[00:00] sarnold: Well, I'm a maintainer of a java game engine and our openGL Backend depends on lwjgl2 (as lwjgl3 still has it's quirks for us). And as I am usually the guy who has to fix things when others give up, that's my joy for this week :D [00:01] Now lwjgl2 depends on AWT (the window toolkit) with a "versioned reference" and this causes glibc to assert when trying to load lwjgl2 [00:05] Darkchaos: so, this is WILD SPECULATION. the https://launchpad.net/ubuntu/+source/lwjgl package shows it's unchanged since xenial -- when it was build with g++ 5.x [00:06] Darkchaos: I wonder if your much newer builds may not be compatible with something so old. I can't recall any c++ abi differences between xenial and bionic but I'm also pretty far removed from those events [00:07] sarnold: That shouldn't be a problem, as AdoptOpenJDK and other Distributions work, only the one from Ubuntu doesn't. Even Debian seems to be working (and it works on the same code, so I could only speculate that gcc 8 fixes it and gcc 7 breaks it) [00:07] Maybe also one of the ubuntu patches of gcc break it [00:09] heh, good point, debian hasn't rebuilt since 2016 either https://launchpad.net/debian/+source/lwjgl [00:10] well lwjgl only seems to trigger the bug by depending on "libjawt.so" with a versioned reference [00:10] https://bugs.launchpad.net/ubuntu/+source/openjdk-lts/+bug/1838740 is the issue [00:10] Launchpad bug 1838740 in openjdk-lts (Ubuntu Focal) "libjawt.so inconsistency lets JVM crash" [Undecided,Confirmed] [00:10] They have a different way of triggering the bug, but still [00:13] i'm trying to configure and run samba ad and i'm running into an issue [00:14] it seems to be failing on the mit kdc daemon [00:17] Darkchaos: it'd be worth tacking on a note with your findings; the one guy is suggesting too-new gcc is a problem, heh [00:17] sarnold: Well, I am the last one commenting there, I just did have this name registered already with NickServ [00:18] aha! [00:32] Greetings, on a freshly installed Xubuntu 18.04.3 system... and installed realmd, sssd-ad, and oddjob-mkhomedir... and joined the system to a Microsoft Active Directory domain... and I can authenticate... but it fails to create a directory for me when I login. Are there some other packages that need to be installed for that to work? [00:34] dowdle: do you get any errors logged to the journal or log files? [00:34] dowdle: how about the other hosts on the network? [00:37] sarnold: This is the first ubuntu-based system I've tried to do this on... so I'm not sure about other hosts. I'm more familiar with CentOS and Fedora... and so what works on those, is working well on Xubuntu except for the home directory creation part. [00:45] sarnold: Thanks for answering... I did a "journalctl -f" while logging in... and I didn't see anything that looked relevant... no errors about home directory creation failure. [00:46] dang :( [00:46] Unfortunately I have to run... but I'll do some web searches later and get it figured out. [00:52] every minute, the lock screen clock bleeds through to my desktop when the number changes. Stays there until something draws over it [00:53] very distracting .. any ideas? [00:59] sarnold: I figured it out... I had to run pam-auth-update and enable "Create home directory on login". [01:00] dowdle: awesome! [01:00] dowdle: thanks for reporting back :) [01:03] sarnold: Sure thing. Most of the instructions I found from web search had a lot of file editing for not good reason. All one needs to do is: "reamld join -U $user-who-can-join $DOMAIN.NAME" [01:04] Then edit /etc/sssh/sssd.conf and to turn off full domain name requirement... and then that pam-auth-update part... and it is actually pretty easy. Yeah, one file to edit but that just lets one not have to do fulldomain\username and instead username [01:04] Anyhoo... [01:04] Later [01:26] 18.04.4 was supposed to be released a couple of days ago but I can't find it [01:26] should i install docker.io or docker-ce on Ubuntu 19.10? [01:26] entourage: https://lists.ubuntu.com/archives/ubuntu-release/2020-February/004897.html [01:28] sarnold, thanks [01:33] Spotify is not working for me [01:33] The browser version keeps telling me to enable DMR [01:33] and the desktop version is slower than a turtle [01:34] not sure what to do === PowerTower_121 is now known as PowerTower_120 [01:51] I installed ubuntu and checked the ZFS option now after an update my bios does not detect a bootable device so I assume GRUB is damaged. The entire drive is apparently zfs and I have no idea how to fix it. [01:54] Some say use mount others say fuse, bsd and solaris each have their separate version of ZFS. And neither nautilus nor gparted or any other tool seems to interact with zfs. [01:56] are you booting UEFI or BIOS? did you install UEFI or BIOS? [01:57] sarnold: I installed and boot UEFI, but I tried both methods since it has happened in the past. [01:57] I even tried disabling the hypervisor and several other bios options. [01:58] (is zfs even supported without UEFI?) [01:58] what is correct way to run webserver on ubuntu [01:58] do i just add /var/www/html to nginx group? [01:58] then add my user to nginx group [01:58] ? [01:58] good question; my one BIOS system that uses ZFS doesn't have root on ZFS, just a storage pool.. and my ZFS-on-root systemis UEFI [01:59] vion: before zfs was added to the installer, community members put together a guide on how to do it by hand [01:59] @celphi doesnt nginx has a sites-enabled file in /etc/ ? [02:00] vion: at the end of the guide is a very quick troubleshooting piece https://github.com/zfsonlinux/zfs/wiki/Ubuntu-18.04-Root-on-ZFS#troubleshooting [02:00] So is GPU offloading supposed to work with intel/nvidia setups (using the proprietary nvidia drivers) in Ubuntu 20.04? [02:00] vion: probably skip the cryptsetup step, but the zpool export / import / zfs mount commands ought to be useful [02:01] silg: that file exists but im trying to add html files to my www dir and im getting permissions error-- so im wondring right way to do this [02:01] sarnold: I think I will just get my files and call it quits on saving the system, but it would be nice if grub worked [02:01] do i just add my user to nginx group and add /var/www/html to nginx group and make it 774? [02:01] celphi: I think it's a bad idea for the webserver to have write access to its files, but that's the debian way -- run the daemon as www-data and have the files owned by www-data [02:02] yeah just check your file permissions [02:02] sarnold: then how would you save files sarnold ? [02:02] sarnold: I want to redo my zfs root installation with cryptsetup at some point... require a USB key inserted to mount the root fs [02:03] vion: it's probably still worth trying these ~dozen commands, they might let you just run update-grub [02:03] sarnold: hey so I noticed my bios had the drive set to raid and not ahci is that usual? I tried switching it up and it did nothing different [02:03] celphi: it depends upon how you want to manage the files; maybe you create a new group for them, maybe you create a new user for them, maybe you just use your user account already.. it depends upon what you're trying to do to, how many people need to be able to change them, etc [02:04] vion: I think ahci ought to be the usual choice [02:04] im just trying to run a webserver out of /var/www/html.. i usually do it as root but im new to ubuntu [02:04] i usually just make my files but i get a bunch of permission issues [02:04] sarnold: I checked the machine over hoping i didn't some how install grub to an sdcard or flash stick by accident lol [02:05] i can easily do chown newuserhere html/ [02:05] but is that irght way? [02:06] celphi: avoid using cp and use mv instead to avoid all that permission hassle [02:06] celphi: chown -R probably, but yes, that's certainly a fine way to do it if you just want to keep it simple [02:06] also there's cp -a [02:06] vion i cant even make a file- [02:06] have to sudo everytime [02:06] ty sarnold [02:07] celphi: touch nameoffile [02:07] same thing error [02:07] im just going to do usermod -a -G nginx [02:08] celphi: hold up is your user not in the sudo group? [02:08] and add my user to nginx group [02:08] it is [02:08] celphi: I like that nginx user idea even better :) [02:08] celphi: what about that wheel group [02:08] not wheel but sudo is there [02:09] if you are doing it graphically then needs wheel group I think [02:09] yeah im not sure i want to give my user the wheel group access [02:09] nginx group seems like a better option [02:11] is there a reason why a webserver directory would need X [02:11] permissions [02:11] for everyone [02:12] can i just do 774? [02:12] nvrmind [02:12] only works for 775 [02:12] nginx is some kind of proxy server that has all manner of porthole for plugging [02:13] one would add the user to www-data .. [02:13] oerheks: i went with nginx instead [02:13] celphi: you using the load balancing or cache features? [02:13] possibly [02:13] celphi: because nginx is running as user www-data, it needs to access the docroot via the 'other' permissions [02:13] mostly for reverse proxy [02:14] i run it as user nginx [02:14] oh snap lol. [02:14] ubuntu has it different [02:23] how come ubuntu is not adding my user to group? [02:23] im doing `sudo usermod -a -G "www-data" usernamehere` [02:23] you need to log in in order to get new supplemtnary group memberships [02:24] do what? [02:24] I saw some guide that suggested to setup a crontab for when I reboot my server to restart irssi but I've setup irssi under a user 'irssi'. How do I do that? [02:24] i have to logout and log back in ? [02:24] celphi: the easy thing to do is log out and log back in again [02:24] @reboot su irssi screen -S irssi [02:24] robertparkerx: sudo -u irssi crontab -e ... screen -S irssi something [02:25] ah lol [02:25] Ty [02:27] @reboot screen -S irssi [02:27] is that valid [02:27] sarnold: didnt work [02:27] still same groups [02:27] i typed `groups` and it's not listed` [02:28] celphi: hrm, try adduser, it's less typing: adduser celphi www-data [02:29] thats weird [02:29] it says im already a member [02:29] but it doesnt show it listed [02:30] you may need to logout, switch to a tty, login there, run "loginctl terminate-user $USER", log out there if needed (but you should get logged out automatically), switch back to the graphical login, login. [02:30] so, very convenient ;) [02:30] check this out [02:30] tomreyn: what's the loginctl bit for? [02:30] sarnold: destroying lingering systemd user sessions, i think [02:30] https://usercontent.irccloud-cdn.com/file/Ti6Qx7qb/image.png [02:31] https://askubuntu.com/questions/1045993/after-adding-a-group-logoutlogin-is-not-enough-in-18-04 [02:31] Does that image make any sense to you? [02:32] if i `su kerafyrm` i see the group i added. [02:32] but if i exit back to kerafyrm it shows a different set of groups [02:33] that's unrelated, but makes sense to me, yes [02:33] https://usercontent.irccloud-cdn.com/file/v8rHqMv9/image.png [02:33] better image [02:33] how is this possible? [02:33] you seem to have incorrect expectations about what "su kerafyrm" does. [02:34] doesnt it switch user? [02:34] maybe you wanted to run "sudo -u kerafyrm -i" [02:34] `su` doesnt mean switch user? [02:35] sudo only executes one sudo command [02:35] "su" is short for "switch user", yes. but this is not a precise definition of what it does. [02:35] you'd still see differences about group membership when using sudo as suggested. [02:35] It means "substitute user [02:36] Nowhere in the man page for su does it mention the word "switch". [02:36] whoops, i think nelgin is right there [02:36] celphi: hmm, what does 'getent initgroups kerafyrm' report? [02:36] how is it that with my `whoami` i see kerafyrm and only two groups but then when i exit and do `whoami` again how do i get more groups? [02:36] 33 sarnold [02:37] nelgin: ... nor the word "substitute" ;) [02:37] celphi: oh shit. [02:37] celphi: do you still have a root shell somewhere? [02:37] hi everyone! [02:37] celphi: do you still have a shell with sudo in the groups output? [02:38] I have something peculiar to show y'all [02:38] It does in the CentOS man page ;) [02:38] root shell somewhere? [02:39] celphi: ASAP run sudo adduser kerafyrm sudo [02:39] https://imgur.com/a/llpQWtK [02:39] I get this same error every time I try to boot ubuntu [02:40] I boot into emergency mode with no option to use recovery mode or a shell. Attempting to recover from a live disks has so far proved ineffective. [02:41] thenori: which error exactly, the one in red on the bottom line? [02:41] "+ shellcheck 'src/*.sh' - src/*.sh does not exist" [02:42] Frankly I'm baffled. I've never encountered shellcheck as a part of the boot process before and I'm unsure why it's looking for "src/*.sh". [02:42] it feels like there's a shellcheck src/*.sh command stuffed in your /etc/*profile* or something [02:42] That seems rational [02:43] celphi: have you added yourself back to the sudo group yet? [02:43] As I've gotten a live disk running at last, testing such a hypothesis should be trivial [02:49] Aha! [02:49] A grep has got me to grok it [02:49] thenori: whoohoo, where was it? :) [02:50] > /etc/bash.bashrc:shellcheck src/*.sh [02:50] that'll do it ;) heh [02:51] it's nearby some other commands [02:51] > set -euo pipefail [02:51] >set -o xtrace [02:51] and that's why it dies, heh [02:52] what do those do? [02:52] I don't know how it got there, frankly. I don't touch /etc/bash.bashrc [02:52] set -euo pipefail is a common way to try to make shell scripts more useful [02:52] set -e will die on errors [02:53] but commands in the middle of a pipeline that die don't trip the set -e [02:53] so pipefail tries to catch those failing too [02:53] set -u makes unset variables painfully loud [02:54] set -o xtrace writes tracing information to the output, which is how you spotted the error at all. without that this probably would have been a thousand times more confusing [02:54] so, nothing out of the ordinary [02:55] but combined they mean that the shell is not going to work well at boot [02:55] which reminds me [02:55] celphi: have you added yourself back to the sudo group yet? ASAP run sudo adduser kerafyrm sudo [02:55] because if celphi logs out or reboots or something, he or she's going to be stuck doing the same thing you';re doing, trying to use the rescue environment to get the computer back === zbenjamin_ is now known as zbenjamin [02:57] rescue environment didn't work at all [02:57] I've never encountered 'emergency mode' before [02:58] thenori: cripes [02:58] I am using sublime text which borrows the system font for its left panel. How do i change the system font on ubuntu 18.04? [02:58] just a recursive loop [02:58] thenori: you've made me wonder if bash will read those if you use init=/bin/bash even :) [02:58] there are probably other shells [02:58] i mean i just did [02:59] thenori: ah I thought you went to rescue disk [02:59] thenori: how much csh do you remember? :) [02:59] none [02:59] erf, tomreyn, how much csh do you remember? :) [02:59] i hear fish is popular [02:59] about as much as thenori [03:00] i know people use zsh too [03:00] with trolls, yes [03:00] the "make nothing work right" shells, heh [03:02] gnome-tweak-tool can handle font settings [03:02] Is anyone here familiar with na_record from speech-tools? I can't get it to capture anything. Just exits immediately. [03:03] not sure if sublime text in confined snap behaves differently [03:03] Anyone using ubuntu on 4k? I could use some help configuring font sizes [03:03] 2560x1440+0+0? [03:05] kyle__, ? [03:05] !info gnome-tweak-tool [03:05] gnome-tweak-tool (source: gnome-tweaks): adjust advanced settings for GNOME - transitional package. In component universe, is optional. Version 3.34.0-2 (eoan), package size 1 kB, installed size 12 kB [03:06] oerheks, that did not help [03:06] cgi: Is that close enough to 4k? I haven't looked at monitor marketing terms in ages. [03:06] oerheks, I set all my fonts large - and sublime text still renders small fonts. The "X" button on windows is still very tiny. The spacing between the words on a window top is still messed up [03:07] With that monitor, I use st built with a monospace font at 12 point, antialiased. [03:07] for 19.10 there are some tweaks, http://ubuntuhandbook.org/index.php/2019/10/how-to-enable-fractional-scaling-in-ubuntu-19-10-eoan/ [03:07] Works great. [03:07] kyle__, its 3840x2160 or something like that [03:07] Ahh ok. [03:07] I am on 18.04LTS - no fractional scaling here [03:08] cgi: Theoretically, on a properly designed graphics system, it knows the resolution of your monitor, and the physical dimensions, and a 12 point font will be the same size regardless of what monitor you're on. [03:08] i've never encountered any issues with text scaling in vim [03:08] But after 30 years I think Apple is still the only company that actually does that :/ [03:08] company, OR organization. [03:09] kyle__, maybe time to ping canonical/gnome with a suggestion? [03:10] how about a wayland session? your issue is linux wide, i believe [03:10] anyways, tyvm sarnold :) [03:10] gonna go reboot and get this bb running [03:11] btw: if anybody ever shittalks ubuntu to you, let it be known that that is the only live disk that worked [03:11] not kali, not nothing [03:11] oerheks, have you used wayland ? [03:12] wait was thenori running a blackbird?? [03:13] cgi, sure, but not with newer hardware and 4k [03:19] celphi: have you added yourself back to the sudo group yet? ASAP run sudo adduser kerafyrm sudo [03:27] okay sarnold [03:27] thenori: "bb" .. is that a blackbird? :) [03:28] or just "bionic beaver"? [03:28] "bad boy" also wouldn't surprise me.. but I want to believe :) [03:30] https://imgur.com/a/Ljbm34h [03:31] i am not sure hat you're referring to but i am oe to notice that my uu key is still disabled on this box [03:32] hmm; did you remove the set -euo pipefail and set -o xtrace lines? [03:32] what else was in that file? [03:32] nope, all i removed uuas shellcheck [03:33] remove those two lines as well [03:33] https://paste.ubuntu.com/p/zBdng7sFgR/ [03:33] there's my /etc/bash.bashrc -- I can't recall what I changed vs what was there already, but it ought to be close to what you've got... [03:33] maybe this will help you spot other strange lines that ought not be in that file [03:34] alright, I uuill do so [03:35] before i begin another episode of voyager, though, i uuill reset this box so that i may regain control over my uu key [03:35] hehehe [03:35] fareuuell [03:35] see ya [03:38] whee [03:38] uuelcome back [03:38] i can type two `u` at once [03:40] gdit i can't turn around for a minute without something falling apart around here [03:40] no sound [03:40] bother [03:50] Hi I'm trying to setup a systemd service to fix the screen size automatically and it's not working. I tried the following at https://askubuntu.com/q/1208933/835021 [03:51] I did some further digging and found that by default xrandr doesn't have access to the DISPLAY environment run as root. If I explicitly set it up in the systemd service as :0.0 it's still unable to set the display size. [03:51] lunorian: there's got to be something way better than that. I don't know what. [03:51] lunorian: it'll also need an .Xauthority for the running server [03:51] For the time being on every login I have to run /usr/bin/xrandr --output eDP-1 --scale 0.66x0.66 manually [03:52] I assumed that since systemd is running as root it wouldn't matter [03:53] I'd also like the login screen to have a pretty resolution [03:54] setting the display size with xrandr is perfect aside from the context issues making the systemd service not work [03:54] the manpage for the xauth(1) command describes how the whole authentication thing works; it's best to NOT USE xauth to try to solve the problem though [03:54] but I seriously hope there's a way to get the same change to your X server via xorg configuration changes instead [03:55] Won't ubuntu overwrite those on system updates? [03:55] That's why I figured it'd be best to just setup a startup service [03:55] there's no configs by default, so nothing to overwrite [03:57] so meanwhile [03:58] what's another way I can solve this [03:58] add another environment variable XAUTHORITY to your service that points to the Xauthority file created when launching the X server [03:59] Could you show me an example? [04:01] lunorian: it'd be something like: Environment="DISPLAY=:0.0 XAUTHORITY=whatever" [04:01] lunorian: and the 'whatever'is wherever the X11 server's Xauthority file has been stored [04:08] alright sarnold we're back in action [04:09] thenori: sweet! [04:09] however the action is not 'booting up' it is 'grepping' [04:09] thenori: not sweet! [04:10] i grepped a few of the lines that were in the screenshot [04:11] skel/.bashrc:[ -x /usr/bin/lesspipe ] && eval "$(SHELL=/bin/sh lesspipe)" [04:11] that should be fine [04:11] /etc/skel/ isn't executed during boot [04:12] hm. I was inclined to start commenting stuff out. [04:12] it's used to populate homedirectories when you create new users [04:13] idk where should I start with this then? https://i.imgur.com/nveiBE6.png [04:13] I removed the two lines at the end of my /etc/bash.bashrc [04:14] aside from those, they're standard [04:15] thenori: did you remove the set -u ? [04:15] near the pipefail and xtrace .. [04:15] yeah, it was like set oup pipefail [04:16] removed both lines [04:16] hmm. I wonder why this stuff is still showing up [04:16] hm? it's not, i just haven't rebooted [04:16] OH [04:16] then you're probably good to go :) [04:16] if you say so [04:17] yeah, it's worth checking to see if these bits are fixed [04:17] I'm not sure why youi're getting into emergency mode in the first place, maybe this is it, maybe it isn't [04:18] if you say so o7 [04:26] heyyyy we have recovery mode [04:27] yay! now we get to solve the original problem ;) [04:27] very limited keyboard control for some reason, arrow keys aren't working proper [04:29] the screen is quickly flooding with input [04:29] oops looks like i'm in emergency mode [04:30] welp [04:30] can you snag another photo of it as early as possible? [04:30] sure, you won't like it though [04:30] heh === Taco is now known as DetectiveTaco [04:33] https://imgur.com/a/2hDNoyq [04:34] bet that isn't what you thought i meant when i said the screen is flooding with input lol [04:35] I saw something vaguely similar the other day [04:35] what was that.. something like an errant directory in a zfs mountpoint kept 90% of my zfs datasets from mounting.. [04:36] "input overrun" sounds a bit like a key is stuck on the keyboard === Napsterbater_ is now known as Napsterbater [04:54] thenori: time for me to bail, good luck [06:05] Is it possible to have ubuntu/kubuntu desktop but run it in command line and do NOT load any UI? [06:06] sure, sudo systemctl set-default multi-user.target # and you will boot in tty [06:06] sudo systemctl set-default graphical.target # would move back [06:11] I have a program (EAGLE specifically) that seems to black-screen after some time of being installed. This is the 2nd time it's happened. When I start it, the window is completely black, but it responds to mouse clicks as if the window elements are there. In fact, it can open new windows just fine, and those aren't black. [06:11] the 1st time was on another computer, also running Ubuntu [06:11] the way I "fixed" that was by downloading a newer version of EAGLE [06:12] Intelo: to switch from multi-user.target (or most other targets) to graphical.target use: sudo systemctl isolate graphical.target [06:12] that version is the same version that I'm currently experiencing the issue with [06:12] so the version isn't the problem [06:12] also, installation consists of uncompressing a .tar.gz [06:12] oerheks, thanks. How much rough difference in a)memory load b) hard disk space c) cpu is there in terminal based ubuntu and UI based? [06:12] so it's not installing anything into system directories [06:14] so is there anywhere that an application could be caching data that would make it not render? [06:14] again, the program works. It's just completely black [06:15] I haven't been able to find anyone else experiencing the same issue [06:15] but it's happened to me twice on different computers [06:16] I recently ran dist-upgrade, so maybe a graphics driver was updated? [06:16] but I have integrated graphics and none of my other programs have this issue [06:17] m5w: we can't really support suoftware which is not in ubuntu here, especially if it wasn't installed using a supported software packaging format - you'll best seek support from its developers. generally, the following storage locations could be used (amongst other): there is ~/.cache /run/user/$UID /tmp/ /var/tmp/ ~/.config/ ~/.local/ [06:19] tomreyn, ^ [06:20] Intelo: if you install any ubuntu desktop flavour then all of the graphical desktop is installed, so the same amount of disk space will be allocated whether or not you start the graphical desktop. about memory allocation, this can differ by the graphical desktop you choose (not) to install/run, check minimum system requirements for each. [06:21] tomreyn, How can I get a rough idea on numbers? [06:22] Intelo: for a CLI-only installation (i.e. ubuntu server / ubuntu-minimal) you'd see less than 10 GB disk space allocation (probably less than 4 GB really), and less than 1 GB RAM allocation. see ubuntu-server minimum requirements. [06:22] ah, so my problem was that DuckDuckGo doesn't always return the best search results [06:22] searched with google and found a bunch of people posting about this [06:22] apparently you need to run it with LIBGL_ALWAYS_SOFTWARE=1 [06:23] m5w: this would enforce software rendering, disable (hardware backed) graphics acceleration. [06:24] Intelo: cpu resource allocation should be very minimal while idle on CLI, may be higher on a desktop if you have integrated graphics. [06:25] Intelo: are you asking for a concrete use case or just for fun? [06:25] tomreyn, concrete [06:25] tomreyn, idle apache server with one dummy page. i)ubuntu desktop ii) ubuntu server [06:26] well if you'Re running a server, run headless. [06:27] how do I scroll up in a console based terminal? [06:27] shift pgup/pgdown [06:27] tomreyn, thanks [06:28] tomreyn, I was planning to upload the virtual image to a vps and then run it. There are some applications that I dont what to do heavy configurations every time [06:28] tomreyn, is that a good idea? [06:29] it's a good idea to separate server and desktop use. [06:29] i do not know what "the virtual image" is [06:32] Hello, I am having difficulty getting ubuntu cloud image 18.04 to run on CentOS 8 using KVM and cloud-init [06:33] I can download the images, resize it, setup my cloud-init ISO, however when the machine boots there is no console access and it does not get an IP address via DHCP [06:41] Please can anyone offer any suggestions? [06:45] dsofeir: try #ubuntu-server === abirkill- is now known as abirkill [07:01] tomreyn, virtualbox* [07:19] Intelo: computer [07:26] How to troubleshoot a hanging boot? I decrypt disk, then just get a blinking cursor. [07:34] tomreyn, I meant to upload vdi to vps [07:49] does anyone know offhand if there is a ppa that has more current versions of qemu and virtualization software for ubuntu? [07:49] (i.e. virt-manager, et cetera)? [07:54] not that are up-to-date, afaict [07:54] at least not on launchpad [08:01] thx ducasse...looks like I'd need to move to 19.10 if I want at least qemu 4.0 [08:06] ElectroStrong: or wait two more months and install 20.04 (or wait some more months and upgrade to it) [08:06] or brave the beta and install 20.04 now [08:07] oh wait, it's before alpha still [08:10] might have to build it from source - I have vfio enabled for my setup but I'm running into wonderful USB centric issues with high bandwidth devices (and extreme DPC latency even after a lot of configuring) - it's workable, but not ideal - if I can experiment on another disk I may do so :) [08:11] hello how to split a file with minimum sized chunks that are evenly sized otherwise [08:11] bsld: 'man split' [08:12] I tried every command [08:12] every option I mean [08:12] I need them to have a minimum base size and then be evenly sized [08:13] I'm trying to mimic how jslinux file system is split [08:13] so I can create my own file system [08:13] '-b, --bytes=SIZE' 'put SIZE bytes per output file' [08:14] yeah but it leaves as the last chunk below the minimum [08:14] I want to be evenly sized starting at a base size [08:14] so pad it up to the minimum [08:15] that's how filesystems do it [08:15] what could be a possible command I'm looking for [08:15] to pad the last file [08:16] https://unix.stackexchange.com/questions/196715/how-to-pad-a-file-to-a-desired-size [08:16] first hit on a web search [08:17] thanks [08:31] I installed vpnc to connect to my corperate VPN. When I connect I expect my DNS to be sent to my company DNS server so I can resolve names of hosts on the office but on this fresh installation of Ubuntu 20.04 it doesn't do that. Have a missed a setting? [08:33] s/20.04/19.10 [08:42] so how do I compile a linux ext2 file system to file for use by jslinux [08:42] I want to have java and gcc [08:43] I used buildroot but it doesn't have build tools [08:47] i am having a very strange error caused by upowerd in syslog pertaining to a usb device it is happning even if no usb device is plugged in, and is affecting the ability of my weather station to operate, here is the log output https://paste.ubuntu.com/p/WSTPyYXg2z/ [08:48] i am wondering if there is a fix to stop the spammage [08:57] Hello, i'm trying to start meshlab, it crashes with the following error : http://dpaste.com/38RS8B1. I'm on Ubuntu 18.04. I tried soln mentioned here, doesn't work : https://unix.stackexchange.com/questions/192642/wkhtmltopdf-qxcbconnection-could-not-connect-to-display [08:59] i think everyone's spending the next several hours stareing at the insides of there eyelids [10:28] how would i clear out a ldap setup to start over? [10:28] i screwed something up [10:33] I hate this kernel panic after VM crash... too much ways to be able to fix it, none works :S [10:37] HiddenDjinn: apt purge the packages you installed, double check all of their configurations are gone from /etc, undo configurations to other packages you may have made. === dingir_ is now known as dingir [10:47] Would kernal logs store to reason behind a lockup ? [10:48] dmesg shjould [10:49] What do the decimals represent ? [10:49] well if the kernel (not "kernal") locked up then dmesg wouldn't be useful. [10:50] ex [ 4663.887296] [10:50] C0nundrum: you can try checking the journal for the previous boot after reboot [10:52] C0nundrum: seconds since boot, use -T for human readable timestamps [10:52] journalctl -b -1 for log from previous boot, but there, too, if the kernel locked up anything after the even and the event itself was probably not logged. [10:53] it's not clear what exactly locked up so far, though [10:53] Really, thats odd that it jumped from 58.3 to 1376.9 [10:54] Is that a lockup period or can logs really be that quiet for that long ? [10:54] if you're referring to dmesg, the latter. [10:55] they can be that quiet, yes [10:55] tomreyn that depends on what happened upfront [10:55] C0nundrum: dmesg is just the kernel. if there are no special events, nothing is logged. [10:56] Kiwis: earlier events would likely still be logged, yes. [10:57] tomreyn which can be very useful [10:57] OK so what should i be looking for to determine what might have casued the lockup in the journalctl logs ? [10:58] before they stop just looks like normal service stuff happening [11:00] C0nundrum: how long had it been running (journalctl --list-boots)? does journalctl -b -1 -p3 list anything you wouldn't expect? [11:00] do you suspend + resume? [11:01] recently, no. Only on a seperate computer that is used for cifs [11:02] kernal logs stop after Feb 7 14:22:58 amd-server kernel: [ 4663.887296] CIFS VFS: Free previous auth_key.response [11:03] journalctl logs end 14:29:26 [11:04] so guess around the time it crahed mostl jhust cloudstack logs [11:04] which hardware is it, how do you use it? [11:05] ah apache cloudstack, so it's a virtualization host? [11:06] It was going to be but not just that at all right now. Just ubuntu 18.04 install with docker run couple services ( repository, some dbs) with some cifs mounts on the host [11:06] Wondering if its a bad memeory slot or a badly seated gpu [11:07] did you test the hardware before setting it up? [11:07] I ran a 24 hour meme test on the sticks with over 10 passes and had no errors so i know the sticks are good [11:07] Just wondering if any of those issues would manifest themselfs in a log somewhere [11:07] is this an epyc server then? [11:08] na just a AMD FX(tm)-8150 Eight-Core Processor repurposed as a small server [11:08] oh. you'd see "mce" or "edac" records in case you have eec ram and error correction configured. [11:09] ... and the relevant modules loaded [11:09] which i'm guessing isn't on by default ? [11:09] * ECC RAM [11:10] o is that a mb feature ? [11:10] DIMM, mainboard, firmware, OS [11:11] which is why i was asking about your hardware ;) [11:11] AMD FX is a desktop CPU, though, so probably no ECC RAM [11:14] tomreyn: amd cpus from am2 onwards support ecc [11:15] yea Dual Channel Non-ECC Unbuffered DDR3, 2 DIMMs [11:15] ducasse: even so, you still need a board that supports it as well. that's also somewhat rare. [11:15] ducasse: all of them, incl. desktop? i was thinking this was newly introduced wiuth zen. [11:15] tomreyn: i have a server with ECC RAM that's an AMD turion. [11:15] so yea they exist [11:16] slow but it has ECC [11:16] will ubuntu boot with a video card ( i do have a gui installed )? [11:16] without a video card or unboard graphics / apu [11:17] C0nundrum: yes, but usually you have serial console for these systems. [11:17] I ask because i couldn't connect to ssh once i took out my video card [11:17] ryuo: i see, good to know. [11:18] tomreyn: pretty much the only systems like this that i know of are the HP/HPE microservers [11:18] this is a microserver G7, ancient [11:19] oh those had OOB as well [11:19] iLO [11:19] not the microservers [11:19] no iLO here [11:19] ok [11:21] But ssh should still work right ? [11:21] if it doesn't come up then it means the system didn't boot completely [11:22] So why would not having a video card cause that ? [11:22] assuming of course that its network configuration is any good [11:22] maybe the board doesn't POST without one. [11:23] ryuo: The Gen8 had full on iLO4. Excellent machines, cheap and can be managed like a real server. Sad they went away with it in the updated Gen10, so back to Like G7 etc. [11:23] plundra: i see. i've only had a gen8 ml10 v2 [11:23] it has iLO4. [11:23] Downside with Microserver Gen8 is the artificial limit of ram to 16GB, afaik :-/ [11:23] Is there a standard feature name for booting without graphics ? [11:24] is that a rare feature [11:24] They were sold for like $200 in Sweden. Extremely good price for what it was. [11:24] C0nundrum: headless? [11:24] it is rare in consumer motherboards ? [11:24] is it* [11:24] no idea. they may not boot if there's no video card but you don't necessarily need one connected. [11:24] To be able to boot without what? A graphics card? [11:25] yea to be able to booth without one [11:25] Pretty sure some sort of graphics is required for a regular PC, as in. But you are of course free to not use it. [11:25] most x86 servers have at least a basic GPU. [11:25] an onboard 2D one used for KVM [11:25] Yeah I don't think BIOS can function without graphics. Maybe possible with UEFI but most things tend to be able to boot in legacy mode. [11:26] As in, some sort of chip looking like or behaving like graphics. No matter if you use it or not. [11:26] hm that sucks. so a bad pci slot just burns the whole system. [11:27] But are you talking Consumer motherboard, with no graphics card or ports + cpu with graphics, no, won't work. [11:27] consumer board , board has no graphics, cpu has no graphics, and no gpu [11:27] I have/had a laptop where the Nvidia graphics card blew up. I've managed to boot into Xubuntu using nomodset [11:28] now, it just blew up the CPU due to a energy power peak -_- [11:29] although my laptop was a Acer Aspire and *probably* had some sort of built-in graphics card? no idea, tbh [11:30] Most certanly if it was sold the last, what, 10+ years? [11:30] +/- [11:31] it was a good laptop nonetheless. at least I could save the HDD and put it into another laptop ^^ [11:32] hi, any idea why my hash key on my keyboard won't work please? [11:32] im on uk [11:38] CraigSuddo: probably the keyboard layout isn't set properly? [11:38] did you tried to do: sudo dpkg-reconfigure keyboard-configuration? [11:40] just tried it PeGaSuS [11:41] did you logged out and logged in again? usually you have to do so in order to your shell to use the new configuration, CraigSuddo [11:41] ah ok [11:41] i'll try now [11:41] thanks [11:48] Also another random question [11:48] Random answer! [11:48] * PeGaSuS grins [11:48] Tried running windows host in kvm but the mouse isn't ligned up correctly and mouse acceleration is crazy screwwy. Couldn't even get though the install promts [11:49] what causes that ? ( i did make this attempt though vnc ) [11:50] the client side mouse isn't synchronized with the server side mouse. [11:50] it's normal. [11:50] Is there a solution ? [11:51] a good vnc client would probably hide the mouse cursor on your end [11:51] it's more like the server side mouse isn't synchronized with the host mouse [11:51] sorry [11:51] service side mouse isn't synchronized with the vm instance mouse [11:52] not sure if that's due to the mouse input on the host going though vnc ? [11:54] running xfce as a frontend [12:08] ugh [12:09] virt manager doesn't seem to be fully compatible with ubuntu 18.04 [12:09] still tries to use ifup [12:11] o i can just add the package [12:11] virt-manager is Section: universe/admin [12:13] neat broke network. Guess i have to fetch a real monitor now x.x [12:18] Can I somehow override the gcc package pbuilder uses, so that it uses the gcc from debian and not from ubuntu? [12:19] !frankenubuntu | Darkchaos [12:19] Darkchaos: it is usually a very bad idea to mix packages from different releases (or Linux distributions), and it is completely unsupported [12:19] Well not to my living distribution, but into the virtual build environment from pbuilder. I want to find out if the patches ubuntu makes to gcc are the source of the problem [12:20] I tracked it so far that both were using the same gcc version, but ubuntu had -ubuntu1 [12:20] <__Milencho> ElectroStrong, thanks for the help yesterday, i've reinstalled the ubuntu (deleted OS windows) and that's it ;] [12:21] Darkchaos: the source of the problem, as you want to help contribute to ubuntu? [12:22] lotuspsychje: Exactly, I am seeing that the package is "broken" only on ubuntu and I now finally tracked it down to even using the "same" gcc version. Something has to be different in the environments [12:23] sigh [12:23] !contribute | Darkchaos [12:23] Darkchaos: To contribute and help out with Ubuntu, see http://community.ubuntu.com and https://wiki.ubuntu.com/ContributeToUbuntu [12:23] ...so tried to create a network interface with virt manager and it broke local interface. logging into xfce worked fine with vnc but since the network is down, i';m trying to login from the pc itself [12:24] At the login screen when i enter my password the screen just flashes and then the password field is cleared [12:24] I know the password it correct because if i enter the wrong one, i get a password incorrect promt [12:24] What do you think could be the issue ? [12:25] Howdy all [12:27] C0nundrum: sounds like some kind of a bug, or bad configuration. is the system full yupdated? is this xubuntu desktop? [12:28] can you inspect the logs? [13:20] Where is the power option to have the system power off when not in use? [13:21] vion, you mean 'sleep mode' ? [13:22] sure sleep or power off which ever [13:22] vion, you need to press 'alt' [13:22] no I mean unassisted [13:22] power off when idle [13:23] hibernate? [13:24] vion: you can use xidle [13:24] that would work too [13:24] I'm running gnome right now is not there an option already? [13:25] IdleAction / IdleActionSec, see logind.conf(5) [13:26] vion, it's not clear to me what you really need exactly [13:26] "Note that this requires that user sessions correctly report the idle status to the system." i do not know whether this is the case, but assume it can be. [13:27] oh nice and that would work with command line too? [13:28] i have no first hand experience. any shell would run within a user session, though. [13:29] idleaction=poweroff idleactiohnsec=90min [13:29] idleaction=poweroff idleactionsec=90min [13:29] that should work I hope [13:30] vion, consider that 'idle' is very ambiguous; input devices might be idle, but the system might not [13:32] are you trying to say this is waiting for logins? [13:32] nvm [13:32] hi [13:32] my mouse keys mouse is moving too slowly [13:32] iroria: I should configure this through the UI you are right [13:33] littlekimmy: mouse keys mouse? [13:33] mouse keys [13:33] the mmouse pointer is moving too slowly using keys [13:34] ubuntu 19.10 [13:35] littlekimmy: there are a few tweaks on mousekeys in dconf-editor if you like [13:39] littlekimmy: oh you got one of those little nubs attached to the keyboard. Are there any key combos to change the sensitivity? [13:40] vion: in settings- that's what I am looking for [13:44] dconf what? [13:45] littlekimmy: sudo apt install dconf-tools [13:45] littlekimmy: then you can run dconf-editor [13:47] where is the mouse keys settings sensitivity adjust [13:48] littlekimmy: dconf-editor has a nice search function [13:48] i did [13:48] what is the best java to use on 18.04? do most people use openjdk? [13:50] !discuss | cgi [13:50] cgi: Want to talk about Ubuntu, but don't have a support question? /join #ubuntu-discuss for non-support Ubuntu discussion, or try #ubuntu-offtopic for general chat. Thanks! [13:55] can i run a script as root without using sudo on startup [13:55] ~/.bashrc is non-root and user, i need systemwide [13:57] littlekimmy: you can use cron or drop a script into the autorun directory which I forget where is at [13:58] littlekimmy: the .bash_profile and .bash_logout can run scripts on logon/logout too [14:01] .bash_profile is fine, but it's not root [14:02] I don't use cron. more a general method. say /etc/profile [14:02] but i need it to be root [14:02] but how do i get root without writing the password [14:02] as it can't be in plaintext [14:09] littlekimmy: add commands to /etc/rc.d/rc.local [14:09] ubuntu has rc? [14:09] nvm old info [14:10] littlekimmy: try this https://smallbusiness.chron.com/run-command-startup-linux-27796.html [14:12] it seems ubuntu still has /etc/init.d and /etc/rc0.d [14:18] what is /etc/init/whoopsie.conf doesn't seem systemd [14:25] littlekimmy: error reporting service notifies you about crashes [14:41] With the ubuntu livecd environment how can I mount a systems drive that is entirely ZFS [14:41] said system is another ubuntu install [14:44] vion, I'm not sure you can [14:45] what is it the default to use encryption? [14:46] vion you just need the zfs tools IIRC. modprobe the driver, make sure you have the tools. You can apt-get install from the live cd. [14:47] As far as encryption goes, for most filesystems it's luks. No idea if linux based zfs uses that or not. [14:49] kyle: zfsutils-linux ah okay so we use openzfs, the installer should really include that information since there are so many versions of it [14:50] vion: livecd environment doesn't include everything. They have constraints. Hell, it doesn't even include firmware for some (fairly common) 10G network adapters. [14:51] kyle__: No I mean like when you click the option to use zfs it doesn't say its openzfs so I've been punching in bsd, solaris, and older commands [14:54] vion: Oh. I wasn't aware there were any production ready-ish versions of zfs OTHER Than openzfs for linux [14:56] kyle__: isn't zfs the usual way to setup raid? I thought it had been in full use for many years [14:56] No. Not for linux. For software based raid, it's been mdadm for over a decade. [14:57] If you use zfs or btrfs, you (generally) use their internal systems instead of mdadm, but in general, for software raid on linux, you use mdadm. [14:58] or lvm [14:59] ducasse: the only raid-like feature built into lvm is lvm-mirroring. Which is kindof a PITA. [14:59] i know, but people still use it [14:59] * kyle__ shudders [14:59] I've used it, but I really didn't like it. [15:00] kyle__: this is a green sauce kind of thing isn't it [15:00] LVM mirroring that is. LVM itself is pretty great. [15:01] report from #ubuntu-packaging [15:01] HI all i found something that should be fix gespeaker package. it need to have installed python-dbus to work but python-dbus is not a dependency of gespeaker. [15:01] where would be best to report this? [15:02] !bug | willdor [15:02] !bugs | willdor [15:02] willdor: If you find a bug in Ubuntu or any of its official !flavors, please report it using the command « ubuntu-bug » - See https://help.ubuntu.com/community/ReportingBugs for other ways to report bugs. [15:02] willdor: If you find a bug in Ubuntu or any of its derivatives, please report it using the command « ubuntu-bug » - See https://help.ubuntu.com/community/ReportingBugs for other ways to report bugs. [15:02] thx [15:03] willdor: Way to go dodging installing that package for so long, regular Neo up in here. [15:04] dodging installing? [15:04] willdor: its a rather common dependency, an easily missed oversight [15:08] a little off topic i am look for placement do you know if canonical or anyone that work in open source commity has any opening that would work for that? [15:11] willdor: try finding an appropriate channel with !alis or ask in #ubuntu-offtopic [15:12] thx [15:33] is there a gui tool to look at system info? [15:35] togo hardinfo [15:36] jeremy31, thanks, ill give i a wirl... [15:39] anybody around doing any AI deep learning experimentation? [16:06] https://en.wikipedia.org/wiki/Mouse_keys what is that equation action_delta*.. = what? [16:07] https://www.devprojournal.com/technology-trends/ai/need-a-linux-distro-for-deep-learning-applications-try-ubuntu/ [16:07] togo: hi [16:07] did you make vinegar ? [16:08] littlekimmy, Please stay on topic. If you have a specific Ubuntu support question, please ask it. For chat and general discussion, please visit us in the #ubuntu-offtopic channel [16:09] togo, you as well. Please use #ubuntu-offtopic for non-ubuntu support related topics [16:20] Hi [16:20] How to install mkcert on ubuntu [16:22] Intelo, I'm not aware of any packages called mkcert [16:32] action_delta is in ubuntu mouse keys [16:36] Hi [16:36] I installed Zawgyi font accidentally [16:36] Then the keyboard is disagreeing with Unicode font [16:36] How can I uninstall the Zawgyi keyboard? [16:37] How did you install Zawgyi font? VegetarianFalcon [16:39] pragmaticenigma: http://naingyeminn.com/posts/burmese-keyboard-for-ubuntu-1404/ [16:40] $ wget https://github.com/naingyeminn/mm-kb/archive/master.zip -O mm-kb.zip $ unzip mm-kb.zip $ cd mm-kb-master/ $ sudo make install $ ibus-daemon -rdx $ im-config -n ibus $ gsettings set org.freedesktop.ibus.panel show 0 [16:40] Ooh. Sorry [16:40] Hi all! After upgrading kernel to 5.3.0-28 my old notebook Aspire 6930 doesn't hybernate normally.I don't any idea how to resolve it. Where i should start? [16:44] VegetarianFalcon, Are you running ubuntu 14.04? [16:44] CDigger, I used to have that laptop [16:45] CDigger, the most frequently offered solution is to not use Suspend-to-Disk or Suspend-to-RAM [16:46] pragmaticenigma: Ubuntu 18.04.3 LTS [16:47] VegetarianFalcon, if you installed with 'mae install' you should remove with 'make uninstall' ; cd mm-kb-master and sudo make uninstall [16:47] *make [16:47] VegetarianFalcon, So why are you following instructions for a much older version of Ubuntu? (Unfortunately, i don't understand the language used on the page in the link provided.) If you installed the font using apt, then it should be as simple as using the same command with "remove" instead of "install" [16:49] ioria: Idk where to find the package of installation [16:49] pragmaticenigma but why? On 5.0.0-37 Suspend-to-disk worked good for me. === cyp3d_ is now known as cyp3d [16:50] VegetarianFalcon, sudo updatedb && locate mm-kb-master [16:51] and now, when i boot in 5.0.0-37 kernel still working [16:52] ioria: what should I do next? [16:53] VegetarianFalcon, cd mm-kb-master and sudo make uninstall [16:53] CDigger, For starters, if you are using a solid state drive, you are significantly shortening the life of the drive with Suspend-to-Disk [16:53] ioria: it says "no such file or directory" [16:54] I deleted the local files [16:54] Hi [16:54] From home [16:54] CDigger, beyond that, the features have always been hit or miss in Linux. There are lots of different hacks to make it work, but there is no reliability. Which is why many suggest not using it [16:54] VegetarianFalcon, you're are in the wrong directory [16:54] I have put Ubuntu Server on my Raspberry Pi, I did touch /boot/ssh so that the Ubuntu Server on the Pi sets up SSH. I also set the IP address in netplan. [16:55] However when I try to log into the pi via SSH, I get connection closed [16:55] I don't even know the default credentials for Ubuntu Server for the pi. [16:55] Helenah: you following some sort of guide? I recently got a Pi myself was thinking about setting up [16:56] vion: No [16:56] But I Googled around [16:56] VegetarianFalcon, if you deleted the local file, just curl it again [16:57] ioria: you mean to install it again? [16:57] VegetarianFalcon, nope [16:57] Helenah: I have never heard of anyone doing touch /boot/ssh why not apt install openssh [16:57] ioria: what does "curl" mean? [16:57] Helenah, did you remember check to make sure the firewall settings? [16:57] chown root:u file ; works [16:57] why [16:58] vion: How can I do that when I an not configuring Ubuntu Server via the Pi but via the file system by mounting? [16:58] VegetarianFalcon, curl is a different tool similar to wget [16:58] clearly root is user and u is group [16:58] I intended to use the Pi as a server so I didn't buy the small HDMI adaptor that the Pi 4 needs. [16:58] but u:root too works [16:58] Helenah: load it up in virtualbox and create a new image maybe [16:58] so root cannot be a group that has u [16:58] VegetarianFalcon, curl, wget, whatever you used to download the zip pkg [16:58] Helenah: probably an easier way, I'm not sure of the method. I would just plug right into the Pi to dump the files on [16:59] littlekimmy, without any context, it's really hard to know what you are asking and what you need help with [16:59] I just want to know what the issue is, I'll go into the file system and sort it myself. Do I need to set a password for the user? [17:00] Helenah, We only know what you have told us. Far as I know, no one here is telepathic and can see what you see, or even talk to your machine directly [17:00] Helenah: is it an already setup image? Who provided you with it [17:00] I downloaded from the Ubuntu website. [17:00] preinstall [17:00] Helenah, It helps to know several things, What version, where did you obtain it from, how did you flash it, what you have actually done so far to setup SSH [17:00] Helenah: okay for Pi specifically? Or is it just ARM image of ubuntu server? [17:00] pragmaticenigma so i thnk i need some hacks... [17:00] vion, https://ubuntu.com/download/raspberry-pi [17:01] thanks I will test this out right now [17:01] pragmaticenigma: I followed this https://ubuntu.com/download/raspberry-pi [17:01] @prag [17:01] flashed to the SD card the way recommended [17:01] I used Ubuntu 19.10 64 bit [17:01] I did touch /boot/ssh to initialise SSH [17:02] Because someone else here the other day told me to do that [17:02] where did you find instructions? [17:02] oh... that explains it... Helenah the "touch /boot/ssh" is for raspian, not Ubuntu [17:02] Okay [17:03] Helenah, I'm not familiar with the RPi image, but if it is similar to any other Ubuntu image... you need to install the openssh package [17:03] that will then get your ssh server up and running [17:03] ioria: some of it has gone. But some are still running [17:03] OpenSSH is installed, I thought the error I got made that clear? [17:03] It closes the connection [17:03] I'm going to mount the file system and investigate further. [17:04] Thank you for your time [17:04] Helenah: nah if you had banner grabbed you would see that it is running [17:04] I was dreading having to emulate the Raspberry Pi [17:04] Does anyone have a guide on how to do this in qemu? [17:05] Helenah, if openssh was installed... you wouldn't get an immediate closed connection [17:05] Helenah: do you have experience with qemu? Why not just plug into the Pi [17:05] vion: Just not qemu-system-art [17:05] er... arm [17:05] I also did this. [17:05] https://ask.ubuntu-mm.net/59/how-to-install-myanmar-font-unicode-in-ubuntu [17:06] Before I installed the font switcher [17:06] I want to uninstall it too [17:06] vion, pragmaticenigma, Does Ubuntu spawn a getty over UART? [17:07] I can access that way [17:09] Helenah: ok so I'm guessing you stuck the Pi in some huge contraption and don't have physical access it. Have you considered an extension cable for one of its ports so you can interface with it? [17:09] VegetarianFalcon, if you used a ppa, first remove those packages, then remove the ppa [17:10] vion: The things I got access to are Ethernet and UART, I have a USB to UART cable. [17:11] But I need to know if a getty is spawned by default on the serial port [17:11] ioria: how can I find ppa? [17:11] VegetarianFalcon, ls /etc/apt/spources.list.d [17:11] VegetarianFalcon, ls /etc/apt/sources.list.d [17:12] VegetarianFalcon, or grep -r -i ppa /etc/apt/ [17:12] Helenah: the Pi comes with a fart port... ok. I doubt its spawneed by default because that would be a security liability and I think Ubuntu lost a challenge in the past for a similar physical access compromise [17:13] It's also why I don't think Ubuntu enables or installs an SSH server by default [17:13] vion: The compromise is good, can't people not secure their own systems up or something? [17:13] VegetarianFalcon, but first you need to remove the pacages installe from it [17:14] Helenah, Ubuntu philosophy is to start with a locked down system, and let users who know what they're doing open it up [17:14] Helenah: I am aware you are running a different version of ubuntu, but it also says its the server edition [17:14] I'm going to open it up then. Thanks guys. [17:14] vion: That is what I want [17:15] vion: I installed it from terminal. And Idk where to find the package [17:15] Helenah, I don't believe the server edition enables SSH by default... again, I suspect most of this would have been resolved by installing the openssh package [17:15] VegetarianFalcon, here maybe ? https://launchpad.net/~ubuntu-mm/+archive/ubuntu/ppa [17:16] VegetarianFalcon: I am the font guy yes. I make you loot gorgeous, yes? [17:17] vion: sorry. What do you mean? [17:17] VegetarianFalcon, or better install ppa-purge and directly remove all of them [17:18] ioria: can you tell me the command? [17:18] !ppa-purge | VegetarianFalcon [17:18] VegetarianFalcon: To disable a PPA from your sources and revert your packages back to default Ubuntu packages, install ppa-purge and use the command: « sudo ppa-purge ppa:/ » – For more information, see http://www.webupd8.org/2009/12/remove-ppa-repositories-via-command.html [17:19] VegetarianFalcon: What language is that? It looks like pokemon. [17:21] I've only ever done this in sysvinit, but how do I enable a getty on the UART interface? [17:22] Helenah, perhaps someone in the #ubuntu-server channel can help you with that [17:25] ubottu: it doesn't work for me [17:25] VegetarianFalcon: I am only a bot, please don't think I'm intelligent :) [17:26] Oh [18:00] VegetarianFalcon, what it does not work ? [18:01] ioria: it worked now [18:01] It needed to install ppa purger first [18:01] ok [18:01] Thanks [18:01] yes [18:29] ubuntu 19.10 is not LTS , I don't like that [18:29] littlekimmy: only support questions in here please [18:37] if I ran grep / file , how do I see that file in the logs [18:37] that cmd [18:38] littlekimmy, what logs are you referring to [18:39] the cmd logs [18:39] the random terminal cmd [18:39] those logs [18:39] you mean bash history found in .bash_history ... which keeps track of the last few hundred commands you've executed? [18:39] say I do echo 'something' | festival --tts [18:39] yes that too [18:40] but those terminal cmds are getting loggged elsewehre too ? [18:40] such as when I grep is it logged elsewhere too apart from bash history [18:41] littlekimmy, nope, the programs you execute are not logged anywhere. A program may create it's own logs, but that would be dependent on the program you're executing [18:42] so does grep , qemu log? [18:42] and how do I know if my program stores logs [18:43] they are still logged by bash in history [18:43] isn't history too a type of log [18:43] no... a log provides information about what is happening... bash_history is just a list [18:44] I use some cmds, like qemu, I want to be sure no one upon forensics can know what I do [18:44] to find out if a program does any logging you read the documentation about that program [18:45] I read man bash, it says it doesn't log [18:45] but it does record what is happening [18:45] I am resetting Ubuntu into factory version with Resetter [18:46] The laptop seems frozen [18:46] In the middle of process === seopaert is now known as littlekimmy2 [18:46] VegetarianFalcon, There is no such there as a factory reset... I have never heard of resetter and will assume that is not a package provided through official Ubuntu software repos [18:46] :( [18:48] in what way does apt record my activity where is that ? and how can I disable it [18:48] not just logs but any information that is recorded [18:50] littlekimmy2, there isn't a way that you can do that with Ubuntu. Apt relies on it's logs to know what it has done and how to undo the things it has done [18:50] use full disk encryption, don't share your credentials, don't enable others to access your system locally or remotely. that's the only way you reliably prevent others from reviewing what you did. [18:51] sarnold: btw I narrowed down the environment differences to 755 LOC of diff :D [18:51] and maybe reconsider your approach to it. if you're supposed to comply with some company policies then your choice is basically (a) comply, (b) challenge them, and if this fails, find another employer. [18:52] littlekimmy2: ^ [19:00] hi there [19:01] where do I find memsest_s memmove_s in eoan? I included string.h #define __STDC_WANT_LIB_EXT1__ 1 before but it does not seem to be in the lib?! [19:08] paddyez, You might want to ask that question in the #ubuntu-app-devel channel ... this channel is more suited for desktop support [19:09] pragmaticenigma: thanx [19:13] hello, does anyone have experience with using multimon-ng? I am trying to use it to read DTMF radio signals and decode them. I've installed the program successfully and it will run in my terminal, but when i use my phone and provide it with some DTMF tones (dial pad) it seems as though multimon-ng does not hear it via the internal mic of my laptop. I've verified my mic is working and I know the [19:13] program works because I've used Audacity to create a DTMF tone recording saved file and sent it to multimon-ng where it successfuly worked but I'm having trouble using the mic for real time decoding. Any help would be greatly appreciated. Here is the git hub link to multimon-ng https://github.com/EliasOenal/multimon-ng/ [19:15] TheSonOfPeter, that seems like a really specific problem that I would be reaching out to forums or channels that are focused on that program [19:15] sarnold: Looks like the problem stems from using a newer binutils (2.32-7ubuntu4 over 2.31.1-16), but it starts to become increasingly difficult [19:40] exit [19:42] I have been fighting and searching for days on this. I recently installed Windows 10 and have another computer on my network running Ubuntu Server 18.04. Under the windows machine my server is not visible on my network tab. I have set the workgroup name on bothe machines [19:42] AurorAWOL, it's easier and more reliable to access the Ubuntu system with its IP address [19:43] AurorAWOL, If you want a friendlier name ... add an entry to your windows hosts file [19:44] Thanks for the input pragmaticenigma [19:44] AurorAWOL, alco W10 recognized mDNS, so running avahi (you may already be) will allow you to use "host.local" [19:45] Ah... Thanks rfm [20:06] Hi frens [20:45] Hi [20:47] I have a problem with an Ubuntu 14.04 [20:47] Apt say i cannot install new package because 404 Error, why? [20:48] you know is eol [20:48] !14.04 [20:48] Ubuntu 14.04 LTS (Trusty Tahr) was the 20th release of Ubuntu. !End-of-life was April 25th, 2019. Paid support (ESM) is available. See also !esm, !eol, !eolupgrade [20:48] ioria: what is eol? [20:48] read above [20:49] end of life, now unsupported [20:55] I wondered if someone using a recent ubuntu 18 or 19 could tell me if the wl driver is included by default without using the broadcom-sta-dkms [20:55] i.e. if you can run "sudo modinfo wl" on your system [20:56] it's not, on either of them [20:56] didn't think so, thanks for checking [20:56] There are lots of drivers in the form of wl[0-9]+.ko [20:56] SO maybe you need a more specific name? Dunno [20:57] Broadcom BCM4360 (PCI IDs 14e4:43a0, 14e4:4360) [20:58] is the device I was seeking to see if it was supported by default.. in debian its profivided by the wl driver via broadcom-sta-dkms [20:59] broadcom-sta-common broadcom-sta-dkms broadcom-sta-source [20:59] Those _all_ show up on my 18.04 box when I apt-cache search broadcom-sta [20:59] :) [21:00] yeah well, those would show up on any debian type system.. doesn't mean they're installed and the driver is built :P [21:00] that just means they're available for install :P [21:00] Well yeah. You wouldn't want all that crap installed unless you needed it. [21:01] well according to this user, their wifi is working in ubuntu without additional steps so idk [21:01] its a damn macbook air.. has no ethernet, so it relies solely on that broadcom wifi [21:02] which channel should I go to ask questions about replacing a failing harddrive? [21:03] rublind: what do you need to know exactly? [21:03] Is wmii still maintained? Some other best alternative nowadays? [21:04] nvz: I set up a raid years ago, and one of the drives failed. the raid is now inactive, but whenever I try to boot, it drops me into an emergency shell. I'm pretty positive that the drive is failed, but I don't know how to replace it such that it actually boots... [21:06] converge: https://github.com/0intro/wmii seems its been updated within the last 6mo there.. but it has been removed from testing in Debian https://tracker.debian.org/pkg/wmii [21:06] converge: there is i3 for example, is a tabbed wm [21:06] I3 sounds like a better option, thanks [21:07] rublind: you can't just disable a raid and use drives individually.. [21:07] nvz: sure, but I should be able to mark a drive as failed and then replace it. but when I try doing this with mdadm, it fails (don't have the error on me) [21:08] idk, but sounds to me that if you're using ubuntu that this is the place to ask because the issues you're having seem to be OS related, not hardware related [21:08] converge: i3 is sort of the successor to wmii [21:08] nvz: but it's more that the system will not start normally because it's waiting for some drive to be available. so even if I put in a _new_ drive, I don't know how to tell it to forget about that old one [21:11] rublind: idk you're being rather vague in your descriptions.. if you are using a real raid, the OS doesn't see individual disks, if youre NOT using a raid, then it is seeing them.. in either case the system likely references VOLUMES regardless if they're raid or not by UUID and any change to the fs, or creation of a new fs changes the UUID and the fstab and such must be updated accordingly as well as the [21:11] bootloader [21:12] rublind: in any case, you'd need to boot a rescue media and chroot in to begin to troubleshoot and fix this [21:12] I'm only being vague because I'm not in front of that computer right now, so Idon't have exact details. [21:13] when I checked in dmesg, it was complaining about /dev/sdd - and when I checked the raid with `mdadm --detail --query /dev/md0` it showed as inactive with 2 spares (which is incorrect...). I do see the UUID referenced at times too. [21:13] mdadm doesn't show /dev/sdd but it does show /dev/sdc and /dev/sde [21:17] in any case, I've ordered a new drive. so this is all premature anyway. I'll pop back on once it arrives. [21:31] Hi Guys, how can I most easy fix a "kernel panic not syncing vfs unable to mount root fs on unknown-block(0 0)" with a liveCD ? I cannot boot into rescue mode [21:32] chrooting doesn't work that easy as well [21:38] Kiwis: (1) got any idea what the cause for the kernel panic may be? (2) what changed between when it booted fine and stopped doing so, if you can tell? (3) what do you mean by "chrooting doesn't work that easy as well" (4) which ubuntu version is installed? (5) which live system are you trying to recover from? (6) which hardware is it, if you know? [21:39] tomreyn my host crashed while the VM was running (and doing something) [21:39] nothing special it did actually [21:39] 18.04 is installed [21:39] Vbox [21:40] so this is an ubuntu 18.04 LTS VM running under an unknown version of VirtualBox under an unknown host OS? [21:41] ^ (7) [21:41] (8) did you make changes to possibly relevant configurations between when it worked fine and stopped working? [21:41] oh vbox version doesn't matter much, why would it, last 6.x version... host does not matter atl all but you like to know too much... W10 [21:42] tomreyn nope, no specific config changes... I had it before a while ago, was able to fix that one tho [21:43] sshd.service fails to start after changing port from default 22: fatal: Missing privilege separation directory /run/sshd [21:44] Kiwis: ok. please also answer questions (3) and (5). [21:44] When trying to connect from a client I get: kex_exchange_identification: read: Connection reset by peer [21:44] tomreyn when I chroot it cannot find some udev lib which is there when I want to run any usable command like apt-get or install-grub2 [21:45] Kiwis: that leaves us with (5) [21:45] tomreyn oops sorry, Ubuntu WS [21:45] just 18.04 [21:45] Kiwis: please also describe which steps you took for chroot'ing [21:46] and the output you receive there [21:47] tomreyn https://askubuntu.com/a/48516/681225 [21:51] Kiwis: and the error message when chrooting is? [21:51] I need ot look it up but something like libudev.so.0 not found [21:53] Kiwis: okay, i suggest you chroot again to reproduce this error message. also, have you done a forced fsck on the root file system (while it is not mounted)? have you tried booting a different kernel version from the grub menu? if not, do it, if you did, what happened then? [21:54] tomreyn yes I did a fsck, all fine... same happens for other kernel, also in rescue :( [21:55] Kiwis: so you *can* boot into rescue mode? [21:56] tomreyn nope, where did I say that [21:56] you said "same happens also [in rescue]" but it's not clear what "same" is. [21:59] I mean, can't boot into it :) [22:00] Kiwis: okay, feel free to ping me when you got the chroot error message after vhrooting as described at https://askubuntu.com/a/48516/681225 [22:02] tomreyn if /boot is in a seperate sdaX and the whole install is on LV so a mapper... what do to with /boot ? mount it after chroot to .boot ? [22:02] /boot [22:07] Kiwis: you don'T chroot to /boot, you chroot to where you mounted the root file system (and any other relevnt file systems below that). [22:07] tomreyn but /boot is on a seperate sdaX, sure it's there when you chroot, I didn't see it earlier [22:08] Kiwis: if relevant file systems are wrapped in LVM you'll need to enable the LVM from the live system before chrooting, then mount file systems as needed, then chroot [22:09] tomreyn but /boot is empty after I chroot [22:09] as it's ona seperate sdaX [22:09] Kiwis: well you need to mount it [22:10] after I chrooted ? [22:10] or before [22:10] i'd do it before but it should not matter for /boot [22:10] yes OK, fine again :) [22:11] OK< chrooted, what would you do ? reinstall grub ? [22:15] Kiwis: i'd want to double-check all relevant file systems are mounted (fdisk -l, lvs), i'd compare blkid against /etc/fstab. i'd do a cursory review of /etc/lvm/lvm.conf, i'd run apt update && apt full-upgrade. given no warnings or errors are encountered, i'd update-grub. once all of this seems to be fine, i' run update-grub and, given no warnings or errors are encountered, grub-install /dev/XXX [22:16] Kiwis: ...where XXX would be the disk you want / need grub on. or you can omit it if you're uefi booting. [22:16] tomreyn I get: /usr/sbin/grub-probe: error while loading shared libraries: /lib/x86_64-linux-gnu/libudev.so.1: invalid ELF header [22:17] i listed update-gub twice above, please ignore one of them. [22:17] Kiwis: is the recovery system the same OS architecture as the installed system? amd64 vs i386? [22:18] tomreyn as far as I know yes... checking out again.. sec [22:18] yap, both 64 bits 18.04.3 [22:18] file /lib/x86_64-linux-gnu/libudev.so.1 [22:19] says what? [22:21] symbolic link to libudev.so.1.6.9 [22:21] Kiwis: disable windows defender, try again [22:21] and yes, host OS matters [22:21] sec [22:23] well this can be a problem on WSL, but then you'Re saying you're using virtualbox, not WSL [22:23] so maybe it's a different problem [22:23] Yesterday I talked about an issue where I was unable to setup a systemd service to automatic do screen resizing due to it not having access to adjust the current Xserver despite being root (root is supposed to not have any permission restrictions so this bothers me: I am administrator, program: lol I don't give a f*ck you don't have permission). Anyways I created a startup script with Ubuntu's built in startup [22:23] mostly solves the problem aside from a weird scaling before I login. I was curious if I'd be able to launch the display adjustment script on the login screen's session instead. It's a simple xrandr command. I'm using Ubuntu 18.04 LTS on the Thinkpad X1 Carbon 3rd Gen [22:24] tomreyn doesn't change a thing [22:26] Kiwis: okay, re-enable defender and review its logs to see whether anything was quarantined recently. then head back to the VM, and report the output of file /lib/x86_64-linux-gnu/libudev.so.1.6.9 [22:27] nothing in Q also, symlink is still there [22:27] Hi! Can someone help me understand why my swap is still being used after setting vm.swappiness=5? https://termbin.com/w0zg [22:28] Kiwis: you ran a different command than i suggested, though [22:28] ? [22:29] i suspect you ran file /lib/x86_64-linux-gnu/libudev.so.1 whereas i suggested running file /lib/x86_64-linux-gnu/libudev.so.1.6.9 [22:29] oih sorry [22:29] ASCII tyext, with very long lines [22:30] Kiwis: apt list --installed libudev1 2>/dev/null [22:30] no output [22:31] Kiwis: dpkg -S /lib/x86_64-linux-gnu/libudev.so.1.6.9 [22:32] dpkg-query: error [22:32] I think this install is just broken [22:32] Kiwis: "error" is the only output? [22:32] no some status near line 0, etc [22:33] if you're still willing to try and fix this i'll need full output. but it's unusual that this package is not installe dbut the file is present and is in an unexpected format. [22:34] tomreyn I start over, much easier, thanks tho! [22:35] Kiwis: do you have any idea what may have contributed to this situation, though? such as use of third party software, foreign packages or package versions, unpackaged software installations, manual changes to root owned files? [22:36] Is there a way to run a script after the login screen has loaded? [22:37] tomreyn wait! [22:37] I think there was some fsck issue tho [22:38] Kiwis: you said you ran a forced fsck while the file system was not mounted earlier. did this not detect and report any errors then? [22:39] tomreyn no but it did now (again) still kernel issue tho... checking further [22:41] Kiwis: if you ran e2fsck -f on the unmounted root file system earlier and it reported no corruption then but it did so now, while runnign it again in the same scenario, then something is wrong with the underlying block / storage devices, i.e. in or below windows. [22:42] I'm not sure what is going on [22:54] lunorian: try using ~/.xinitrc instead [22:55] tomreyn: I specifically need to run code on the lockscreen. Before login. Where my home folder wouldn't be available. [22:55] so right when ubuntu boots and starts up Xorg or whatever [22:56] the splash screen is very breif and UEFI overwrites part of it with the Lenovo logo anyways [22:57] lunorian: this sounds like you want to modify gdm's options [22:57] how would I go about that? [22:57] see /etc/gdm3/custom.conf [22:58] what would I add to it? [22:59] or maybe it should be in /etc/gdm3/PreSession/Default [22:59] or maybe you just want a custom /etc/X11/xorg.conf [23:00] lunorian: what does your script do exactly? [23:01] It runs /usr/bin/xrandr --output eDP-1 --scale 0.66x0.66 to set the scaling factor to about 150% [23:01] this looks perfect on my thinkpad display [23:02] lunorian: hmm gdm runs via wayland by default, though. [23:03] yeah so, what are my options to set scaling in GDM [23:08] lunorian: try experimenting with gsettings' "org.gnome.desktop.interface scaling-factor" setting [23:08] does that take effect before the login screen loads? [23:08] also ubuntu's settings only offer 100% and 200% [23:09] no in between [23:09] at least not in 18.04 LTS [23:09] 19.10 gives fractional scaling per 25% .. but you know this from yesterday session. [23:09] * with a tweak [23:10] i think wayland does support fractional scaling as an experimental setting in 18.04. [23:10] The new Ubuntu distros always seem to have annoying bugs for the first year or two so I prefer to stay on LTs [23:10] I won't be adopting 20.04 LTS until at least 2022 [23:10] and I'll probably have a new laptop by then [23:11] https://www.foell.org/justin/simple-hidpi-monitor-scaling-with-wayland-in-ubuntu-18-04/ [23:11] i assume this would also apply to gdm, but have not tested it. [23:12] alternatively you ocould choose to edit /etc/gdm3/custom.conf setting WaylandEnable=0 anduse xorgs' scaling multiplies. [23:13] lunorian: the proper answer would be that 18.04 LTS does not support fractional scaling, though. [23:14] hope something gets backported to support it [23:14] that is what i find too [23:14] it's not too big of a deal as long as the session itself is easily readable === monkeystance__ is now known as monkeystance [23:23] how about !HWE, does the newer kernel and drivers give you the 25% scaling feature? [23:23] !hwe [23:23] The Ubuntu LTS enablement stacks provide newer kernel and X support for existing LTS releases, see https://wiki.ubuntu.com/Kernel/LTSEnablementStack [23:24] it wont upgrade gnome or glib so i guess not [23:25] nor gdm [23:28] Just checked [23:28] found out there were already installed [23:28] I guess I'm just gonna have to deal with a broken lockscren [23:28] maybe 20.04 will be different [23:28] guess I'll find out in a few years [23:28] login screen != lock screen