[00:39] oerheks I just saw your message. Unfortunately I've hit a brick wall with mkchromecast. https://github.com/muammar/mkchromecast/issues/357 I have the same issue as the OP, but I'm a newbie when it comes to computers and I couldn't figure out the fix. [00:39] Issue 357 in muammar/mkchromecast "Invalid MIT-MAGIC-COOKIE-1 Cannot open display :0.0+0,0" [Open] [02:00] I am trying to upgrade from ubuntu 18 to 20, but when I execute `sudo do-release-upgrade` it says "Please install all available updates [...] before upgrading."  The 2 updates are for libgnutls30 and libsrt1.4-gnutls [02:01] sudo apt update && sudo apt dist-upgrade # and reboot [02:01] When I try to update/upgrade those, it just says they have unmet dependencies and recommends they stay at the current version. [02:01] Hm ok, I will try that [02:01] apt install -f # should be in that error message, no? [02:01] !yy.mm | Guest46 [02:01] Guest46: 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 [02:02] 18.04 to 20.04, my bad [02:03] oerheks I tried using -f, no that was not in the error message [02:03] Sounds like you maybe installed some things from external repositories [02:04] ..if so you can ppa-purge to remove those, upgrade, then add them back in after [02:04] hm, maybe. Is it possible to trace that? [02:05] Have y'all been hearing about apt getting unauthorized github URLs? [02:05] apt-cache policy [02:05] sem, no. [02:06] hear-say is mostly FUD [02:06] I am going to try and install the fixed version from the ppa, but wondering what the procedure is for getting the fix in the normal repos? Maybe it is already fixed? [02:06] oerheks, it is affecting me [02:06] https://paste.ubuntu.com/p/pM5BTwCkCX/ [02:06] now you name PPA, that is beyound our responcibility [02:06] hey relax, i'm just asking a question [02:06] i haven't used any ppa yet [02:07] i mean besides the one that apt is failing on [02:07] you should join #debian ? [02:07] this is ubuntu [02:07] here i'll find the bug ticket [02:07] sem: ask whoever runs surface linux what the new archive is? [02:07] I haven't read this yet: https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1950095 [02:07] Launchpad bug 1950095 in apt (Ubuntu) "[github] 20.04: Apt fails to download URLs with non-encoded querystrings" [Undecided, Fix Released] [02:08] sem: you need to remove all of those 3rd party repos [02:08] leftyfb, is that a requirement for asking a question in #ubuntu? [02:08] sem: your issue is with those repo's, not with ubuntu. Remove them and the issue goes away [02:08] sarnold, looking at the surface-linux bug, they haven't changed the github URLs but someone has patched apt [02:09] sem, did you install apt-transport-https ?? [02:09] https://github.com/linux-surface/linux-surface/issues/625 [02:09] Issue 625 in linux-surface/linux-surface "Ubuntu: Apt update fails with Error 401 Unauthorized" [Open] [02:09] oerheks, no haven't installed it [02:09] not manually anyway [02:09] then you will never get https souces.. [02:09] c/sources [02:10] !info apt-transport-https [02:10] apt-transport-https (2.3.9, impish): transitional package for https support. In component universe, is optional. Built by apt. Size 2 kB / 163 kB [02:10] I'll try adding that [02:10] what release are you on? [02:10] by focal: [02:10] This is a dummy transitional package - https support has been moved into [02:10] the apt package in 1.5. It can be safely removed. [02:11] I forget how you find out, is it "usb-release" ? [02:11] lsb_release -a [02:11] Ubuntu 20.04.3 LTS [02:11] sem: cat /etc/os-release | nc termbin.com 9999 [02:12] sarnold, I see. Have you taken a look at the launchpad bug? (If you're interested) I am going to start reading it now [02:14] to all the ubuntu and larger linux community, thank you. i had to run a proprietary windows program for a customer, new customer, first windoze contact in a year. [02:14] fresh win7 install fully updated @ 4-6 hours. thing locks up constantly [02:14] It looks like I can try the ppa for apt, and see if I get any regressions ;) [02:15] after the last aborted shutdown i said silent praise for ubuntu [02:15] sem: just now read it. bummer :(* that looks gross [02:15] it is weird for sure! [02:15] sad that i have to use windoze for the customer [02:15] !support | jwash [02:15] jwash: The official ubuntu support channel is #ubuntu. Also see http://ubuntu.com/support and http://ubuntuforums.org and http://askubuntu.com [02:16] !ot | jwash [02:16] jwash: #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! [02:37] if I have no wifi card can I mask wpasupplicant? [02:38] I don't think you need wpasupplicant if you don't use wifi [02:38] but I am not sure what mask means [02:38] disable the server completely [02:38] service* [02:38] I am pretty sure you can do that, but not an expect [02:39] *expert' [02:39] I'm pretty sure too [02:40] probably it's fine [02:44] Hey, channel. [02:44] Is there an Ubuntu package to install gdb with ARM support on an x86-64 system? [02:44] I'm running Ubuntu 20.04.1. [02:44] sidenote: 20.04.3 is current, update properly please [02:45] btw what does it mean for gdb to 'have arm support' when running on a non-arm platform? [02:46] !info gdb-multiarch [02:46] gdb-multiarch (11.1-0ubuntu2, impish): GNU Debugger (with support for multiple architectures). In component universe, is optional. Built by gdb. Size 4,369 kB / 17,178 kB [02:46] sarnold, It means I want to debug a binary running on an external target, connected with a debugger. [02:47] gdb-multiarch reports "unknown architecture: arm" [02:47] afaik there's no 'arm' arch, there's armel, armhf, aarch64 [02:47] ubuntu doesn't do armel [02:47] F14t_Looks: is it connected via jtag? serial? ssh? I've never tried anything this fancy :) [02:47] sarnold, +1 [02:48] You know how to get GDB to print a list of its supported architectures? [02:49] USB-connected debugger. "single wire debug" protocol, SWD. (The debugger connection works fine. I can use it to program the target.) [02:51] oerheks: No info for gdb-multiarch. And gdb-multiarch is installed (so says apt -qq list gdb-multiarch). [02:56] man there's way fewer files in gdb-multiarch than I expected [02:56] thanks for the single-wire debug :) I've not heard of that before [02:56] uh huh...seems like, doesn't it? [02:56] I saw a report that cross-GDB now has to be installed by hand now. [02:56] https://itectec.com/ubuntu/ubuntu-how-to-install-arm-none-eabi-gdb-on-ubuntu-20-04-lts-focal-fossa/ [02:56] I was really hoping somebody in this channel would tell me that wasn't the case... [02:58] sarnold: yeah. Know what's easier than hooking up 6 different connections to 6 IC-specific pins for debugging? "single wire" is a serious selling point. (it also needs a clock line. truth in advertising only goes so far I guess...) [03:01] hrmph. I see 'set architecture arm' in this old bug report https://bugs.launchpad.net/ubuntu/+source/gdb/+bug/1233185 near the debian/rules that's doing the configure step.. [03:01] Launchpad bug 1233185 in gdb (Ubuntu Trusty) "gdb-multiarch cannot read ARM cores: 'wrong size gregset struct in core file'" [High, Fix Released] [03:01] sorry, you had that right :) [03:04] --enable-targets in the buildlogs has a long list of architectures https://launchpadlibrarian.net/491828005/buildlog_ubuntu-focal-amd64.gdb_9.2-0ubuntu1~20.04_BUILDING.txt.gz [03:05] $ gdb-multiarch --configuration [03:05] This GDB was configured as follows: [03:05] configure --host=x86_64-linux-gnu --target=x86_64-linux-gnu [03:05] ... but it doesn't seem to show up in the built binary? [03:07] Right. I saw that. I think the gdb command was "show configuration" [03:22] So, what's the moral of the GDB story? [03:22] If I want cross-GDB I have to install the arm.com binary package by hand? [03:22] https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnu-toolchain/gnu-rm/downloads [03:22] Or build it myself? (Hmm. There's no gdb-multiarch source package for focal.) [03:23] F14t_Looks: it's built from the gdb source package [03:24] F14t_Looks: it's probably worth a bug report, too; it sure doesn't feel like there's something simple that we're overlooking [03:29] when checking for dependencies, what's the difference between Depends and |Depends? [03:30] Guest68: | are alternations -- one of the choices within the commas will satisfy that block [03:30] Hi! I'm running Impish and the 6.1.26 version of VirtualBox won't allow me to look for a ISO image to add it to the machine because nothing is clickable in the fil chooser window. [03:31] They haven't released a deb for Impish yet so I can't upgrade to 6.1.28 to check if the problem is solved. [03:32] Unless there's a Nightly PPA somewhere for VirtualBox. [03:32] sarnold: Thanks. [03:35] sarnold: thanks [03:36] Umeaboy, there is no PPA, just wait for the .28 build === frank_ is now known as ffendland [03:44] oerheks: Can you confirm that you also have that problem with 6.1.26? [03:46] I'm pretty sure oerheks uses qemu/kvm through libvirt, not virtualbox [03:47] i cannot confirm it, indeed [03:48] Here's the output I get when the file chooser window appears to find the file: https://pastebin.com/xfAuicY5 [03:48] is this on wayland or xorg? [03:59] Seems like I'm using Xorg. [04:00] At least according to about:support in Firefox. [04:04] Umeaboy: ' echo $XDG_SESSION_TYPE ' . [04:13] Bashing-om: x11 [04:16] Umeaboy: Then yeah - confirms on Xorg :) [04:17] What's next? [04:17] probably time to file a bug with virtualbox folks [04:18] I could try to build Virtualbox locally. [04:18] Just to see if the issue is already fixed. === Ricardus_ is now known as Ricardus [04:37] If packages A, B, and D are installed, and C is not installed, and then I do  aptitude why A  and it answers B: B depends C | D, and C depends A.  Then isn't that wrong?  I've got D, why does it need A when I don't even have C? [04:41] because package B [04:41] else give a usefull example, not AB C D [04:42] but | means or. So, D should satisfy B [04:42] B depends C, you wrote [04:43] C | D  (that's the output of aptitude, but, anyway yea without a literal output you cant see it clearly enough i understand === gschanuel3 is now known as gschanuel === stoned is now known as Hash [05:27] can someone recommend software like hamster to track time spent on tasks [06:02] brandoneliza, sudo snap install hamster-snap [06:02] https://snapcraft.io/hamster-snap [06:03] and.. read the tips ! [06:03] and.. read the tips ! === EdFletcher is now known as EdFletcherT137 === EdFletcherT137 is now known as EdFletcher === diskin is now known as Guest3597 === diskin_ is now known as diskin [06:46] oerheks, thanks I will try. [06:59] oerheks, it works [07:12] !info draai [07:12] draai (20201215-2, impish): Command-line music player for MPD. In component universe, is optional. Built by draai. Size 46 kB / 144 kB === Abrax- is now known as Abrax === gschanuel8 is now known as gschanuel === zhuo_ is now known as piposhena === kostkon_ is now known as kostkon === pah_ is now known as pa [09:25] which openssh server should I install? apt gives me 4 options   openssh-server:i386 1:8.2p1-4ubuntu0.3, openssh-server 1:8.2p1-4ubuntu0.3, tinysshd 20190101-1build, lsh-server 2.1-12build3,  dropbear-bin 2019.78-2build1 and I just want to standard one [09:26] 5 options sorry [09:26] openssh-server is the "regular" one, if you can call it that [09:29] thx, now apparently I broke my packages and can't fix it https://dpaste.org/hRXZ what should I do here? [09:32] Hi [09:32] I've just upgraded to 21.10, but i cannot switch workspaces anymore? [09:32] sounds like two different, third-party, apt-sources are colliding with conflicting packages [09:38] mmm does it means, 2 different repos offer same package and apt does not know which one to use? [09:40] masber: what does this output? apt-cache policy containernetworking-plugins kubernetes-cni [09:40] https://dpaste.org/Ln6h [09:41] why is opensuse in there? [09:42] tail -n+1 /etc/apt/sources.list /etc/apt/sources.list.d/*.list [09:43] yeah, didn't know opensuse provided deb packages [09:47] https://dpaste.org/J9qt. [10:06] hi folks, when i boot from a live usb stick i am able to use the ethernet but it is not recognised when i boot from the disk [10:06] any suggestions on what to look at [10:06] 20.04 [10:08] is this an UEFI machine? fastboot enabled? [10:09] secureboot should be oke, still you can disable this to see if this makes a change [10:09] ok i need to check what the above means [10:10] should i output some grab config? [10:10] or is it grub [10:12] no need for the grub config [10:12] !uefi [10:12] UEFI is a specification that defines a software interface between an operating system and platform firmware. It is meant as a replacement for the BIOS. For information on how to set up and install Ubuntu and its derivatives on UEFI machines please read https://help.ubuntu.com/community/UEFI [10:12] see your bios [10:18] fast boot disabled [10:18] not sure how to see in the bios if it is UEFI [10:18] oh wait i see storage boo option control -> UEFI Only [10:18] so i guess it is UEFI [10:19] oke, does disabling fastboot solve it? [10:19] it is not enabled [10:25] gebbione: did you interrupt any updates, e.g. did you forcibly shut down while unattended upgrades were running? If so, you should "chroot" into your installation from the live cd, and run `apt full-upgrade` to complete the updates [10:26] alkisg, i cloned a disk from NAND to NVME and using it but with both disks the ethernet does not work [10:26] also not sure why one block of ram is not recognised, it is not the RAM, it is the board because if i swap them the block works [10:27] but always only one [10:27] i see the board has two BIOS updates [10:27] maybe they will fix the problem? [10:28] i ll reboot from a live USB and see but for the ethernet it should just work really [10:28] gebbione: what is the output of `uname -a` from the live cd, and what is it from the booted disk? [10:29] checking , will take a few minutes === Andrew_ is now known as AndrewYu [10:33] from disk -> Linux machineName 5.4.0-90-generic #101-Ubuntu SMP Friday Oct .... x84_64 ... GNU/Linux - i guess the time is off as it is thursday here [10:34] hi. I'm using Ubuntu 20.04 LTS. I am interested in changing the login screen if that's possible.  I have found different methods searching online, though I'm not sure what is most current or "canonical" method (pun intended).  FWIW I am using i3wm as well. [10:39] from usb -> 5.11.0-27-generic #29~20.04.1-Ubuntu ... augusti 11 === oakyy is now known as oaky [10:45] alkisg, so different kernels, does that make really a difference? [10:46] gebbione: yes, it's possible. Try to use the same kernel. [10:47] i need to connect it to update it :) [10:48] maybe that bios update fixes things.. [10:48] but it is weird the usb gives a higher kernel [10:56] is evolution really the only open source mail client for linux that works with exchange/oauth2? [10:57] what on ubuntu handles creating the /etc/resolv.conf file? [11:06] do not edit /etc/resolv.conf, but /etc/resolvconf/resolv.conf.d/head , after restarting the service it will be populated , one of the many howto's https://www.tecmint.com/set-permanent-dns-nameservers-in-ubuntu-debian/ [11:06] alkisg, if i run an update on my ubuntu it does not see the new kernel versions and does not offer to upgrade it. Shouldn't this happen automatically? [11:07] gebbione: if that's 20.04, run: sudo apt install linux-generic-hwe-20.04 [11:07] It'll get you the 5.11 kernel after reboot [11:08] thank you, lets see if that fixes the ethernet [11:08] i plugged in the wifi to have connection from it for now [11:09] about the ram not being recognised, one slot works but not the other in dual channel, what can i check apart from updating the BIOS ? [11:09] nothing, update that bios. it is of no use looking further untill you do. [11:10] about the kernel will it become effecting only after reboot? [11:10] or wait it is asking for it [11:10] going to run that first [11:12] kernel update fixed it [11:12] thank you [11:18] hey, how can I kill my kde desktop session via ssh? basically, like logging out? [11:20] kquitapp5 plasmashell seems to work half-way [11:28] cbreak: systemctl --user stop user@XXXX. ? [11:39] Is it possible to boot with wifi off by default? I have done this with bluetooth by modifiying /etc/bluetooth/main.conf but can't figure it out with wifi. Using 20.04 [11:43] Guest573: there's a service that remembers the setting in rfkill and sets it, so "rfkill block X" then after a reboot that setting would be applied [11:44] so is it acceptable to have a script run at shutdown that would do that, or is there a more orthodox approach? [11:53] so i update my MOBO bios, now the full ram is recognised but ethernet is not working again and WIFI settings are forgotten :/ [11:53] damn [11:59] restarting what service? [11:59] I didn't say I needed to change resolv, I need to know what sets it so I can refresh it entirely to reflect new dhcp settings [11:59] gebbione: ethernet not working as in the pci device is not detected, the driver not loaded, or configuring the interface fails somehow (how)? [12:00] gebbione: and which wifi serttings are being lost? [12:07] ist here a partiuclar reason all latest ubuntu relases misuse systemd-resolved, forcing the `~.` search domain to be set on all DHCP managed interfaces? === rexbinary_ is now known as rexbinary [12:13] open [12:14] is there a mailing list where I could ask that question in a more directed manner? [12:20] Hi Everyone, I have Dell XPS 15, Ubuntu 20.04, when I lock screen, the WIFI turns off, how can I disable this? [12:24] tomreyn, using network manager (ie click on the right top) i have no ethernet connection or device showing. Also Wifi I had set it up in a previous session, added the password and it would not connect on its own when i re-enabled wifi. FYI it is a Gigabyte B550 AORUS ELITE AX V2 and i just updated the BIOS, the memory was not fully recognised before but now it is fine. As suggested by @alkisg i had already run sudo a [12:24] pt install linux-generic-hwe-20.04 and this had fixed the ethernet temporarily but it is not turning on again now [12:25] bluetooth is on by default when i boot, so another thing i might disable in the future [12:25] i wonder if something wireless related gets in the way [12:27] gebbione: NM by default is disabled from managing wired ethernet devices ! [12:30] TJ-, just for clarity. I have never had the ethernet disabled when i boot up unless a connection problem exists and it would auto-connect. I call the tooling that does this and shows its status network manager but I might be wrong ... it is just the on the top right where the networking information is displayed but not present right now [12:32] gebbione: on desktops Ubuntu uses NetworkManager but on 'server' installs uses systemd-networkd. For NetworkManager Ubuntu ships a file that disables managing wired links: /usr/lib/NetworkManager/conf.d/10-globally-managed-devices.conf [12:35] so , I googled and found this command `apt rdepends ` I want to read about it to understand what it does. Neither man apt not apt --help shows that the rdepends switch? why? [12:35] i just run - lash -C network and shows the Realtec controller as unclaimed, investigating [12:36] might need to install an extra driver [12:41] masber: rdpends comes from and is part of "apt-cache" - see "man apt-cache" - it reports reverse dependencies of a package (that is, Depends, Recommends, Suggests) [12:41] Hi Everyone, I have Dell XPS 15, Ubuntu 20.04, when I lock screen, the WIFI turns off, how can I disable this? [12:41] gebbione: aha! that'd cause a problem for sure. You've likely got a DKMS out-of-tree driver package for it that didnt' build successfully with the latest kernel [12:42] Guest71: That does sound abnormal but, it could be due to some power-saving settings [12:42] Guest71: have you added any tools like laptop-mode-tools, or changed any power settings recently? [12:43] TJ-, should i install the driver? the driver says up to kernels 5.6 but i have a 5.11 now [12:43] if its so clever than apt rdepends means apt-cache rdepends... then, why is it not so clever with man or --help? I just dont understand this nonesense [12:43] actually i have a 5.4 [12:43] not sure why [12:43] i will rerun the kernel update [12:43] it is just not consistent [12:43] TJ- ok got it, I have TLP, i'll check, thanks [12:44] masber: you'd have to ask the Debian apt team [12:44] would that be a wise thing to do? ie run - sudo apt install linux-generic-hwe-20.04 - again? [12:45] gebbione: no, first identify how it should be configured. Did that interface work under an earlier kernel for example? If so, that tells us there should still be a driver available [12:46] TJ-, it works only when i run a live usb disk, then i did run the install of the above package and it worked suddenly. But after i update the bios (and this to me is totally unrelated to the driver) it is back to not working and the kernel is 5.4 [12:46] i could try installing the driver from reattach [12:47] gebbione: does it have any networking, via Wifi for example, so we can collect some info from it? [12:48] driver installation worked [12:48] i had yes but looks fine now with the driver from reattach [12:48] i will just reboot that machine to double check it stays ok [12:50] nice all sorted [12:51] going to attempt fast boot configuration now [12:51] fastboot is a no no, secureboot should be oke [12:51] !uefi [12:51] UEFI is a specification that defines a software interface between an operating system and platform firmware. It is meant as a replacement for the BIOS. For information on how to set up and install Ubuntu and its derivatives on UEFI machines please read https://help.ubuntu.com/community/UEFI [12:51] i ll check [12:51] at the moment it prompts me to select the OS [12:51] don't want that [12:52] is XMP a good thing to enable for RAM ? [12:53] TJ-: thanks, that probably would have worked. I just went around and killed all the things :) [12:53] apparently the ram is going slower than its declared speed ... 2667 instead of 3600 [12:56] gebbione: 3600 is much faster than DDR4 spec allows [12:56] you'd need overclocking, or at least XMP for reaching those frequencies [12:56] did you update your bios yet??? [12:56] and even then it's a gamble if your CPU / Mainboard / ... supports it without crashing [12:57] if not, have fun with all sorts of issues. [12:57] enabled it, oerheks yes it is now recognised so i have the full 32GB [12:57] 2x16 [12:57] I've not seen good (ECC) ram that goes that high [12:58] the first update might even unlock features .. seen that before [12:58] this is what i have btw CMW32GX4M2Z3600C18 [12:58] smoke crack? [12:59] VIA, move to -offtopic please. [13:01] dmidecode -T memory shows ... Speed 3600 MT/s [13:08] bios is still seeing slower speed ... not sure what is the true output [13:10] where to send requests for package bumps ? [13:11] https://packages.ubuntu.com/impish/bats desperately needs a bump, parallel processing is broken with the current version [13:11] !sru [13:11] Stable Release Update information is at https://wiki.ubuntu.com/StableReleaseUpdates [13:11] see that wiki, for update request [13:15] i have set up GRUB_TIMEOUT=0 [13:15] but i m still getting the prompt from grub [13:15] :/ [13:20] i have also run update-grub but it won't just boot ... still prompting for boot options [13:26] 'Morning folks === global_elites_ is now known as global_elites === Allie_ is now known as Allie === Razva_ is now known as Razva [14:06] Is there a way to have the disk unlock screen show up on a plugged in monitor during boot? [14:06] I have to open my laptop every time I boot to unlock the drive. [14:07] echo $home [14:07] merpnderp: that comes from the initialramfs code, via plymouth, at which point the output may not be available or known [14:08] mabye hitting escape would work, though? [14:08] i.e. entering it on the text screen [14:10] if this also doesn't work: all you really need to know there is that you're definitely at the passphrase prompt, maybe there could be another way to signal that (LED, audio etc.) [14:10] merpnderp, does your BIOS give an option, what screen/ext screen to be primairy? [14:11] merpnderp: you could add a key-file to the LUKS container, put it on an external USB device, ensure a keyscript reader is used, and then you just plug it in (Can do that easily with a Yubikey generating a key too) === Guest1118 is now known as Znevna [14:29] exit === kostkon_ is now known as kostkon [14:56] Can anyone explain me line 38 to 61 here on focal : https://github.com/eilandert/dockerized/blob/master/base/Dockerfile-focal ? [14:57] Good morning. I recently upgraded from xubuntu 18.04 to stock ubuntu 20.04.3. I seem to be getting this random "boop" sounds. How do I figure out where they're coming from? [14:59] Hello, how is /tmp/ handled in ubuntu? [15:00] puff: 'boop' ?! [15:02] mithrin: quite well. [15:03] mithrin: it's a tmpfs by default, if that's what you mean [15:04] mithrin: systemctl cat systemd-tmpfiles-clean.timer [15:04] mithrin: systemctl cat systemd-tmpfiles-clean.service [15:04] mithrin: basically, cleaned on a scheduel === ^wuseman is now known as Guest8052 [15:16] TJ-: Yup, just happened again. It's not the google chat notification sound, it's a bit deeper than that, maybe 2-3 tones but very quick/short/jammed together. There's no popup that I can see. [15:16] puff: probably a notification sound [15:17] TJ-: When I get the boop whenI happen to be in this terminal window, where I can see the Ubuntu launch bar (or whatever you call it) on the left edge of the screen, sometimes I think I see a flicker of motion on the launch bar, but I don't know what. [15:17] TJ-: Is there any sort of centralized spot where I can see all of the notifications? [15:18] for terminal, there is only the terminal bell in its own settings [15:19] puff: Not sure, I don't use Gnome. What IRC client are you using - it could be notifying you of something, like your nickname being mentioned in a channel that doesn't have the focus [15:20] Hm,g ood point. [15:20] Emacs erc, running on a virtual server (in a screen session) via terminal window &ssh. [15:21] puff: so not complicated at all then :P [15:22] Hm, okay,s o that's one, yeah... except that I just got a second boop. [15:23] It booped, I alt-tabbed back here to see your comment, and then it booped again, almost immediately. So unless it's going off twice per comment or something. [15:23] tomreyn: I haven't seen /tmp/ being tmpfs by default on debian/ubuntu, while I think it is in some other distributions. Is there a new default somewhere, e.g. in -server, that I"m unaware of? [15:25] Is "launch bar" the right term for the menu bar at the left edge of the screen? [15:25] alkisg: hmm, i just assumed that's a default because i have that on a (previously release upgraded) 18.04 LTS system, but i may actually have set this up on my own, let me check what a fresh install does. [15:25] Yeah, definitely getting boops from some other app. [15:26] boops? [15:27] tmpfs isn't the default on /tmp/ - I always have to configure that myself. It was about 10 years ago I think [15:28] BluesKaj: Like beeps only deeper sounding. [15:30] so, yes, that was wrong, sorry, /tmp is not a tmpfs by default [15:30] alkisg: ^ [15:35] puff, ok, I don't bother with pc noises, I have them turned off [15:36] puff: maybe Betty Boop has got into your PC? [15:36] heh [15:38] puff, sorry, you have sounds and dont want them, or you want more? [15:38] Hello all, I have an issue with 'agetty' service on Ubuntu 20.04. It looks like a Hardware problem as everything checks out. I have a few devices and one has the issue throwing an error to /var/log/auth.log: "agetty /dev/console not a tty" I did an strace of the getty command and got an error "ioctl(3, TCGETS, 0x7ffcb52c1670) = -1 EIO (Input/output error)" === pah is now known as pa [15:40] jwhitmore: what is trying to start agetty on /dev/console? [15:40] The driver doesn't like the TCGETS request, [15:40] TJ-, systemd I assume as the kernel parameter list in /boot/grub/grub.cfg has console=ttyS0... [15:41] jwhitmore: hmmm, I have devices that do that and haven't see that error [15:43] I have multiple devices that are happy with exactly the same config but one is not happy... I can't see the issue so I'm assuming it's a Hardware issue. I'm only asking in case I've missed something not obvious to me [15:43] jwhitmore: you've got a " systemctl status seria-getty@ttyS0.service " with a failure? [15:43] jwhitmore: and similar for the problem device? [15:43] hhhm, mistype there! serial not seria ! [15:43] console-getty.service is showing the error [15:45] Like I say looks like the driver isn't happy with I/O request to TCGETS to get the serial port details [15:46] But during boot syslog has "Serial: 8250/16550 driver, 32 ports IRQ sharing enabled [15:46] jwhitmore: you could edit the console-getty.service and try removing the TTYReset=yes [15:47] jwhitmore: as in "systemctl edit console-getty.service" [15:47] TJ-, Cheers I'll try that. And thanks for the "as in" suggestion that wasn't what I was going to do. [15:48] how do I refresh the resolv.conf file? [15:48] jwhitmore, sy, why grub.cfg has console=ttyS0 ? [15:48] I've updated the dhcp options on the router [15:48] I need the system to regain them? [15:48] ioria: serial port console on headless device [15:49] oh, ok [15:49] e.g. " root=/dev/mapper/VG_OS-bullseye ro debug systemd.log_level=debug console=ttyS0,115200n8 " [15:51] TJ-, your "as in" systemctl edit console-getty.service gives me a blank file? [15:52] no typo there? [15:52] jwhitmore: yes, you add the entries you need (but it should show some # comments lead-in) [15:52] oh actually that#s normal [15:52] cluelessperson, systemctl restart resolvconf.service [15:52] 'edit' usually puts the original file in there, with each line commented out [15:53] so I should just add the line "TTYReset=no" and save that [15:53] jwhitmore: add [Service] and TTYReset=no then save [15:53] jwhitmore: it has to have the Section headers for the configs you override [15:54] jwhitmore: when saved systemctl will do an automatic daemon-reload internally, so you just need to restart the service and check its log [15:54] tomreyn: thanks for verifying! :) [15:54] jwhitmore: "systemctl restart console-getty; systemctl status console-getty" [15:55] FYI anyone wanting to know how to set up a headless boot with serial console, with GRUB and Linux accessible: https://termbin.com/9nhz [15:56] oerheks, that doesn't exist [15:57] cluelessperson: you shouldn't be touching /etc/resolv.conf [15:58] cluelessperson: that should be pointing to the systemd stub resolver. Whatever is managing the network should inform systemd-resolved of any changes [15:58] TJ-, "I've updated the dhcp options on the router and I need the local system to update them" [15:58] cluelessperson: is it a Desktop (NetworkManager) or server (systemd-networkd) install ? [15:59] cluelessperson: I saw that, so you need to tell the local network management tool to redo the lease [15:59] Desktop [16:00] TJ-, :( no joy still have the same error [16:01] cluelessperson: if it is NetworkManager, then "nmcli con" to list the links, then "nmcli con down " and "nmcli con up " [16:01] jwhitmore: hmmm! what is the device behind /dev/console for that? [16:02] well /proc/consoles ttySO -W- (EC p a) 4:64 and tty0 -W- (E p) 4:1 [16:03] so I assume ttyS0 is the default that /dev/console is using [16:03] hello, i have a taichi x570 mobo and the usb type-c port on the rear panel doesn't seem to work. The other usb 3.2 type-a port does work [16:03] i'm testing with a cable which i verified works, what steps can i do to try and debug this? [16:04] TJ-, do you by chance know if there's a method to simply renew the lease without downing the entire connection? [16:04] and I assume the 4:64 is the 8250/16550 driver [16:04] cluelessperson: I don't /think/ NM has a way, "nmcli con reload" only reloads the local configuration file. Have a look in "man nmcli" at the nmcli connection ... options [16:05] TJ-, Thanks a million for your help. Like I say this works on other Hardware so I'm assuming this is bad hardware and move on. Can't think of another reason why this unit be different [16:05] jwhitmore: could this be beause another getty already has it exclusively? [16:05] LordDoskias: lsusb to check which root hubs there are, and when you have a device connected, whether it's listed there. but this can also be just a hardware question - it sounds like one. join #hardware for those [16:06] TJ-, yes that's possible as systemd starts getty for each serial port [16:06] tomreyn, the usb hub is there, i have 2 ports on it - type c and type a, type a seems to work - testing with an usb device, type c2 doesn't though [16:06] LordDoskias: "not work" how? [16:06] tomreyn, doesn't enumerate the device, doesn't deliver power either [16:07] the device is my mobile phone :) [16:07] LordDoskias: power supply or firmware issue, i'd guess. [16:07] but it doesn't enumerate it either? [16:07] and of course this works on my laptop which also has a usb type c port... [16:07] and is running ubuntu 18.04 [16:07] (my pc runs 20.04 ) [16:08] LordDoskias: could it be disabled in firmware ? [16:08] TJ-, that's what i thought, updated my bios, looked around the menus i didn't see anything usb-related apart from boot order stuff which is irrelevant in this case o\ [16:09] jwhitmore: I'm intrigued as to why the console-getty.service is starting. On my headless units it isn't - is there something on the kernel command line other than the console=ttyS0 and so on? [16:09] can i derive the number of ports via lsusb -vv ?> [16:09] TJ-, actually doesn't look like it the services getty@ttyS0 is masked as is gett@ttyS6 service [16:10] LordDoskias: there may be some clue in the kernel log. [16:13] boot kernel parameter line is - /vmlinuz-5.4.0-47-lowlatency root=/dev/mapper/vgdata-lvroot ro console=tty0 console=ttyS0,115200n8 rootdelay=5 ipv6.disable=1 [16:14] TJ-, my understanding was that if the kernel parameter list contains console then systemd auto starts the service to manage it [16:14] jwhitmore: didn't you say that system is headless? [16:15] Sorry I never said but yes this is headless [16:15] jwhitmore: I'm comparing it against my own headless routers that also use serial console [16:15] jwhitmore: then why the "console=tty0" ? there is no tty0 [16:15] LordDoskias: do you have the 8-pin 12V power connector connected? [16:15] jwhitmore: take a look at my config: https://termbin.com/9nhz [16:16] you can have two the second one is the default but I think kprintf will go to both [16:16] tomreyn, i would have to open the case to see, but i'm sure i do because my videocard is somewhat power hungrey [16:16] jwhitmore: but my point is there is no hardare device to provide /dev/tty0 is there? [16:16] the thing is the usb a port works, and it's on the same root hub [16:17] this leads me to believe the hub itself works ... [16:17] TJ-, Those two console commands are working on other units. Hold on I look at your config [16:17] i'm beginning to the thing the usb type c port is fucked i.e faulty solder ... [16:17] huhz [16:17] jwhitmore: OK ... this area always makes me feel a bit fuzzy in the head! but I'm trying to spot obvious issues. [16:18] jwhitmore: have you closely inspected the system boot log ("journalctl -b") for clues? [16:19] I'll go through it again thanks a million for your help again. I'll go through the logs with a strong coffee :) [16:22] jwhitmore: I think I'll join you with the coffee!! === edro is now known as edr [16:33] Hi, it's been some times (happened around the 27 or 28 October I think) now that I can't properly boot, it's hanging before the logging screen. If I tap esc a bit before it hangs, it shows the lines as expected then at some points erase all and just let a _ and hangs again. I can go to command line through recovery mode, or even boot if I take an older kernel version (5.13.0-19, what is strange is that 5.13.0-20 worked at some time, but [16:33] not any-more, and new update to 5.13.0-21 doesn't solve). I use Nvidia drivers. Does someone knows of any similar issue? [16:42] ldsh: how did you install nvidia drivers? [16:43] What I see is that my laptop (Ryzen 7, 5800) with a GTX 3060 running 21.10 gets incredibly slow with an external HDMI monitor attached and closing my laptop lid. [16:44] tomreyn, Update manager, extra drivers (sorry, not sure how they translate in english), when I open from kernel 5.13.0-19, it shows metapackage driver nvidia-driver-470 [16:45] hello [16:46] ldsh: okay, i guess that should have worked then, so it's probably a bug [16:47] I recently installed Ubuntu 21.10 with ZFS on a laptop with 8G of RAM. I know ZFS is memory hungry but from what I understand it should be scaling based on available memory. However, I had Firefox open and had some expensive tabs open using about 2G of RAM and saw my swap go to 100% usage and everything chug. When I closed everything I had open, my RAM usage hardly dropped. I'm now wondering if [16:47] Ubuntu just requires 4-6G of RAM to run or if there's something else I should be looking at? [16:47] apt install sospspwpspspsosodjdncndsij [16:47] Hmmm same slowness happens when i only select HDMI as display. Slowness is unbearable [16:47] Guest11: do you have an ubuntu support question? [16:49] tomreyn, It worked for years like that, I have no idea what created this issue. I don't recall having made anything special else than the normal updates. Just cleaned inside my computer (but I doubt I created problems as Windows works ans kernel -19 works) [16:50] ldsh, are you fully updated ? sudo apt update && sudo apt full-upgrade [16:51] ioria, I didn't try the sudo apt full-upgrade, only the sudo apt upgrade. Is it fine to do from the -19 kernel or should I run it from command line in -21 kernel? [16:51] ldsh, no problem in that [16:51] ldsh, just post the output [16:52] ioria, it says nothing to install [16:53] ldsh, dpkg -l | grep nvidia | nc termbin.com 9999 [16:54] ioria, https://termbin.com/k7i0 === beaver is now known as pong [16:55] javaJake: 4GB RAM is the minimum https://help.ubuntu.com/community/Installation/SystemRequirements [16:56] ldsh, apt -s autopurge | nc termbin.com 9999 [16:56] javaJake: with zfs and a modern web browser and multiple tabs / expensive rendering you can probably hit 8 GB somewhat easily. [16:56] ioria, https://termbin.com/i71z [16:57] sorry, it needs sudo, I rerun [16:57] ioria, with sudo: https://termbin.com/ymus [16:59] ldsh, no it's ok; so what i'd suggest is to remove All the nvidia leftovers and reinstall 470 [16:59] ioria, this list are all installed? https://termbin.com/i71z [17:00] ldsh, what ? nope [17:00] ldsh, maybe you mean this : https://termbin.com/k7i0 [17:01] ioria, yes, sorry, wrong link, I meant https://termbin.com/k7i0 [17:01] ldsh, yes, but not only installed ; also removed but not purged === thegodsquirrel is now known as thebotsquirrel [17:02] ioria, Ok, I agree this needs some cleaning. [17:02] I however have a question, when I'm using older kernel (choosing older kernel at boot), which graphic driver does it use? [17:04] ldsh, which kernel is now in use ? uname -r [17:05] because graphic drivers also have some links with the kernel I think (there are always some dkms, and initrd work when there is driver update), so maybe I should also keep the one used with -19 kernel (as this works at least) [17:05] ldsh, uname -r [17:05] ioria, 5.13.0-19-generic, while last installed one is 5.13.0-21-generic [17:06] ldsh, sudo lshw -c video | grep driver [17:07] configuration : driver=nvidia latency=0 [17:07] configuration : depth=32 driver=i915 latency=0 mode=1280x1024 visual=truecolor xres=1280 yres=1024 [17:07] ldsh, with -19, nvidia module is in use [17:07] (second line is the intel IGPU I guess) [17:07] yes === rr is now known as rosros [17:08] ldsh, apt -s purge nvidia* | nc termbin.com 9999 [17:08] ioria, So I just purge all nvidia drivers then re-install the last one? (no reboot inbetween? On the igpu, as the nvidia would not be available) [17:09] ldsh, we'll see, but yes : apt -s purge nvidia* | nc termbin.com 9999 [17:09] ldsh, without sudo [17:09] ioria, https://termbin.com/r95b [17:11] "Les paquets suivants seront ENLEVÉS :" => this is the list of package to be removed [17:12] ldsh, sudo apt install --reinstall nvidia-driver-470 [17:12] I don't purge them all first? [17:12] nope [17:12] not now [17:13] Reinstallation of 470 done [17:13] ldsh, sudo update-initramfs -u -k all [17:15] (running) [17:15] ioria, done: https://pastebin.com/EaDE7v6A [17:16] The work has been done with the 3 kernels [17:17] ldsh, so, reboot on -21; if it fails , reboot on -19 and we'll purge nvidia this time [17:17] hello can someone help me about to move a /home ? [17:17] ioria, Ok, Hope to see you soon for the good news ;) [17:18] ok === Trieste_ is now known as Trieste [17:27] ioria, still hanging in -21, still working in -19 [17:27] ldsh, sudo apt-get remove --purge '^nvidia-.*' [17:27] I tried to look at the last services to start before hanging, and I saw: gdm.service then virtualbox.service [17:29] (purging on the run, I guess will take a bit of time) === mihaiadrian1 is now known as mihaiadrian [17:31] ioria, Should I try reboot after purge (hoping the open drivers will take the lead), or better to "sudo apt install nvidia-driver-470" ? [17:31] purge done [17:31] ldsh, ok; dpkg -l | grep nvidia | nc termbin.com 9999 [17:32] https://termbin.com/h71d [17:32] ldsh, sudo apt purge nvidia* [17:33] ioria, it just asks me to run "autoremove". [17:34] ldsh, run : sudo apt autopurge [17:35] https://pastebin.com/cnsF0Q4y => list of packeged to be purged [17:35] packages* [17:35] ldsh, nope, wait [17:36] ioria, Sorry, already done :s [17:36] ldsh, it's ok; again : dpkg -l | grep nvidia | nc termbin.com 9999 [17:36] dpkg -l | grep nvidia | nc termbin.com 9999 => https://termbin.com/v5ap [17:36] seems some libnvidia remains [17:37] ldsh, sudo apt purge libnvidia* [17:37] sudo apt purge libnvidia* ? [17:37] yep [17:37] should be all of the list now [17:38] check it [17:38] Still one: https://termbin.com/3ug19 [17:39] purge it : sudo apt purge ... [17:39] strange, this is even a i386 package [17:39] sudo apt purge libnvidia-compute-390:i386 [17:40] done, empty list === Nokaji_ is now known as Nokaji [17:40] are you sure ? check again [17:41] dpkg -l | grep nvidia return nothing [17:41] ok ; sudo updatedb && locate -i xorg.conf | nc termbin.com 9999 [17:42] In the software update extra driver, I see a 495 driver, strange that I was still with the 470 [17:42] ioria, I don't have updatedb [17:43] (in that extra diver page, nouveau is activated) [17:44] ldsh, do you have 'mlocate' installed ? [17:44] ioria, not yet, should I install it? [17:45] ldsh, we are looking lo a xorg.conf file ... [17:45] ldsh, usually in /etc/X11/ but not always [17:46] hi :) what's the best remote desktop software for Ubuntu focal for remote control via LAN? still Remmina -> SSH -> tightvnc? [17:46] I have on in /etc/X11 [17:46] also an other one in /etc/X11/xrdp [17:47] hello. i have some massive data stored on my hard drive. I don't want to change the location but I want to run a program that looks for it at a certain place. is it possible to create a symbolic link for this purpose? [17:48] ApostleInTriumph, if by "place" you mean a path in the filesystem, a symlink would be a good first try. most programs accept that. [17:48] Sven_vB yes it's a path in file system [17:48] can you briefly mention how to? [17:48] ldsh, ls -l /etc/X11 | nc termbin.com 9999 [17:49] my massive file is located at /media/Disk1/massiveData. i want to make it available at /home/User/Project/MassiveData [17:49] ApostleInTriumph, ln --verbose --no-clobber --symbolic --no-target-directory -- /path/to/actual/file /path/to/alias/to/be/created [17:49] ioria, https://termbin.com/lulp [17:49] Sven_vB also holds if my "massiveData" is a directory [17:49] ApostleInTriumph, the actual file path can also be relative to the alias location [17:50] ldsh, sudo rm /etc/X11/xorg.conf [17:50] ApostleInTriumph, yes, the symlink doesn't care about the target type. the --no-target-directory just means that your alias path will be treated verbatim [17:50] Sven_vB thanks! [17:51] ioria, I renamed into backup2, just in case [17:52] ldsh, ok : ls /usr/share/X11/xorg.conf.d/20-intel.conf [17:52] https://termbin.com/n88b [17:52] soyy [17:53] ldsh, ls /usr/share/X11/xorg.conf.d/20-intel.conf [17:53] ls /usr/share/X11/xorg.conf.d => https://termbin.com/ore3 [17:54] No such file [17:54] no such folder* [17:55] ldsh, so you don't have this file /usr/share/X11/xorg.conf.d/20-intel.conf ? [17:56] Indeed [17:56] ldsh, sudo apt install nvidia-prime [17:57] ioria, nvidia-prime_0.8.17.1_all installed [17:57] (thanks to the command) [17:57] ldsh, sudo prime-select nvidia [17:57] turns out, I already had x11vnc installed. and once I connect to it, xrandr status of my HDMI port magically changed from disconnected to connected. yay. :D [17:58] ioria, done, but will it work if I didn't install the driver yet? [17:59] ldsh, wait ... ; grep blacklist /etc/modprobe.d/* | grep nvidia [18:00] ioria, /etc/modprobe.d/blacklist-framebuffer.conf:blacklist nvidiafb [18:00] ldsh, move (or rm) that file somewhere else [18:01] ioria, I need the framebuffer drivers? [18:02] ldsh, what you mean ? [18:02] ioria, this file: /etc/modprobe.d/blacklist-framebuffer.conf:blacklist nvidiafb is for framebuffer drivers, not the normal ones [18:03] it sounds normal to me that they are all blacklisted [18:03] ldsh, the file is /etc/modprobe.d/blacklist-framebuffer.conf [18:03] yes [18:04] # Framebuffer drivers are generally buggy and poorly-supported, and cause [18:04] # suspend failures, kernel panics and general mayhem. For this reason we [18:04] # never load them automatically. [18:04] ldsh, move it in another directory (you can move it back, in case) === kostkon_ is now known as kostkon [18:06] ioria, done [18:07] ldsh, grep blacklist /lib/modprobe.d/* | grep nvidia [18:07] ioria, no output (but nouveau is in the blacklist) [18:08] ldsh, ls /lib/modprobe.d/ | nc termbin.com 9999 [18:09] https://termbin.com/34xwt [18:11] ldsh, ubuntu-drivers list | nc termbin.com 9999 [18:12] piping didn't work, here are the results: https://pastebin.com/WDcmHmMh [18:15] ldsh, we want to try 495 ? [18:16] ioria, I'm wondering if this would not solve the issue. I have no idea why it didn't update to 495 automatically [18:16] Could -21 be not compatible with 470? [18:19] ldsh, sudo apt install nvidia-driver-495 (keep an eye on random errors) [18:20] ioria, I'm chacking compatibility with my graphic card (1050 Ti) [18:21] ioria, Should be fine: https://www.nvidia.com/download/driverResults.aspx/181274/en-us [18:22] yes [18:22] My 21.10 install no longer boots, it just stops and plymouth/login never appears, no error messages(maybe in logs but haven't found any yet). If I boot to root shell and run systemctl status plymouth-start I get 'ConditionKernelCommandLine=splash was not met', can thaqt be solved? [18:23] ioria, 66 new packages are being installed [18:24] Slartiba1t: can you still F1 at booting to see text boot? [18:25] Slartiba1t: system up to date from recoverymode/root shell also? [18:27] ioria, Done, no error as I can see [18:27] lotuspsychje: Yes, I get text boot, there's a lot of '[ OK ] Info this, info that' lines. ..which is as expected. But they stop coming and the login screen doesn't popup :-[. Yes, updated packages. I tried editing grub files to add splash, but it seems I didn't do it correctly, nothing changed. [18:27] ldsh, try a reboot [18:30] ioria, I don't get well the thing with the framebuffer kernel modules. I don't know what they are for, why they were blacklisted, and what are the consequences to remove this blacklist list. [18:31] §I like to understand what I do and try to learn, but that step I don't know which impact it has) [18:32] ldsh, might cause problems in some cases [18:33] ldsh, you can easily restore it btw [18:33] Yes, I saved the file, but I don't know what these modules do. [18:35] Hi all. I'm using a router in Wi-Fi extender mode (B) between my computer (A) and my main WAN router (C) which is the gateway for the network. If I connect to B from A, does the connection go A-B-C-B because C is the gateway? Or does it just go A-B? [18:35] ldsh, the result is that nvidia module won't be loaded at boot time [18:36] I'm rebooting now [18:40] ioria, This time booting went fine, i'm on -21 kernel [18:41] ldsh, ok, good ; test the resolution if correct [18:42] ldsh, play a 1080 yt video and such [18:44] ioria, Works well, but on the IGPU (nvidia GPU is 0% use, but IGPU is consuming power) [18:45] strange, because prime was on performance. I'm switching to On-Demand [18:46] ioria, folding (computing) on GPU works [18:46] ok [18:55] Hello is there a version of ubuntu that I can install that does not have a graphical installation? [18:56] I'm trying to install ubuntu on a old macOS with 4GB ram and intel i3, but only lubuntu boots and only after ~30 mins [18:56] I'm assuming that it works much better after installation but usb's simply don't boot! No idea why [18:59] euouae: you can use debootstrap? [18:59] i have a rudimentary documentation that might be helpful: http://www.aiei.ch/gnustep/install.txt [19:00] tarzeau what is debootstrap? [19:01] After you boot lubuntu after 30 minutes, run `dmesg` and check what's wrong, in order to avoid it and make things fast... [19:01] https://wiki.debian.org/Debootstrap [19:02] tarzeau why should I use that? Do you want me to debootstrap on the macos? [19:02] euouae: no you can debootstrap ubuntu or whatever debian based thing you want, you just need to boot any linux with the tool and get network to work [19:03] So you're saying that I could install any basic linux distro and then debootstrap? === genii-core is now known as genii [19:05] alkisg Do you have a suspicion of what it could be? [19:05] euouae: no, debootstrap is the installing part [19:06] tarzeau I don't understand what you're saying, I have a macOS right now [19:07] euouae: no, it can be any number of issues. E.g. unhandled IRQs that may go better with irqpoll [19:07] euouae: another option would be to remove the disk, put it elsewhere, install there, then move the disk back to your mac [19:08] alkisg It's hard to do these things because I'm video chatting a friend and instructing them [19:16] Hmmm steam client also stopped working for me after doing a apt upgrade today on 21.10 [19:17] do you get an error scanno ? [19:18] no. it just does not start anymore [19:18] running from cmd just says:... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/70a7ebe5f40ac20a1df86a4b3a7dbc929eb724ca) [19:18] and it quits [19:21] scanno I don't think this is the right channel to ask about, but maybe https://github.com/ValveSoftware/steam-for-linux/issues/7813#issuecomment-861475454 will help you [19:21] Issue 7813 in ValveSoftware/steam-for-linux "Steam crashes on auto-login with utlmemory.h assertion hint" [Closed] [19:22] scanno note for instance https://github.com/ValveSoftware/steam-for-linux/issues/8222 seems to be exactly what you have [19:22] Issue 8222 in ValveSoftware/steam-for-linux "does not start at all" [Open] [19:22] Well this morning it worked fine [19:22] ¯\_(ツ)_/¯ [19:37] Also something is off with X11/Gnome because when my I only use my HDMI connected monitor (laptop lid closed or just set to the HDMI monitor), every keyboard / mouse event takes 2 seconds to show effect. Weird [19:38] Also started today [19:38] Solved it to by switching from NVIDIA On-Demand to Performace mode [19:38] But that should not happen ofc. [19:52] does focal have some middle ground between chroot and AppArmor profile? I have a config updater program that I want to run in lots of different directories, and it shall only be able to read/write in subdirectories of where I started it, and also read some selected additional directories. as I understand AppArmor, I'd need a separate profile for each project directory in which I want to run the config updater, right? [19:53] should have been "(sub)directories". the directory where I started it, it may read/write as well. === pah_ is now known as pa [20:02] Is this the right place to ask about a display issue I'm having with Ubuntu Desktop? [20:04] Sven_vB linux namespaces? [20:05] euouae, yeah they could probably do it. I'm looking for ways to easily set them up. [20:05] hippoid, as long as it's a version still in support (listed in the topic). [20:06] Sven_vB firejail? [20:09] euouae, I'll try it [20:09] I'm running Desktop Ubuntu 20.04.2 LTS. If the computer is idle for about 20 minutes, the monitor turns off, and the light on the monitor goes from blue to yellow. When I press the keyboard or mouse, that light on the monitor goes to blue for a moment, then back to yellow. The monitor then displays that there is input signal. [20:10] The computer is still on, and I can ssh into it, but I can't get the monitor connected by HDMI to turn back on. The HDMI is plugged into the graphics card, an NVIDIA RTX 2070. How to fix this? [20:10] Sven_vB https://firejail.wordpress.com/documentation-2/building-custom-profiles/ [20:10] s/input signal/no input signal === Guest8052 is now known as ^wuseman === ^wuseman is now known as wuseman [20:48] Curious if anyone else on 20.04.3 has issues with Firefox crashing when creating a new window. Normal or private. I seem to only reproduce it on Ubuntu but it happens for me and a friend. [20:51] pjs: if you have it installed from snap, that means it's upgrading itself and restarting [20:52] pjs: try https://winaero.com/how-to-disable-firefox-background-updates/ [21:01] leftyfb: No, I ripped snapd out a long time ago. [21:08] How can I upgrade to the 5.15 kernel on 20.04? Without compiling. === peer is now known as rails [21:21] tahc: step #1: stick around long enough to hear the answer [22:06] How do I set the splash flag(switch?) so that 'ConditionKernelCommandLine=splash was not set' goes away? [22:08] I tried editing things in grub, but that doesn't seem to have made any difference at all.. [22:09] Slartibart: that messages comes from one or more of the plymouth services (plymouth is the start-up splash screen handler) [22:10] Slartibart: what does "cat /proc/cmdline" show? does it have "splash" ? [22:10] TJ-: Exactly. It's shown when I run systemctl status plymouth-starat [22:10] Checking [22:10] TJ-: No, it doesn't [22:11] Slartibart: aha! so now check if the GRUB config has it: "grep -n splash /etc/default/grub{,.d/*} " [22:11] Slartibart: you'd expect to see it in a line with GRUB_CMDLINE_LINUX_DEFAULT [22:15] TJ-: /etc/default/grub has it: GRUB_CMDLINE_LINUX_DEFAULT="initctl_debug no_console_suspend acpi_osi= acpi_backlight=native RESUME=UUID=dbbe5783-b735-495f-8321-b4e06ad5a0ec splash" [22:15] ..and GRUB_CMDLINE_LINUX [22:16] I added splash at the end there [22:16] Slartibart: OK, so that should be added to /boot/grub/grub.cfg when "sudo update-grub" is executed. Check if it is there right now with "sudo grep -n splash /boot/grub/grub.cfg" [22:17] Ah. *shrinks* I think I forgot about updating. But I'll check to be sure. [22:19] Well, look at that :). Maybe a bit too good, now splash appears twice in some lines in grub.cfg [22:19] Is it ok or should I fix that ? [22:20] tias, perhaps? [22:21] Slartibart: that is because you've put it on GRUB_CMDLINE_LINUX as well as GRUB_CMDLINE_LINUX_DEFAULT [22:21] Yeah, I guessed that :). But which one should be removed? [22:22] Slartibart: the 1st is combined with the 2nd for 'default' entries, but only GRUB_CMDLINE_LINUX is used for recovery boot entries [22:22] Slartibart: usually it should only be in the DEFAULT entry [22:22] you don't want a splash screen getting in the way of a recovery boot [22:34] TJ-: Good point. Interesting, now boot splash (that I thought belonged to bios?? The one with the logo of the computer manufacturer) remains during boot, that hasn't happened before. Would that be because of quiet option in grub? It still *seems* to be stopping for something, but since I didn't see any boot text I'm not certain. No login screen appeared, but plymouth-start has status "Active (exited)" [22:34] which is better than before.. Looking into removing quiet (from GRUB_CMDLINE_LINUX, that's the last item in it) [22:35] Slartibart: unless GRUB does a mode-change the mobo's splash can remain; on most Ubuntu installs GRUB's menu is set to hidden so GRUB doesn't try to draw anything [22:36] Slartibart: but then I'd expect plymouth splash to take over once the initialramfs starts [22:36] 'quiet' controls the kernel log level [22:36] the opposite is "debug" [22:38] Could it be useful to enable debug for now, to try and solve this? Removing quiet didn't bring back the text but instead an odd text about UEFI secure boot was shown, I've never seen it before. [22:41] Slartibart: what is the issue now? you want the Ubuntu splash screen but aren't seeing it/ [22:43] Can anyone explain me line 38 to 61 here on focal : https://github.com/eilandert/dockerized/blob/master/base/Dockerfile-focal ? [22:44] TJ-: I'd very much like the login screen. Currently Ubuntu never leaves text mode, and on this computer it used to do that.. The mobo splash does disappear now, but the 'OK this&that' lines haven' returned. Now I get 'EFI stub: UEFI Secure Boot is enabled.'. Maybe I should check bios settings?? [22:45] Slartibart: did you change the mode in /etc/default/grub ? [22:46] pycurious: yes - it's all about the "usrmerge" program (where the previously stand alone /bin/ and /sbin/ directories are now all under /usr/ --- and Debian has also now merged those as well so those comments are out of date [22:47] TJ-: is there a reason to have those lines ? Or can i get rid of it? Or perhaps i should read more about "usrmerge"? [22:48] TJ-: I edited it so that there was just one splash. I tried changing quiet to debug, but it didn't show anything extra so I guess that's for the logs. Changed back from debug to quiet now, the mobo splash is back, the drive led flashes for a short while and then it stops doing so, the mobo splash is still there. Alt+F2 gives me a terminal. [22:53] Wee! I disabled Secure boot in BIOS. Success! I have no idea how Secure boot was enabled a few days ago when this started though. [22:54] OK lines while booting still not back though, that's odd.. [22:56] Doesn't plymouth(?) work with secure boot? ..or perhaps something before plymouth borked? [23:02] Slartibart: The splash screen should work in any config [23:03] I don't see it for very long running an SSD [23:04] jeremy31: Yeah, that's what I thought. So I guess something before plymouth in the boot sequence got stuck because of that Secure boot. I don't know what is was though. [23:04] jeremy31: I can imagine that ;) [23:12] Slartibart: Secure Boot only gets involved if you try to use out of kernel drivers [23:12] holas [23:15] jeremy31: Still weird though, I don't know what that driver was then, I tried booting without any devices connected to the computer. But I guess Secure boot was "force enabled" then, since it was set to Enabled(not "Auto" or anything) in BIOS? [23:25] TJ-: Thanks a lot for helping :). === caverton2 is now known as caverton === wille6 is now known as wille === doubleqp5 is now known as doubleqp