=== HappyHotDog is now known as SmugHotDog === ledeni_ is now known as ledeni [01:19] What's the easiest way to create an encrypted bootable ubuntu usb drive? Is there anyway to do this without tricking the OS? [01:20] and removing the hard drive? [01:20] full encryption/lvm is an install option, but an USB will be horribly slow. [01:21] like a boot of 5 min.. [01:21] tails runs pretty good, was hoping for an ubuntu version of that [01:24] oerheks: you are talking about installing from one usb to another right? I am worried I might mess up my bootrecords on the device in the process [01:26] yes, usb to usb, and you would get the question where to put the bootloader, for sure [01:26] :-) [01:27] hmm....ok, I'll try that [01:41] At what point do I select encrypt home folder? I already created a user and selected require password to login...does that mean the home folder is encrypted (i am guessing no)? [01:44] depending upon release, that feature may not exist any more [01:44] https://computingforgeeks.com/ubuntu-focal-fossa-installation-with-screenshots/ [01:44] yes, advanced features https://computingforgeeks.com/wp-content/uploads/2020/04/install-ubuntu-20.04-step-by-step-04.png [01:49] I have to select Advanced Features to do that? [02:10] is there no way to remove a value in nmcli interactive? I know theres mod -option val but that doesnt seem to work with set. === zbenjamin is now known as Guest67924 === zbenjamin_ is now known as zbenjamin === NeverLoads2 is now known as NeverLoads === jnsgruk08 is now known as jnsgruk0 === denwa6 is now known as denwa === guillaumedsde012 is now known as guillaumedsde01 === Linkandzelda_ is now known as Linkandzelda === jeddo- is now known as jeddo === graingert is now known as graingert[m] === graingert[m] is now known as graingert [04:04] Good evening everyone. I use a VPN which requires sudo to engage via command line. Is there a way to allow a non-sudo user to start and stop the VPN? [04:23] hi [04:23] how to stop X from listening tcp | local and unix ??? [04:23] like: -nolisten tcp -nolisten local -nolisten unix [04:25] i ran: sudo ps lf -C Xorg , and this return do show nothing about listening or not === Patrick__ is now known as ptah === EddyKid is now known as MJCD [06:13] what does it mean if someone say "signatures deployed on a firewall" [07:07] Hey party people, I have a problem I have been trying to figure out for a while and was hoping maybe someone here has experience with it. [07:07] I am attempting to create a signed apt repository for my apt/debian packages. [07:08] I was able to copy my old Bintray repo and get it to work but when I tried updating the gpg key and packages/release files I'm now getting this error. [07:08] The repsitory is not signed. [07:13] Think I might have just figured it out answer at bottom of this doc: https://blog.packagecloud.io/eng/2014/10/28/howto-gpg-sign-verify-deb-packages-apt-repositories/ === mycookie8 is now known as mycookie [08:16] I have 700% - 900% load on swapd process sometimes. Any idea how make swapd use less cpu? I am on a ryzen with 12 cores / 24 threads [08:17] Hi guys, im on ubuntu desktop LTS 14, im trying to connect to my synology nas. but i cant i realise its my smb version on my ubuntu desktop. how to upgrade smb version on my ubuntu desktop? [08:17] https://i.imgur.com/njDR3qd.png my swapd process [08:18] leumashm: how about you get 20.04? [08:19] cbreak but i have a ton of them deployed outside 100+ runing 14. and hardware can only support up to 14 [08:19] that sounds idiotic [08:20] do you even get security updates for that? [08:20] cbreak yeah it is. but will be changing all of them to vms after we virtualize all of them. thne wont have this issue. [08:21] cbreak any idea how to upgrade smb on ubuntu desktop to protocol 3? so that my ubuntu desktop can connect to my synology nas? [08:21] you want to change 100 desktops to VMs? [08:21] as I said: I'd just update ubuntu. Newer versions come with newer samba clients [08:35] Hello when I run pip3 I get this error: https://termbin.com/gws1 I have python-pip3-whl installed from apt [08:36] nevermind I needed to delete the local version that somehow I obtained [08:48] hey guys. Rather beginner Linux user here. I have a couple of game servers around the globe running Ubuntu. Since recently a bash command doesn't give the output expected anymore and I have no clue why. This is an example which worked and now just isn't giving me any results anymore (the string is existing, I have done this with different values dozens of times): [08:49] find . -type f -name "*.profile" | parallel -k -j150% -n 1000 -m grep -H -F 'sid="1111"' {} [08:49] maybe someone could tell me what changed or what could be the reason that it's not working anymore? [08:49] thanks for the help [08:53] JackMayol: my advice is, make a small testcase === jelly-home is now known as jelly [08:58] p0a, just made a testcase with "just" 5000 files in a dir (instead of 290k) but same issue. Tried just with find . -type f -name "*.profile" | grep -H -F 'username="DFGHJRTYU"' but still nothing [09:00] does find work? [09:00] JackMayol: does it output anything [09:01] I don't see why the grep pipe is there if `find' doesn't work; bisect your command until you get to the point of interest [09:01] this here works: find . -type f -name "*.profile" | parallel -k -j150% -n 1000 -m grep -H -F "DFGHJRTYU" {} [09:02] so "DFGHJRTYU" instead of 'username="DFGHJRTYU"' [09:02] JackMayol: does your find output contain word username nor uid [09:04] part of my output containing this string: How about a grep for '"DFGHJRTYU"'? [09:06] works too [09:06] Continue adding 1 character after another until it fails [09:07] it fails with the equal sign [09:07] If it's XML, why are you using grep? :P Use an xml parser [09:08] yea I know, but we don't have an admin tool for it, so I just make stuff through bash command >_< [09:09] So it fails at the equal sign, open up the find output with an editor and search for ="..." [09:09] if the search fails, see why. Inspect it yourself [09:10] My understanding is, you want to make sure the shell is not getting in the way, and that the file actually contains the data you are searching for [09:10] try to rule out 1) and 2) [09:14] I searched for username="DFGHJRTYU" in the output result file and it finds it. As I said, I have used this search command this way since a long while and never had issues until today (first time I used it again since about 2 weeks) [09:15] find* [09:16] JackMayol: what's the output of echo '="foo"' ? [09:16] is it what you'd expect? [09:16] ="foo" [09:17] yea [09:17] Write a bash script and store the search string 'username="foo"' in a variable $x that you pass like "$x$" to grep [09:19] ouch, I know that it's not hard but still too hard for me though. Never wrote a bash script myself *hides* [09:20] JackMayol: try the command without parallel [09:20] JackMayol: on the 5k testcase, does it work? [09:21] tried it already, no output: find . -type f -name "*.profile" | grep -H -F 'username="DFGHJRTYU"' [09:22] try without -H and -F [09:25] no output either [09:25] Is there an easy way of getting systemd-resolved to stop listening on port 53, and still have DNS work on the machine? Like, can I have it use a socket or something? On ubuntu server 20.04. I'm asking because I'm installing powerDNS, and that wants to use the same port, but I don't want to use it for the server's own lookups. [09:27] JackMayol: reduce the problem to 1 file, make sure find outputs to stdout instead of stderr [09:27] JackMayol: at some point, you're telling us that grep is not working. I am not believing that ;P [09:29] limbo_: port doesn't make sense here, depends on /etc/resolv.conf where system asks for dns [09:33] limbo_: to overwrite systemd dns settings, you could configure /etc/systemd/resolved.conf [09:33] after that you restart systemd-resolved [09:34] to check queryed dns servers, type "systemd-resolve --status" [09:37] pasiz: what I've tried so far was using DNSStubListener=no in resolved.conf and restarting systemd-resolve. This led to it not listening on 0.0.0.0:53 anymore, and DNS lookups failing. (I tried curl example.com) changing the config back and restarting made it work again. [09:38] of course it would, if you don't specify where it should query [09:39] there is option DNS and FallbackDNS [09:39] you need to configure those [09:40] pasiz: you are right, I'm an idiot. Thank you. [09:41] limbo_: no, you are too fast to make changes without reading documentation ;) [09:53] How do I check if a package is installed ? what is the command terminal? [09:54] any idea how I look in logs using journalctl but ignoring sshd service? (I want to see logs for all other services [09:55] Peyam, you can use apt to list the packages installed already [09:55] Peyam: apt policy package [09:55] meandrain, if you cat the log | grep -v ssh | less [09:56] summonner: there is journalctl -u service to list a service. I need a similar command but to ignore that service instead [09:56] with grep I don't get coloring [09:56] grep --color -v [09:56] also I want to use journalctl like tail [09:57] summonner: Interesting, I did not knew that [09:57] man is your friend [09:57] man grep [09:58] for anyone that just got triggered, it's short for manual - in the "old" days, when analog modems were used, every keypress in unix was precious [09:59] although i'm always getting traceroute and tracert mixed up on windows/unix [10:01] interesting, journalctl -f works with grep (the tail-like command) [10:02] Morning, I have 2 questions today: [10:02] 1. when i configure a sftp access for one folder(structure) is the rest of my system safe or are there possibilites to access to the data outside the sftp area? === cjoke_ is now known as cjoke [10:04] Ublx, sftp stands for... what is your understanding? [10:04] 2. gnormalize does not work for me ... all track files are broken or contain just the first seconds. is there a one liner for command line to get cd tracks to mp3 ... without any special naming? [10:04] summonner: safe and secure? :) but what about the rest outside the sftp folder structure? [10:05] Ublx, i'd recommend a short googling [10:05] because, if you're going to open your system up, you should understand the tools and their capabilities [10:06] i did. for both. many times. but i didn't find a good answer. [10:09] in that case, probably a book? you're touching on a complex topic that is difficult to convey on an ubuntu problem-fix channel [10:11] ok, thought about a helpful start by knowing little bit about security. asking isn't a mistake - i thought. [10:16] Ublx, stick around in the channel then, perhaps someone will pick up the question === yawkat` is now known as yawkat [10:21] yup. of course. [10:21] JackMayol, did you get a solution? [10:21] unfortunately not :-/ [10:22] if at least the exact same command wouldn't have worked for years before, I would think bah it's maybe just me... [10:22] just reading the scrollback [10:23] forgive my questions if you've already answered, so you're in the folder with the profile files, yes? [10:25] and you run grep username=\"DFGHJRTYU\" *.profile does that give you anything === apw_ is now known as apw [10:27] also for your other commands, just as a suggestion, escape the * or other characters with \ try not to use quotes, it makes it harder to figure out where the string closure can be (unless you're using an editor like nano which will colorize the strings automatically) [10:28] does anyone know if there will be a update from Ubuntu Touch to 20.04 this year? [10:29] chriss_toni: ask in #ubports please [10:29] chriss_toni: ubuntu-touch is not community driven and renamed to ubports [10:29] now [10:30] yes i doe. but i ask many questians at this channel. no problem. thx. [10:31] :) [10:35] summonner, grep username=\"DFGHJRTYU\" *.profile works [10:36] grep 'username="DFGHJRTYU"' *.profile as well, though [10:38] does: find . -type f -name \*.profile" | parallel -k -j150% -n 1000 -m grep -H -F \"DFGHJRTYU\" {} work? [10:38] correction: find . -type f -name \*.profile | parallel -k -j150% -n 1000 -m grep -H -F \"DFGHJRTYU\" {} [10:40] summonner, yes [10:44] and JackMayol your original find/grep was looking for the sid string, yes? [10:44] summonner, yes [10:46] JackMayol, and if you run grep sid=\"1111\" *.profile in that folder? it works? [10:51] summonner, in the orginal folder (284k files, half of them *.profile files) I get this: -bash: /bin/grep: Argument list too long [10:51] while in the 5k files test folder, it works [10:54] JackMayol, what about if you bash -x before the parallel, do you get /usr/bin/parallel: line 22: use: command not found [10:58] summonner, yes [10:59] JackMayol, also, try this: save the output from the find into a text file. typically find *.profile > test.txt then use parallel but instead of pipe and {} replace that with .... like parallel -k -j150% -n 1000 -m grep -H -F sid=\"1111\" (whatever you name the file) [11:05] JackMayol, AH [11:05] doesn't output anything in the file [11:06] JackMayol, found the cause. so, looking at your original command: find . -type f -name "*.profile" | parallel -k -j150% -n 1000 -m grep -H -F sid=\"1111\" {} [11:06] introduce an echo into the parallel [11:07] so you see what is happening with the perl script. ie. instead of grep -H -F do echo grep -H -F etc [11:09] and the output, what you see, is what parallel is running - which is clearly not correct - none of the quotes are being escaped correctly, they're being lost. so you need to keep escaping until you see the correct command - then you can remove the echo [11:10] does that make sense? It now works on my machine [11:11] and if this command used to run until today, then you now have a file name, or contents inside the file, which are causing the escaping to fail. probably there's a quote in there somewhere, that isn't being closed. [11:13] if you want to find the culprit, you're going to have to find all recent files, and check them, such as writing some sort of debug script to check each file and compare count, and when you find difference, a missing result, then you can narrow the problem file down and escalate that to whomever writes the files - there's obviously a bug in the code which produces these files [11:13] that was fun === Amol is now known as Amol|afk [11:52] Hi. I typed a non existant command in my terminal with bash as shell and quickly did Ctrl-C after pressing enter. Then I'm getting some python errors saying frozen_importlib KeyboardInterrupt. Does this mean than python is used somewhere in bash? [11:53] Or is this python error coming from somewhere else? [11:57] summonner, thanks for your help, much appreciated! Some files might indeed be truncated. At least now using /" instead of " helps [12:02] * nicoz augura buon pranzo a tutti.. [12:13] famubu, the command-not-found tool (that makes suggestions about installing a package that contains said command) is packaged in python ... the specific shell error is routed to it [12:14] if you managed to create a traceback your timing must have been really lucky to hit it at the right moment or your computer is really slow 🙂 [12:17] is there a way to set status "away" and "active" on all the chat clients at once? Right now I use: hexchat, skype, teams, I'll probably install slack, maybe discord [12:18] on bionic [12:19] JackMayol, yw === Amol|afk is now known as Amol [12:29] Hey, I'm trying to run Robot OS using VirtualBox/Vagrant. Issue is I'm on a ARM M1 mac. I see on the ROS site that it's compatible with 18.04 Bionic ARM, but I'm only finding images for Ubuntu Server or raspberry Pi. Would one of these work or am I SOL? === viaSanctus_ is now known as viaSanctus [12:53] 'Morning folks [12:56] meowning === dtux_ is now known as dtux === kinghat2 is now known as kinghat === waxfire2 is now known as waxfire === cryocaustik7 is now known as cryocaustik === vioan_ is now known as IonutVan_ === DrKK`` is now known as DrKK` === dvdmuckle_ is now known as dvdmuckle === genpaku_ is now known as genpaku === themill_ is now known as themill === WaV2 is now known as WaV === krisd_ is now known as krisd === OliPicard_ is now known as OliPicard === Ankhers_ is now known as Ankhers [13:11] ogra: Oh.. Didn't know that. Thanks! [13:12] ogra: No my computer is a bit slow.. :-D [13:12] I can easily get that traceback. [13:14] How is bash configured to use command-not-found? Is it some configuration in the .bashrc or .profile files? Couldn't find it there on quick grep. === shann_ is now known as shann === osse_ is now known as osse [13:46] famubu: there is a package installed, command-not-found, see "dpkg -l | grep command-not-found" [13:49] famubu: sudo updatedb ; locate command-not-found === ace_me^^ is now known as ace_me [13:54] famubu, you should have echo $PATH, where the command is. [13:55] you could have two command-not-found, but only was executed the first in PATH [13:56] famubu, /etc/bash.bashrc [14:02] Hello, I've got a networking scenario with multiple IPs on one interface and I want to make all OUTGOING traffic from either a specific process or a system user's processes (whichever is easier) to use a specific source IP. Summarised here, would appreciate if any one with ideas could take a look: https://serverfault.com/questions/1056688/forcing-outgoing-traffic-via-specific-ip-by-process-or-user [14:02] found one potential per-user solution via iptables already, though there's a dns problem I'm investigating [14:18] loco41211: maybe stick the user/services in an lxd container. Binding a bridge for that container should be pretty easy [14:19] leftyfb, thanks for taking a look - seems similar to a docker style container, if I understood correctly? [14:20] loco41211: yep. docker would also work [14:20] would mean dockerising the process(es) though... would prefer not to add that complexity layer [14:21] the iptables trick seems similar, just need to find why it breaks dns resolution [15:05] Ello all [15:05] Does anybody know how to add addition mount points to snaps? [16:05] ogra: Found it in /etc/bash.bashrc [16:06] Had lost connection but found messages thanks to the log. :-) [16:06] 🙂 === mIk3_09 is now known as mIk3_08 [17:10] hi [17:10] i am facing [17:10] https://paste.ubuntu.com/p/NYYbKRZknm/ [17:11] hi, i have a problem with a ubuntu-windows dualboot. I have upgraded the ubuntu with apt get upgrade. While that process i had to set an (UEFI?) password for secure boot. Now after reboot i had to validate that password, but when starting windows now, it can't unlock the bitlocker volume anymore [17:11] how can that be? [17:13] Rockwood: from where did you get this error log? [17:13] at time build react [17:14] sudo npm run-script build on this command [17:16] is there a way to set my VM to autostart through the Oracle VM Virtualbox Manager GUI? [17:18] juli: that is a #windows question [17:18] leftyfb: why? It happened because i upgraded ubuntu... [17:18] Mibix: maybe read the Oracle VM Virtualbox Manager GUI documentation? [17:19] Rockwood: what happens if you try to set the symlink manually? [17:19] juli: that might be true, but there's nothing to do in ubuntu to fix it on the Windows side [17:20] juli, its auto setup [17:20] leftyfb: True, i just want to understand because of what that could happen, so that i know whether i can solve this somehow [17:20] Rockwood: sure, but it fails, because you don't have permissions to set that symlink '../node-sass/bin/node-sass\' -> \'/var/www/criminalimpulse/theme/node_modules/.bin/node-sass\'' [17:20] i dont see it in there [17:21] juli: to be honest, I don't know because I disable EFI and secure boot whenever possible. But at the moment, your issue is getting Windows to work. This is not the correct place to get help with that [17:22] juli, how to set permissions then ? [17:22] Mibix: https://kifarunix.com/autostart-virtualbox-vms-on-system-boot-on-linux/ [17:22] of course you cant do it in gui [17:23] facking linux [17:23] Mibix: that is an Oracle issue, not linux. "linux" did not create the application [17:25] Rockwood: you have first to figure out, why you don't have the permissions there, especially when using sudo (with sudo you should normally have all permissions) [17:26] juli, i am confuse [17:27] bcz VM not permanent permissions if once i give it [17:27] Rockwood: ? [17:27] https://paste.ubuntu.com/p/j2jhCZsjYr/ [17:28] always user change on restart [17:28] back to root [17:28] instead of rockwood [17:28] my current user is rockwood [17:29] Rockwood: can you show the output of ls -la ../node-sass/bin/node-sass? [17:29] and are you sure u used sudo? [17:30] Rockwood: what version of ubuntu is this? [17:30] rockwood@cyberia:/var/www/criminalimpulse/theme$ sudo ls -la ../node-sass/bin/node-sass [17:30] ls: cannot access '../node-sass/bin/node-sass': No such file or directory [17:30] ubuntu 20 [17:31] Rockwood: 20.04 or 20.10? [17:31] Rockwood: "20" is not a valid version of ubuntu [17:31] how to chk [17:31] ? [17:32] Rockwood: lsb_release -d [17:33] No LSB modules are available. [17:33] try with sudo [17:34] https://paste.ubuntu.com/p/5YXPpjfbXq/ [17:34] its 20.04.2 LTS [17:34] Rockwood: you are pretty outdated [17:34] yeah [17:34] why? [17:34] i just download this [17:34] very recently [17:35] just installed [17:35] oh wait, nevermind. My bad, sorry [17:35] leftyfb: its LTS so it should be still supported? [17:36] sorry, got confused with my machine running 18.04 which is on 18.04.5 [17:36] :) [17:37] sudo do-release-upgrade -d -f DistUpgradeViewGtk3 should i do that? [17:38] i'am not very familiar with node/npm and so on, but from what i can read there is just a permission error, and you don't need to upgrade your whole system [17:38] Rockwood: no, you do not need to do that [17:38] ok [17:38] then [17:38] Rockwood: I would look for support from whoever wrote the application you are trying to install [17:39] i am creating project [17:39] in react [17:39] my old OS was dead [17:39] so i just installed new version [17:39] 19 [17:39] was my old version of ubuntu [17:39] Rockwood: can you show the log with the command you issued and the path in which you are in? [17:39] !yy.mm | Rockwood [17:39] Rockwood: Ubuntu version numbers are: YY.MM (YY=release year,MM=release month). Each year sees two releases, so just specifying YY is imprecise. See also https://www.ubuntu.com/about/release-cycle [17:40] Rockwood: maybe ask in #react for support writing your own projects? Maybe there's good practices for locations/permissions you are missuing [17:41] ok [17:42] suppose if i wanna do rockwood user and all kind of permission on parent folder my user rockwood [17:42] lol that guide doesnt work leftyfb [17:42] https://forums.virtualbox.org/viewtopic.php?f=7&t=90767 [17:42] same think happened to me grrr [17:42] what kind of command helps? [17:52] Hello anytime know how to set up 3proxy? [17:54] AndroUser10, please don't crosspost [17:58] AndroUser10: also, we can only support official Ubuntu packages here. 3proxy is not one of them [17:58] oerheks: trying to setup 3proxy on ubuntu [17:58] already setup 3proxy but not working satisfactorily [18:00] AndroUser10: you are receiving help in ##linux. Good luck [18:00] what?! 3proxy not official package?! please confirm. Its listed on "apt list" [18:06] ok got it 3proxy not in "apt list", so no support here for the same? where else can the query be posted? pls guide [18:09] !info 3proxy [18:09] Package 3proxy does not exist in focal [18:10] AndroUser10: you are being helped in ##linux. Please pay attention there [18:11] Thank you [18:13] Hello everybody! I'm in trouble. While starting my PC I'm stuck in the grub menu, I mean I can't naviguate and choose an option. There's also no any timer. Please help!!! [18:16] PS: i'm running ubuntu 20.04 on a partition and Win10 on an other [18:25] I feel stupid for asking this question, but in Unity desktop, is there a way to see the full application name when the app launcher comes up? [18:27] I have a 4K screen and the app finder is trying to save screen space by truncating the application names [18:30] the fann too rumorous [18:30] how can i silence it? [18:30] clip the wires === theloudspeaker_ is now known as The_LoudSpeaker === arooni_team_b is now known as arooni === Luca_ is now known as MrLuca [19:02] Hello everybody! Do you know how I could reinstall GRUB [19:06] !recovergrub | aleksei-hash [19:06] aleksei-hash: Lost GRUB after installing Windows? See https://help.ubuntu.com/community/RestoreGrub [19:08] vesafb.invalid=1 enables nomodeset with 1915 driver [19:08] grub [19:08] i915 [19:14] Hey. I suffered weird behaviour, like FireFox altering the url to redirect for the ubuntu firefox website. I figured out that this setting was within FireFox and set it off. Anyway: I remembered than openSUSE had some option in the setting what kind of header/userAgent to identify. Now I stuck in the consideration where in the OSI model this is [19:14] something a OS can alter, and if such altering could be done systemwide or for some specific network. Any inspiration what to look up for? [19:19] The user agent is layer 7 [19:21] I know that the Application can determ such, but like mentioned, in openSUSE you could set some value in the settings that I think was the user agent. So I was wondering what it could be I remember and figured I ask here before installing some openSuse in a VM and look it up again. [19:23] Maybe I should ask in opensuse instead ^^ I was just one step further already and thinking of checking what set in ubuntu at this place [19:23] #opensuse [19:29] quaklo: I assume the application (firefox) would be using tools like lsb_release or uname or /etc/issue to determine OS. I doubt you want to change those just so your browser gives off a different OS. [19:31] leftyfb you are exactly seeing what I try to do. Just that I do not care about FireFox, because I can alter that program within itself. But I consider stuff like wget. I just saw that wget has some httpconfig, but what about other system connections that use the http protocol. I think I want to mess with it ... [19:32] quaklo: it might not be worth the headache [19:32] quaklo: I would do it per application (wget, curl, php, firefox). They should all have methods to set the useragent [19:35] tons of firefox addons, User-Agent Switcher [19:35] its about http on the system level. [19:36] quaklo: there is no http on a system level. http is a service provided by an application [19:36] show us the opensuse solution? never heard of http on system level too [19:37] its about altering it on a "system" level. There are programs you can not alter, like Zoom, Skype, Teamviewer or whatever. [19:37] all configs that are not in a .conf file, are reachable with dconf-editor [19:38] !info dconf-editor [19:38] dconf-editor (source: dconf-editor): simple configuration storage system - graphical editor. In component universe, is optional. Version 3.36.0-1 (focal), package size 306 kB, installed size 1212 kB [19:38] but for mozilla i expect some about:config [19:39] I was more into hacking the network adapter and tell him to overwrite clear http headers. seems like I found my workaround: proxy, collect it in squid, filter request_header_replace header_name [19:40] that is not the goal of this channel [19:40] quaklo: I don't think you can accomplish what you're looking for. Other than changing what your OS actually is. Or guessing at which mechanisms an application might use to determine it's OS. As for changing what that application sets as a client, I don't think that's happening at all. Certainly not a "system level" thing [19:42] quaklo: ah right, a proxy would work. You'll have to configure each application to utilize the proxy though. [19:42] yeah, and most applications offer to set some proxy, but not to alter the http header :P [19:45] oerheks what do you think about the idea of determing what http header is send by os process http requests? There is no need for any true http header to enable features, but there may be a interest of a user changing it for security concern. Wouldnt it be great if thats a variable that could be altered at a centralized place, so users have to [19:45] freedom to chose about it? [19:47] quaklo: that's what the proxy is for [19:47] leftyfb wouldnt work for https. [19:48] its a hacking solution rather than a clean setting solution! [19:48] quaklo: other than that, all http traffic on Ubuntu would need to be routed through an installed proxy by default to take advantage of said feature. I can tell you there are an order of magnitude of people who do not want that over those who want what you are looking for [19:49] quaklo: good luck in your goal [19:50] ? im talking about changing the source code and implementing of some variable instead of a hard coded value. This should be more than likely to be in the interest of a modern and modular os :P [19:50] quaklo: feel free to discuss further in #ubuntu-offtopic [19:51] im talking to you... [19:52] Anyone know how to display a notification to a remote users GUI? 20.10 [19:52] Mr_Queue, message of the day, comes in mind [19:53] Mr_Queue: use notify-send via ssh [19:54] also good, leftyfb , e.g. ssh user@host 'DISPLAY=:0 notify-send "TEST MESSAGE." [20:15] notify-send isn't working.. [20:15] no errors. [20:36] ok, for real though, how do you upgrade packages non-interactively? I've got environment variables, apt flags, DPKG:: options. My god, the incantation is incomprehensible, and it would seem still doesn't cover all cases. Good grief. What is the right way to do this? [20:37] arch-nemesis: investigate how unattended-upgrades does it [20:38] alright, i have two xubuntu installations on separate devices, one on i686 laptop and other on arm64 mobile device. [20:38] now i am trying to launch app from mobile device using the laptop's x server [20:39] and i get this [20:39] Unable to init server: Could not connect: Connection refused [20:39] (xfce4-terminal:17910): Gtk-WARNING **: 23:19:29.281: cannot open display: 192.168.1.102:0 [20:39] i did `xhost +mobile_device_hostname` on laptop, i even did `xhost +` [20:39] gravitos: did you enable tcp listening? [20:39] same results [20:40] sarnold: how do i? [20:40] that was disabled by default a decade back.. [20:40] gravitos: grep for tcp in /etc/X11 -- I use xdm on my system, and it looks like it's configured in /etc/X11/xdm/Xservers on my machine [20:42] gravitos: whatever display manager you're using may or may not have it configured there, but conventions are powerful :) [20:43] `cat /etc/X11/default-display-manager` gives me `/usr/sbin/lightdm` [20:43] so i use lightdm, i guess [20:44] sarnold: Thanks, that'll help. [20:46] okay, i found a file, `/etc/X11/xinit/xserverrc`, it has only two lines in it, one is stating that this script runs in /bin/sh, and other is `exec /usr/bin/X -nolisten tcp "$@"` [20:46] this could be... something [20:48] gravitos: that's used when you start your xserver by running 'xinit' by hand [20:49] that used to be popular 25 years ago but most people use graphical display managers these days [20:52] nothing interesting in /etc/lightdm, hm [20:52] check the manpage? [20:57] found some info about passing a line into lightdm.conf, BUT for some reason there are no lightdm.conf in /etc/lightdm, only an empty folder named lightdm.conf.d [20:57] aha, then you can probably create a new file in that directory with the one line you want [20:58] and reboot? === corrupto is now known as CorruptO [20:59] * gravitos rebooting, brb === remy is now known as Remy [21:04] `ps ax | grep X` now shows that X are -listening to tcp [21:04] and now running any app on mobile device while using lap X server gives me this `No protocol specified` in addition to that two lines before [21:05] and no app actually appears [21:10] gravitos: yay, that's progress anyway :) 'no protocol specified' on its own doesn't ring any bells though :( [21:11] this is literally all i get https://www.irccloud.com/pastebin/fYkMz6HV/std [21:13] oh [21:13] xhost wipes settings on reboot [21:13] that's what the problem was [21:13] after another `xhost +mobile_device_hostname` everything works [21:13] thanks [21:16] gravitos: woot [21:16] gravitos: do be careful with such wide-open settings, the XAUTH framework is designed to be safer than letting anyone and everyone connect to your X server [21:18] if i dont `xhost +` in autorun, everything will be safe enough === nicoz- is now known as nicoz === nicoz- is now known as nicoz [21:57] i did a apt-get upgrade to ubuntu, rebooted and now i cannot start my vm .. error is: error: Failed to activate service 'org.freedesktop.machine1': timed out (service_start_timeout=25000ms) [21:57] has anyone seen this before and knows a fix [21:57] os is ubuntu-focal [22:27] I followed this guide https://www.tecmint.com/enable-http2-in-apache-on-ubuntu/ but it still return http 1.1 [22:28] admin0: systemctl status libvirtd ... maybe another reboot, this error has to do with dbus [22:29] pavlos, https://pastebin.com/raw/r8mBqScT [22:29] i will do another reboot and check [22:31] robertparkerx, same steps as https://dev.to/ayekpleclemence/enabling-http-2-in-apache-on-ubuntu-4pe8 [22:31] why does it not work? [22:31] robertparkerx: pastebin the whole terminal log? [22:32] and how do you test? [22:33] curl -I --http2 -s https://www.exactprecisions.com/ | grep HTTP [22:34] apachectl -M says module is enabled [22:34] i have added line to virtualhost and restart apache2 [22:34] it doesn't change [22:36] pavlos, https://pastebin.com/raw/zKcPSGgt [22:37] libvirtd is up .. but the error does not go away [22:38] admin0: I think there is a conflict btween qemu 64 bit and libraries. I guess the upgrade hosed something [22:39] hi - i seem to install the wrong guest services on my virtual machie and now i get a black screen after typing in password [22:39] anything i can do? [22:40] i really dont want to reinstall everything [22:40] Celphi: can you get to a tty? [22:40] not unless theres some trick to not loading gui in virtualbox [22:42] like normal recovery mode? hold shift on boot [22:42] !recovery [22:42] If your system fails to boot normally, it may be useful to boot it into recovery mode. For instructions, see https://wiki.ubuntu.com/RecoveryMode [22:43] ok shift worked [22:43] how do i remove my last package? [22:44] look into dpkg log? [22:44] im new to ubuntu [22:44]