[01:58] <kenwoodfox> Hey im trying to get minecraft to use java 8 and i have it installed under /usr/bin/java but minecraft cant see it?... i installed via flatpak maybe thats the issue?
[01:59] <kenwoodfox> Im actually not very familer with flatpak
[02:01] <sarnold> kenwoodfox: iirc flatpak has some sandboxing things, perhaps you have to install a java through flatpak to get it 'inside' the sandbox? or do something to expose the java you *do* have ot the sandbox?
[02:01] <oerheks> me neither, we do supprt snaps, and java 8 .. isn that EOL? anyway, try https://snapcraft.io/mc-installer
[02:02] <kenwoodfox> Yeah but 1.12.2 runs only on 8 so we need it, it breaks forge compatability :c
[02:02] <kenwoodfox> Im not flatpak familer, shoot this is kinda tough....
[02:02] <kenwoodfox> can i get inside the flatpak?
[02:02] <kenwoodfox> https://github.com/flathub/com.mojang.Minecraft/pull/50
[02:02] <kenwoodfox> so confused..
[02:04] <kenwoodfox> basics of flatpak how do i install this? org.freedesktop.Sdk.Extension.openjdk8
[02:05] <oerheks> 'sudo update-alternatives --config java' .. flatpak install flathub org.freedesktop.Sdk.Extension.openjdk8 ... but not sure it is dead
[02:05] <oerheks> good luck!
[02:05] <kenwoodfox> what is dead?
[02:06] <kenwoodfox> I dident know :/ i thought everybody used java 8 for forge
[02:06] <oerheks> i don't use flatpak.
[02:06] <kenwoodfox> Ooh
[02:06] <kenwoodfox> ok
[02:06] <kenwoodfox> yeah i could ditch it too i think :c
[02:56] <de-facto> How do i install ubuntu without snap?
[02:56] <sarnold> sudo apt purge snapd
[02:59] <oerheks> install mint?
[03:00] <de-facto> hmm no i want the original but without snap
[03:01] <oerheks> one can purge after install, there are no guides during install
[03:02] <snake> hi, proxmox seems to have blocked my ip address, so i want to start over with just an ubuntu server. however, i want to install with full disk encryption using TPM so i dont need to type the password on boot
[03:03] <snake> is there a guide for doing that? or will the installer just handle it if i have secure boot enabled
[03:05] <oerheks> snake depends what TMP?
[03:05] <snake> oerheks, its TPM not TMP
[03:06] <snake> oerheks, would it say in my bios
[03:06] <snake> i wasnt aware there are more than one tmp but isuppose that makes sense since there are different cpus and motherboard
[03:07] <sarnold> snake: I don't think we've got a TPM-measured FDE available yet
[03:07] <sarnold> it's possible all the pieces are there for you to assemble, but I don't know off-hand how to do it
[03:08] <oerheks> dmesg | grep tpm
[03:08] <oerheks> if so, this guide is usable https://incenp.org/notes/2020/tpm-based-ssh-key.html
[03:09] <snake> sarnold, i see
[03:12] <snake> oerheks, that guide is for ssh not fde
[03:14] <snake> i'm going to try this for now: https://aboutcher.co.uk/2020/06/fedora-linux-luks-encryption-with-tpm-unlock/
[03:14] <snake> if that works, i will just wait for the day it is supported on ubuntu
[03:15] <sarnold> snake: yeah, that should work
[03:16] <sarnold> snake: you may also be able to do this on fedora, though perhaps it'll be a release before it works http://0pointer.net/blog/unlocking-luks2-volumes-with-tpm2-fido2-pkcs11-security-hardware-on-systemd-248.html
[03:28] <alystair> ok so I was told to ask my quesion in #ubuntu-on-windows and 5+ hrs later no one else has responded, all I want to do is automatically start a script as root when ubuntu loads
[03:29] <snake> alystair, what script?
[03:29] <snake> alystair, you could use AHK perhaps
[03:33] <ivo_cavalcante> alystair: that's all, running a script as root when Ubuntu starts?
[03:33] <ivo_cavalcante> on a "metal" or WSL ?
[03:35] <alystair> what do you mean, it's via WSL2
[03:35] <ivo_cavalcante> hum
[03:35] <alystair> '/usr/local/lsws/bin/lswsctrl start'
[03:35] <alystair> that's it.
[03:35] <ivo_cavalcante> on a real system, systemd was your guy
[03:35] <ivo_cavalcante> on WSL, not sure systemd runs
[03:36] <ivo_cavalcante> do you happen to know?
[03:36] <ivo_cavalcante> if it runs, its simple
[03:36] <alystair> there's a systemd folder in /etc so
[03:36] <ivo_cavalcante> are you in front of the machine?
[03:37] <ivo_cavalcante> if yes, type systemctl
[03:37] <alystair> yeap
[03:37] <ivo_cavalcante> let's see what happens
[03:37] <rander2> hello
[03:37] <alystair> "System has not been booted with systemd as init system (PID 1). Can't operate.", I tried running it as my normal user, not root
[03:37] <alystair> guess it's using something else?
[03:37] <ivo_cavalcante> doesn't matter
[03:37] <ivo_cavalcante> it is not running
[03:37] <ivo_cavalcante> yeah
[03:38] <ivo_cavalcante> unfortnatelly, I'm not familiar with WSL
[03:38] <alystair> is there an 'on boot' for cron?
[03:38] <ivo_cavalcante> there is on systemd
[03:38] <ivo_cavalcante> :)
[03:38] <ivo_cavalcante> but we can try something else
[03:38] <alystair> ... oh it's part of that? lol
[03:38] <tomreyn> !wsl
[03:38] <rander2> I'm using ubuntu 20.04LTS , and I have problems installation rocm , I follow this instructions : https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html#supported-operating-systems
[03:38] <alystair> there's a ton of RC1 through 6/S in etc as well
[03:39] <alystair> er rc0
[03:39] <ivo_cavalcante> I see
[03:39] <ivo_cavalcante> you could try rc.local
[03:39] <ivo_cavalcante> long shot, but might work
[03:39] <alystair> I tried locating that file, it doesn't exist
[03:39] <rander2> wsl run on a microsoft made kernel with virtualization
[03:40] <ivo_cavalcante> create it
[03:40] <ivo_cavalcante>  /etc/rc.local
[03:40] <alystair> ok!
[03:40] <rander2> and it lock the VT-X instruction
[03:40] <ivo_cavalcante> make it executable
[03:40] <ivo_cavalcante> chmod a+x /etc/rc.local
[03:40] <OceanGao> I tried, that didn't work.
[03:40] <ivo_cavalcante> so.....
[03:40] <OceanGao> chmod?
[03:40] <ivo_cavalcante> chmod
[03:41] <ivo_cavalcante> yes
[03:41] <ivo_cavalcante> doesn't work?
[03:41] <alystair> from a safety perspective the script it's running should also be owned by root right
[03:42] <ivo_cavalcante> yes, it should
[03:42] <ivo_cavalcante> you'll make it a shell script
[03:42] <OceanGao> well, I tried to use rc.local but that didn't work
[03:42] <ivo_cavalcante> inside it, you put whatever you want
[03:42] <sarnold> alystair: dang :( I'm sorry that didn't work out.. something you said just now reminds me of this bug report .. https://github.com/microsoft/WSL/issues/5324
[03:42] <ivo_cavalcante> OceanGao: then, don't know how to help
[03:42] <sarnold> alystair: there is a @reboot specifier for cron, but I don't know if that works in wsl either
[03:42] <de-facto> just trying out ubuntu, deinstalled stapd, now trying to install chromium-browser wants snapd as dependency?!
[03:43] <de-facto> where is the normal chromium-browser?
[03:43] <ivo_cavalcante> de-facto: on latest Ubuntu, Chromium is a Snap package.
[03:43] <sarnold> de-facto: you could download something from google, I think they offer a non-snap version
[03:43] <de-facto> omg seriously?
[03:43] <ivo_cavalcante> seriously
[03:43] <de-facto> wow
[03:44] <ivo_cavalcante> did you try?
[03:44] <alystair> oh wait couldn't I just have the rc.local just be `su '/usr/local/lsws/bin/lswsctrl start'` without an intermediary script?
[03:44] <ivo_cavalcante> many people use it
[03:44] <de-facto> how can i purge anything from snap from the software choices?
[03:44] <tomreyn> chrome is non-snap, but also ungogoled-chromium. but there's also a chromioum ppa by now.
[03:44] <alystair> ok rebooting the VM, fingers crossed
[03:44] <ivo_cavalcante> de-facto: if you really want to purge everything Snap, better try Mint
[03:45] <raccoon_dog> At this point I'd say, if you want to do that, maybe you shouldn't be using Ubuntu .
[03:45] <ivo_cavalcante> de-facto: but I believe you should try it first
[03:45] <ivo_cavalcante> I really don't understand this "Snap" haet thing
[03:45] <ivo_cavalcante> hate*
[03:45] <alystair> .... huh I can't use 'shutdown' to restart it LOL
[03:45] <de-facto> i really dont want snap
[03:46] <ivo_cavalcante> de-facto: try Mint, then. Canonical is very commited to Snaps.
[03:46] <alystair> nvm I Got it
[03:46] <OceanGao> alystair try the script below
[03:46] <OceanGao> echo "sudo su" > auto_root.sh
[03:46] <OceanGao> chmod +x auto_root.sh
[03:46] <OceanGao> cp auto_root.sh /etc/init.d/
[03:46] <OceanGao> cd /etc/init.d/
[03:46] <OceanGao> update-rc.d auto_root defaults 90
[03:46] <tomreyn> i disagree about recommending other Linux distros over Ubuntu here just because of snapd.
[03:47] <de-facto> i know mint its nice, but i am used to ubuntu and i want its experience without anything snap
[03:47] <ivo_cavalcante> tomreyn: that's my opinion. Ubuntu won't get rid of Snaps in the foreseable future. If OP want's a Ubuntu based distro, completely Snap-less, that's the way
[03:48] <OceanGao> alystair https://paste.ubuntu.com/p/DRhg2dTWFG/
[03:48] <tomreyn> !discuss
[03:48] <ivo_cavalcante> de-facto: mind me asking why not Snaps?
[03:48] <OceanGao> !wsl
[03:48] <de-facto> ivo_cavalcante my personal experience with it was not good, its just against how i like to use my programs
[03:49] <de-facto> i dont want to go into details, i just decided i dont use snap
[03:49] <alystair> wait why would I want auto_root
[03:49] <Lutin> Does anyone know if it's possble to install PHP without the version folder ? So /etc/php/fpm ? instead of /etc/php/VERSION/fpm ?
[03:50] <tomreyn> Lutin: you can set a symlink, but why would you need to?
[03:51] <Lutin> tomreyn symlink to what ? Because of docker builder script. It's not nice for my folder path
[03:51] <Annoyed> Ah. Life in here... Question about RAID. Ubuntu 18.04 server, with 2 6 TB drives set up in a RAID mirroring array. Going to rebuild the OS soon. Although I have a backup of the contents of the RAID drives, can I just set them up again in the new OS install, if i copy the /etc/mdadm/mdadm.conf over, might I expect to see the data intact?
[03:52] <Lutin> Annoyed if you have a seperate /home md then just remount it
[03:52] <Kevin> Annoyed: yes. you don't actually need the mdadm.conf to import them cleanly (although it would help make it automatic), since the metadata for array configuration is stored on each drive
[03:52] <Lutin> Annoyed what I always do is... take one drive out, or disconnect, install with one disk in fail on the connected disk... copy data over and then rebuild using the old disk
[03:53] <Annoyed> No hardware changes being made; just clean install of 20.04.1
[03:53] <Lutin> Annoyed even then... you have a "offline" backup of your data until you rebuild :D
[03:54] <Kevin> .. note that the installer doesn't understand that raid1 can have 1 drive. things I learned today fighting with it. so you'd kinf of have to do that part yourself
[03:54] <Annoyed> I just didn't want to wait a day or so while 6TB copies... one of the box's roles is a media server.
[03:54] <Kevin> you can use the array while it's rebuilding
[03:55] <tomreyn> Lutin: you want /etc/php/fpm to exist, but what's there is really at /etc/php/VERSION/fpm. so you do    ln -s /etc/php/VERSION/fpm /etc/php/fpm
[03:56] <Lutin> tomreyn yeah but that doesn't work in my docker builder script or I do need to copy over all config for all versions to entrypoint
[03:56] <Lutin> which can do
[03:56] <Annoyed> Oh, I don't plan on getting the installer involved at all. I have this unholy mistrust of automatics.
[03:57] <Annoyed> Thanks for the info. Sounds like I can save some time. Have a good evening.
[03:57] <Annoyed> Or morning... whatever it is
[03:57] <Lutin> Kevin it can, trust me... you need to go cli for that and build raid array there and reinitialize the installer part for disks
[03:58] <tomreyn> Lutin: i'm not into your docker builder script, and i don't think i should be.
[03:59] <Lutin> tomreyn you can do whatever you want on entrypoint so also remove what you don't need :)
[03:59] <Lutin> tomreyn old installs of PHP also leave the configfiles there
[04:00] <tomreyn> yes, a long time ago
[04:03] <Lutin> tomreyn nah... if you don't purge I think it still leaves them there ?
[04:04] <tomreyn> oh, i may have gotten you wrong
[04:04] <Lutin> can happen, you can always keep me satisfied with beer :P
[04:08] <de-facto> are those all the packages with dependencies on snapd? https://paste.ubuntu.com/p/k4CcqvVVhj/
[04:09] <nikolam> This is really terrible.. after a half a day of surfing and many tabs open, I went to sleep leaving computer working. Even with 16GB of RAM, Firefox managed by itself to fill in all the RAM and I have seen upon waking up that system is dog-slow and were even happy Firefox can be killed (but often it is not the case and all system gets locked)
[04:10] <de-facto> probably used swab file then, i always disable those so programs get killed if asking for too much memory
[04:11] <nikolam> What is the problem with Linux in general of not being able to handle memory-leaking applications, Ubuntu specifically of having such uncontrollable events of system locking when RAM is filled and applications in detail that always expect to fill the whole RAM and are leaking
[04:11] <lotuspsychje> nikolam: are you on an ssd or a spinner?
[04:12] <nikolam> System is on SSD with Btrfs, /home is on ZFS on Spinner
[04:12] <nikolam> because.. raid1
[04:12] <lotuspsychje> nikolam: leaving a browser open when you're afk isnt the best idea in my opinion
[04:13] <lotuspsychje> why make the system under constant work, when you're not doing anything
[04:13] <nikolam> Well, having some sessions open yada yada..
[04:13] <nikolam> It would go to sleep, eventually..
[04:14] <lotuspsychje> perfect for browser hijacks
[04:14] <nikolam> But problem is Linux not handling apps that are eating RAM, Ubuntu not controlling apps, Firefox leaking..
[04:15] <nikolam> User experience is runed and it doesn;t matter if 4, 8, 12 or 16GB of RAM, I experienced mostly same
[04:17] <lotuspsychje> nikolam: a browser open eats a lot of things, makes your system work, doesnt prove its leaking
[04:17] <lotuspsychje> nikolam: if you do can prove its really leaking, open a new !bug against firefox
[04:18] <nikolam> Maybe I could disable Linux memory overcommit? But I used to do that before, and somehow worse experience when RAM is filled
[04:18] <nikolam> SOmetimes RAM is filled in active use with the browser, more RAM just make it happen later
[04:20] <nikolam> lotuspsychje, how to prove it? Except ff eats more and more ram over time and everytig swapped out after few hours
[04:21] <lotuspsychje> nikolam: open FF from terminal and leave a journalctl -f open all night and htop or any other monitor that could capture things
[04:22] <nikolam> thanks lotuspsychje
[04:22] <lotuspsychje> nikolam: these days ubuntu has an own secret life inside, doing a lot of snap things automatic, make sure its not something else eating memory
[04:28] <Lutin> tomreyn those config files should not harm either
[04:29] <de-facto> can firefox play mp4/H264?
[04:29] <lotuspsychje> Lutin: tomreyn is afk atm, might wanna re-ask your issue in the channel
[04:30] <Lutin> lotuspsychje are you a bot or his guidance angel ?
[04:31] <lotuspsychje> Lutin: i'm a volunteer, informing you
[04:31] <Lutin> lotuspsychje thanks no need :)
[05:31] <golden_ticket> Is there any project like open-sharedroot for ubuntu? That is, a a diskless shared-root cluster
[05:52] <yates> how do i find out which package provides a specific file?
[06:04] <Bashing-om> yates: My goto: https://packages.ubuntu.com/ .
[06:07] <illmortal> Hey guys, something strange about my RDP session from Windows to Ubuntu 20.04. When I RDP in, I have limited authorization and have to keep entering my password. And even when I need to make changes, for example, trying to change my ethernet's IP address to a static IP, the settings don't save. Even after rebooting. It just uses the same IP address.
[06:46] <arahael> I think that's more to do with the RDP client you're using.
[06:49] <illmortal> Yeah its windows RDP
[06:55] <illmortal> I'd VNC in but that's been the typical outcome - black screen upon connecting. So went the RDP route.
[07:21] <arahael> Sounds like the VNC is misconfigured.  But not sure why I'm here - dunno how to configure it right in ubuntu.
[07:21]  * arahael doesn't actually use ubuntu.
[07:23] <illmortal> lol
[07:58] <ducasse> yates: try apt-file search filename
[07:59] <ducasse> (you need to install it and run apt-file update first)
[08:06] <td_> hey quick question.. i'm running ubuntu 20.04.1, when playing video at 1920x1080 60Hz, there is no horizontal screen tearing but everything is too small for my eyes to operate at this resolution all the time. When switching to a comfortable 1600x900 there's horizontal screen tearing at both 59.9Hz and 120Hz. Is there any way to prevent this horizontal screen tearing at the lower resolution?
[08:33] <td_> so it seems the solution is to use a custom 20-intel-graphics.conf in (newly created folder) /etc/X11/xorg.conf.d/
[08:33] <td_> if anyone looks back on this later :)
[08:34] <td_> found answer at: https://askubuntu.com/questions/945895/solution-to-intel-graphics-screen-tearing-flickering-causes-excessive-fan-use
[08:35] <td_> you all have a great night
[08:43] <nmatrix> Is there a kvm support channel?
[08:44] <EriC^^> nmatrix: #kvm
[10:32] <Industrial> Hi!
[10:32] <Industrial> When I pick the NVIDIA display driver from Additional Drivers and reboot I get a black screen
[10:32] <Industrial> https://i.imgur.com/vadKvA1.png
[10:33] <Industrial> I have a MSI GT76 with a Geforce RTX 2070 Super Mobile / Max Q
[10:33] <Industrial> I want to play World of Warcraft with wine and dxvk and vulkan
[10:44] <nidua42> dow down
[12:00] <lord4163> why doesn't my script run through cron? :((((
[12:01] <lord4163> this all worked fine on 18.04, I don't understand what's wrong
[12:01] <ThinkT510> what does your script do?
[12:01] <lord4163> It makes a backup
[12:02] <lord4163> and I can run that script just fine, but it doesn't get executed properly it seems via cron?
[12:02] <JoeBk> I don't have a problem with cron under 20.04.1
[12:03] <JoeBk> you might need to set PATH and stuff like that in the script.
[12:05] <RonaldsMazitis> so I got Intel i5 HD 3000 gpu, AND I have an idea I can't do "AI VIDEO rescaling"
[12:05] <lord4163> JoeBk: why should the PATH be different?
[12:06] <RonaldsMazitis> not on windows, not on ubuntu linux ether
[12:06] <RonaldsMazitis> right?
[12:06] <lord4163> JoeBk: oh well you're right, PATH=/usr/bin:/bin
[12:20] <tomreyn> RonaldsMazitis: what's "AI VIDEO rescaling"? (no search results on a web search)
[12:28] <BluesKaj> Hi folks
[12:28] <lotuspsychje> Industrial: how many times did you ask this same question already here?
[12:32] <RonaldsMazitis> tomreyn: using artificial intelegence to upscale old videos
[12:38] <tomreyn> RonaldsMazitis: given you have a software which can do this, i don't see why you couldn't do it. but it's certainly going to take a loooong time with an integrated gpu.
[12:40] <RonaldsMazitis> tomreyn: I'm not scared of waiting
[12:40] <JoeBk> lord4163, Is it working now?
[12:41] <RonaldsMazitis> but I can't find a tool that works
[12:41] <lord4163> JoeBk: yes
[12:55] <hansh> i need gparted >= 1.1.0 on my 20.04, but ubuntu ships 1.0.0, how should i get it? (or specifically, i want a gparted with this bug fixed: https://gitlab.gnome.org/GNOME/gparted/-/issues/16 - it was fixed in the 1.1.0 release )
[12:56] <lotuspsychje> !info gparted focal
[12:57] <tomreyn> you have a <=40KiB capacity drive?
[12:57] <hansh> yup.
[12:58] <tomreyn> that's so special, you could probably work with alternative CLI tools
[12:59] <hansh> yup, parted works fine, for example
[12:59] <tomreyn> so there is no real world problem then?
[13:00] <hansh> it's easier to inspect with gparted than with parted, a time-saving thing.
[13:00] <tomreyn> so you need to work with <=40KiB capacity drives on a regular basis?
[13:01] <hansh> i'm developing the drive back-end code, so.. kindof
[13:03] <lotuspsychje> hansh: your link says parted doesnt have an issue with it, did you try that?
 yup, parted works fine, for example
[13:04] <lotuspsychje> oh missed that :p
[13:05] <tomreyn> hansh is just looking for a gui
[13:05] <lotuspsychje> hirsute has same version, so not sure backports or proposed could help?
[13:06] <tomreyn> hansh could probably backport the hirsuite version.
[13:06] <hansh> what's hirsute?
[13:07] <tomreyn> !hirsute
[13:07] <hansh> oh, 21.04
[13:08] <hansh> !info gparted testing
[13:09] <hansh> what's the testing branch? 0.32 sounds ancient
[13:09] <tomreyn> here's a(n unsupoported) backport https://launchpad.net/~savoury1/+archive/ubuntu/gparted?field.series_filter=focal
[13:10] <hansh> oh thanks
[13:11] <tomreyn> note "You can update your system with unsupported packages from this untrusted PPA"
[13:11] <hansh> yeah got it
[13:26] <hansh> dang, seems 18.04's do-release-upgrade didn't replace /etc/apt/sources.list's "# deb-src http://no.archive.ubuntu.com/ubuntu/ bionic main restricted"  with focal
[13:26] <hansh> is that intended?
[13:27] <tomreyn> the commented out lines? yes, that's normal.
[13:27] <hansh> yes the commented-out deb-src lines
[13:27] <tomreyn> i assume a developer can manage search and replace. ;-)
[13:29] <tomreyn> sudo sed -i 's/bionic/focal/g' /etc/apt/sources.list
[13:29] <hansh> right, but was lucky that i noticed it, i almost missed it when needing them
[13:31] <tomreyn> i agree it would be nicer if the upgrader would also modify those urls. on the other hand, if those were pointing to third party repositories whch don't actually support the new / target release, you'd run into non obvious error messages.
[13:32] <tomreyn> or non self-explanatory
[13:45] <hansh> building gparted from source was rather easy, albeit it'll probably get overwritten by apt next time ubuntu updates gparted :|  - git clone https://gitlab.gnome.org/GNOME/gparted.git; cd gparted; sudo apt-get build-dep gparted; ./autogen.sh; ./configure; make -j $(nproc); make install;
[13:45] <hansh> (unfortunately the last step seems mandatory, without it the local gparted binary will just complain about "/usr/lib/udisks2/udisks2-inhibit: 26: /usr/local/sbin/gpartedbin: not found", whatever that means)
[13:47] <lotuspsychje> !compile
[13:55] <B0g4r7> Ubuntu 20.04 should still install on a system with 1GB of RAM, shouldn't it?
[13:56] <tomreyn> B0g4r7: server: yes, desktop, no.
[13:57] <B0g4r7> I've tried both, and can't make it happen for some reason on this system.
[13:57] <JoeBk> I think the minimum for desktop is 2GB
[13:57] <lotuspsychje> B0g4r7: what are the other specs of your machine?
[13:57] <JoeBk> not sure
[13:57] <B0g4r7> Desktop just freezes hard, and Server says "installation failed".
[13:57] <tomreyn> server is 1 GB https://ubuntu.com/server/docs/installation
[13:58] <tomreyn> desktop is 4 GB
[13:58] <tomreyn> https://help.ubuntu.com/community/Installation/SystemRequirements#Ubuntu_Desktop_Edition
[13:58] <B0g4r7> lotuspsychje, there's not much to it.  It's got an older AMD Sempron processor and a 2.5" 160GB hard drive.
[13:59] <lotuspsychje> maybe a very tweaked lubuntu 20.04.1
[13:59] <lotuspsychje> im running one on an older netbook, with 2GB ram
[13:59] <lotuspsychje> but as tomreyn says ubuntu desktop with gnome surely likes 4GB and higher
[14:00] <hansh> i have a recipe for a *very* lightweight desktop though, using some 130MB ram more than server, but you'll need to know what GPU you're dealing with and replace xserver-xorg-video-dummy with xserver-xorg-video-YOURGPU (eg for Intel, xserver-xorg-video-intel or for AMD xserver-xorg-video-amdgpu or for VMWare: xserver-xorg-video-vmware ): apt install --no-install-recommends lxde-core lxde-common; apt install lxdm xserver-xorg xserver-xorg-video-d
[14:00] <hansh> ummy lxterminal policykit-1;
[14:03] <lotuspsychje> B0g4r7: server with i3 could do some magic too
[14:09] <elmomani> Hi, my Kali linux (installed on SSD) and Ubuntu (installed in HDD) both of them take too long to boot. I've done many things recently which could cause this issue: 1)I updated BIOS from windows update (insyde firmware) 2)installed Hackintosh 3) (most recent) removed swap partition(on SSD), but I also commented the swap line from /etc/fstb for both kali and ubuntu. finally, I fixed the problem by adding "nosplash" for linux args on GRUP, but this is
[14:09] <elmomani> very strange and I want to know whay's happening?
[14:11] <tomreyn> this channel supports only ubuntu. which release are you running? how long is "too long", how do you measure it? how long is it when it is "fixed" by using "nosplash"?
[14:12] <gartt> elmomani: They're in VMs? Kali and ubuntu can be pigs and take lots of RAM
[14:12] <tomreyn> based on the (yet) rough description, it could be related to uefi boot graphics switching
[14:13] <gartt> How much RAM are you giving them? I've had to use them in courses and they were excrutiatingly slow unless given lots of RAM and other resources. Far more than if a WM like fluxbox was used though
[14:14] <tomreyn> there should be no problems there as long as you comply with the minimum system requirements
[14:16] <lotuspsychje> gartt: maybe you should wait until the user responds before assuming VM's and advice?
[14:16] <gartt> lotuspsychje: You're right, I'm assuming a lot
[14:19] <tomreyn> elmomani chose to continue this conversation on the two (?) other channels they started it on simulataneously.
[15:21] <sols> ping
[15:21] <lotuspsychje> welsome sols how can we help you?
[15:31] <Guest40> hello, is this the right channel to talk about dns server setup and recommendation on ubuntu 20?
[15:32] <tomreyn> For (actual) Ubuntu 20.04 LTS or 20.10, you can get community support here, yes
[15:33] <tomreyn> There is also #ubuntu-server
[15:34] <Guest40> thanks, i was wondering first what is the recommendation for dns server software? i look up and mostly get results about resolvers which require pointing to another existing dns server such as 1.1.1.1
[15:35] <Johnnyd> how well does macbook pro 13 m1 chipset work with Ubuntu?
[15:36] <tomreyn> Guest40: so you're looking for software to run an authoritative DNS
[15:37] <Guest40> yeah, i have a clean vps with ubuntu 20, that i want to use as that, whilst from my home connection use dnscrypt to that
[15:37] <tomreyn> Guest40: the classic option for that is bind, but there are others, such as knot, powerdns etc
[15:37] <Guest40> i want to use for adblock filtering and whatnot as well
[15:39] <Guest40> yeah, bind9 is usually in the results, seems like the "classic" one indeed. does it provide same security/privacy as knot, etc?
[15:40] <tomreyn> probably, if configured properly
[15:40] <lotuspsychje> !discuss | Johnnyd
[15:43] <Guest40> tomreyn is this the same? https://github.com/m13253/dns-over-https
[15:43] <Guest40> as in, does it act and behave as knot/powerdns/bind9 or is that something else
[15:43] <Guest40> i'm trying to get something like this setup https://github.com/AhaDNS/dns-server-setup/ while using oisd.nl
[15:47] <Guest40> huh, powerdns admin ui looks just like pi-dns :P
[16:01] <magic_ninja> what is the procedure to take a piece of source code and set it up with dkms?
[16:01] <magic_ninja> right now I have to compile it with every kernel update.
[16:05] <jeremy31> magic_ninja: Make a dkms.conf file in the directory
[16:05] <gravitos> so, my laptop refuses to connect to any bluetooth devices and points me that device or resource is busy whilst it clearly is not, any ideas?
[16:06] <jeremy31> magic_ninja: Is that source code online?
[16:07] <gravitos> laptop is on ubuntu 18.04 xfce if that matters
[16:10] <gravitos> btmon's log when trying to connect to my phone with bluetoothctl: https://www.irccloud.com/pastebin/6C8jusjz/btmon
[16:10] <magic_ninja> I don't have it on me. It is at work.
[16:12] <jeremy31> magic_ninja: source code for a driver?
[16:14] <magic_ninja> yea
[16:14] <magic_ninja> you make the config, then what?
[16:14] <jeremy31> magic_ninja: what driver?
[16:14] <DJones> 0.1pp
[16:26] <osse> Hi, guys. I have installed 20.10 elsewhere on my computer and I want to copy old $HOME from the 20.04 install to the new $HOME. Are there anything I should definitely exclude? I am thinking stuff like .cache or .gvfs
[16:26] <osse> Maybe conflicting gnome config files and such
[16:28] <ExcessiveAggro> morning
[16:29] <magic_ninja> osse, you could back up the home folder on the 20.10 machine, just copy over your entire new home folder, and see how it goes.
[16:29] <magic_ninja> If you have issues with a config for some application, delete it.
[16:32] <osse> magic_ninja: very good point
[16:32] <osse> i will do that
[16:33] <magic_ninja> osse, if gnome doesn't fire up, just delete your gnome configs.
[16:48] <Hash> How to get the latest web browser in ubuntu
[16:48] <Hash> everythig is old in 20.4
[16:48] <Hash> How do you get the latest version of chromium?
[16:49] <Hash> Ok thank you
[16:50] <Hash> Even the firefox is old it says old version cannot start profile
[16:53] <coconut> Hash, sudo apt update && sudo apt upgrade
[16:53] <JoeBk> don't fprget autoremove
[16:54] <Hash> that installs a snap though.
[16:55] <JoeBk> hash, doing update is not optional.
[16:55] <Hash> Hmm.
[16:56] <Hash> I don't know if LTS keeps up with new versions of chromium
[16:56] <Hash> Version 87.0.4280.141 (Official Build) snap (64-bit)
[16:56] <Hash> https://www.chromium.org/getting-involved/dev-channel/
[16:56] <Hash> I am confuse!
[16:56] <coconut> Hash, yeah chromium is a snap, but firefox will just update with apt
[16:57] <Hash> I don't like a firepoop!
[16:57] <Hash> I don't like chrom either. I like chromium. It's cleaner
[16:57] <Hash> And smaller too
[16:57] <JoeBk> this is the version of chromium I have: Version 87.0.4280.141 (Official Build) snap (64-bit)
[16:57] <Hash> I was afraid it wouldn't pick up my ~/.config/chromium directory, the snap
[16:57] <Hash> But it did, so it's alright now
[16:58] <Hash> This way I don't have to login anywhere all over again with 2FA etc. etc.
[16:58] <Hash> I just carry my encrypted/packed chromium directory
[16:58] <Hash> unencrypt and unpack on target machine, install chromium and voila, got all mylogins.
[16:58] <Hash> That's why I do this. ok thanks
[17:03] <Hash> Your bookmarks, history, and other data have been transferred automatically.
[17:03] <Hash> file:///snap/chromium/1444/firstrun/snap-en.html
[17:03] <Hash> Uhm... What?!
[17:03] <Hash> It copies my chromium data to inside the snap?
[17:03] <Hash> So if the inside snap chromium directory is modified, it won't affect my outside one, so it won't be in sync.
[17:04] <Hash> That's gonna be a problem.
[17:07] <Hash> Is that gonna be a problem? coconut
[17:16] <coconut> Hash, dunno no experience with chromium here
[17:16]  * coconut always uses firefox
[17:20] <slop> hi all <3
[17:25] <slop> Hi all - anyone use Citrix Workspace? Have any luck with getting it to work properly. Getting unknown error 1000119
[17:31] <Toadisattva> does anyone know of a ubuntu application to convert cdi to iso format? search engines aren't turning up very helpful results
[17:32] <Toadisattva> will bchunk do that?
[17:42] <JoeBk> Toadisattva, did you try imgburn?
[17:42] <Toadisattva> found one a command line tool called iat
[17:42] <JoeBk> imgburn is a windows application tho
[17:45] <Maik> Disks can create them too iirc
[17:46] <Maik> it's installed by default in ubuntu
[17:47] <JoeBk> imgburn says it will do it.  that's one of the few reasons I keep a working windows installation.
[17:54] <Maik> JoeBk: all good and well but he asked for a Ubuntu app didn't he?
[17:55] <JoeBk> yes, was thinking maybe he has a windows installation available.
[17:56] <Maik> never count on it and recommend a ubuntu app ;)
[17:58] <Maik> Toadisattva: there's somthing called cdi2iso in the repo's
[19:16] <bog3n> anybody here knows how to change top bar of the window to these old ones before 19.04?
[19:23] <LordDoskias> hello i'm running 20.04 HWE (kernel 5.8) and I can't get my ax 200 wifi adapter from working, it can't connect to my 2.4ghz network. I get No beacon heard and the session protection is over already. I had the same issue on 5.4 kernel as well
[19:23] <LordDoskias> the adapter is the on-board my motherboard asrock x570 taichi
[19:23] <LordDoskias> i have attached the external antenna
[19:36] <Maik> LordDoskias: HWE 5.8 isn't officially released as of yet for 20.04.1, it should come out in february afaik with 20.04.2. A lot of users who mysteriously got the 5.8 kernel already amongt the regular updates reported having all sorts of issues with it.
[19:36] <zutat> how to populate /boot/efi/EFI on 20.04? i did not use the installer. i have installed the grub efi packages, but those did no touch the efi partition.
[19:37] <Maik> LordDoskias: I'm still on 5.4 on my Ubuntu MATE 20.04.1 install and didn't see any 5.8 kernel coming in.
[19:46] <LordDoskias> Maik, fair, but i was having the issue on 5.4 hence why i ugprade, i can easily downgrade, but this still won't resolve my wifi question :(
[19:49] <jeremy31> LordDoskias: Have you checked for newer firmware?
[19:53] <LordDoskias> jeremy31, there is newer firmware, which when i put in /lib/firmware
[19:53] <LordDoskias> i first get uCode file size XXXX does not match expected size and the laoded firmware is 48
[19:53] <LordDoskias> i can se there is 59
[19:54] <jeremy31> LordDoskias: Is it using iwlwifi-cc-a0-55.ucode
[19:54] <LordDoskias> no it's using cc-a0-48
[19:54] <LordDoskias> (i'm now back on kernel 5.4) on kernel 5.8 it was using 55
[19:55] <LordDoskias> i have also disabled all the power saving options for both iwlmvm and iwlwifi as well as 11ax (since my router doesn't support). I've also hardcoded channell width to be 20mhz as well as disbaled 40mhz channels for 2.4ghz in cfg80211 module
[19:56] <jeremy31> LordDoskias: did you set your country code?
[19:57] <LordDoskias> so iw reg get returns global country: 00 .... phy#0 (self-managed) country BG - which is correct
[19:59] <jeremy31> LordDoskias: and it won't find anything on 2.4 in>  iwlist scan
[20:02] <LordDoskias> nope, it does find my AP
[20:02] <LordDoskias> it says signal level =-87dbm - is that high or low ?
[20:02] <jeremy31> It is a bit low
[20:04] <LordDoskias> when i start iw event in a windows and then start authenticating to an ap via NM i guest "new station" then "del station" and auth timed out eventually
[20:07] <jeremy31> LordDoskias: try this once>  sudo modprobe -r iwlwifi && sleep 10 && sudo modprobe iwlwifi 11n_disable=8
[20:09] <LordDoskias> same thing
[20:10] <LordDoskias> "no beacon ehard and the time even is over already" message is printed and  get timeouts while authenticating
[20:11] <jeremy31> LordDoskias: Is your router using WEP or TKIP?
[20:11] <LordDoskias> WPA2 PSK (CCMP)
[20:11] <jeremy31> no beacon would be something else
[20:12] <jeremy31> LordDoskias: Does power management show enabled in>  iwconfig
[20:12] <LordDoskias> no, it says - power managent:off
[20:12] <LordDoskias> mode: managed frequency : 2.462ghz access point : not associated
[20:13] <jeremy31> So you have the frequency set to a single channel on the router?
[20:13] <LordDoskias> yes, channel 11
[20:13] <LordDoskias> the router is running openwrt
[20:14] <jeremy31> LordDoskias: post URL for>  dmesg | grep iwl | nc termbin.com 9999
[20:14] <LordDoskias> well i don't have network on that machine :D
[20:14] <LordDoskias> but ok
[20:15] <LordDoskias> https://termbin.com/wrbe
[20:16] <LordDoskias> that's from journalctl -b | grep iwl
[20:17] <LordDoskias> that file size does not match is for firmware 50
[20:18] <LordDoskias> which i have manually copied over to /lib/firmware from upstream linux-firmware repo, otherwise the file didn't exist
[20:20] <LordDoskias> that built-in wifi also requires an external antenna which i have attached
[20:20] <LordDoskias> dunno if i need to do anythign special to detect it, i guess not ...
[20:21] <jeremy31> LordDoskias: But those connectors are tiny if it is M2
[20:21] <LordDoskias> ok the antenna is definitely detected becuse without it iwlist scan finds only 3 aps and with the antenna it finds a lot more
[20:22] <jeremy31> It might help to rename that 48 firmware file so it will use 46
[20:22] <jeremy31> Do a complete shutdown rather than a restart
[20:25] <LordDoskias> hm
[20:25] <LordDoskias> running couple of iwlist scan consecutevely
[20:26] <LordDoskias> i can see last beacon interval is rather large
[20:26] <LordDoskias> like 15 seconds ago ..
[20:26] <LordDoskias> and sometimes it doesn't find any APs o_O
[20:29] <giaco> not strictly ubuntu related, but maybe ubuntu can help me tracking down the freezes I've been experiencing on my laptop. Basically suddenly everything hangs: mouse doesn't move, ctrl-alt-del won't do anything, ctrl-alt-f* won't switch to console, soundcards replays last audio buffer endlessly, all this while I was writing code and cpu was at 3%
[20:29] <LordDoskias> ok with fw 46 and a full shut even from th power grid - same thing ...
[20:29] <giaco> "sudo journalctl -b -1" shows nothing relevant as last lines
[20:30] <jeremy31> LordDoskias: can you increase beacon time on the router?
[20:31] <LordDoskias> i did try
[20:32] <LordDoskias> it was 100 i changed it to 200 to no avail
[20:34] <jeremy31> Go the other way, try it at 50
[20:37] <LordDoskias> same shit, with 50 and 20 and i get auth: time out in iw event ....
[20:38] <LordDoskias> i have other devices which have worked perfectly fine with those settings though ...
[20:38] <jeremy31> LordDoskias: restarted router?
[20:39] <GrosLalo> hello I am using 20.10 and gnome terminal and often when i insert a pipe character (|), i get some non-printable chars after the char. Example, "ls -al| grep x" -ls -al|M-BM- grep x $> ". I am using to key AltGr to access the pipe symbol and after that immediately press spacebar and i suspect that is the reason for the non-printable chars. Is it
[20:39] <GrosLalo> possible to reduce the timing for key-combos?
[20:39] <LordDoskias> i will but that won't help ...
[20:44] <LordDoskias> jeremy31,  no luck ..
[20:51] <jeremy31> LordDoskias: Might want to post at https://community.intel.com/t5/Wireless/bd-p/wireless
[20:55] <jmspeex> Hi, my 20.04 workstation recently started crashing frequently. This seems to be correlated to the following error, which started appearing after an upgrade from kernel 5.4 to 5.8:
[20:56] <jmspeex> kernel: pcieport 0000:80:03.0: AER: Multiple Uncorrected (Non-Fatal) error received: 0000:00:00.0
[20:57] <jmspeex> Anyone has seen that before? Also what's the best way to downgrade kernel (Ubuntu has already erased my old kernel)
[20:57] <jeremy31> jmspeex: Use grub to boot into 5.4 kernel using Advanced options
[20:57] <jmspeex> jeremy31: 5.4 is no longer installed. How do I get it back?
[20:58] <jeremy31> jmspeex: in terminal try> sudo apt install linux-generic
[20:58] <jeremy31> Should get 5.4.0-6?
[20:58] <oerheks> sure the previous kernel is still installed
[20:58] <jmspeex> 5.4.0.62
[20:59] <jmspeex> I mean 5.4.0-62
[20:59] <jmspeex> what's the linux-generic package exactly?
[20:59] <jeremy31> jmspeex: That package should always keep a 5.4 kernel installed
[21:00] <jmspeex> OK. I'll try that. I have not completely ruled out hardware issues either, but at least having 5.0.4 should reduce the search space
[21:00] <jmspeex> Any idea what that error is BTW?
[21:01] <jeremy31> Some advance error reporting, might not even be important
[21:01] <jmspeex> The crashes seem to be triggered from using my nvidia card for deep learning, but it did not happen until these errors started appearing
[21:02] <jeremy31> jmspeex: I doubt the Nvidia dkms was fixed to work in 5.8 yet
[21:02] <jmspeex> Last time I saw similar errors, the pcie number corresponded to an nvme disk that seemed to have a faulty adapter (causing crashes when using the nvidia card). I removed the adaptor and the errors disappeared.
[21:03] <jeremy31> jmspeex: reboot and use Grub/Advance Options to boot a 5.4 kernel
[21:03] <jmspeex> I think I need to hold the left shift key to get the prompt, right
[21:04] <jeremy31> jmspeex: I think that works
[21:04] <jeremy31> Usually shift or esc
[21:05] <jmspeex> fyi, the nvidia drivers worked for a few days without crashing. They only started crashing last night
[21:06] <tichun> Hi, is there dev/debug/testing channel?
[21:07] <coconut> tichun, #ubuntu+1 ?
[21:07] <jeremy31> jmspeex: see what happens in 5.4, it will be more stable than getting a major kernel upgrade every 6 months
[21:08] <jmspeex> jeremy31: 5.4 booted fine. MTBF when crunching with the GPU is about 1-2 hours, so we'll see
[21:09] <jmspeex> Still weird that 5.0.4 got replaced automatically without my knowledge
[21:09] <jeremy31> jmspeex: if you don't want it to happen again>  sudo apt remove linux-generic-hwe-20.04
[21:09] <arkurious> How do I add a password prompt to https://www.gnome-look.org/p/1009736/ ?
[21:10] <jmspeex> A week ago my drivers started complaining about mismatched versions, I just updated them and rebooted and suddenly I was under 5.0.8 (which I hadn't reaslized until I looked carefully today to track what changed)
[21:10] <jmspeex> What's linux-generic-hwe-20.04 exactly?
[21:11] <jeremy31> jmspeex: It is a kernel package, the one that switched you to the 5.8 kernel
[21:12] <oerheks> arkurious, add a password to a site??
[21:12] <jmspeex> Will the 5.0.4 i just istalled keep being updated for security fixes and all?
[21:12] <jmspeex> And how do I make it the default?
[21:13] <arkurious> oerheks: To a plymouth theme.
[21:13] <zutat> do the changes in the -oem kernel eventually appear in -hwe?
[21:14] <oerheks> arkurious,  i still have no clue, do you have that theme installed?
[21:14] <oerheks> b.t.w. the source is questionable, use it at your own risk
[21:15] <jeremy31> jmspeex: The 5.4 kernel will get security patches until 2025
[21:16] <jeremy31> jmspeex: post URL from terminal for> dpkg -l | grep linux-image | nc termbin.com 9999
[21:17] <arkurious> oerheks: Yes, I have encrypted drive and it prompts for a password during boot, but when I boot with this plymouth theme it does not show a password prompt, I can still input the password but it does not show a prompt like it does on the default plymouth theme.
[21:17] <jmspeex> jeremy31: https://termbin.com/0h1b
[21:17] <oerheks> arkurious, that must be a bug in the theme?
[21:17] <oerheks> contact the maintainer?
[21:18] <oerheks> well, 2016 is a long time ago, maybe it is a dea project
[21:18] <jeremy31> jmspeex: in terminal>  sudo apt remove linux-image-5.8.0-36-generic linux-image-5.8.0-38-generic linux-image-generic-hwe-20.04 && sudo apt autoremove
[21:18] <jmspeex> BTW, I suspect it's harmless (not new), but my logs are flooded with "irq 4: Affinity broken due to vector space exhaustion."
[21:21] <jmspeex> Any way to default to 5.0.4 but to keep 5.0.8 for debugging purposes?
[21:22] <jeremy31> jmspeex: The problem with keeping a 5.8 is that Grub will boot into it by default since it has a higher version number
[21:24] <jmspeex> I see. So you think it could be the nvidia drivers not being properly updated (because they load and appear to work)?
[21:26] <jeremy31> jmspeex: It is always a possibility when the LTS gets the kernel from the non LTS release
[21:26] <oerheks> livepatch could have done it?
[21:27] <jmspeex> I know it's tempting to blame the nvidia driver by default (I hate this thing too), but I want to make sure it's not another problem with the kernel
[21:28] <jeremy31> jmspeex: If it is a problem with 5.4 and 5.8 worked better, you can always reinstall the linux-generic-hwe-20.04 package
[21:28] <jmspeex> As I said, the last time I got crashes, the nvidia card was the trigger, but the real culprit turned out to be a bad nvme adapter
[21:29] <jmspeex> I mean that even assuming 5.0.4 works fine, it could still be a bug with 5.0.8 that's unrelated to nvidia -- in which case it would still be good to report/fix it
[21:29] <jeremy31> I think you will have fewer issues using the 5.4 kernels over the long run.  The 5.8 kernel will no longer be supported by August
[21:31] <jmspeex> I mean someday I might want to install a new distro so if the kernel devs broke support (e.g.) for my board, then it would be good to make sure it's fixed in 5.0.x (x>8)
[21:31] <jmspeex> I mean 5.x
[21:32] <jmspeex> my setup isn't exactly common I suspect: 2x Xeon E5-2640 with an nvidia 1080Ti card and an AMD 5500 XT card
[21:33] <jeremy31> jmspeex: I am sure the 5.8 kernel works great in Ubuntu 20.10 as all the drivers are patched for 5.8
[21:34] <jmspeex> how can the driver be patched in 20.10 and not in 20.04? Aren't they part of the kernel tree? (or are you just talking about the nvidia blob?)
[21:35] <jmspeex> Sorry for all the questions. I'm trying to understand how the whole thing fits together. I've been running Linux for 25 years now, but haven't kept up with all the changes
[21:35] <jeremy31> jmspeex: 20.10 and 20.04 have separate software repositories
[21:36] <jeremy31> It does take some time to get the fixes backported from the newer version to the LTS
[21:37] <jmspeex> jeremy31: you're talking about backporting the ubuntu-specific patches on top of the vanilla kernel?
[21:37] <jeremy31> jmspeex: to the nvidia source code
[21:38] <jmspeex> I thought there was no source code?
[21:38] <jeremy31> jmspeex: the deb files you get from the repos are compiled from source code
[22:11] <Lutin> Is this something new in 20.04 ? Package 'locales' has no installation candidate
[22:12] <oerheks> sure it gives locales
[22:12] <oerheks> !info locales
[22:13] <oerheks> installed by default
[22:17] <mrtrousers> Hello, how can i find out what com port is my usb?
[22:18] <mrtrousers> Dont even know if that question makes sense, help please..
[22:18] <mrtrousers> I got an electronics board where i need to reflash the firmware, for that i need to provide the command with the com port or something..
[22:20] <jmspeex> jeremy31: no crash yet, but got another strange clue. It seems like on at least two occasions I've seen these PCIE bus errors, they came along with messages from my AMD card like this: https://termbin.com/fk3ne
[22:20] <ixdap> mrtrousers, so "com port! is kinda loose term
[22:20] <oerheks> mrtrousers, lsusb or dmesg | grep tty # could give a clue
[22:21] <ixdap> on dos it was the name for the serial port, which was probably 5V uart (?)
[22:21] <oerheks>  /dev/ttyUSB0 or something like that
[22:21] <ixdap> if you really need to go usb to uart then you want something like an ftdi
[22:21] <jmspeex> it's mentioning the GPU "resuming" and in both cases it coincides with the display being turn off (display locked). The AMD card is relatively new (before the upgrade to 5.8, but still just 2 weeks), so I'm wondering if it could be related
[22:21] <Lutin> oerheks but why that error then ?
[22:22] <oerheks> one better look for a linuxupgrade guide for *that* board
[22:22] <oerheks> Lutin, what error?
[22:22] <oerheks> it is installed, so ..
[22:22] <Lutin> oerheks READ
[22:22] <jeremy31> jmspeex: The AER ones?
[22:23] <jmspeex> jeremy31: yeah
[22:24] <mrtrousers> hi im back
[22:25] <jmspeex> I didn't paste the earlier messages but they were much earlier. Then suddenly the amdgpu messages arrive and the PCIE errors follow within the same second
[22:26] <mrtrousers> lsusb lists the devices but doesnt give me the port name
[22:26] <jmspeex> the pcie errors don't always crash my machine but in several cases, they were the last messages I saw (aside from the vector "space exhaustion" spam)
[22:27] <mrtrousers> dmesg | grep tty shows only either ttyUSB0 or ttyUSB1
[22:27] <oerheks> mrtrousers, and  'dmesg | grep tty ' ?
[22:27] <oerheks> !paste
[22:29] <jeremy31> jmspeex: I think there is a parameter you can add to grub to disable the AER messages
[22:29] <mrtrousers> Thats what the help section says: /dev/ttySy    - serial (RS232) port y (Linux)
[22:29] <mrtrousers>                    remark: has to be a valid system device name
[22:29] <mrtrousers> ?
[22:32] <mrtrousers> Oerheks dmesg outputs the same wheter the usb device is connected or not..
[22:36] <ixdap> mrtrousers, yes /dev/ttyS0 would be your first serial port if you have the hardware. (or at least it used to be called that).
[22:36] <mrtrousers> And what would be the name for usb ports?
[22:37] <ixdap> mrtrousers, I am trying to think how that works!
[22:39] <ixdap> mrtrousers, in a simple world every device would have a file in /dev, but it has not been that way as long as I can remember. ethernet devices have names but not /dev/ files.
[22:40] <jmspeex> jeremy31: Why would I want to kide them? I'm pretty sure they're the ones signalling the problem
[22:40] <shinobi> Does ubuntu send email to root?
[22:40] <jeremy31> jmspeex: it is always the case
[22:40] <mrtrousers> https://pastebin.com/raw/fjPFf1Vk
[22:40] <ixdap> when you plug a usb device, often it will get a /dev file then according to its type.
[22:41] <jeremy31> jmspeex: you might have to find someone with more experience with graphics drivers to know exactly
[22:41] <ixdap> so, I plug a flash stick, I get /dev/sda and so on.
[22:41] <ixdap> If you plug in a modem, you might well get a device name like /dev/ttyUSB0 or something.
[22:42] <ixdap> I think you might get that with an ftdi.
[22:42] <mrtrousers> Thats the command im trying to run: https://pastebin.com/raw/dmR6mXGH
[22:43] <ixdap> mrtrousers, so you have two boxes cabled together by usb, yes?
[22:43] <mrtrousers> Mmm, yes..
[22:44] <ixdap> and your running your flash program at one end to reflash the other?
[22:44] <ixdap> And at the program end, you need to figure out the port name.
[22:45] <Lutin> oerheks /usr/sbin/dpkg-reconfigure: locales is not installed
[22:45] <cjoke> Im following a tutorial where Im going to set up the network bridge (kvm/virt-manager) but I dont find any /etc/network/interfaces. how do I put up the good stuff ?
[22:45] <ixdap> I would use lsub to see the vendor:product number
[22:45] <ixdap> and then look for the name being assigned in dmesg or syslog.
[22:46] <ixdap> maybe there is an easier way, but I am old.
[22:47] <oerheks> Lutin, interesting, one of your locales is not configured, it is installed >>> apt-cache policy locales
[22:47] <oerheks> so, without proper output on paste.ubuntu.com of what you are doing, i cannot give answers anymore
[22:47] <Lutin> oerheks yes so... this happens in a docker build, never had before
[22:48] <oerheks> see, now it is a docker build..
[22:48] <ixdap> mrtrousers, I was doing something similar a few months back. bought an ftdi to connect to the serial console on an embedded box.
[22:48] <ixdap> but it is lost in the mess right now.
[22:49] <ixdap> I have a fair idea it came up as ttyUSB0, but I think the default name depends on the driver?
[22:49] <Lutin> oerheks doesn't matter... if I want to install locales it cannot find it
[22:49] <Lutin> oerheks it's just apt-get in a 20.04 image
[22:51] <ixdap> mrtrousers, okay just saw your first pastebin. looks like you are on the right track.
[22:53] <lusrx> https://mirror.duvaliden.com/ubuntu-releases/20.04.1/ubuntu-20.04.1-desktop-amd64.iso
[22:53] <lusrx> is this down for you too?
[22:53] <oerheks> Lutin, then yoou should add this example to your build script, locales is indeed not standard in docker.
[22:53] <oerheks> https://leimao.github.io/blog/Docker-Locale/
[22:54] <oerheks> you missed apt update, to get fresh lists?
[22:55] <ixdap> mrtrousers, nice tool. I plugged in a flash stick and got more than a page of too-much-detail. syslog was easier for me to read.
[22:56] <ixdap> mrtrousers, but I bet one day I will find it useful.
[22:56] <Lutin> oerheks yes updated list
[22:56] <mrtrousers> Dont know, im a bit lost, just trying to work out a command to reflash those electronics..
[22:57] <Lutin> oerheks mhh this is weird... seems to be some bugg shit in multiple runs
[23:04] <ixdap> mrtrousers, those /devices/ paths look to be under sys. I have some vague idea that /sys exposes some kernel structure to userland.
[23:07] <ixdap> oh god, linux 2.5, that takes me back.
[23:08] <ixdap> what was that before udev? devfs ...
[23:08] <ixdap> anyway, I have survived all that by reading syslog.
[23:09] <orp> lol
[23:10] <ixdap> or is that systemdlog now?
[23:11] <ixdap> I needed an extra 1.6 million lines of code.
[23:33] <cjoke> for the record (time 23:45) I tweeked /etc/NetworkManager/system-connections/nameoffile.nmconnection. and the virtualnetwork got up after a systemctl reload network-manager.service
[23:49] <shinobi> is /etc/ssl/certs/ca-bundle.crt supposed to be installed by default?
[23:53] <oerheks> i think ubuntu uses /etc/ssl/certs/ca-certificates.crt
[23:53] <oerheks> so, you are not using ubuntu?