[00:12] <tomreyn> hmm, Sbur3 seems to be gone for good. I'll head to bed as well. If they return, please tell them they should be good to upgrade, after backup.
[00:13] <Matrox> do i need to use this in my bashrc [   export GPG_TTY=$(tty) ]  to be able to use gpg?
[00:18] <sarnold> Matrox: I think I've heard that is very handy if you're doing gpg operations on remote hosts
[00:38] <Sven_vB> finally I was able to observe my Ubuntu focal's wifi issue while it happened. :-) https://paste.debian.net/plainh/d5bda9f7 It seems to not be an actual wifi issue, because connectivity is immediately restored when I run "sudo ifconfig wlxe894f6XXXXXX 192.168.1.5/24". Is there an easy way to make NetworkManager do that for me automatically? or is there a way to avoid the disappearance of the IPv4 address in the first place?
[00:45] <ApOgEE> Sven_vB: you can set it in netplan
[00:45] <ApOgEE> Sven_vB: https://linuxize.com/post/how-to-configure-static-ip-address-on-ubuntu-18-04/
[00:47] <Sven_vB> ApOgEE, there's currently a static IP in /etc/NetworkManager/system-connections/wifi.ini that NM correctly applies when I (re)start NM. do you mean NM might apply the netplan IP in a situation where it doesn't apply its own config?
[00:58] <Hallcyon> Do you ever run anything other than LTS on a server?
[01:00] <sarnold> Hallcyon: sure, the decision of what to run is more individual than a clear proscription
[01:00] <Sven_vB> only if I have to
[01:01] <Hallcyon> 20.10 has official support for raspberry pi's, should I upgrade those from the current 20.04 lts, would that make sense for a wireguard server?
[01:01] <Hallcyon> I will keen my KVM hosts running 20.04
[01:02] <sarnold> Hallcyon: I understand eg microk8s isn't supported or isn't built for 20.10 or something similar to that
[01:03] <Sven_vB> Hallcyon, what would be your motivation to upgrade?
[01:03] <sarnold> oh, actually, thanks for the idea. I've got an rpi3b+ that's on 20.04 at the moment, and the one application I run on it throws some C++ warnings while building it...
[01:03] <sarnold> it *might* be a toolchain problem, and upgrading the machine would get a new toolchain pretty quick :)
[01:04] <Hallcyon> Sven_vB https://liliputing.com/2020/10/ubuntu-20-10-released-and-its-the-first-version-with-full-raspberry-pi-support.html and a newer kernel so maybe backporting wg isn't required?
[01:05] <leftyfb> Hallcyon: it's the first version with the full desktop. There's been an official server version for a while now.
[01:05] <Hallcyon> ah I see.
[01:05] <Sven_vB> Hallcyon, oh ok. before I didn't gather that 20.04 isn't officially supported
[01:05] <Hallcyon> I guess I will stick with 20.04
[01:05] <leftyfb> Hallcyon: https://ubuntu.com/download/raspberry-pi
[01:06] <leftyfb> Sven_vB: it is
[01:06] <Sven_vB> leftyfb, thanks!
[02:11] <hays> hmm. no one knows about my AMI question anywhere...
[02:37] <Felishia> why did ubuntu suddenly stopped recognizing my host?
[02:38] <Felishia> it modified the /etc/hosts file all on its own
[07:40] <DrNostril> hey can anybody help me increase my swap space?
[07:41] <rk4> sure, your choices are swap file or swap partition
[07:41] <EriC^> DrNostril: what's the output of 'swapon -s'
[07:43] <DrNostril> one moment
[07:44] <DrNostril> no output as i am on a livedisk
[07:44] <DrNostril> https://termbin.com/msjp
[07:44] <DrNostril> maybe that will give an idea
[07:44] <DrNostril> the target is made with LVM and is encrypted
[07:44] <DrNostril> THANK YOU eric
[07:45] <EriC^> DrNostril: i'd just make a swap file then
[07:45] <DrNostril> im worried about ruining my main partition when i resize it
[07:45] <EriC^> DrNostril: yeah, no need to resize
[07:45] <DrNostril> wait so i can just make a file within the main partition to use as swap?
[07:46] <EriC^> DrNostril: how much bigger do you want the swap?
[07:46] <EriC^> yesh
[07:46] <EriC^> *yes
[07:46] <DrNostril> at least 8GB i think
[07:46] <DrNostril> maybe less is more :p
[07:46] <EriC^> type "sudo fallocate -l 8G /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile"
[07:46] <DrNostril> im trying to run a windows OS in vmware, and it suggests that i need more swap for vmware
[07:47] <DrNostril> so with 64 GB of ram, i'm a bit confused
[07:47] <DrNostril> i feel that i shouldn't need it as i do not use hibernate
[07:47] <EriC^> yeah, i think so too
[07:48] <DrNostril> would it be better to do this after logging into the hardware-native system?
[07:48] <EriC^> DrNostril: doesnt really make any difference
[07:49] <DrNostril> okay then one last question..... how do I notify my system that this swapfile exists?
[07:49] <DrNostril> swapon .../swapfile ?
[07:49] <EriC^> fstab so its permanent, did you run the fallocate command?
[07:49] <ducasse> first run mkswap on it, too
[07:51] <DrNostril> yes i did
[07:51] <EriC^> DrNostril: ok, type "sudo chmod 600 /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile"
[07:51] <DrNostril> i ran fallocate, did not touch fstab yet
[07:51] <EriC^> then "sudo mkswap /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile/media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile"
[07:52] <DrNostril> okay chmoded
[07:52] <EriC^> finally add a new line with "/swapfile swap swap defaults 0 0" to /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile/media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/etc/fstab"
[07:52] <DrNostril> mkswap: cannot open /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile/media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile: Not a directory
[07:53] <DrNostril> ffffff
[07:53] <EriC^> * to /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/etc/fstab
[07:53] <EriC^> sorry typo
[07:53] <DrNostril> oops i saw as soon as i pasted
[07:53] <EriC^> sudo mkswap /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile
[07:54] <EriC^> then add "/swapfile swap swap defaults 0 0" to to /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/etc/fstab   and you're done
[07:54] <DrNostril> mkswap: invalid block count argument: '/media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/etc/fstab'
[07:54] <EriC^> DrNostril: sudo mkswap /media/mint/782c5c8f-994f-4f2b-a9d7-f9a2f8c76bb3/swapfile
[07:55] <DrNostril> Setting up swapspace version 1, size = 8 GiB (8589930496 bytes)
[07:55] <DrNostril> no label, UUID=2f65ca2b-0074-4fa5-b8a9-8090e86abfc5
[07:55] <DrNostril> nice
[07:55] <DrNostril> you rock
[07:56] <DrNostril> okay im editing fstab
[07:57] <DrNostril> should i leave the current swap entry in place?
[07:57] <DrNostril> just add another line?
[07:59] <DrNostril> i might add that this fstab entry does not look like the others........
[08:01] <DrNostril> well..... here goes nothing..... i guess i reboot x|
[08:07] <DrNostril> Eric you still around?
[08:08] <EriC^^> DrNostril: sorry back
[08:08] <EriC^^> DrNostril: yeah leave the current swap line and add a new line with /swapfile ...
[08:18] <DrNostril> i booted and everything worked great
[08:18] <DrNostril> i modified your fstab line a little bit
[08:18] <DrNostril> thanks a bundle
[08:18] <EriC^^> DrNostril: great, no problem
[08:19] <DrNostril> eric what's your favorite stuff to work on? why you here?
[08:37] <xBfrog> hi, one of the nice features of Ubuntu is dark mode But I noticed that Ubuntu help guide is white backround unless you change to "all Help" then its dark. can this be made permenant?
[08:38] <xBfrog> ok is it a bug?
[08:53] <DarkTrick> Hello, I try to format my SD card with "disks", but it tells me "failed to probe the device mmcblk0 error-quark, 0. Any ideas?
[08:54] <xBfrog> try gparted it worked for me
[08:54] <DarkTrick> didn't work. Gave me hundreds of popup errors in a loop
[08:56] <Maik> download Raspberry Pi Imager and try to erase the SD card with that
[08:56] <xBfrog> after you insert the drive in files an you right click and choose format?
[09:00] <DarkTrick> xBfrog, what do you mean with "insert the drive in files"?
[09:01] <xBfrog> can you insert the drive in usb port?
[09:01] <xBfrog> or in a sd port?
[09:02] <DarkTrick> sd port
[09:02] <xBfrog> ok then use files to see the drive and right click on it. do you get a format option?
[09:02] <DarkTrick> xBfrog, I guess you mean nautilus?
[09:03] <xBfrog> where ever you can see your files
[09:03] <DarkTrick> drive doesn't appear in nautilus
[09:03] <DarkTrick> also not in thunar
[09:03] <DarkTrick> it's not mounted, I guess
[09:04] <xBfrog> i dunno either
[09:05] <gebbione> is gio trash a ubuntu specific application? is there something amongst gio trash, trash-cli, gvfs-trash  that is ubiquitous in all distros?
[09:06] <mgedmin> gio is part of gnome, so it's not specific to ubuntu, but I'm not sure it would be right to say it's ubiquitous to everything
[09:06] <mgedmin> I think there's a freedesktop spec for trash, so it should be more-or-less universal
[09:06] <mgedmin> (except for people who use twm and rm -rf and don't care for all the new-fangled desktop thingys)
[09:07] <xBfrog> Normally you don't need to do anything special. Could you update the question to include the version of Ubuntu you're using? Also, you can open a terminal (ctrl-alt-t), type "tail -f /var/log/syslog", insert an SD card, and see what messages pop up. If you don't know what they mean you could paste the lines that appear when you insert the SD card in your question too.
[09:10] <ThinkT510> for multiple lines it is recommended to use a paste site
[09:15] <xBfrog> yes true, pastbin.com works well
[09:15] <xBfrog> *pastebin
[09:17] <Deano59> xBfrog: see the topic in the channel. use ubuntu's paste site, pastebin is riddled with ads.
[09:19] <Deano59> xBfrog: this; https://paste.ubuntu.com/
[09:19] <xBfrog> ok thanks, i've never had issues but ok
[09:19] <Maik> Deano59: not if you use a proper ad blocker but i agree that https://paste.ubuntu.com/ is the way to go :)
[09:20] <Deano59> Maik: shouldn't have to use an ad-blocker but k.
[09:20] <xBfrog> i guess thats it then, i have adblockers
[09:25] <mgedmin> eh pastebin dot com is annoying even with adblockers
[09:26] <mgedmin> it also mangles whitespace in some circumstances
[09:26] <Deano59> mgedmin: indeed.
[09:45] <Toxic_Waste> morning guys :) was wondering if anyone has got ubuntu 20.10 running in virtualbox already
[09:46] <Toxic_Waste> the guest additions don't seem to play nicely on groovy gorilla
[09:46] <WoC> what would i use to create the initrd for a manually compiled kernel ?
[09:52] <WoC> k, update-initramfs
[09:54] <tomreyn> Toxic_Waste: virtualbox as distributed by ubuntu?
[09:54] <tomreyn> Toxic_Waste: if not, head over to #vbox
[09:54] <Toxic_Waste> fair point, I suppose that is more a vbox thing :)
[09:58] <Bustin> Hey there, trying to install the newest ubuntu 20.10 on my laptop, which has an nvidia rtx 2070, sporting dual external monitors, one via display port, one via hdmi. Prior to this version, worked out the box fine. Now, when booting, I get EXTREME lag, can barely move the mouse, click on things, etc.
[10:00] <Bustin> disconnecting the dual monitors, lets the OS function fine, but as soon as I reconnect, back to extreme lag.
[11:05] <dka> I have a user on Ubuntu 20.04 LTS, he destrroyed the system (with an apt autoremove). He forgot the root password, the password he his giving me for him is not working, he tells me he used that password all the time and he seems to be confident the password should work
[11:15] <EriC^^> dka: what's your objective? recover access to the pc?
[11:23] <dka> yes
[11:23] <dka> EriC^^, we are trying password over and over, I am 100% with the username as he saved some log in his usb backup drive and I can see the username being "alexander"
[11:24] <dka> He never wrote the password anywhere, he says the password he use, but I have tried it with root user and alexander, both didn't work
[11:24] <dka> he does not have the live cd, he has one, but he gave me one from debian, tought it is obvious it is not debian running here but  Ubuntu 20.04.1 LTS
[11:24] <dka> I am downloading the Ubuntu iso so I'll make another USB drive, i'll probably try to chroot
[11:25] <dka> There's no grub boot loader, instead there's a splash screen
[11:25] <dka> I personnally use debian, how can I access the grub boot loader instead of the splash screen ?
[11:27] <dka> I was able using ESC to access the Grub Boot Loader. Can I recover the access for this computer?
[11:28] <dka> the older kernel is stuck on "Loading initial ramdisk"
[11:28] <EriC^^> dka: hold shift to get grub
[11:30] <EriC^^> dka: if he hasnt set an actual root password, just his username password, then you can go to advanced > recovery and from there you can press on "drop to root shell" and type "mount -o remount,rw /" and then do "passwd alexander" to change his password
[11:30] <BluesKaj> Hi all
[11:32] <EriC^^> dka: if he's set an actual root password, then you can boot to a shell by adding "systemd.unit=multi-user.target" to the linux /boot/vmlinuz....., after pressing "e" in grub to edit the Ubuntu entry and pressing F10 to boot the edited entry
[11:32] <dka> I finally recovered the user password and root password
[11:33] <dka> he was giving me a wrong username from a wrong old from a wrong computer, such missleading informations...
[11:47] <tux__> Hi, I am trying to make systemd-boot use shim. has anyone done this successfully? If yes, can you share your layout of /boot? I have my ESP mounted at /boot
[11:58] <Metamorphosis> Hello. I wonder why is the highly praised software "GNU PSPP" has been removed from the official repositories. I'm having trouble installing it on 20.04.
[12:07] <BluesKaj> Metamorphosis, pspp is in the 20.10 universe repos
[12:12] <Metamorphosis> BluesKaj Oh? Didn't knew that. What can I do? I'm on Kubuntu 20.04 and I don't really want to upgrade at the moment.
[12:27] <BluesKaj> Metamorphosis, check this,  https://askubuntu.com/questions/1234137/why-is-the-pspp-program-not-present-in-the-ubuntu-20-04-lts-focal-fossa-reposito
[12:36] <mgedmin> maybe there's a ppa?
[12:48] <rotaerc> The computer screen flickers occasionally after installing Ubuntu 20.04 for Mac RRO.
[12:49] <rotaerc> What should I do with it?
[12:53] <ttytwister> Hi, I've installed both ubuntu and win 10. But I didn't manage to encrypt "ubuntu's drive" during installation. What is the propper way to encrypt it after installation?
[12:56] <tomreyn> ttytwister: ideally, you'll do it during installation. it's quite involved and error prone later on. see the cryptsetup-reencrypt man page.
[12:57] <Deano59> ttytwister: just re-install again but this time encrypt the drive, less hassle.
[12:57] <ttytwister> Deano59, it is easy (and I always do so) when you install ubuntu only
[12:58] <tomreyn> right, it won't work (using just the GUI) for a desktop installation which shares a drive with windows.
[12:58] <Deano59> do it through the install instead of trying *after* less hassle.
[12:59] <ttytwister> I need two OS and one of them encrypted :) So, you say it is not possible during installation?
[13:00] <tomreyn> at least that's what i seem to remember. i duggest you double-check, I don't mean to spread FUD.
[13:00] <tomreyn> * suggest
[13:01] <tomreyn> I'm sure it should work if you install Ubuntu to a different physical storage.
[13:05] <tux__> Hi, I am trying to make systemd-boot use shim. has anyone done this successfully? If yes, can you share your layout of /boot? I have my ESP mounted at /boot
[13:05] <tomreyn> pop os is not supported here
[13:12] <][o4|-o_5o_PYcki> \ (•◡•)
[13:15] <ttytwister> tomreyn, well, it is exactly my case. I want to install to different hard drive. But when I choose to install along with win 10, encryption is not offered
[13:20] <tomreyn> ttytwister: i assume your windows installation is configured for uefi boot mode. if you'll install ubuntu the same way, to a separate storage, without choosing to install side by side with windows, but with encryption, i would assume you will be able to boot either.
[13:23] <ttytwister> tomreyn, in this case, should win 10 hard drive be connected during installation?
[13:23] <ttytwister> or it is not necessary?
[13:27] <tomreyn> ttytwister: yes, needed, because it'll contain the efi system partition which both windows' and ubuntu's  boot loaders need to be placed on.
[13:28] <tomreyn> you could also copy it later, though, if you want to prevent windows' being overwritten (i don't expect that to happen, though)
[13:29] <ttytwister> There is nothing important there on windows hard drive. So I can experiment
[13:29] <ttytwister> So, at installation type page, I should choose "Something else" right?
[13:30] <tomreyn> what are you installing?
[13:30] <ttytwister> ubuntu
[13:30] <ttytwister> 20
[13:30] <tomreyn> !YY.mm | ttytwister
[13:30] <ttytwister> 20.04
[13:31] <tomreyn> server? desktop? some other flavor?
[13:31] <ttytwister> desktop
[13:31] <tomreyn> glad we got all the basic info gathered.
[13:31] <tomreyn> then yes.
[13:31] <ttytwister> can you guide me?
[13:32] <tomreyn> select advanced when it asks you about partitioning
[13:32] <ttytwister> done
[13:32] <tomreyn> then encryption
[13:32] <ttytwister> no
[13:33] <ttytwister> wait a sec
[13:33] <EriC^^> i like the idea of putting the efi partition on the ubuntu hdd so that it's a standalone install, not dependent on the windows hdd to boot
[13:33] <ttytwister> I've chosen "something else". At next page I can see my hard disks
[13:34] <ttytwister> I haven't installed both OS before
[13:34] <tomreyn> EriC^^: are there mainboards which can handle more than one ESP?
[13:35] <tomreyn> ttytwister: "something else" was wrong then, take a step back
[13:35] <tomreyn> i'll have a look at the screens, give me a minute
[13:35] <EriC^^> tomreyn: i dont think there's a limit on the efi partition in the uefi entry to have the same partuuid in all entries there
[13:36] <ttytwister> tomreyn, ok, Im on prev page. 3 options: 1. Install ubuntu alongside win 10. 2. Erase disk and install ubuntu (Warning: this will delete blah-blah...) and button Advance features. 3. Something else
[13:37] <tomreyn> EriC^^: hmm, nice idea. but i guess this means you'll need to update the second ESP to match the ID of the first, right?
[13:37] <EriC^^> it should be fine i guess, never seen any complain, but havent seen so much i guess
[13:37] <ttytwister> if I choose 2 option, I don't what disk it is going to erase
[13:38] <EriC^^> tomreyn: nope, just install to 2nd hdd as usual, it'll add the entry in the uefi list to grub's efi file, and ubuntu's grub will pick up the bootmgfw.efi on the windows efi
[13:38] <ttytwister> *don't know
[13:38] <Deano59> what's the package name for lxqt themes??
[13:38] <Deano59> *not lubuntu*
[13:39] <EriC^^> tomreyn: the board doesnt really have a efi designated kinda thing, just a list of stuff like this for instance, Boot0000* ubuntu	HD(1,GPT,472ed800-a2c2-4f94-bbb0-85207dba31f3,0x800,0x96000)/File(\EFI\ubuntu\shimx64.efi)
[13:40] <EriC^^> the hex stuff is the PARTUUID, i guess that's how it knows what to mount and look for /File.... and just boot it
[13:40] <Deano59> nvm
[13:40] <Deano59> :D
[13:41] <EriC^^> ttytwister: i think it would erase windows
[13:41] <tomreyn> EriC^^: hmm, nice idea, i hadn't though of this approach.
[13:41] <ttytwister> EriC^^, I do think so too. So what should I do :(
[13:41] <tomreyn> ttytwister: I guess you're better off following EriC^^'s suggestion
[13:42] <EriC^^> ttytwister: either manually partition, the easier thing would be to just remove the windows hdd and let it do its thing
[13:42] <EriC^^> after its done plug in the windows hdd, boot into ubuntu, and run 'sudo update-grub' windows should be in grub now
[13:44] <Veltum> hey am i registered
[13:45] <EriC^^> ttytwister: however it's important you make sure ubuntu is booted in uefi mode, it's easy to check if you can get a shell and run 'ls /sys/firmware/efi' before starting the actual installer
[13:45] <tomreyn> Veltum: you must be. this channel is cmode +r.
[13:45] <Veltum> EriC^^,
[13:45] <Veltum> ok
[13:46] <ttytwister> EriC^^, can't I see it in BIOS?
[13:47] <NIKOFIGOXX> hey again
[13:47] <tomreyn> NIKOFIGOXX: for irc help, please /join #freenode
[13:51] <ttytwister> EriC^^, so, what output should it give?
[13:51] <ttytwister> I can several folders and files
[13:51] <ttytwister> *can see
[13:53] <Voltum> yo
[13:53] <Deano59> I have this installed: linux-image-5.9-rpi-lobo-arm64 is already the newest version (5.9.1~lobo-arm64+) but when I check uname -a: Linux ayescott89-desktop 5.8.0-1006-raspi #9-Ubuntu SMP PREEMPT Fri Oct 16 12:55:30 UTC 2020 aarch64 aarch64 aarch64 GNU/Linux -- why is that?
[13:53] <Voltum> my irc so broken
[13:53] <Voltum> any good client to try?
[13:53] <Voltum> irssi or weechat
[13:53] <ttytwister> Voltum, I use hexchat
[13:54] <Voltum> mirc ?
[13:54] <Deano59> I use weechat. :D
[13:54] <Voltum> mirc used to be good
[13:54] <Deano59> anyone know how to change it?
[13:54] <BluesKaj> mirc is a windows client
[13:55] <Voltum> we used to use mirc before 12 years
[13:55] <BluesKaj> hexchat is good on gnome or kde/plasma desktops
[13:55] <Voltum> i thought it was the first irc client
[13:57] <Deano59> can you guys to to /join #ubuntu-offtopic
[13:58] <tomreyn> Deano59: linux-image-5.9-rpi-lobo-arm64 is not an ubuntu kernel image, the kernel you're now probably is one, though.
[13:58] <Deano59> tomreyn: yeah but it's defaulting to the ubuntu one and I want to know why. the other kernel is installed and up to date...
[14:01] <Deano59> am I missing something, tomreyn?
[14:01] <Deano59> :D
[14:02] <tomreyn> Deano59: it's probably a matter of file names in /boot
[14:11] <ttytwister> EriC^^,  'sudo update-grub' didn't help
[14:13] <Deano59> tomreyn: Ignoring old or unknown version 5.9.1~lobo-arm64+ (latest is 5.8.0-1006-raspi)
[14:13] <Deano59> how can I force it?
[14:15] <waveform> Deano59, the kernel that gets loaded (along with initrd, etc.) is the one installed to the boot partition; that's copied there from its installed location by the "flash-kernel" tool - you can try running "sudo flash-kernel" too see whether it notices the new kernel
[14:16] <waveform> however, my guess is it won't because it's the wrong "flavor" - on the Pi it expects kernels to have the "-raspi" (or historically "-raspi2") suffix. You can override this by adding an "overlay" entry to the flash-kernel database
[14:16] <waveform> (let me just go check something as it's been a while since I messed around with flash-kernel's innards)
[14:17] <Deano59> waveform: yeah it doesn't show the kernel :(
[14:17] <Deano59> it flashes 5.8
[14:17] <waveform> right, under /usr/share/flash-kernel/db/ you'll find "all.db" which is flash-kernel's "database" (it's just a text file)
[14:18] <Deano59> yeah
[14:18] <Deano59> see it
[14:18] <waveform> open that in $your_text_editor and have a look for the "Raspberry Pi 4" entry
[14:19] <Deano59> found it.
[14:19] <Deano59> Kernel-Flavors: raspi
[14:19] <leftyfb> Deano59: why do you need linux-image-5.9-rpi-lobo-arm64 (unsupported)?
[14:19] <waveform> you want to copy that chunk to /etc/flash-kernel/db (which is the local-administrator controlled file which overrides flash-kernel's built-in one)
[14:19] <waveform> and once you've copied it, you've probably guessed already, just delete that "Kernel-Flavors" line
[14:20] <Deano59> waveform: copy all the pi4 lines?
[14:20] <waveform> yup
[14:20] <waveform> that whole block
[14:20] <waveform> (delimited by blank lines))
[14:20] <leftyfb> Deano59: why do you need linux-image-5.9-rpi-lobo-arm64?
[14:22] <Deano59> delete the line from where waveform ?
[14:22] <Deano59> > /usr/share?
[14:22] <waveform> no, in the /etc copy
[14:22] <waveform> that /etc one overrides the /usr/share one
[14:22] <Deano59> there's nothing in the etc one :/
[14:23] <leftyfb> Deano59: why do you need linux-image-5.9-rpi-lobo-arm64?
[14:23] <Deano59> leftyfb: please stop.
[14:23] <leftyfb> Deano59: please answer
[14:23] <waveform> yes, you need to copy the raspberry pi model 4 block to the /etc one, then delete the kernel-flavors line from the /etc one
[14:24] <Deano59> oh :P
[14:24] <Deano59> it's still doing the old kernel
[14:25] <waveform> in that case the new kernel may not be installed in the places flash-kernel expects to find it
[14:25] <Deano59> I added them db files to locations it said it was missing
[14:25] <leftyfb> waveform: this is a lot of effort to support installing an unsupported kernel on a platform with it's own support channel to someone who refuses to explain why they need it in the first place
[14:27] <waveform> leftyfb, I wouldn't call a single entry in an /etc config file a "huge" effort (assuming it worked), and it's also pretty trivial to rollback (remove the override, re-run flash-kernel). I'll admit I'm curious as to why someone might want a later kernel, but I can think of plenty of legitimate reasons one might - doesn't seem like an obvious X/Y problem to me?
[14:29] <Deano59> waveform: this is in etc: https://paste.ubuntu.com/p/tzHynDsGzp/
[14:31]  * mgedmin is curious if the rpi becomes bricked if you flash the wrong kernel and how hard it is to recover
[14:31] <Deano59> mgedmin: if this works (which I hope it does) it *will* benefit other rpi4 users.
[14:32] <waveform> Deano59, yup - that's fine - if it's not working then as mentioned above it's likely that f-k isn't finding your kernel, which probably means it isn't installed in the places it expects to look - at this point I'd say if you *really* want to stick with it you still want to have f-k work (because it's the "right" way to do this and offers an easy recovery mechanism, but you'll need to read /us/share/flash-kernel/functions to figure it out
[14:32] <waveform> mgedmin, in this case "flash" really is a misnomer - it's just copying stuff to the boot partition
[14:33] <waveform> which is just a FAT partition; furthermore when it overwrites a file it always places a .bak next to it so, at worst, recovery is just "copy the .baks over the originals"
[14:34] <waveform> (however, personally I loathe that mechanism - would much rather have separate dirs for old / new state so switching becomes near-atomic - it's on my todo list, but it means a lot of changes to the boot setup)
[14:34] <EriC^^> ttytwister: ?
[14:35] <EriC^^> did you install ubuntu and plug back the windows drive?
[14:46] <wyoung> win 12
[14:48] <dka> I did wanted to upgrade skypeforlinux ubuntu, and it offered me to do apt auto remove
[14:48] <dka> since, I am not able to boot, but I have access to the console with terminal
[14:48] <sunrunner20> what's the current correct way to mount an image to a loopback device upon boot in ubuntu? (well rpi os)
[14:49] <leftyfb> sunrunner20: please join #raspberrypi for support with rpi os
[14:49] <sunrunner20> all I can find is a pre systemd SO post
[14:50] <sunrunner20> its going to be the same
[14:51] <leftyfb> sunrunner20: rpi os is based on debian and has nothing to do with Ubuntu. We can only support Ubuntu here. Not other OS's.
[14:51] <leftyfb> sunrunner20: there's no reason you can't join the support channel for the OS you're running and ask for support with it there
[15:39] <dkfkdn> i've updated and now lightdm fails to start, what to do?
[15:40] <dkfkdn> i have to login manually and use startx (and sound does not work that way)
[15:49] <sqpp> Hi
[15:50] <sqpp> What can I do if, I get Dummy Output in Sound settings?
[15:52] <dkfkdn> https://wiki.archlinux.org/index.php/PulseAudio/Troubleshooting
[15:53] <Deano59> waveform: how long does the server image take to boot?
[15:54] <sqpp> I am not using PulseAudio
[15:55] <waveform> Deano59, on a pi4 it's pretty quick (usually sub 1-minute, though there's wide variation in SD card quality), but the first time boot you do have to wait a good minute (sometimes a bit more) for cloud-init to do its thing, and snapd's pre-seeding takes up a fair bit of IO bandwidth. Obviously those are one-time things though
[15:57] <Deano59> waveform: yeah weird thing is happening, I'm stuck at it trying to configure cloud init, shows the ubuntu loading screen and repeat. Usb boot.
[15:57] <sqpp> Also as a reminder
[15:57] <sqpp> I want to use 5.1 (DTS) on this machine, not sure if possible?
[15:57] <Deano59> Keeps saying a load of lines saying okay then back to the ubuntu loading screen for a second and that's all it does lol waveform
[15:58] <waveform> Deano59, this is the groovy server image? (I've been so focused on desktop  I don't think I've booted a server image with a screen attached for weeks - didn't realize we even had a loading screen!)
[15:59] <Deano59> Yeah server image
[15:59] <waveform> hmm, okay - let me find a spare drive here and see if I can replicate...
[15:59] <Deano59> Keeps saying a load of lines with okay and back to a purple loading screen and that's all it does xD
[16:01] <Pr070cal> Anyone use advanced features zfs on the install
[16:01] <Pr070cal> Of Ubuntu 20.10
[16:02] <Deano59> No.
[16:02] <sqpp> So I guess noone is into DTS who uses Ubuntu?
[16:02] <mgedmin> what's DTS?
[16:02] <dTal> audio codec
[16:02] <Pr070cal> Should I use zfs on 20.10 or overlay FS for read only root
[16:03] <sqpp> Digital Audio
[16:03] <Deano59> waveform: where do you want this video uploaded to of what's happening
[16:04] <Deano59> Imgur?
[16:05] <waveform> Deano59, no need - I've got a spare drive here - I'll try and replicate it once it's flashed
[16:05] <waveform> Am I right in assuming this is basically a fresh flash and you're running it on a pi4? (just want to be sure I'm replicating the same hardware)
[16:06] <Deano59> Yeah fresh flash with the server image with rpis imager
[16:06] <Deano59> https://streamable.com/hbuv4t
[16:06] <sqpp> audio worked for a while, but once I followed some guides on Google to get DTS working it caused the issue
[16:06] <sqpp> but DTS didnt't work just to inform you
[16:06] <sqpp> only stereo sound
[16:07] <Deano59> waveform: video above explains it :)
[16:07] <Deano59> Usb boot if that matters
[16:07] <Austin__> hi all, my new aptxHD capable headphones are cutting out when paired to my laptop. (im using 20.10, new laptop). im suspecting that its because aptxHD codec isnt supported out of the box by 20.10. any pointers?
[16:07] <Deano59> September 3rd eeprom
[16:08] <Deano59> Desktop image is fine, weird.
[16:09] <Deano59> Yeah desktop is fine just not the server image
[16:10] <waveform> Deano59, huh - yup - I'm seeing the same
[16:12] <waveform> well, this definitely warrants an issue
[16:12] <Deano59> Happening to you also?
[16:12] <Deano59> XD
[16:13] <waveform> yup - going to grab a serial console and see if I can figure this out (console's enabled by default on server)
[16:13] <Deano59> Nice :)
[16:13] <Deano59> I'm not alone!!! XD
[16:14] <waveform> oh hang on ...
[16:15] <Deano59> Hang on? Figured out why?
[16:15] <waveform> ha! got it! I shall now make a prediction: you're on a pi4 and you've got a pimoroni fan-shim installed :)
[16:15] <Deano59> Yep
[16:16] <Deano59> How did you know??
[16:16] <Deano59> You have to?
[16:16] <waveform> the fan-shim "abuses" the serial pins and basically fires garbage down them all the time; two options: remove the fan-shim or disable the serial console
[16:16] <Deano59> What's the fix lol xD
[16:16] <Deano59> How can I disable it?
[16:16] <Deano59> I like the fan lol
[16:17] <waveform> in config.txt on the boot partition, change enable_uart to 0
[16:17] <Deano59> Thanks!
[16:17] <Deano59> Gonna try now
[16:18] <waveform> argh, I had that in the release notes on focal and forgot about it on groovy because on the desktop we disabled the serial console precisely because I anticipated *loads* of people using the desktop would have a fan-shim
[16:18] <sqpp> Pls someone :D
[16:23] <Deano59> Thanks waveform
[16:23] <Deano59> :))))
[16:27] <jmspeex> Hi, I have a minor issue that Ubuntu (and probably other distros) turns into a disaster. I suspect the fix would be a one-liner or so, but I just can't find where that would be.
[16:28] <jmspeex> Basically, there's an issue with my battery where it sometimes stops reporting its charge.
[16:29] <jmspeex> As in /sys/class/power_supply/BAT0/ stops existing for a brief moment
[16:29] <jmspeex> That normally wouldn't cause any problem *except* that some daemon somewhere decides that it means my battery is critically low and the system has to be shutdown immediately
[16:30] <jmspeex> Anyone knows which of the many interacting pieces of software is responsible for making that decision?
[16:30] <mgedmin> hmm, upower?
[16:31] <jmspeex> That was my first bet, so I cloned that, but all I can see is layers of redirection and I've been unable to find where that decision is taking place
[16:31] <jmspeex> So either I missed it (totally possible) or upower is just somehow propagating values from somewhere else
[16:32] <mgedmin> I see a bunch of options in /etc/UPower/UPower.conf that hint that it is upower making these decisions
[16:33] <mgedmin> hey can you check the journal of the previous boot to see why the laptop went into shutdown?
[16:33] <mgedmin> journalctl -b -1 -e
[16:38] <jmspeex> OK, all I can really see is: systemd-logind[966]: System is powering down.
[16:39] <jmspeex> which I realize isn't quite informative
[16:43] <sqpp> ok..
[16:44] <jmspeex> OK, so I see something around UP_DEVICE_LEVEL_ACTION. I assume that "action" here means shutdown, right?
[16:44] <jmspeex> In upower/src/up-daemon.c
[16:56] <sqpp> :(  really noone had issues with intel HDA?
[17:02] <tomreyn> !audio | sqpp: this is a bit dated, but many of the debugging approaches still apply
[17:03] <TR2990WX> hi, does a package get updates when the app is installed through snap ?
[17:05] <tomreyn> TR2990WX: i whoever published the snap pushes updates, yes, otherwise, no
[17:07] <mgedmin> TR2990WX: yes
[17:07] <mgedmin> TR2990WX: you can see what apps were automatically updated when with snap changes
[17:07] <mgedmin> and you can see what apps have available updates with snap refresh --list
[17:08] <mgedmin> and you can ask all snaps to update right now with sudo snap refresh, but normally that happens automatically in the background
[17:08] <mgedmin> (except when a snap-ified application is running, that pauses updates for that app for up to 7 days)
[17:09] <sqpp> This didnt help
[17:17] <sqpp> I just still getting Dummy Output
[17:17] <sqpp> :(
[17:28] <TR2990WX> Sysinfo for 'legion5': Running inside KDE Plasma 5.20.1 on KDE neon 5.20 powered by Linux 5.4.0-52-generic, CPU: AMD Ryzen 7 4800H with Radeon Graphics at 1276-1676/2900 MHz, RAM: 4970/15370 MB, Storage: 6/916 GB, 333 procs, 0.52h up
[17:40] <Deano59> TR2990WX: show off. :P
[17:41] <Bustin> Is there a bug in the latest nvidia driver with the latest kernal on Ubuntu 20.10? 20.04 worked fine, doing a fresh install of 20.10 external monitor support seems broken. Getting EXTREME lag with the desktop, can barely move the mouse, or click on anything. One display is Display port connected, the other is HDMI. Disconnecting both, then using
[17:41] <Bustin> the laptops display, it seems to function proper.
[17:44] <Deano59> waveform: just a fyi.... I couldn't get that kernel flashed :( gave up and went ahead and installed the server image :) no more gnome! also, not sure if you actually know or not but kms doesn't work. I just get a blinking cursor. fkms works just fine, though. :)
[17:46] <waveform> Deano59, hmm, sounds like there's been some regressions in kms - last time I tried it (months ago) it at least did *something*; anyway, it's still the "future" option for $when_its_ready - we'll keep an eye on it :)
[17:47] <Deano59> waveform: good to know. :) I love playing with this rpi4. :D
[17:55] <Pr070cal> Anyone used overlayfs
[17:56] <Pr070cal> With read-only root
[17:56] <Pr070cal> So changes to root are stored somewhere else
[17:57] <Pr070cal> Then if you install a new version of the os wiping root your changes are stored somewhere else all you /etc conf file etc you firewall your ids
[17:58] <Pr070cal> I need help with this I had it working once but forgot how
[17:59] <Deano59> ask what *you* need help *with* then wait Pr070cal :)
[18:00] <oerheks1> "With read-only root" .. interesting.
[18:00] <Pr070cal> I need help with overlayfs read only root
[18:00] <Pr070cal> Yeah man
[18:00] <Pr070cal> Fancy doing it with me
[18:00] <Deano59> what sort of *help*? you're not being clear.
[18:01] <oerheks1> and on what ubuntu version?
[18:01] <Deano59> 20.10 iirc.
[18:01] <Deano59> he's been here for a while now oerheks1...
[18:04] <Pr070cal> 20.10
[18:04] <Pr070cal> i started with https://dev.to/napicella/how-are-docker-images-built-a-look-into-the-linux-overlay-file-systems-and-the-oci-specification-175n
[18:04] <Pr070cal> and came up with this
[18:04] <Pr070cal> sudo mount -t overlay overlay-example -o lowerdir=/:/media/user/500GB/lower,upperdir=/media/user/500GB/upper,workdir=/media/user/500GB/workdir /media/user/500GB/tmp
[18:05] <Pr070cal> but its not working
[18:05] <Deano59> define not working. again, you're not being clear. it's not hard. :D
[18:05] <Pr070cal> the example works if you create a file the folders are stacked and merge folder shows files created
[18:06]  * Deano59 lurks
[18:06] <Deano59> :D
[18:11] <Pr070cal> so fancy trying it with me
[18:12] <Pr070cal> oh its not working because root isnt readonly and changes to root dont get written somewhere else
[18:15] <gambl0re> hi everyone, are there any issues running docker on fedora?
[18:16] <Deano59> gambl0re: this is a ubuntu channel... ask in fedora's channel.
[18:16] <gambl0re> i mean ubuntu...lol
[18:16] <Deano59> I'm running docker fine.
[18:16] <Deano59> 20.10. also used 20.04...
[18:18] <mgedmin> I have this one problem where a containerd shim randomly dies and then the container it was managing blocks forever trying to write to stdout
[18:18] <gambl0re> oh i thought you were in #docker
[18:19] <mgedmin> I'm _everywhere_ muahahaha
[18:19] <Deano59> lol
[18:40] <pmitros> Does anyone know of a good Bluetooth headset for WFH/LFH? All the ones I have don't have working mikes in Ubuntu (and yes, I did flip between A2DP and HSP/HSF)
[18:41] <pmitros> My #1 concern is just works. #2 is adequate microphone to be understood clearly (I don't need great quality). #3 is price.
[18:41] <lotuspsychje> !hardware | pmitros
[18:42] <lotuspsychje> pmitros: see also #ubuntu-discuss if you want to meet other users and discussing ubuntu friendly hardware
[18:44] <pmitros> I'm not seeing much on headsets on Ubuntu's wiki. I went down a rabbithole of understanding Bluetooth codecs and things, but...
[18:45] <mgedmin> huh weird my Street HL-430 just works[*]
[18:45] <mgedmin> [*] most of the time
[18:46] <mgedmin> audio quality of HSP/HSF is terrible, you're better off using a separate microphone and A2DP for the headphones
[18:46] <mgedmin> *HFP
[18:47] <pmitros> mgedmin: Part of the point is to be mobile. Kid activities involve running around the house. Adults want to fold laundry during meetings.
[18:47] <Pr070cal> oh i didnt do the docker bit in the tutorial you dont need it just the mount
[18:48] <Pr070cal> for overlayfs
[18:48] <pmitros> But terrible audio quality is probably fine, so long as it works
[18:49] <mgedmin> pmitros: I was on the receiving end, and I have to say the quality was much better when my coworker used his thinkpad's builtin mic compared to his BT headset's mic over HSP/HFP
[18:49] <pmitros> But streetz hl-430 is not coming up on Amazon.
[18:50] <mgedmin> I don't know that I would recommend this particular model, but it was cheap and had a nice carrying case, and there was a sale at a local electronics dealer like a year ago so *shrug*
[18:50] <mgedmin> have you checked that the mic's volume was non-0 in audio settings?
[18:50] <mgedmin> incidentally when you choose the headset as the input source it automatically switches to HSP/HFP
[18:51] <pmitros> mgedmin: Yes. I spent an inordinate amount of time futzing with it.
[18:52] <pmitros> mgedmin: From what I can find online, there are issues with codecs and compatibility, but very little information on how to debug.
[18:55] <mgedmin> bluetooth is the worst technology ever, worse than printers
[18:56] <mgedmin> the hoops I had to jump through to make it work in a dual-boot situation
[18:56] <mgedmin> (both OSes need to use the same private link key or the headset gets confused)
[18:57] <pmitros> Sounds like I should get a dedicated bluetooth receiver, and go over analog, I guess.
[18:57] <Industrial> Hi! `lsof -i -P -n | grep :80` shows `chrome` having bound this port
[18:57] <Industrial> I don't even have chrome, I have chromium-browser
[18:59] <mgedmin> chromium also uses 'chrome' as its process name
[18:59] <Industrial> I see.
[18:59] <Pr070cal> ok i found this https://spin.atomicobject.com/2015/03/10/protecting-ubuntu-root-filesystem/
[18:59] <mgedmin> also are you sure it's a bound port?  outgoing connections to port 80 also show up in lsof
[18:59] <Industrial> Well, why does it take up that port? I want ot launch an nginx server on port 80
[18:59] <Industrial> hmm
[18:59] <mgedmin> bound ports show up as `TCP *:80 (LISTEN)`
[19:00] <mgedmin> it's not suspicious at all to see a web browser talk to remote web servers on port 80
[19:00] <mgedmin> although ideally it would talk to port 443 because encryption is good
[19:06] <Industrial> mgedmin: okay, well, this is the error I'm getting
[19:07] <Industrial> ERROR: for nginx  Cannot start service nginx: driver failed programming external connectivity on endpoint platform_nginx_1 (f00434420580053d8820bef493810706a4d2fe9af19cf2bb200600c09b724140): Bind for 0.0.0.0:80 failed: port is already allocated
[19:07] <Industrial> I'm trying to start an nginx on ubuntu (desktop) with docker and docker compose
[19:07] <Industrial> on port 80
[19:07] <TJ-> Industrial: check what is already there: "sudo ss -tnlp sport = 80"
[19:09] <Industrial> oh. lmao. It's a Traefik proxy that I'm using in the same project >_<. I added nginx because I'll also use that in my kubernetes cluster on production because atm I'm not sure how to configure the traefik for kubernetes and I thought just throwing up a nginx quickly to deploy my `web` container at / and my `api` container at `/api` would be the quickest solution towards golive :D
[19:09] <mgedmin> or sudo lsof -i TCP:80 without the confusing grep
[19:24] <Industrial> TJ-, mgedmin; got it working with Traefik instead of Nginx :-)
[19:26] <Bustin> Is there a bug in the latest nvidia driver with the latest kernal on Ubuntu 20.10? 20.04 worked fine, doing a fresh install of 20.10 external monitor support seems broken. Getting EXTREME lag with the desktop, can barely move the mouse, or click on anything. One display is Display port connected, the other is HDMI. Disconnecting both, then using
[19:26] <Bustin> the laptops display, it seems to function proper.
[19:29] <jmspeex> mgedmin: So I think I managed to understand how to "fix" upower. Now, I need to recompile it. How can I figure out the right configure options to make sure it's "compatible" with the existing binary I have on my system?
[19:33] <mgedmin> jmspeex: best fetch the source package and build on top
[19:34] <jmspeex> how do I do that?
[19:35] <mgedmin> pull-lp-source upower focal (or whatever your distro is)
[19:35] <mgedmin> there should be a guide somewhere online...
[19:37] <MrMobius> hi, I just started using Ubuntu. I'm following some instructions and I should type % export PREFIX = "$ HOME / opt / sh-elf-2.32-9.2.0"
[19:38] <MrMobius> but I get the error bash: fg: %: no such job
[19:38] <mgedmin> jmspeex: the tools that will help you are apt-get build-dep upower; dch -i to increase the version number in debian/changelog to note that your local package build doesn't match the ubuntu version (I tend to add mg1 to the ubuntu version number), and debuild -i to build the patched package; then finally apt install ../*.deb
[19:39] <MrMobius> what am I doing wrong?
[19:39] <mgedmin> apt-get build-deb needs deb-src lines in your /etc/apt/sources.list, these are commented out by default
[19:39] <mgedmin> jmspeex: you also need to have build-essential and ubuntu-dev-tools installed
[19:39] <mgedmin> MrMobius: % is the default zsh prompt character, don't type it in
[19:40] <MrMobius> woops..
[19:40] <mgedmin> also I don't think you're allowed to put spaces around the = in export VAR="value"
[19:40] <mgedmin> and definitely not in "$HOME/opt/sh-elf-2.32-9.2.0"
[19:46] <Pr070cal> It looks like overlayroot is what I want it works but stores changes to tmpfs
[19:46] <Pr070cal> In ram
[19:47] <Deano59> you still going on... lol
[19:47] <MrMobius> mgedmin, ya google translate added the spaces
[19:47] <MrMobius> how can I get the terminal to tab complete a file name? tab doesnt always work
[19:47] <Deano59> I just type and tab, works here.
[19:49] <MrMobius> I must be doing something wrong. it just beeps at me
[19:49] <MrMobius> https://ibb.co/JsDqmxp
[19:50] <mgedmin> use alt-/ to complete pathnames anywhere
[19:50] <mgedmin> by default tab-completion is smart: the first word is the command so it completes all available commands in your $PATH
[19:50] <mgedmin> filenames in . are not in your $PATH so aren't completed
[19:50] <danielb> MrMobius, try typing ls binu and then press tab.
[19:51] <danielb> tarballs are not executable files, so the shell isn't expecting you to try to run them as commands.
[19:51] <MrMobius> mgedmin, that did it
[19:51] <tomreyn> it is not executable so just "binutils" would serve nu purpose and isn't tab completed
[19:52] <MrMobius> right. I was trying tar binu first and that doesnt work either
[19:52] <MrMobius> as mgedmin explained
[19:52] <tomreyn> because "tar filename" is improper syntax
[19:53] <MrMobius> ah
[19:53] <tomreyn> "tar l filename.tar.gz" would have worked
[19:53] <MrMobius> so its smart enough to know I should put a flag first?
[19:53] <tomreyn> yes
[19:54] <MrMobius> neat
[19:55] <mgedmin> bash-completion is nice
[19:58] <tomreyn> MrMobius: sorry, i meant "tar lf filename.tar.gz"
[20:00] <Pr070cal> how do i use overlayfs with a real drive and not tmpfs it says use dev= but doesnt work
[20:02] <Pr070cal> sorry i mean overlayroot
[20:02] <Pr070cal> not overlayfs
[20:09] <Bustin> Is there a bug in the latest nvidia driver with the latest kernal on Ubuntu 20.10? 20.04 worked fine, doing a fresh install of 20.10 external monitor support seems broken. Getting EXTREME lag with the desktop, can barely move the mouse, or click on anything. One display is Display port connected, the other is HDMI. Disconnecting both, then using
[20:09] <Bustin> the laptops display, it seems to function proper.
[20:10] <Deano59> Bustin: you check if there's an active bug?
[20:10] <Deano59> checked*
[20:11] <Bustin> Deano59, sorry, where do I check for active bugs? I did a quick google search, and since the release is so new, there aren't many (but there are a few, 2 actually, 3 including me), saying they have the same problem.
[20:14] <garrettkajmowicz> I have a 16.07 machine running an i386 install. It only has 4GiB of RAM so I didn't bother switching to x86_64 when I moved to a compatible processor. Now I need to upgrade to a newer version of Ubuntu. How can I make this work?
[20:14] <Deano59> garrettkajmowicz: fresh install if you *don't* want problems. :)
[20:14] <dust> restore thunderbird... in 20.04 had a snap install and now at 20.10 is normal package... how to restore from snap to normal install? its super important!
[20:15] <Bustin> Deano59, that's not true, I am attempting a fresh install, and having nothing *but* problems lol.
[20:15] <Deano59> Bustin: idk it off by heart but sounds like you found it already
[20:15] <garrettkajmowicz> Deano59: That's why I run LTS distributions. And why I'm asking about upgrading.
[20:15] <Deano59> Bustin: nvidia have always had issues with drivers on linux.
[20:16] <Deano59> garrettkajmowicz: what's your specs?
[20:16] <dust> restore thunderbird... in 20.04 had a snap install and now at 20.10 is normal package... how to restore from snap to normal install? its super important!
[20:16] <Bustin> 20.04 was fine, so was Pop!_OS ootb, prior to this current release (they just released their new update yesterday too, following Ubuntu's).
[20:16] <Bustin> leads me to think it's a new driver, or Kernal issue
[20:16] <Deano59> Bustin: I done 3 fresh installs, including a rpi4 (one issue) and they've been fine.
[20:16] <FunGUI> What is the number one method to staying secure and private on Ubuntu?
[20:16] <garrettkajmowicz> Deano59: -4.75 diopters.
[20:17] <Deano59> guntbert: encrypt your hdd.
[20:17] <Deano59> garrettkajmowicz: excuse me?
[20:17] <garrettkajmowicz> Deano59: Bad pun. (Specs as in spectacles, -4.75 diopters is the principle part of a lens prescription)
[20:17] <Bustin> Deano59, well you're lucky the. I mean, for the most part, older hardware is usually, for the most part, issue free, but I'm on *somewhat* newer hardware.
[20:17] <Bustin> then*
[20:18] <Deano59> garrettkajmowicz: funny, I mean specs of your system.
[20:18] <dust> restore thunderbird... in 20.04 had a snap install and now at 20.10 is normal package... how to restore from snap to normal install? its super important!
[20:18] <Deano59> if it's nvidia like Bustin's could probably have the same bug. :)
[20:19] <Deano59> dust: we saw, wait like everyone else.
[20:19] <garrettkajmowicz> Deano59: Minor server. Uses a headless embedded motherboard (ASUS C60-1) with a NIC plugged in, and the RAM maxed out at 4 Gib. A MD array of 2 SSDs mirrored. Nothing remotely spectacular.
[20:19] <Deano59> so a clean install would be *best* for you imo.
[20:20] <garrettkajmowicz> It would be inconvenient. That's the whole point of upgrades. (Not that Ubuntu has ever managed to have upgrades work correctly)
[20:20] <Deano59> yeah so a fresh install would be best. you asked. :)
[20:20] <Bustin> haha
[20:21] <Deano59> I've never upgraded from 1 ubuntu to another, I don't like fixig issues.
[20:21] <Deano59> fixing*
[20:21] <Bustin> Likewise, always fresh install.
[20:21] <Deano59> yeah but you're out of luck, Bustin :D
[20:21] <Deano59> you could always try the latest gpu drivers though.
[20:21] <zeroes> someone know how to fix usb capacity? It is "Disk /dev/sdb: 956 MiB", too low.
[20:21] <Deano59> sec let me see if I can find it.
[20:22] <tomreyn> zeroes: who much did you expect to see?
[20:22] <zeroes> 8 GB
[20:23] <tomreyn> zeroes: can you pass the url returned by     sudo fdisk -l /dev/sdb 2>&1 | nc termbin.com 9999
[20:24] <Deano59> Bustin: when you're booted with your laptop screen attatched, trying adding this, apt update and reboot: sudo add-apt-repository ppa:oibaf/graphics-drivers
[20:24] <Deano59> may fix it, may not Bustin.
[20:24] <tomreyn> wouldn't it be better to try and understand what the problem is beforehand?
[20:24] <zeroes> tomreyn: https://termbin.com/5ck3j
[20:25] <Deano59> tomreyn: go ahead. I'm still learning. :)
[20:25] <tomreyn> zeroes: this is an external slash drive, 1 GB capacity, i t would seem.
[20:26] <tomreyn> zeroes: you could check how it was initialized by searching through    journalctl -b    for     sdb
[20:27] <zeroes> tomreyn: yesterday it was way bigger than than 1 GB. I tried to write an os to usb stick at capacity got messed up.
[20:27] <tomreyn> Bustin: are you able to install ubuntu 20.10, yet, or did you already install?
[20:27] <TJ-> garrettkajmowicz: step 1: replace 32-bit i386 kernel with amd64 kernel so it is running a 64-bit kernel + 32-bit userspace. step 2: add amd64 as a foreign architecture if not already. step 3. create a chroot with the new amd64 release userspace in it using 'debootstrap --foreign ...' . step 5. use 'debfoster' to get a list of all the 32-bit top-level packages that are installed, then install those
[20:27] <TJ-> in the amd64 chroot. step 6. copy over the service configs from /etc/ on the i386 to the amd64 chroot. step 7. generate a 64-bit initrd.img in the chroot, copy it to the 32-bit /boot/ directory along with the matching vmlinuz-$VERSION. step 8. on the 32 bit host, run 'update-grub' to add the 64-bit kernel+initrd.img to the boot menu. step 9. reboot and in GRUB choose the new 64-bit entry, edit its
[20:27] <TJ-> 'linux ...' line and add "break=mountroot" then boot into the initrd shell. step 10. inside the initrd busybox shell mount the rootfs, create a sub-dir /old/ in the root-fs mountpoint $m, and do "mv $m/usr $m/var $m/old/" then mv $m/new/usr $m/new/var $m/" and other minor fixups. step 11. 'exit' to continue the boot into the 64-bit userspace and new release if you're lucky
[20:28] <TJ-> garrettkajmowicz: or just do a config backup and install the new amd64 release from scratch :)
[20:28] <tomreyn> zeroes: let's see this:   sudo hdparm -I /dev/sdb 2>&1 | nc termbin.com 9999
[20:29] <garrettkajmowicz> Hrrrm. Now to see if I can find a CD (not DVD) install ISO I can use.
[20:29] <zeroes> tomreyn: https://termbin.com/z9j1r
[20:29] <TJ-> garrettkajmowicz: you could potentially do a net install via PXE
[20:30] <garrettkajmowicz> This is my server which would be the PXE server.
[20:30] <TJ-> garrettkajmowicz: you could use your laptop/PC as a temporary TFTP+NFS host for PXE booting
[20:30] <tomreyn> zeroes: try again, this time with lower case I
[20:31] <tomreyn> zeroes: if this also reports 1 GB, you can try installing f3 (available via apt) and testing with that: http://oss.digirati.com.br/f3/
[20:31] <zeroes> https://termbin.com/k86k
[20:32] <tomreyn> zeroes: you ran it with duso?
[20:32] <tomreyn> * sudo
[20:32] <zeroes> ran as root
[20:33] <tomreyn> well, try f3, and watch journalctl -f
[20:34] <tomreyn> you could also try to just write a new partition label to it (loosing all the data stored on it)
[20:34] <tomreyn> partition table, that is
[20:34] <jmspeex> mgedmin: where should I run dch -i? and how do I add a suffix?
[20:35] <tomreyn> zeroes: i.e. using fdisk or parted
[20:35] <zeroes> tomreyn: which f3 command should I use?
[20:35] <zeroes> there is nothing in /dev/sdb
[20:36] <tomreyn> start with f3probe
[20:38] <Bustin> Deano59, I personally believe it's a kernal issue, 20.04 was 5.4? 20.10 is 5.8? I mean, as I said, prior versions seem to work fine with external monitor support. One of the monitors is plugged in via USB C to display port, and I think 5.8 adds some sort of support? Will have to confirm.
[20:41] <tomreyn> usb c to DP has been working for a good while. at least with open source drivers, not sure about nvidia
[20:42] <Bustin> Yea, just odd it's happening on a fresh install, on Ubuntu 20.10, AND now on Pop!_OS's release yesterday, which follows Ubuntu's.
[20:42] <Bustin> prior version of both worked fine OOTB
[20:42] <Deano59> wait for the bugs to be ironed out then Bustin :(
[20:42] <TJ-> Bustin: maybe the nvidia driver kernel wedge hasn't kept up with the mainline Linux kernel
[20:43] <zeroes> tomreyn: https://pastebin.com/AXuEhnE4
[20:43] <tomreyn> zeroes: so i guess it was fake 4 GB yesterday, now is what it really is, 1 GB
[20:44] <tomreyn> sorry, 8 GB yesterday
[20:49] <zeroes> Don't understand this anymore. USB stick is cheap, but I am interested in fixing it :)  Long time ago I got similar problem but USB stick vendor told me to run their program to fix. Unfortunately, their program is not open source.
[20:50] <zeroes> Something similar seems to have happened: https://askubuntu.com/questions/289971/usbs-storage-capacity-reduced-to-2-mb-from-16-gb
[20:50] <zeroes> tomreyn: thanks for helping anyway. Learned new tool f3. and termbin.com
[20:53] <tomreyn> zeroes: fdisk seems to have confirmed that it's not just a matter of unallocated storage that you'Re seeing. fdisk reports a total stroage of 1 GB. and f3 confirmed it.
[20:53] <tomreyn> f3 actually writes different patterns of data to the stick and reads it back, ignoring any capacity limits
[20:56] <tomreyn> * total capacity (not storage)
[21:07] <zeroes> tomreyn: happened again, https://pastebin.com/JzR786vv :D
[21:08] <zeroes> from 14.9 GiB => 3.8 GiB ^^;
[21:08] <zeroes> I have only one DataTraveler G2
[21:09] <tomreyn> zeroes: this is a different stick now, though, right?
[21:09] <zeroes> This is different stick.
[21:10] <tomreyn> so it didn't happen again.
[21:10] <zeroes> first stick from 8 GB => 1 GB, second stick from 14.9 GiB => 3.8 GiB
[21:10] <zeroes> first stick and second stick are different stick.
[21:12] <tomreyn> these are either fake capacity sticks or faulty hardware (or both)
[21:49] <dust> restore thunderbird... in 20.04 had a snap install and now at 20.10 is normal package... how to restore from snap to normal install? its super important!
[21:52] <TJ-> dust: find out where the user's snap .thunderbird/ directory lives so you can get the profile
[21:52] <dust> and where is that?
[21:53] <TJ-> dust: 20.04 installed thunderbird from the archive too; if you installed from a snap that would have been a manual choice
[21:53] <dust> thats my question...
[21:53] <TJ-> dust: I don't use snaps but maybe $HOME/.snap/ ?
[21:54] <tomreyn> i think the snap data folder isn't hidden, so it's just $HOME/snap
[21:54] <ttytwister> tomreyn, yeah, installed ubuntu, and then plug in win 10 drive
[21:55] <TJ-> tomreyn: ahhh, right
[21:55] <tomreyn> ttytwister: were you responding to <EriC^^> did you install ubuntu and plug back the windows drive?
[21:55] <ttytwister> yes
[21:55] <dust> in home7snap7thunderbird are no mails
[21:56] <dust> there is no thunderbird profile...
[21:56] <TJ-> dust: did you look for 'hidden' directories? e.g. "find $HOME/snap/thunderbird -ls"
[21:57] <dust>  /snap/thunderbird’: No such file or directory
[21:59] <TJ-> dust: you do have $HOME set? because the path should have been /home/$USER/snap/thunderbird
[22:00] <dust> yes but in that folder is no profile... i can only use the deja dup backup to restore or i find a way to mount the broken zfs of 20.04... thats the reason i did a fresh install on another disc of 20.10
[22:02] <dust> seems the snap profile of thunderbird isnt in home and deja dup didnt do a backup because of this
[22:03] <TJ-> dust: snap is a weird monster best avoided"
[22:03] <davros> is it ok to ask here about setting up samba in a linux install (18.04)
[22:03] <dust> tell that the ppl who decided on ubuntu side to use snap...
[22:09] <compdoc> davros, you can ask, or try #samba
[22:11] <audio> why can't i install python3-venv on 16.04, 18.04, nor 20.04? Did sudo apt-get update and everything
[22:11] <davros> ah let me try #samba first. to sumarise, i see what i setup with smbclient -L localhost, but from other machines using the IP adress of my desktop machine, it always say NT_STATUS_CONNECTION_REFUSED  or NT_STATUS_UNSUCESSFUL or something.. )..   SSH is working fine, i'm using SHH to make the other machine attempt to connect to samba
[22:11] <audio> "No installation candidate"
[22:16] <audio> hello?
[22:16] <audio> 1171 people in here and no one is talking
[22:19] <tomreyn> audio: run     apt policy python3-venv 2>&1 | nc termbin.com 9999
[22:19] <tomreyn> it will return a url you can share here
[22:19] <krytarik> My guess is 'universe' is disabled..
[22:24] <tomreyn> 1 audio and not one is talking
[22:25] <audio> tomreyn: i fixed it
[22:25] <audio> needed another update/upgrade
[22:26] <tomreyn> congratulations.
[22:52] <ttytwister> tomreyn, what if I install ubuntu first, then windows 10, and then somehow repair grub. Will it work?
[22:53] <WoC> the other way around is alot easier
[22:53] <tomreyn> ttytwister: i don't know the context (you weren't discussing this with me), but generally, i'd suggest the other way around.
[22:54] <tomreyn> or better yet, don't dual-boot at all
[22:54] <WoC> vm ftw :)
[22:54] <WoC> if you really have to have win
[22:54] <ttytwister> not my case
[22:55] <ttytwister> I need win on top of hardware not on top of ubuntu
[22:55] <WoC> gpu ?
[22:56] <WoC> Anyhow, for multi-boot, very simple as long as you install win first
[22:57] <ttytwister> WoC, yes, but it is simple if don't need encryption
[22:57] <WoC> afaik, you can do that too, during install
[22:58] <WoC> and installer set it up for you
[22:58] <ttytwister> It seems to me I can't. I can do it if I agree to erase everything including win 10
[22:58] <ttytwister> I tried yesterday
[22:59] <WoC> ah, fulll disk encryption ?
[22:59] <WoC> for multi-boot, that would be a bad pick
[22:59] <ttytwister> I have two hard drives. One for win, another one - for ubuntu. Yes I want the hd for ubuntu to be fully encrypted
[23:00] <ttytwister> WoC, so, should I install both OS on one hard drive?
[23:00] <WoC> two disks should be fine
[23:01] <ttytwister> https://www.mikekasberg.com/blog/2020/04/08/dual-boot-ubuntu-and-windows-with-encryption.html
[23:03] <WoC> Hmmm, I to make it easy, d/c the linux disk and install windows, then d/c the windows disk and connect the linux disk and install linux, then you reconnect both and use the uefi menu for boot menu until you have grub updated
[23:04] <WoC> anyhow, that's how i would
[23:05] <tomreyn> it's also what was suggested to ttytwister some hours ago
[23:05] <WoC> oh
[23:08] <WoC> i hope they fix the timezone thing some day...
[23:12] <WoC> Good thing there's rc.local :)
[23:14] <ttytwister> WoC, I didn't manage to do it. I'll try again
[23:16] <WoC> for the linux disk, it will still wipe the whole disk
[23:16] <ttytwister> it's ok, because I'll use the whole disk for linux
[23:17] <WoC> just making sure
[23:18] <WoC> you d/c'ed the win disk completely right ?
[23:18] <ttytwister> what is d/c? detach?
[23:18] <WoC> dis-connect = d/c
[23:19] <ttytwister> WoC, No, I think I disconnected linux disk when I was install windows
[23:19] <ttytwister> *installing
[23:20] <WoC> right, then linux the same way (w/o win disk)
[23:20] <ttytwister> WoC, this is what I'm going to try
[23:20] <WoC> k :)
[23:20] <ttytwister> and when I finish, what should I do with grub?
[23:21] <ttytwister> 'sudo update-grub' ?
[23:21] <WoC> should work, as long as you remember to use the uefi boot menu to choose os
[23:21] <ttytwister> ok
[23:21] <WoC> aye update-grub2 should detect otheros
[23:22] <WoC> and you can use efibootmgr to set default disk
[23:22] <WoC> so you dont really need win in grub
[23:23] <ttytwister> ok, I got it
[23:23] <ttytwister> thank you
[23:23] <WoC> np :)