the internet just in case if you have anything sensitive on your machine [23:33] https://www.unrealircd.org/docs/Using_AppArmor_with_UnrealIRCd [23:34] no im running it on a vps [23:38] This article is hilariously written, but it's contains a ton of useful info, this was a great link. http://www.muppetlabs.com/~breadbox/software/tiny/teensy.html [23:39] > Of course, half of the values in this file violate some part of the ELF standard, and it's a wonder that Linux will even consent to sneeze on it, much less give it a process ID. [23:39] Anyways, thanks for the link :D [23:40] LadyCailin: I *really* enjoyed those things when I was working on an ELF project ages ago [23:41] how to open ports with ssh ?> [23:43] Genius: ubuntu does not have a firewall enabled by default. If you are having trouble accessing services, you need to contact your VPS provider for support [23:43] the unrealircd is running but im unable to connect [23:44] Genius: I might add though, given your current level of expertise, I strongly recommend against proceeding further with setting up an IRCd in a public manner until you are a bit more established when it comes to linux and security [23:44] its just the opening ports thing because the linux is fresh its amazon aws [23:45] Genius: ok, then you need to look into your EC2 security groups [23:45] it's lightsale bro [23:46] the firewall is inactive [23:46] i did ufw status verbose [23:46] Genius: your firewall is out of band. There is no firewall enabled on ubuntu by default [23:46] but bro how can i open 6667 port [23:47] it should be enabled udp or tcp ? [23:48] Genius: first, stop with the "bro". Second, you read up on the VPS service you are using on how to manage it properly. Thirst, read the documentation for UnrealIRCd and it's requirements [23:48] lol why so serious [23:49] Genius: because you are not qualified to be spinning up a server which is going to be owned in a matter of minutes and you won't even know. Then the rest of us have to deal with the garbage that will be spewing from it [23:50] Genius: you have shown zero interest in properly researching and understanding what is involved and should be required in building something like this [23:52] Im trying to research myself too, and no need to be that aggressive, im not offending anyone or anything [23:52] so calm down mate [23:52] Hi, I have a problem. Ethernet works during the live session, but after installation, ethernet no longer works. I'm running 20.04.2. These are my specs: https://www.cnet.com/products/dell-latitude-3150-11-6-pentium-n3540-4-gb-ram-500-gb-hdd/ My ethernet is Realtek PCIe GbE Family Controller. I booted from USB, and I used rufus with the .iso. Someone suggested that the kernel upgrade during installation may be my issue. Also, now that [23:52] it's installed, when I turn on my computer there is the error "Invalid partition table!" Then I press enter and then it boots. A kind fellow suggested that I try using the GA kernel option, and explore the partition table from live media. Not entirely sure how to follow those two suggestions, but I know the first step is reinstalling from the usb. [23:52] you asked [23:54] Genius: ssh can open ports with -L. See man page for details [23:55] cbreak: I assume you're referring to a local ssh tunnel, which isn't what Genius is referring to [23:56] he didn't say what he'd want to open the port for [23:56] cbreak: he's trying to spin up an IRC server in AWS [23:56] and since he said ssh and not sshd, this is my closest guess [23:57] that makes no sense. [23:57] cbreak: by "with ssh" he means from command line [23:57] he'd want ircd to open a port for that, not ssh [23:57] hmm :/ [23:57] cbreak: he's very green and doesn't know what he's doing [23:57] seems like it, yes [23:58] yeah honestly i dont know how to do it [23:58] which is why he shouldn't be doing this. Not in a publicly accessible way anyway [23:58] Genius: I HIGHLY recommend playing with this in a VM or even an lxd container on a linux machine on your local network. Get it working and get a feel for how to set it up and lock it down properly [23:59] one question, when i allow a port i should choose udp or tcp [23:59] * leftyfb sigh