[00:24] I have around 70GB of memory on my computer (xeon processors), but when I run "free" I only see about 36GB of memory in total. Is this listing the amount of memory available per processor? [00:27] not really, the mobo may be limited to using 36 GB [00:28] eelstrebor, mobo? Sorry I don't know what you mean [00:29] mobo == motherboard [00:29] or mainboard [00:29] where your processor and memory is installed [00:29] eelstrebor, it's a server mobo [00:30] take a look at the mobo specs and see how much memory it is capable of [00:31] Different dimms maybe incompatible? [00:32] Greyztar, any way to check? [00:32] my mobo is limited to 32 GB [00:34] Mine aswell but he appears to have multi cpu,to check id just power it down and oull them out and compare amazoniantoad [00:34] pull* [00:34] Thanks [00:34] Yeah I have a mobo with two processors guys. [00:36] No problem,i dont know but it might be worth a shot amazoniantoad im off then 01:35 here good luck [00:36] Greyztar, I'm almost 100% sure that in a previous ubuntu installation it had my memory listed as 72gb. Do you think that it might be the OS? [00:36] I can't take the computer apart atm [00:38] grep -c “Time has been changed” /var/log/messages [00:40] amazoniantoad: i dont think the os would do that,ive never experienced that other than the was it 2 or 3 gb limit with windows xp due to 32bit /64bit,i think its either bios changes or some modules incompatible or loose or so [00:40] thanks [00:41] amazoniantoad: im no expert so might also try the ##linux channel also :) [00:46] if I leave my machine overnight, it dumps a lot to swap, when i come back in the morning, it swaps back in to RAM and takes forever [00:50] Greyztar, I just restarted my computer and now it says I have 69GB of memory. What gives? [00:50] any ideas? [01:00] My Wifi keeps disconnecting... even though it doesn't show as disconnected [01:01] Like I can't even ping 192.168.1.1 [01:08] i can run a service as an microsoft AD account cant i? I have auth working and whatnot [01:10] back sorry [01:10] So like Is aid [01:10] my wifi connection is very.... unstable [01:10] the connection to the internet goes down.... [01:10] AD is not much fun. [01:10] and I can't even so much as ping the Gateway [01:10] ACTIVE directory, right ? [01:11] The Wifi Symbol shows up with a "?" over it [01:11] all the time it seems [01:11] even when the connection to the internet is established [01:11] but when it goes down [01:11] I have no option, but to reboot my PC [01:11] whcih usually works [01:11] I am running a VPN [01:11] but disablign the vpn doesn't seem to work either [01:12] raidghost: yeh. have everything working but having a svc run [01:12] and trying to connect to a different server basically results in a "can't connect" message === TheHonorableKitt is now known as ThKitten === wodencafe is now known as Guest25302 === Guest25302 is now known as wodencafe [01:58] Why does Ubuntu seem to have two separate clipboards? [01:59] e.g. in the Terminal app, pressing Shift+Insert and Shift+Ctrl+V will paste different things. [02:00] It's really confusing. [02:01] k_sze[work]: maybe that's a Gnome thing; on Xubuntu/XFCE4 they paste the same thing [02:01] k_sze[work]: see e.g. https://superuser.com/questions/68170/how-can-i-merge-the-gnome-clipboard-and-the-x-selection). [02:10] k_sze[work]: you can disable the terminal's shortcuts by editing the terminal's profile === nickNolte1 is now known as nickNolte [02:56] k3b no medium present, it does this no matter what :/ === deathonater is now known as Smeef [02:58] pi0: try another tool to see if it is K3B or the drive/disk that is the problem: e.g. "xorrecord dev=/dev/sr0 -toc" [02:59] !info xorriso | pi0 [02:59] pi0: xorriso (source: libisoburn): command line ISO-9660 and Rock Ridge manipulation tool. In component universe, is optional. Version 1.4.8-3 (bionic), package size 269 kB, installed size 310 kB [03:00] sudo apt install xorriso? [03:01] also TJ- xorrecd [03:09] Hi, guys. My ubuntu has been really slow lately, like it stutters. Typing gets interrupted, opening files takes A LONG TIME. I try using the process manager to see what's going on. When I close everytime, I see that some CPU take up 10-20% CPU usage but the process tab sorting by '% CPU' shows nothing [03:10] hell, even trying to call a calculator or log takes over 30 seconds to load [03:10] -_- [03:15] moredrowsy: sounds like I/O errors; possibly a failing storage device [03:15] moredrowsy: check "dmesg -w" see if there are I/O errors reported [03:15] oh [03:16] can a power surge cause this? [03:16] moredrowsy: power surges can cause lots of things [03:16] i had a storm going on and my power flicker on and off for awhile and restarted my cojmp [03:16] and after i restated...this happened [03:17] i typed 'dmesg -w' [03:17] but it's stuck [03:17] well, there's a lot of stuff and afterwards it's stuck [03:18] moredrowsy: no, it isn't stuck, the "-w" makes it wait for more messages from the kernel [03:18] oh okay [03:18] moredrowsy: leave it running do some 'stuff' see if you get any messages [03:18] asus_wmi: Unknown key 10005 pressed [03:18] i get a lot of these even when not typing [03:18] moredrowsy: do things that usually take ages, if you see I/O errors, ata_xxx and lots of weird lookign technical info, then we'll be interested [03:20] no errors reported [03:20] but my mouse and keyboard is stuttering [03:21] i wonder if this is my computer and not linux. i [03:21] moredrowsy: how are they connected to the PC? USB? [03:21] i'm gonna reboot into windows and see if this slowdown/stutter in windows too [03:21] moredrowsy: OK, then it doesn't seem like I/O errors which is good [03:21] both are in connected to usb [03:21] moredrowsy: yes, a comparison in Windows is a good idea [03:21] brb 5 mins. [03:24] okay...it's happening in windows [03:25] must be a hardware issue after the power surge [03:25] gosh, wat the heck [03:25] thanks for the help === capella|away is now known as capella [04:53] what is a good app to purge deleted items [05:25] Hi, Can anyone please tell, how can I install security-manager package on ubuntu 16.04 ? [05:28] Amit_T: there is no package of that name in Ubuntu [05:28] Checking for module 'security-manager' [05:28] -- No package 'security-manager' found [05:29] this is something I see when configuring one application framework on ubuntu16.04 [05:31] TJ- https://git.automotivelinux.org/src/app-framework-main/tree/README.md#n25, for instance this [05:32] Amit_T: you'll need to ask the people that develop that [05:33] ok [06:49] Hello. I am using Ubuntu 18.04. Do four finger trackpad gestures work out of the box? I can swipe up or down on GNOME to switch workspaces, but I don't have touchegg installed. [06:50] Or any other trackpad/gesture utility that I'm aware of. When I was on Ubuntu 16, I remember having to set this up manually. === e_motion-afk is now known as e_motion [07:08] If your system locks up, and you have to power down hard, how can you find out why it happened once the system has been rebooted? This is 18.04.1 "journalctl -xe" isn't giving me any smoking guns. Neither is /var/log/syslog [07:26] SuperLag: check /var/log/syslog ? check temperatures of the computer inside, and watch the memory usage for next time [07:29] SuperLag: 'journalctl -b 1' should show you messages from the previous boot [07:47] ducasse: had to look at that closer. That's actually the very first boot. :) [07:47] ducasse: journalctl -b -1 gives you the previous boot [07:47] and the only telling thing I can see is when I plugged in my phone to charge it [07:47] SuperLag: sorry, my bad, typo :) [07:47] but I can't see *that* making my system hang... [07:48] It's a Pixel 3XL [07:49] If it matters at all, it's an Intel NUC. 32GB of RAM, 1.5TB of SSD (1TB NVMe stick, .5TB in 2.5" SSD) [07:49] nice little machine [07:50] which gen cpu? [07:52] how do i check if my cd rom drive is working [07:55] im looking for a way to unlock a luks encrypted root partition during boot using a usb stick with a keyfile, all guides i can find only use a usb device and write on the first blocks but no guide using a keyfile - is there a reason this is the case? [07:55] If syslogs clean and no gremlins in dmesg and computer still hard reset i might be looking at faulty hardware then? [07:57] Greyztar: or a firmware issue, have you looked for bios updates? [07:58] ducasse: ummmm.... i7-7567 [07:58] not sure which gen that makes it [07:58] any suggestions [07:59] SuperLag: first number, so 7 [07:59] Greyztar: if that's the case, that suuuuuuucks [07:59] Greyztar: it's waaaaay out of warranty [07:59] ducasse: oh, I thought you were asking like which *Lake version it is. [08:00] or something along those lines [08:00] ducasse: ... I just forgot,i recently updated bios ,and ofc all the fun stuff started after,ugh,kinda need that patch though so computer doesnt "melt" down. Had all but forgotten i had updated hehehe [08:01] SuperLag: thanks for help also [08:02] Greyztar: I helped? [08:02] SuperLag: in general, the nucs work really well with linux. are you running the latest firmware? [08:03] ducasse: I *think* so, it may be a little bit behind. Not sure how to update it now that I'm not running Windows 10 on it anymore. [08:03] SuperLag: actually yes,reminded me of the update because of cpu hehehe [08:03] ducasse: if there's still a means to update it while on Linux... I'd like to get it current. [08:03] SuperLag: you can probably update from a usb stick [08:14] pi0: Put a disk into the drive and see if it gets read and mounted? [08:15] Hello, trying to send a mail in a script sh using this command : printf "$BODY" | mail -s "${SUBJECT}" --attach="${FILE_PATH}" "${EMAIL}" [08:16] but I receive the email with empty body [08:16] something wrong in the command ? [08:21] i do not see it mounted [08:21] shoot [08:22] any errors in dmesg? [08:23] what is the cmd for that just dmesg [08:23] sudo needed? [08:23] no, just 'dmesg' [08:24] cdrom: Uniform CD-ROM driver Revision: 3.20 [08:25] ducasse: sure enough. Now I'm current. [08:27] how do i burn iso to usb drive? [08:27] SuperLag: good, see if it still locks up [08:28] ducasse: yeah, now it's just a waiting game. [08:28] neure: Linux ISO? [08:28] yes [08:28] neure: and are you doing it from Linux, or another OS? [08:28] from linux [08:28] Ubuntu [08:29] i want to create a file that will store all my online user account passwords. whats the best way to do this? [08:29] gambl0r3: I may depart from the hive mind here... but I'd say just use a password manager. LastPass, KeePass, etc. [08:29] gambl0r3: use a program like keepassxc [08:29] neure: you may use dd command: dd if=/path/to/file of=/dev/sdX be carefull though [08:30] Hello why is no possible install mbed TLS in ubuntu 14.04? [08:30] are these open source? whats the recommended one to use? [08:30] neure, https://fossbytes.com/create-bootable-usb-media-from-iso-ubuntu/ [08:30] neure: what Greyztar said. You can check the tail end of dmesg to see what location your USB stick got recognized at [08:30] like /dev/sda /dev/sdb..... and so on [08:30] gambl0r3: keepassxc is open source [08:31] So I have /dev/sdb1 [08:31] should i write to sdb1 or sdb? [08:31] sdb [08:31] ? [08:33] ~/dell/dell-xps-9560-ubuntu-respin$ sudo dd if=linuxium-v4.20-rc4-ubuntu-18.04.1-desktop-amd64.iso of=/dev/sdb [08:33] dd: failed to open '/dev/sdb': No medium found [08:34] how come? [08:34] neure, are you looking for command or gui app? [08:35] mircx1: the exact packagename please? [08:35] either will do [08:35] how do i see a list of all the PPA's i added to my system? [08:35] im using kubuntu [08:35] !sources | gambl0r3 [08:35] gambl0r3: The packages in Ubuntu are divided into several sections. More information at https://help.ubuntu.com/community/Repositories - See https://wiki.ubuntu.com/RecommendedSources for the recommended way to set up your repositories. [08:35] neure, install etcher [08:35] I suppose because i ejected the drive from nautilus [08:35] then the drive disappeared [08:36] the drive no longer shows up with sudo fdisk -l [08:36] neure, then reinsert the drive [08:36] however, Disks application still shows it, at /dev/sdb [08:36] this is strange [08:37] neure, it isnt mounted, reinsert [08:37] i thought i could dd to sdb even when sdb1 was not mounted [08:37] neure, and disks app will tell if it is sdb..etc [08:37] hello, is it possible to install the glfw packave, version 3.2.1, on my ubuntu 16? I compiled it myself and am having trouble linking against it [08:37] !latest | dadabidet [08:37] dadabidet: Packages in Ubuntu may not be the latest. Ubuntu aims for stability, so "latest" may not be a good idea. Post-release updates are only considered if they are fixes for security vulnerabilities, high impact bug fixes, or unintrusive bug fixes with substantial benefit. See also !backports, !sru, and !ppa. [08:38] i try install mbed TLS in ubuntu 14.04? and i not see nothing [08:38] coz_, so is it okay to dd to sdb while sdb1 is mounted? [08:38] !ppa [08:38] A Personal Package Archive (PPA) can provide alternate software not normally available in the offical Ubuntu repositories - Looking for a PPA? See https://launchpad.net/ubuntu/+ppas - WARNING: PPAs are unsupported third-party packages, and you use them at your own risk. See also !addppa and !ppa-purge [08:38] mircx1: the exact packagename please? [08:38] well glfw 3 2 1 fixes a big bug [08:38] neure, it will work its magic [08:39] trying now [08:39] dadabidet: we also reccomend using packages from the official repos, instead of compiling your own [08:39] dadabidet, are you programming your own code which uses glfw, or something else? [08:39] lotuspsychje, there is a nasty bug in previous versions of glfw [08:39] neure, yes I am coding something [08:39] dadabidet: wich version? [08:40] look at their issues #747 [08:40] https://github.com/glfw/glfw/issues/747 [08:40] from april 2016, so it's an old bug [08:41] dadabidet, if you use cmake in your own project then you can have glfw as subdirectory/git submodule and include it [08:41] dadabidet: what happens on git, ubuntu cant do much about [08:41] dadabidet: can you explain whats your end goal with glfw please? [08:41] dadabidet, there is also #glfw [08:41] I don't use cmake in my project only make [08:44] is there a channel on here for midnight commander ? [08:45] !alis | Kingsy [08:45] Kingsy: Alis is an IRC service to help you find channels. For help on using it, see "/msg Alis help list" or ask in #freenode. Example usage: "/msg Alis list http" [08:46] ducasse, do you use keepassxc? [08:46] ugh, doesnt look like there is [08:47] Kingsy: if its a package on ubuntu, and you have trouble with it, try here perhaps [08:47] do you need a usb drive to use keepassxc? [08:48] does anyone in here know a good way of browsing the network for drives (just like nautilus) or somethinfg could do it but with the terminal? I am trying to us mc but I only see a way of entering the full smb://i.p.address rather than a browser [08:48] gambl0r3: yep, it's what i use. you don't need a usb drive, no. [08:49] ducasse, you mind showing me to set it up? [08:49] its complaining about a yubikey [08:49] gambl0r3: you don't need a yubikey, just don't select that [08:50] Kingsy: can this help? https://help.ubuntu.com/stable/ubuntu-help/nautilus-connect.html.en [08:50] lotuspsychje: no, I want terminal only [08:50] do i have to create a new database? [08:50] gambl0r3: yes, you need one to save your passwords in [08:51] ok nevermind. it works. thanks [08:51] hello. I tried to upgrade to 18.4 from 16.4 last night and it was really a hassle, lots of package problems and crap, but finally the system is up, well almost... [08:52] problem is the context menu and title bars all looking like from the 90s [08:52] is it a gtk package problem? [08:52] this happens when I choose to login with wayland [08:52] the other option doesn't work at all! [08:52] but that's another problem [08:52] Kingsy: smbclient ? [08:53] sentiment: screenshot of the glitch please? [08:53] lotuspsychje: don't you need to know the ip address of the server? [08:55] Kingsy: check the manpage for smbclient [08:56] lotuspsychje: yeah it all talks about specifying a specific server. [08:56] lotuspsychje: I tried print screen but where does it save?! [08:56] I don't know the ip, I just want to "browse" [08:56] the window title bars are non existent [08:56] sentiment: pictures normally or /home [08:56] Kingsy: http://www.giannistsakiris.com/2007/11/12/accessing-an-smb-remote-folder-from-ubuntu-s-command-line/ [08:56] and menus look so ugly, i.e no style at all [08:57] etc [08:57] sentiment: lets have a look :p [08:57] lotuspsychje: --> sudo mount -t smbfs //remote-host/folder-name <-- you need to know the ip address [08:57] lotuspsychje: can't find the pics! [08:58] this is silly [08:58] it should tell me where does it store the screenshots [08:58] and what name [08:58] this is ridiculous [08:59] sentiment: just made one, saves in pictures [08:59] what name? [08:59] How is smbfs different from cifs [08:59] windows have no scollbars [09:00] found it [09:00] https://pasteboard.co/HPxpc5W.png [09:00] lotuspsychje: ^ [09:01] Kyros: same thing [09:01] sentiment: thats just your chat window, what about your whole desktop? [09:01] Kyros: it was renamed from smbfs to cifs [09:01] Fair enough, if it doesnt work you might need to install cifs-utils [09:01] lotuspsychje: desktop is there [09:01] ducasse: ty [09:02] the windowss have issues [09:02] sentiment: wich desktop please? [09:02] it's a gtk thing i suspect [09:02] wayland [09:02] gnome [09:02] ubuntu 18.4 [09:02] sentiment: allright, install gnome-tweak-tool and let us know wich theme is active please [09:02] ok bbl [09:02] thanks [09:02] exit [09:21] Hello. I am trying to install some software, but it says I am missing the libraries needed. I can't seem to figure out how to obtain them. [09:21] libpng12.so.0: cannot open shared object file: No such file or directory [09:21] lets say i have a VPS ubuntu server, and i want to share a root with another partner, but i dont want him to see my stuff and i dont want to see his stuff, so how to do this separation on VPS correctly on ubuntu? [09:21] I am running lubuntu as my distro. Any help on how to get it? [09:24] https://askubuntu.com/questions/978294/how-to-fix-libpng12-so-0-cannot-open-shared-object-file-no-such-file-or-direc Galactor ? [09:25] jluc: I'll give it a look [09:25] that was an easy one ! [09:25] lets hope it'll be ok for you [09:25] gotta turn on my vpn for it tho. yay china internet :P [09:27] it says : wget -q -O /tmp/libpng12.deb http://mirrors.kernel.org/ubuntu/pool/main/libp/libpng/libpng12-0_1.2.54-1ubuntu1_amd64.deb && dpkg -i /tmp/libpng12.deb && rm /tmp/libpng12.deb [09:45] after creating usb flash from iso with dd, how can i verify the drive is good and has the right data? [09:47] nvm, found https://unix.stackexchange.com/questions/75483/how-to-check-if-the-iso-was-written-to-my-usb-stick-without-errors [10:01] Hello why is no possible install mbed TLS in ubuntu 14.04? [10:05] mircx1: https://packages.ubuntu.com/search?keywords=mbed+TLS&searchon=names&suite=all§ion=all you can find it for xenial or bionic, trusty seems to old [10:06] with trusty it had an other name: https://packages.ubuntu.com/search?suite=all§ion=all&arch=any&keywords=PolarSSL&searchon=names [10:07] as you can read it here: https://tls.mbed.org/ :"PolarSSL is now part of ARM Official announcement and rebranded as mbed TLS. " [10:13] Hi - while doing security patches (18.04), Ubuntu added a new package: bubblewrap. It installed something setuid, and the only news I could find about it is that Ubuntu was going to -remove- it a while ago, because it can't be audited. Does anyone have further details? === LabMunkeh is now known as Mechanismus [10:14] q [10:17] !info bubblewrap bionic [10:17] bubblewrap (source: bubblewrap): setuid wrapper for unprivileged chroot and namespace manipulation. In component main, is optional. Version 0.2.1-1ubuntu0.1 (bionic), package size 33 kB, installed size 91 kB (Only available for linux-any) [10:18] seems optional Peanut [10:29] hi [10:29] Hi! [10:29] I updated my kubuntu 14.04 yesterday evening (about 17-18 hours ago) and shut down the machine when it was done [10:30] today, when I booted up my machine it very soon froze because kwin was hogging the display server [10:30] I know that, because killing kwin via ssh revived it [10:33] my question is: is there any known issue? should I expect this to happen again until an update fixes it? === [1]MrMobius is now known as MrMobius [10:34] TomyWork: It seems optional, but then why did it get added to my system when doing patches this morning? [10:34] Peanut can you rephrase the question? [10:35] did you ask a question and mistake my question for an answer to it? [10:37] quick question: what's the easiest way to execute a script depending on WIFI name? (adjust screen layout at home or office). I'm not sure how to get the wifi name in a dispatcher.d script [10:40] solars i once had a script that fingerprinted xrandr output [10:41] but that wasn't very stable with driver updates. might work better for you :) [10:42] https://askubuntu.com/a/117068 [10:42] this might be what you want [10:43] how do i reset usb drive partitions [10:43] i wrote ubuntu installer and i now need to restore full size of the disk [10:44] !details | neure [10:44] neure: Please elaborate; your question or issue may not seem clear or detailed enough for people to help you. Please give more detailed information; for example, we might need errors, steps, relevant configuration files, Ubuntu version, and hardware information. Use a !pastebin to avoid flooding the channel. [10:53] blackflow it's pretty clear already. he wrote an ubuntu installer image to his usb drive. now he can't use the full size of that usb drive [10:53] I don't read that from what they wrote. [10:54] maybe it's cause i work in support *g* [10:54] maybe it's maybelline. [10:59] TomyWork: yeah, sorry about that, it seems I need glasses, and it doesn't help that I had 5 colleagues at my desk all of a sudden. [11:01] hello [11:01] heh [11:17] is there any check/(cheque) scanner application for linux? i will need to find a scanner device that works on linux but i have not found a check scanner manufacturer which has scanning software for linux, with check scanning apart from scanning the image the software needs to read the magnetic text in the bottom of the check [11:18] Check? Wasn't that a type of payment method used in the 80's? [11:18] isene: some people have not moved on [11:19] Wow [11:19] Sorry, don't have an answer, though [11:24] sruli: only thing I could think of is use this "linux scanner compatibility list" in search engine [11:25] SimonNL: i get nothing, related to check scanner, the hardware is not the real issue, i once tested scanning with libsane was able to get an image, i need a way to utilise the scanner to read the actual magnetic data [11:26] OCR ? [11:27] SimonNL: no its magnetic, ocr might read it but it will also read the rest of the page which i dont want, i have 1 workstation with windows for this and am desperate to get rid of it [11:27] sruli: you already have the scanner itself ? [11:27] I'm of no use to you I'm afraid [11:27] tchakatak: yes but dont mind replacing it, its old so i can replace with any [11:28] sruli: did you try to use the native app in wine ? [11:28] maybe more easy than fine a linux app for that [11:28] hm… trying to wite command to do this things as follows, but it doesn't work. i type „sudo apt update && sudo apt upgrade && sudo apt autoremove -y” . sorry for noob question, but what am i doing wrong? it stops after update (there are broken repos i don't want to get rid of yet) [11:28] how do I make USB stick read/write? [11:29] tchakatak: didnt think of wine, (havnet used wine since i tried to get IE to work on it in 2012) but would much rather a lightweight solution, doesnt even have to be gui [11:29] neure: it already should be rw, its your mount option wich is in ro, try to mount it on rw [11:30] why would it be mounted as ro? [11:30] sruli: wine works really nice and is not that bad for memory consumption [11:31] neure: depend what is on the usb stick. [11:31] hello is this a good channel to ask about conjure-up kubernetes with MAAS ? [11:31] tchakatak: will look into that option, thanks [11:31] neure: if you try to dd a iso to a usb stick the fs will be ro [11:31] i repartitioned [11:31] turns out i had to unplug and replug [11:31] .. i get stuck at juju controller is initializing === [1]MrMobius is now known as MrMobius [11:32] .. and logs sit at 'awaiting bootstrapped at ...' [11:33] neure: so it might be that your repart was stopped in the middle [11:33] neure: do you have any datas on it ? or can you repart it again ? [11:33] .. in maas i see a node which got deployed and says in logs 'curtin: Installation finished.' [11:33] it works fine now after i unplugged and replugged it [11:34] neure: ok. [11:36] nautilus still has problems [11:36] but could be permissions [11:36] i put ext4 on the drive [11:36] neure: what kind of problems do you have on the drive ? [11:37] trying to copy files to the drive with nautilus -> "The destiunation is read-only." [11:38] meanwhile cp works just fine === [1]MrMobius is now known as MrMobius [11:39] also chrome has no problems saving to the stick [11:39] so i think this is Nautilus issue [11:40] just realized #juju is better [11:44] neure: try to close all instances of nautilus and restart it [11:47] Hello friends, im in dire need of your help [11:47] something really weird happened: [11:48] i tried to fix my screen tearing issue in my ubuntu-studio, ran some scripts in terminal, download some updates, boot my system [11:48] now im running basic ubuntu [11:49] Azukawa: what de are you using [11:49] tchakatak, excuse my lack of knowledge. What is a de? [11:50] Desktop Environment [11:50] Are you using gnome ? [11:50] or unity ? [11:50] what version of ubuntu are you using [11:51] i dont know, how can i find out? [11:51] Azukawa: type "lsb_release -sd" [11:52] in a terminal [11:52] ubuntu 18.04.1 [11:53] What flavor did you install ? the official ubuntu ? [11:53] the version that i installed and ran happily for 6months was ubuntu-studio, now after the updates its the normal ubuntu [11:53] Azukawa: what does "echo $XDG_CURRENT_DESKTOP" give? [11:54] tchakatak: i installed the canon software on wine, it launches, will connect the scanner soon to test, are there any security considerations? can i block that application from accessing the network / internet? [11:54] sruli: you can use ufw if you want to manage how wine is contacting the network [11:54] EriC^^: Ubuntu: gnome [11:55] EriC^^: composite manager maybe ? [11:55] but this is not the one i installed, i installed and ran ubuntu studio [11:56] tchakatak: thats a whole new level, i dont know ufw and use iptables [11:57] should i just back my things up and re-install ubuntu-studio, or is there a way back? [11:57] sruli: you can use iptables if you want [11:58] Azukawa: you can try this, but it switched by itself from ubuntu-studio to ubuntu ? [11:59] yes! not completely autonomys, i installed some updates and run some scripts in the terminal. [12:00] after the updates it asked me if i want to boot my system, i say yes and all of a sudden im in basic Ubuntu [12:00] maybe the problem is the scripts you run [12:00] never had that system installed on this computer [12:02] Azukawa: maybe just by changing the source.list you might have upgrade from ubuntu-studio to ubuntu [12:02] I was trying to install and later reinstall xorg [12:02] Azukawa: why ? [12:03] Azukawa: then you removed the full DE for ubuntu-studio to ubuntu. [12:03] because i had to modify a .conf file to fix my screentearing and couldnt find the folder where it was supposed to be [12:03] Azukawa: the screentearing is not a xorg (not directly) issue [12:03] i figured there reinstall might help, obviously not. [12:04] and i think you reinstalled the vanilla one. [12:04] i must admit i had no idea what i was doing [12:04] The easyest way would be to reinstall fully ubuntu-studio [12:05] some how the path of trying to find the right .conf file led me to xorg and xorg led me to this. [12:05] I am pretty sure you upgraded to a LTS vanilla. you might have modified dependencies [12:05] Azukawa: What you would need is a composite manager such as compton [12:05] that might be! after certain scripts in terminal it downloaded several hunded mb worth of files [12:05] So, thats why :) [12:07] Azukawa: So, either you can chose the easy or the hard way. === tsglove2 is now known as tsglove [12:07] Ool i try install from terminal but is not download [12:07] Easy, backup and reinstal [12:07] and hard way? [12:08] Azukawa: the hard way would be to change the sources list, check everything apt modified, desinstall all of it, and reinstall everything in command line [12:08] hello. I need some info about the packages that render the UI in ubuntu 18.4 [12:08] sentiment: gnome / wayland by default [12:08] because I have some problems with the upgeade from 16.4 [12:09] it has messed things up ui wise [12:09] tchakatak: there are two menu options beore I login [12:09] one is wayland and it has issues with the rendering of the window parts like scrollbars and title bars [12:09] they are non existent [12:10] sentiment: there is a change from 16.04 to 18.04 from unity to gnome [12:10] the other option (non wayland) totally breaks up because input gets disabled [12:10] Tchakatak: Thank you thousands for your help. What do you think, as someone who got myself in to this mess, should i even consider the harder way? What about the amount of work compared to reinstall? [12:10] Azukawa: harder way will be more rewardfull for knowledge :0 [12:10] tchakatak: I know, but this is about a problem. not a change. [12:11] Azukawa: more work i have to say [12:11] I need to understand what package renders the window in wayland so I can try to fix it [12:11] can you please guide me? [12:11] is it GTK? [12:11] Azukawa: take a look at apt / sources.list and how it interact with the system [12:12] Azukawa: look also the script you used [12:12] you should have a modification on the source.list [12:13] tchakatak: look at this screenshot please https://pasteboard.co/HPyFhoB.png [12:13] see the title bar [12:13] all components of windows are messed up like that [12:14] sentiment, tchakatak is currently helping another person, please be patient if you'd like help [12:14] some package has been messed up during the upgrade process [12:14] sentiment: mhh... first i would use xorg [12:14] not wayland for now [12:14] can't use xorg at all [12:14] it is worse because all input gets disabled [12:14] after I login, that is. [12:14] i would actually fix this :) [12:14] guiverc_d: ok [12:15] I tried reinstalling all the packages that seemed to be related to UI to no avail [12:15] e.g ubuntu-desktop, gnone-shell, xorg... [12:15] guiverc_d: i already reply to the other person :) [12:16] is that a GTK issue in Wayland? (the screenshot) [12:16] because at least I can use the desktop in wayland unlike xorg which is totally borked [12:17] sentiment: just to know, it happend before you upgrade ? [12:17] tchakatak: ran the first two lines of this tutorial before realizing that it was propably a mistake : [12:17] Did you try to re apt update/upgrade ? [12:17] https://www.computersnyou.com/4945/re-install-xorg-xserver-completely-ubuntu/ [12:17] Azukawa: i will take a look [12:18] btw, there was another package issue with python3-gdbm that I just installed OK [12:18] I don't think that has something to do with UI [12:18] tchakatak: it happened after upgrade [12:18] because the upgrade process was not smooth at all, there were various package issues [12:19] tchakatak: I tried re upgrade and update many times [12:19] I see no problems with packages as far as apt-get reports [12:19] apt-get check is ok [12:20] but it was not yesterday [12:20] after that i did sudo apt-get update, and i believe after that i went to the software and drivers guided by another tutorial, and after updating those, i was running Ubuntu instead of studio [12:20] Azukawa: Can you tell me what other script you used ? [12:20] just someone please tell me what package handles window components? is it GTK? [12:21] I could find my way from there [12:21] right now I am focusing on the wayland issue. [12:21] sentiment: GTK is a framework/lib which indeed does it but the primary renderer is the the WM/DE you're using [12:21] because xorg is totally messed up, doesn't accept input at all. [12:22] which is wayland؟ [12:22] sentiment: if I were you, I'd concentrate on solving whatever xorg issue you've got there. There's a reason wayland is not yet the default [12:22] no, wayland is a protocol [12:22] gnome then [12:22] so it is a problem with gtk and gnone? but gtk is gnome [12:22] right, gnome/mutter is a wayland compositor [12:22] so gnome is messed up [12:22] mutter? [12:23] no idea which element in that chain is responsible. [12:23] maybe I should check mutter state. what is it anyway? [12:23] in a few words [12:24] tchakatak: I dont recall running other scripts :/ [12:24] sentiment: ask google, you'll get more info [12:24] ok, I just dpkg -s Mutter [12:24] and it is ok [12:24] also read the description there [12:25] so it IS the window manager right now [12:25] and it is OK installed [12:25] I wonder what package is it then [12:26] OK now I try to reinstall Mutter [12:27] Azukawa: Maybe without knowing [12:27] dpkg: warning: files list file for package 'linux-sound-base' missing [12:27] ^ after I tried reinstalling mutter [12:27] it installs with alot of such warnings [12:28] tchakatak: any clue please? [12:28] sentiment: did anything went wrong during your upgrade ? [12:28] alot [12:28] had to run the upgrader twice [12:28] lots of package dependency problems eetc [12:29] but that's kind of beside the point at this moment [12:29] is it not? === al2o3-cr is now known as Guest39666 === al2o3-cr-tmp is now known as al2o3-cr [12:29] sentiment: dont look like it [12:29] dpkg: warning: files list file for package 'libswscale-ffmpeg3:amd64' [12:29] lots of such warnings [12:30] it look like you have a lot of dependencies wich are not here anymore [12:32] sentiment thats why your imput dont work, and all ui is messed up. [12:32] but the apt doesn't report problems with dependencies [12:33] sudo apt-get purge xorg "xserver-*" [12:33] sudo apt-get purge lightdm plymouth [12:33] sudo rm -rf /etc/X11/xorg [12:33] now I try to reinstall some of those packagtes [12:33] oups [12:33] sorry [12:33] packages* [12:33] sentiment: i would do that [12:33] OK thanks, but in that process I will lose all GUI [12:33] so I'd better run that from a login shell right? [12:34] Dont past what i pasted... what a past mistake [12:35] sentiment: i would look at every warning package [12:35] and reinstall them [12:35] Hello, I need some quick help. In 16.04 i used udev to change the interface name based on the KERNELS ID.. how can i do the same in 18.04 ? [12:36] ok [12:36] bbl [12:36] tellus83: you can still use udev under 18.04 [12:38] is there a command for checking and resolving that kind of errors? [12:38] I am at the login shell now [12:39] sentiment: what did you exactly change and how [12:39] I just purged xserver xorg [12:39] tellus83: are you talking about the NamePolicy=kernel for default links? [12:39] ikonia: I just tried to upgrade to 18.4, I was actually hesitant to make that move. [12:39] now I kind of regret that I did [12:39] sentiment: ok ? [12:39] what's the actual problem rather than the tale of woe [12:40] sentiment: just back up all your data, and reinstall from scratch [12:40] well lots of package file missing for some package warnings [12:40] ikonia: he have no input under xorg, and no ui under wayland [12:40] and problems with xorg and wayland UI [12:40] xorg doesn't accept input at all [12:40] sentiment: package files are missing how ? [12:40] and wayland renders windows badly [12:41] blackflow: thanks but no thanks [12:41] ikonia: lots of package list file missing [12:41] you know [12:41] sentiment: it's fastest and easiest thing to do at this point [12:41] I need to get a grasp of the cause though [12:41] sentiment: you already sad package files are missing [12:42] I used this on 16.04. "SUBSYSTEM=="net", ACTION=="add", KERNELS=="0000:01:00.0", NAME="DMZ"" .. if i compy the file from 16.04 to 18.04 the name dont change. [12:42] ikonia: can you scroll up? [12:42] I pasted some examples [12:42] sentiment: yeah, I'm asking why package files are missing [12:42] as they don't just vanish [12:42] botched upgrade [12:42] well I think I explained that as far as I could [12:43] yes ^ [12:43] tchakatak: if i cant figure which script i ran, then reinstall is the only option? [12:43] blackflow: no .. i dont want the eth0 .. but my own name based on witch interface it is. [12:43] sentiment: lts upgrades only bork, when users have maintained their system a bad way before, adding external ppa's and other goodies [12:43] Azukawa: the easyest yes. [12:44] tellus83: I wasn't talking about eth0 but naming policy (and order of it). you can always rename the interface with the systemd-networkd .link and .network units [12:44] tellus83: just to know, why dont you us ip to do so ? [12:45] well thats what im propably going to do this friday night then!:) [12:45] Azukawa: enjoy :) [12:45] what can i use to change cursor size? [12:45] i remember there was a tool which had numerical setting [12:45] tellus83: like'ip link set blablabla naame blablablabla2' [12:46] neurre: which cursor ? [12:46] sentiment: how about some pastebin on: sudo apt update && sudo apt full-upgrade for us please? [12:46] tchakatak, mouse cursor [12:46] its too small for me [12:46] tchakatak: thank you once again, the helpfullness of you and this whole community is something i find very valuable and rare these days! [12:47] Have a nice weekend everybody!:) [12:47] tchakatak: I'm deploying this on 80 boxes and the mac is difrent. .. chan i use the id and not the mac to change it whit ip command ? [12:48] tellus83: look at the Match section https://www.freedesktop.org/software/systemd/man/systemd.link.html [12:48] tellus83: yup. ip is using the default name of the device [12:48] Azukawa: Your welcome [12:49] neurre: Are you using gnome ? [12:49] look in tweak tools if so i am pretty sure you will find it there [12:49] tweak tools did not have cursor size :( [12:49] but I found it in dconf-editor [12:52] Perfect. Thanks blackflow and tchakatak .. love the ubuntu community. you can always get help :) [12:53] neurre: watch out with dconf-editor. [12:53] neurre: you might broke your system. [13:02] !info nvidia-driver-340 [13:02] Package nvidia-driver-340 does not exist in bionic [13:02] !info nvidia-driver [13:02] Package nvidia-driver does not exist in bionic [13:02] !info nvidia [13:02] Package nvidia does not exist in bionic [13:03] you can pm the bot as well [13:03] !info nvidia-340 [13:03] nvidia-340 (source: nvidia-graphics-drivers-340): NVIDIA binary driver - version 340.107. In component restricted, is optional. Version 340.107-0ubuntu0.18.04.1 (bionic), package size 28665 kB, installed size 139972 kB (Only available for i386; amd64; armhf) [13:07] Hey all [13:07] What could cause "Permission denied (publickey)." rather than different keys? [13:07] I've faced same situation already twice. [13:08] Created keys, copied to server public key to authorized_keys and was able to connect. Later, tomorrow, I've tried to connect, but got "Permission denied (publickey)."!!! [13:08] What could this be? This is insane. [13:08] I lost control already and don't know what to do ... [13:08] trying to do a 3 member scenario: 1 nfs server + 2 clients who mount a share with the same username but with different uids. I just cant get the bloody clients to use the mapped UID so basically have the same owner. any idea on that? [13:08] This damn public key ssh system is crap or something ... [13:09] !language | VjdfMQ [13:09] VjdfMQ: Please avoid any language that may be considered offensive, including acronyms and obfuscation of such - also see https://wiki.ubuntu.com/IRC/Guidelines || The main channels are English only, for other languages, see https://wiki.ubuntu.com/IRC/ChannelList [13:09] VjdfMQ: no usernames or hostnames changed? [13:09] hateball: Users are same. [13:09] What do you mean by hostnames? [13:09] Sorry, lotuspsychje [13:09] Try to imagine that you've lost control for your server with only one door [13:10] This is what happened: [13:10] VjdfMQ: take a look at the .ssh folder in your home directory [13:10] you can use also ssh -vvv to see where the problem can came from [13:10] We've generated keys, copied public key to authorized_keys on remote server and tested connection. All worked. Disabled plain password. [13:11] VjdfMQ: possibly some cloud-init nonsense reset the key in the meantime? [13:11] Later, tomorrow, we've tried to connect to server and got "Permission denied (publickey).". No changed were at ALL! [13:11] s/changed/shanges/ [13:11] changes* [13:11] blackflow: This is own server(not cloud) [13:12] VjdfMQ: Any pupet running on the server ? [13:12] tchakatak: Already, files exist [13:12] are you sure the local key is accessible by ssh? [13:12] VjdfMQ: but is cloud-init installed? (it is on default server ISOs) [13:12] I'll send all permissions and -vv output [13:12] blackflow: What is cloud-init? [13:12] VjdfMQ: ok please use pastebin [13:12] !info cloud-init [13:12] cloud-init (source: cloud-init): Init scripts for cloud instances. In component main, is extra. Version 18.4-0ubuntu1~18.04.1 (bionic), package size 368 kB, installed size 1863 kB [13:12] tchakatak: Can we use termbin? [13:13] termbin's fine [13:13] anything you want [13:13] blackflow: I don't know [13:14] VjdfMQ: you can also use the /var/log/auth.log on the remote server to see what is happening [13:14] and syslog too [13:15] VjdfMQ: you protect ssh with fail2ban? [13:16] ssh output: http://termbin.com/bc96 [13:17] VjdfMQ: echo_1 ? [13:17] lotuspsychje: Already tried to login with different ip. Same. [13:17] oh, sorry: http://termbin.com/3gcn [13:17] File for output was echo_1 ^^ [13:18] tchakatak: So that's would be good, but we can't to login to server because there's only one door "SSH" [13:18] And it's closed [13:18] can't login* [13:20] it looklike it come from local... [13:20] VjdfMQ: do you have some kind of (virtual) console access to the machine? IPMI? [13:20] id_rsa => no such file or directory [13:20] blackflow: Yes, and we're searching for password (for IPMI) right now. [13:21] but your key is : debug2: key: /home/user/.ssh/id_rsa (0x17xba716b95) [13:21] tchakatak: https://hastebin.com/vefeboquma.sql [13:22] well that key was not accepted (But it was offered) [13:22] You mean wrong public key? [13:22] Or private key? [13:22] The thing is that we didn't change any keys [13:22] not necessarily. the reason for rejection is not stated [13:23] blackflow: the local /var/log/auth.log could not state it ? [13:23] could be a number of reasons on the server side where sshd responds with "Next auth method" (meaning, "try again") [13:23] tchakatak: local on the client? no. on the server? might [13:23] I'd look into journalctl -u ssh.service tho' [13:24] some examples where it could go wrong: you've set AllowUser or AllowGroup in sshd_config but didn't add your user to those lists. Or, you did not unlock the user with passwd -u. [13:24] at any rate, you should observe what's going on from the server side. [13:24] blackflow: We were able to connect to it yesterday [13:25] VjdfMQ: maybe you think you did. could've reused whatever existing ssh connection you had and didn't really actually TRY the new user with a new connection? === V7 is now known as VjdfMQ [13:25] ssh connections can be cached to speed things up, on client side. some older ansible configs suggested that, for instance. [13:26] Uh ... [13:26] Btw only user and password should match for default configuration? [13:26] VjdfMQ: yes. [13:26] Oh dear [13:26] what? [13:27] VjdfMQ: your user is the only one who can use the .ssh [13:27] if you have a priv key for user, user2 is not able to use it [13:27] User and passwors^ [13:28] unless it's set in user2 authorized_keys as well (you can share same key with multiple users) [13:28] This is strange ... [13:28] VjdfMQ: Just thinking, did you set disabling the root user to connect ? [13:29] you might use this side then [13:29] Root is disabled [13:29] Only one user is permitted [13:29] ipmi / ilo [13:29] Only one key was generated [13:29] Uh ... men [13:30] VjdfMQ: Well last resort is going to the datacenter :D [13:31] Or IPMI bruteforce [13:31] good kuck [13:32] luck [13:32] VjdfMQ: reboot to rescue system? [13:32] surely the hosting company supports PXE booted rescue environments? [13:33] blackflow: without ipmi... will be impossible [13:33] what would? [13:33] Connecting physically on the server and chrooting it to modify the user ? [13:34] boot the machine into a "rescue mode". can it do that? it's usually done with PXE booted "rescue" OS [13:34] blackflow: yes, but he dont have access to the server itself other than ssh [13:35] tchakatak: well yes, ssh is what I mean. depends on how the server is set up. if it's a hosted dedicated machines, almost all hosting companies support booting into a "rescue mode", which has ssh access. [13:36] blackflow: i know. but he is looking for the ipmi password.. so its his own one. [13:36] we're talking past each other [13:36] Hi folks [13:36] IPMI is irrelevant here, if they can PXE boot a rescue OS [13:36] blackflow: then yes [13:37] BluesKaj: Hi o/ [13:37] hi tchakatak [13:38] My computer often shuts down as soon as I open Chromium. Has anyone seen this? How can I troubleshoot this problem? [13:38] hi, is there a way to find out why a package was installed? (what other package required its installation?) [13:38] dserodio: a full power off ? [13:39] tchakatak: yeah. It's a Dell laptop [13:39] MrKeuner: aptitude could do it.. look in the apt manual [13:40] dserodio: look at /var/log/syslog [13:40] Thank you all btw! [13:40] VjdfMQ_: you found ? [13:41] Nope, actually. 'll drive to server locally and recover an access. (it's over 500km over there) [13:43] info nivida-driver-340 [13:43] !info nivida-driver-340 [13:43] Package nivida-driver-340 does not exist in bionic [13:43] !info nivida-340 [13:43] Package nivida-340 does not exist in bionic [13:43] VjdfMQ_: Ok. If you need a hand dont hesitate to comeback when your in the front of your server [13:43] !info nvida-340 [13:43] Package nvida-340 does not exist in bionic [13:43] !info nvidia-340 [13:43] gigirock ? [13:43] nvidia-340 (source: nvidia-graphics-drivers-340): NVIDIA binary driver - version 340.107. In component restricted, is optional. Version 340.107-0ubuntu0.18.04.1 (bionic), package size 28665 kB, installed size 139972 kB (Only available for i386; amd64; armhf) [13:43] VjdfMQ: so you're absolutely sure you can't PXE boot that machine into a rescue OS? [13:45] Only locally [13:45] VjdfMQ: you could ask someone from the DC to do so ? [13:45] blackflow: VjdfMQ_: did you do a port-scan at the host in case there is some other potential method of access? [13:46] Only mail and ssh [13:46] VjdfMQ: inwhich case I'd honestly recommend you to find another hosting solution, one that allows you to fix things remotely and not drive 500 km if you lose ssh access. [13:47] (even colo services can do that) [13:47] what kind of DC doesn't provide remote hands ? [13:48] TJ-: Online, ovh... [13:48] oh doesnt [13:48] sorry [13:48] :D [13:49] TJ-: if you just host it, and its your own server, and you dont configure the ipmi... [13:49] hello guys: so, I was upgrading my VPS to Ubuntu Bionic via a web console provided by the hosting and at some point the console errored. how can I be sure that the upgrade has completely made, without any mistakes? [13:50] run another dist-install ? [13:50] look at the log ? [13:54] tchakatak: right, but even in colo there's usually a remote hands service jsut for this kind of issue [13:55] how do i make alt-tab NOT group multiple shell windows into single? [13:57] TJ-: They provide the network... not the access itself wich is on the server you own.. especially for ilo / ipmi [13:58] tchakatak: I'm aware, I have CoLo, but in every facility there is a remote hands service for just this kind of issue. Usually involves sticking a rescue image USB in [14:00] Depend your agrement with the dc [14:11] how do i remove alt-tab delay? [14:13] what kind of delay ? [14:14] i press alt-tab - it takes some time before i can see the menu [14:14] maybe 200 ms [14:14] could be even more [14:15] i cant find any such delay with dconf-editor :( [14:45] neurre: in gnome? [14:47] yes, in gnome [14:47] neurre: that's just gnome. its UI is javascript based, so yeah, it's visibly slow in reponse [14:48] no, it is not due to js [14:48] they intentionally put a delay there [14:48] * blackflow shrugs [14:48] so if you alt-tab between two apps, you dont get disturbed by the menu [14:49] however this is not my typical use case, i have lots of windows open all the time [14:49] and I typically switch between windows that are not just the last two [14:50] I would have imagined they made this delay configurable somewhere [14:50] and I do remember fixing it in my previous linux desktop [14:50] now I cant find it :/ [15:02] anyone encounter issues with external 4k monitor over HDMI limiting to 30hz via GTX 960m [15:04] babou_tunt, what graphics drivers are you running? [15:05] I was on the default nvidia closed source driver for 18.10 [15:05] but i enabled PPA for beta and let it upgrade to..... 4... something one second === OpenSorc_ is now known as OpenSorceress [15:05] nvidia's website recommends 410.78 for that card [15:06] mmm... i think im typing the command wrong... "ubuntu-drivers devices [15:06] babou_tunt, running 'nvidia-smi' will display it [15:06] ahh thx [15:07] lordcirth 415.18 [15:09] babou_tunt, you did a full-upgrade and reboot when installing it? [15:11] yesssir [15:13] ah.. one second. i forgot i was logged into wayland... [15:13] brb [15:17] ok, I'm back [15:18] Can we desactivate HyperThreating on a linux kernel without touching the bios ? like loading the kernel with a --nohyperthreading ? [15:19] there's probably a kernel setting for that [15:19] My bluetooth doesn't used to work on ubuntu 14.04, I upgraded to ubuntu 16.04, now the devices can connect to my computer via bluetooth but the conneciton is not persistent. Is it a problem with my laptop's bluetooth radio or a problem with OS? [15:20] Platonides: I do think so.. but can't find it [15:21] adroit_machine: If it connect now.. i would look into the OS more than the radio itself [15:21] on https://bbs.archlinux.org/viewtopic.php?id=49933 they suggest ht=on for the opposite [15:21] maybe try ht=off ? [15:21] Il try that :) [15:22] Dont want to flash another bios to disable it so :) [15:22] Thanks Platonides [15:22] on https://serverfault.com/questions/806770/turning-off-hyperthread-by-bios-or-kernel [15:22] they say that you should simply set maxcpus to the number of physical cores [15:23] I would prefer a kernel htoff :) i want to desactivate it for the latest bug with HT... but i cant use the bios for that :) [15:25] thanks [15:27] another question, is there anyway to start openvpn trough a terminal, sending it user/pass, and closing the terminal without loosing the vpn session ? [15:28] (Without NM / NM-applet) [15:28] ahhhh... so i think my nvidia issue is died to... what is that called.. optimum, optimus, prime displays [15:28] tchakatak: You can pass ovpn config to opemvpn on the cli [15:29] sud0x3: i just dont want to store the user/pass in the ovpn file [15:29] Somehow through accidental keyboard input I have disowned two ssh processes, their parent is now PID 1. Does anyone know which keyboard command did this? And is there a way to re-own these ssh sessions to make sure I don't have unsaved work before ending them? [15:29] and i need a challenge from an otp :) [15:30] ca-on-adam: you could try to play with the ssh_agent in /tmp [15:30] tchakatak: if you run openvpn under screen/tmux you can close the terminal and the process won't die. [15:30] ca-on-adam: i dont want to run in under screen/tmux. I would prefer something integrated to ovpn directly. [15:31] ah. [15:31] maybe a composant with connect trough cli to openvpn and just pass the user/pass info [15:31] i will script it if it dont exist... but i would prefer using something built in for it [15:31] tchakatak: If you dont provide password / username you will be asked for it whem running via terminal [15:32] i dont mind sending them via the terminal [15:32] i just dont want to let the terminal open after i sended the login/password [15:32] tchakatak: the easy way is to create client key without any passphrase required, but if you want the hard way, I understand that that 'expect' scripting language can automate keyboard input for interactive CLI programs. [15:32] I don't know how to write expect scripts ... yet. [15:33] just because you need to be root to execute openvpn. Leave it in a screen would be horrible securitywise for this computer. [15:34] or, in a twist of fate, my keyboard accident could solve your open terminal issue? Perhaps?? XD [15:34] if i want to run it as an unprivileged user, i need to create a tun/tap before wich create complication. [15:34] ca-on-adam: ? [15:34] Hello why is no possible install mbed TLS in ubuntu 14.04? [15:35] And if I understand you correctly, the file [15:35] srw------- 1 adam adam 0 Nov 17 20:22 /tmp/ssh-oXvPIMWCtKRr/agent.1853 [15:35] could help me to communicate with the 2 disowned ssh sessions? [15:35] mircx1: because it was called polarssl in Trusty [15:36] well, for now i switched it back to intel low power mode and still stuck at 30hz most likely due to the prime display issue [15:36] ca-on-adam: yep. [15:36] tchakatak: if you disown the openvpn process, that will let you close the terminal without killing openvpn [15:37] not sure. Never used it this way. but you should be able to reown the connexion itself [15:37] And today I'm here because I accidentally have disowned ssh sessions that I didn't want to disown, through hitting some keyboard combo unexpectedly. [15:37] I don't know how to connect the process back to a tty. [15:38] so I can kill it, and hope I don't lose unsaved work. [15:38] but I can't see what it was displaying anymore. [15:38] just trough ssh-add [15:38] dont kill it. [15:39] ssh-add just changes keyfiles. [15:39] you can try to unlock the agent itself. [15:39] look at the man [15:39] it's not related to ssh protocol, it is the general problem of attaching processes to a terminal after they are disowned. [15:40] ca-on-adam: what i am suggesting is reconnecting to your previous ssh connexion trough the ssh.agent still running on your server [15:40] use the gnome equivalent to yakuake, if you're on gnome, you can exit that terminal without losing your vpn connection [15:41] BluesKaj: There is no other way to communicate trough cli with openvpn without that ? [15:41] hmm, okay, so start a new ssh session, log in to the server, then try to connect to sshd somehow? [15:43] I can see how that could work on the server side. If sshd owns a nano process, I could theoretically re-own nano and save and quit, and voila. [15:43] then I could kill the ssh sessions on my client side knowing I won't lose anything. === pavlushka_ is now known as pavlushka [15:43] ca-on-adam: yep. You should not even have to resupply a key or a password to 'steal' it. [15:44] you can also look at forward_agent [15:44] thanks [15:44] into sshd [15:44] (and desactivate it next time... :D) [15:44] Thats a really really bad things. [15:45] yes there is, tchakatak one can add the openvpnauto file to /etc/init.d/ and the vpn auto connects at boot [15:45] anyone know why ubuntu one in the application store would logout everytime you reboot? [15:45] I still need to figure out what the heck caused the disowning in the first place. I have a "Microsoft Multimedia" keyboard with special buttons in a column on the left and right edges, maybe one of those keys causes the right character code that disowns processes in the terminal. [15:46] BluesKaj: Dont want a auto lunch it but i take a look at openvpnauto [15:46] babou_tunt: your password is not set in the keyring ? [15:46] Hi, is there a way to make 'always on visible workspace' permanent [15:46] dhruvasagar: gnome-tweaks [15:46] tchakatak: I have that installed, where in it [15:46] babou_tunt: that seems like the default behavious, admittedly I haven't used the application store in a long time. [15:47] that might be it [15:47] babou_tunt: maybe there is a "stay logged-in" option that needs to be enabled. [15:47] ca-on-adam: there is a xorg command to see what your imput does in real time... [15:48] tchakatak: I have used compiz in the past (long ago) to achieve this, I don't believe gnome-tweaks has the options to do this, I am on ubuntu 18.10, what are my options ? [15:48] Actually I am pissed-off that Discord logs in forever, and I want it to forget the password on reboot! The only way is to click 'log out' before powering off or rebooting. So insecure... [15:49] dhruvasagar: i dont run 18.10... cant help you with it exactly. but to modify theses options, i would look into d-conf [15:49] ca-on-adam: stop using gnome keyring [15:49] tchakatak, you'll need the openvpnauto script from your vpn provider [15:49] BluesKaj: i am the provider :) [15:50] ahh [15:50] :D [15:50] tchakatak: ok, thanks for your help, i'll see if anybody else has better knowledge of this [15:50] dhruvasagar: lookup online, pretty sure google can help you with that [15:50] is compiz still a thing for gnome3 on ubuntu 18.10 ? [15:50] I think that Discord is doing this internally to itself, and the support staff admit that they did not add an option to forget the password when the application closes. [15:50] Can I specify a number of days for logrotate delaycompress? I'm not seeing it in the man pages. [15:51] tchakatak: actually it hasn't helped much, i've been looking for long, the only thing that comes close are all related to compiz, which I am already aware of but not sure if compiz is still something I should rely on [15:51] dhruvasagar: if you run a stock 18.10 i assume you use gnome/wayland [15:51] tchakatak: thanks for the tip on xorg keyboard input real-time info [15:52] ca-on-adam: noprob iam looking for the command [15:52] tchakatak: it is freshly installed, I believe it is gnome3 [15:52] I'll figure that out and press all the multimedia keys on my keyboard until an unimportant process gets disowned. [15:53] ca-on-adam: not really a good idea :) but up to you [15:53] I will make sure that nothing important is running on my PC when I do that experiment. [15:53] anyway you can usean xk app to desactivate your keyboard or rebind them [15:54] If I can learn to 'disable' that key somehow, that would be enough to prevent this mistake from happening again. [15:54] dhruvasagar: so yep gnome3/wayland by default [15:54] ca-on-adam: how do you start your session ? .xinitrc ? gdm ? [15:55] The thing that really confused me is there is no visual difference in the terminal between a network failure and a disowned ssh process. It just appears to be a dropped network connection. [15:55] ca-on-adam: look at xkbcomp [15:55] Only when I realized that I had myself logged-in 3 times instead of once did I realize there were two disowned processes that didn't die. [15:55] ca-on-adam: another thing [15:55] i dont know on what your working at [15:56] but if your on emacs files or so, you can steal the buffer of the running app. [15:56] hmm ok. No it was just nano so not much extra features. [15:57] mhh... maybe you can have access to nano's buffer [15:57] If I wasn't lazy I would have started nano under screen like I usually do, but as luck would have it, or as Murphy's law would confirm, the oops had to happen when I was unprepared! [15:57] use emacs-nox :0 [15:57] or vim [15:58] vi/vim is on my long list of to-do. [15:59] :) good choice. [15:59] ca-on-adam: i just looked up [15:59] Until then I just use a script to flock nano so at least I won't clobber other people's work when doing collaborative projects. [15:59] on the serverside try to look at /tmp [15:59] you might have the nano buffer there [16:00] nope, my nano has put nothing in /tmp [16:00] ca-on-adam: sounds like you'd benefit from setting something like tmux as the login shell so it's all automatic [16:01] but as you said, using vim instead would be a good choice instead of compensating for nano's intended simplicity. [16:01] or emacs. [16:01] buffer file / backup file by default [16:01] more simple than vim [16:01] (no double modes) [16:02] Okay, I'm going afk now, I have taken notes on your recommendations into a text file, can't promise I'll try these advices in 2018 but eventually, definitely. Thanks a bunch! [16:03] ca-on-adam: running and playing on emacs is like a 10 min tutorial :) [16:03] to have the same features than nano [16:04] vvcjklbnbrrifkfrncneddtbhgngekgnjelvvheugcku [16:04] fu*** cat [16:04] hidden verses after Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn. [16:05] Lovecraft 3 [16:16] I upgraded from 16.04 to 18.04 and I completely messed up my python3 + pip installation which I think I manually upgraded outside of the package manager. is there a good wiki page on how to manage python3 alongside python2 + possible upgrades properly? [16:17] Hello, I'm using c9 ide with ubuntu installed to learn C programming am at the moment trying to get Makefile to work. Problem is that my *.o keep getting recompiled even though I have it set to only compile files that got changed. I tried to make -d to get debug info and 'make' seems to have an option somewhere that it forces to recompile everything. "Making `test.o' due to always-make flag.". I am new to linux and really dont ahve [16:19] I looked in a file called bashrc and looked for environment options with printenv but can't find anything. Anyone have a clue where such options can be found/edited? [16:20] biet1234, is your ide calling 'make' with --always-make, or -B? [16:20] biet1234: show us the Makefile in a pastebin [16:20] https://pastebin.com/M7DTqctW [16:21] Seems fine. I bet it's the IDE doing it [16:21] when I 'make' it just says - gcc -O -c test.c - gcc -O -c link.c [16:21] aww [16:23] biet1234: change "test.o: test.c" to "test.o:" and same for "link.o:" [16:23] biet1234: also, you should declare "clean" as a ".PHONY: clean" [16:24] biet1234: that's a separate, new line, that tells make that the 'clean' target doesn't produce anything [16:24] thanks, will try it now [16:24] biet1234: but it is likely, as lordcirth said, that the IDE is forcing a rebuild with "-r" option [16:25] biet1234: might be useful to capture the command-line the IDE is issuing to make to be sure [16:25] yeah, no luck. if it helps, the ide seems to use GNU bash, version 4.3.11(1)-release (x86_64-pc-linux-gnu) [16:27] biet1234: e.g. create /usr/local/bin/make with "#!/bin/sh" and "echo $@ > /tmp/make.log" and "/usr/bin/make $@", save, then "chmod +x /usr/local/bin/make" then use the IDE to build then look at /tmp/make.log. Also, remove the script with "rm /usr/local/bin/make" [16:28] biet1234: if the ide uses PATH to find make the above interceptor will work; but if IDE uses the exact path to make it will not [16:33] do I just type "/usr/bin/make $@" ? [16:34] i am getting a make: *** No targets specified and no makefile found. Stop. when I try that [16:36] which jabber XMPP client do y'all reccomend to use with cisco jabber [16:36] biet1234: no, those commands would go in the temporary /usr/local/bin/make as I mentioned [16:36] Oh i think i read your explanation incorrectly, im supposed to make a file make and put those command in it right [16:37] sorry terribly new to linux and terminals [16:37] biet1234: yes :) it's a temporary interceptor that should be be found before the real make, so you can capture the arguments sent to it [16:38] biet1234: if you do "echo $PATH" you should see /usr/local/bin is listed before /usr/bin so if a file of the same name is found it gets executed [16:40] I navigated to /usr/local/bin/ and when i try "touch make" or "> make" Im getting permission errors [16:40] I guess the sytem doesnt let me allow to do anything in these folders? [16:41] touch: cannot touch ‘make’: Permission denied [16:41] sudo? [16:42] !! thanks [16:42] You're welcome! But keep in mind I'm just a bot ;-) [16:42] by default /usr/local/bin is not writable by a user [16:42] you shouldnt make it directly in /usr/local/bin [16:44] biet1234: you can modify $PATH var env for this [16:44] biet1234: use sudo [16:44] tchakatak: that will not work [16:45] TJ-: he is trying to make a file in /usr/local/bin. [16:45] tchakatak: the IDE will have a separate env [16:45] tchakatak: yes, for a 20 second test after which it gets deleted [16:45] TJ-: you can always export the path of the env from the ide no ? [16:46] Is it not better to add the path to a dev env ? [16:46] and then execute everything ? [16:46] tchakatak: this way is much easier, drop something in /usr/local/bin/ then remove it [16:46] ok. :) [16:46] tchakatak: I do it this way because usually I'm trying to intercept something an already active process is doing [16:47] ok i think i got it copied into that folder and did sudo chmod +x /usr/local/bin/make [16:47] now do i just make like i normally do, right? [16:47] and then check that tmp file? [16:47] tchakatak: as I said, this doesn't help if that process calls the target binary with absolute path. Then I use dpkg-divert :) [16:47] biet1234: use the IDE to build the program [16:48] biet1234: if you're in luck, it'll end up calling your make script and we'll capture the arguments [16:48] Yeah I just type make when in my c program folder on the terminal [16:48] biet1234: OK, do it whichever way usually causes the rebuild [16:53] nothing seems to get written to make.log as it is 1 byte and the date doesnt change hmm [16:54] just to be sure: https://pastebin.com/7Fw9D7CT this should be in the make file, right? [16:55] biet1234: what are you trying to do exactly [16:55] biet1234: correct, did you "sudo chmod +x /usr/local/bin/make" ? [16:55] tchakatak: the shell-based IDE is causing 'make' to always rebuild all targets [16:55] Ok :) [16:55] hi [16:55] tchakatak: so we're trying to see what arguments the IDE is sending to make [16:55] I can't open nautilus! [16:56] after upgrading to 18.4 [16:56] sentiment: that's to prevent it sinking :) [16:56] it writes this error to console gconf-editor === [1]MrMobius is now known as MrMobius [16:56] oops [16:56] g_type_add_interface_dynamic: assertion 'G_TYPE_IS_INSTANTIATABLE [16:56] yeah i did sudo chmod +x /usr/local/bin/make and then compiled with 'make' the usual way [16:57] TJ-: what? === deathonater is now known as Smeef [16:57] biet1234: so, the IDE must be calling /usr/bin/make and not relying on the PATH [16:57] TJ-: not everyone has read Verne it seems :) [16:58] biet1234: lets' do something a bit more invasive then: "sudo mv /usr/bin/make /usr/bin/make.real" then "sudo mv /usr/local/bin/make /usr/bin" - this puts your script where the real make used to me, then rerun the IDE build :) [16:58] tchakatak: whenever i invoke make it seems to use --always-make option even tho i am not specifying it [16:58] blackflow: apparently not :) [16:58] ohh i like invasive [16:58] kids these days. [16:58] sorry im slow as a turtle. i appreciate your help [16:58] man this upgrade completely fucked my system up [16:59] everythign botched [16:59] just when I thought it's over here is another one [16:59] there's another one [16:59] goddammit [17:00] GLib-GObject-CRITICAL **: 20:27:55.160: g_type_add_interface_dynamic: assertion 'G_TYPE_IS_INSTANTIATABLE (instance_type)' failed [17:00] I mean WTF... [17:00] am I supposed to go and read the source code?! [17:00] for gods sake [17:00] sentiment: have you done "apt -f install" to ensure all packages are fully upgraded? [17:01] I swear I did that a hundred times already, but I'll try again [17:01] sentiment: there was a reason I told you to back up data and reinstall Bionic. you could've been done several times by now :) [17:01] TJ-: no problems [17:01] blackflow: I dunno man :( [17:01] it said "/usr/bin/make: 3: /usr/bin/make: Cannot fork" on the console and when i now check the make.log file it has the -B option for make! [17:02] sentiment: you dont know how to backup your data ? [17:02] sentiment: hmmm [17:02] no, I am not sure if that is the best way [17:02] in the linux world [17:02] biet1234: well that is something [17:02] in Windows world everyone does that [17:02] sentiment: I've been using and installing linux distros for over 10 years. I've built LFS even. I've never been able to recover from ubuntu botched upgrade. ever. reinstalling from scratch has always been the best solution, if upgrade fails. [17:02] virtual machines or (insert best opensource container solution for ubuntu here) [17:03] blackflow: you can, but it's painful [17:03] you are a genius sir/mam. what does this mean? :-) [17:03] biet1234: so now you've confirmed it, remove your script and put things back: "sudo rm /usr/bin/make; sudo mv /usr/bin/make{.real,}" [17:03] blackflow: I don't have an image for Bionic though [17:03] sentiment: you can download it easily [17:03] I have the 16.4 CD thugh [17:03] biet1234: see "man make" and you'll see "-B --always-make" are the same thing [17:03] no not easily, it wastes my traffic, sorta [17:04] do you have network enough to download 18.04 and install it ? [17:04] I have a traffic cap [17:04] biet1234: so you do need to persuade your IDE not to pass -B [17:04] oh.. [17:04] sentiment: then install 16.04 and straight upgrade to 18.04, with nothing else added (especially not any PPAs) [17:04] but I have enough I think for that, it's just that I'd rather go back to 16.4 if push comes to the shove [17:05] blackflow: How likely is that to succeed? [17:05] sentiment: it'd be easier to track down the error; search for bug reports on bugs.launchpad.net for example, or your favourite search engine [17:05] sentiment: also consider that upgrading does not give you full Bionic experience. some things are not done for upgrades. so upgraded bionic != installed from scratch. for that reason I'd always recommend install from scratch. [17:05] oh [17:06] I might buy the cd from a retailer then [17:06] anyway downloading 18.04 or 16.04 and ugrading will be almost the same [17:06] sentiment: my initial research suggests this is due to bad shell extensions [17:06] network wise [17:06] tchakatak: true that [17:06] TJ-: yes, I went that far [17:07] sentiment: here's an example; maybe you can figure out how to disable extensions? https://github.com/andyholmes/gnome-shell-extension-gsconnect/issues/293 [17:07] I really shouldn't have taken the plunge [17:07] So silly of me [17:07] I was reeaaly hesitant to upgrade exactly because of this kind of issues [17:08] sentiment: how is your system at the moment. what is working and what is not ? [17:08] TJ-: ok, thx, gonna give it a shot [17:08] sentiment: search the file-system for suspects with something like "dpkg -S nautilus-python" [17:08] oh ok [17:09] sentiment: that'll clue you in on what (python) extensions are installed... maybe there's one leftover from 16.04 that isn't in 18.04 and wasn't removed [17:09] tchakatak: well the window issues were solved at last [17:09] the title bars and stuff [17:09] it's running stable enough in wayland [17:09] except that nautilus thing and rhythmbox problem with mp3 even though I have all the gstreamer plugins installed [17:10] all the bad and good and ugly that is [17:10] TJ-: ok thanks, let me see... [17:10] you could try to purge completly nautilus and reinstall it [17:10] hmm [17:11] I'd think there'd be greater success removing the extensions :) [17:11] TJ-: by purging the package, you dont remove his extensions ? [17:11] BTW I stil have some package files list warnings in dpkg [17:11] though alot less than before [17:11] did anyone suggest running a loop for apt install --reinstall, for each of the packages from the installed database? and perhaps removal of packages that are still listed from old repos, if they fail to reinstall [17:12] tchakatak: I doubt it, unless also doing "apt autoremove" - dependencies aren't removed [17:12] warning: files list file for package 'libxtables11:amd64' missing; assuming package has no files currently installed [17:12] for instance ^ [17:12] sentiment: OK, so those do need solving [17:13] I tried to reinstall it but it said can't be done [17:13] let me try again [17:13] Reinstallation of libxtables11 is not possible, it cannot be downloaded. [17:13] ^ for all the other such warnings [17:14] TJ-: And using aptitude to resolve the problem ? there was a module for it if i remember well [17:15] sentiment: libxtables11 was updated to libxtables12 in 18.04, but on my 18.04 libxtables11 is also still installed but doesn't cause such errors [17:15] I searched for Reinstallation of libxtables11 is not possible, it cannot be downloaded. [17:15] oh sorry [17:15] I tried sudo dpkg --list *nautilus* and it returned just 6 results [17:16] libnautilus-ex ,nautilus, nautilus-data, nautilus-exten, nautilus sendt, nautilus-share [17:16] TJ-: so it's irrelevant [17:17] I suspect those warnings are irrelevant to this case [17:17] btw, how do I make alt+tab between windows of same application easily switch between them? [17:17] as it is, it switches between different applications only [17:17] and this is really annoying [17:18] sentiment: alternatetab extension, iirc [17:21] removing every extension listed now [17:23] I removed libnautilus-extension1a and it removed ubuntu-desktop too! [17:23] does it matter? [17:25] https://pastebin.com/5HaNKhye why I can't ping 8.8.8.8 with my wireless usb (enx0cb6d2d1a525) ? [17:25] I followed these instructions https://shapeshed.com/linux-wifi/ [17:26] TJ-: thanks for your help. i tried looking around in the IDE but cant find an option to disable it :-( [17:27] at least I learned a few more linux commands :> [17:29] hey everyone [17:29] hello [17:30] o/ EriC^^ [17:31] goodafternoon: route ? [17:32] goodafternoon: you sure you set up th4e device properly? `ip route list` please, also confirmi that you have the proper IP range set on the device. [17:33] you might also need to share the configurations you've set, because it sounds like your network and the network that the wifi works on are different [17:33] (your network meaning the settings you've set) [17:35] https://pastebin.com/FmfuTNAm enp4s3 works properly btw [17:35] teward [17:37] i create use but seems be problem after login to user. PS1 not executed and .bashrc not loaded. [17:38] goodafternoon: the next question is why are you connected to the same network twice? [17:38] because it sounds like you've got IP route conflicts because both are on the same IP subnet [17:38] use one or the other, not both. [17:38] https://paste.ubuntu.com/p/PvgwwgGmBp/ [17:38] the use acctually created by ansible [17:39] tst [17:39] > /etc/network/interface https://pastebin.com/8VY8jnnT [17:39] teward because I want my device to connect with wifi instead of ethernet [17:40] and I use ethernet to connect to it [17:40] goodafternoon: what device ? the usb one ? [17:40] goodafternoon: so unplug your ethernet. [17:40] then test. [17:40] the problem I see is you have two routes going on here [17:40] alright [17:40] and there's a network route conflict [17:42] doesn't work, should I down enp4s3 ? [17:43] goodafternoon: or reboot. the problem is you've got two devices trying to use the same network, over the same route over different interfaces (Which doesn't work) [17:43] Network unreachable now [17:43] you might want to reboot without the LAN connected [17:43] ok I do that [17:43] when you start to do stuff like this with both wifi and ethernet on the same network segment it can cause issues [17:45] did you check my /etc/network/interfaces ? does it look ok for you ? [17:45] https://pastebin.com/8VY8jnnT [17:45] (reboot in progress, it takes minutes lol) [17:53] oh yes it works ! [17:53] thank you [17:56] but now my reverse ssh doesn't work : Warning: remote port forwarding failed for listen port 22222 [17:57] What is the difference between wireless and ethernet ? I thought it wouldn't change anything [18:04] powertop is listing docker as top power consumption. i removed docker and restarted, but it is still there. there are no docker services running. how can i completely remove docker? [18:11] I am running caddy on ubuntu 14.04. So I get why my service crashed, too many attempts, but now how do I start it again? "Job for caddy.service failed. See "systemctl status caddy.service" and "journalctl -xe" for details." [18:11] uh [18:12] 14.04 didn't have systemd [18:16] oops, 16.04 sorry, was mistaken it with another vm [18:16] nomoney4u: caddy does not appear to be an ubuntu package [18:18] nomoney4u: the answer is "Read the logs output from those commands" [18:18] and then fix it [18:18] but that doesn't seem to be a program within the Ubuntu repositories [18:25] teward: I'm not sure I understand. So the command `sudo service caddy start` is reading a file /lib/systemd/system/caddy.service and then throwing the errors in `journalctl -xe` ? [18:26] and possibly in `systemctl status caddy.service` [18:26] but yes, error messages and such get dumped to the journal [18:26] which journalctl and such can access. [18:26] but again, caddy isn't in Ubuntu, so you might be forced to find external support from Caddy's devs to solve any issues you find [18:30] teward: I see, so in general, the .service file would be the source that would tell systemd to stop the server if something was to go wrong? [18:30] you should really read up on how systemd works to understand this better. https://www.digitalocean.com/community/tutorials/understanding-systemd-units-and-unit-files [18:31] will do, thanks [18:36] As far as I can tell, gdm3 is completely ignoring /etc/gdm3/custom.conf. What am I doing wrong? [18:40] It prints the contents to the log, but it neither obeys it nor throws a syntax error [18:54] lordcirth: you're sure you restarted gdm3 properly? [18:54] nacc, I rebooted just in case [19:02] is there a nice way to ban these people? I see a lot of sshd log: `Invalid user ___ from _____ ` is there an "autoban" or something of that nature for these attempts? [19:02] nomoney4u: properly configured fail2ban [19:02] lordcirth, what part of custom.conf did you change ? [19:03] nomoney4u: disable password authentication and use only SSH key auth and it'll help a bit more with security. [19:03] you'll still see attempts but they won't be able to successfully login [19:03] teward: first thing I did was do that, it's just annoying to see it flood the journalctl [19:03] ioria, http://paste.ubuntu.com/p/ht9kn8YHHv/ [19:03] I'll look into fail2ban [19:04] lordcirth, automatic login does not work ìcause a bug [19:04] 'cause [19:04] ioria, I suspected so. So far it seems to have taken less time to switch to lightdm [19:04] https://bugs.launchpad.net/ubuntu/+source/gdm3/+bug/1719128 [19:04] Launchpad bug 1719128 in gdm3 (Ubuntu) "Automatic login works only for wayland session" [Undecided,Confirmed] [19:04] lordcirth, you can try lightdm, yes [19:06] It works perfectly. Just testing that I put it in Salt correctly. [19:06] ok [19:19] Couldn't get size: 0x800000000000000e [19:20] Couldn't parse context === [1]MrMobius is now known as MrMobius === daniel is now known as Guest67454 [19:22] clovermint: we'd need more context - where are you encountering this, what were you doing that triggered the error, etc. === [1]MrMobius is now known as MrMobius [19:26] clovermint: It's a UEFI SecureBoot message caused when there's no MOK certificate db [19:30] anyone use published apps through x2go? it works so great [19:37] * clovermint [19:40] Much quieter :) [19:43] lordcirth: I setup lightdm with salt for a thinsation like setup, http://dpaste.com/0NTX0CF [19:45] sud0x3, thanks. I don't really want it to do anything except autologin. Testing it now [19:48] lordcirth: That was all i need from lightdm also gdm needed too many dependencies and i chose to use openbox as i could easily lock it down. If you want aa look at any of the contents of lightdm config just let me know. [19:55] NVIDIA, No CUDA 10 has no 32 bit libs. rhe .run does does mess with the .deb in ways that will break your system. Blaklisting the Noveaux helps. Steam should be 64 bit by now. bitch, bich.. Ie I don't use steam anymore. [19:56] I have a Titan V and i need the newest updates. If you don't 'ubuntu-drivers autoinstall' should works fine. [19:56] does [20:17] Anyhow the focus now is on varicad. To my mind the only CAD system that cut's the mustard for mmechanical engeneering. The free versions seem like glorified drawing programs. They might cut it for defining for a 3D printer (additive modeling) but seem to lack libs for DIN parts and the like. [20:17] I have about 20 Desktop machines I need to image. I was thinking of using MAAS and then Salt or Ansible to configure and bootstrap the machine. How would I go about automatically adding the keys and stuff though? [20:18] It sells for 1000 EUR which is pretty cheap for a CAD (CADD) [20:19] kur1j, Neither Ansible nor salt can make images for bottstrpping the machine. You need something like werewolf. [20:19] kur1j, I have a PXE server than installs Salt in the d-i preseed. I accept the keys on the salt-master when they finish booting, and state.apply. More automation is possible, but I didn't find it worthwhile [20:20] kur1j, alternatively, you could use Packer's salt-masterless provisioner to build an image, then push it out with clonezilla or something [20:20] lordcirth: I didn't want to try and maintain an image [20:20] Looks like packer only does VM images anyway [20:20] PXE booting is done using a dedicated ip address. It is DHCP that looks at this address and uses TFTP to install the image. [20:20] Withe the correct configuring. [20:21] Yeah, I use dnsmasq to handle all that [20:22] I am a member of the beawolf newsgroup. there is also a mag called HPC (High performance cluster) which has a newsletter. I suggest you follow these to get a better idea of what others are doing. [20:22] lordcirth: what is "d-i"? [20:22] debian-installer, sorry [20:23] You can give it a preseed file which answers all the questions it normally asks, and it automates it. MaaS does a lot of this for you, might be a good option [20:24] lordcirth: I'm going to go ahead and setup MAAS and play with it and see what options it provides [20:24] i think that might provide the answers I need [20:24] kur1j, let me know how that goes, I look at MaaS ages ago and didn't end up using it, but it's probably improved [20:24] For 20 machines saltstack and ansible perform about equal. However satstack scales better. [20:24] I just need it to PXE the machine and then allow me to run configuration on the system [20:25] installing python, java, development tools, etc [20:25] I'll take a look at Salt, I've just use ansible in the past [20:26] how difficult is it to deal with updates and stuff for machines with Salt? [20:26] for example, say I want to run an update to Ubuntu 16.04.4 to 16.04.5, can you "manage" those 20 systems and have it do that upgrade [20:26] Is it a heterogeneous environment or are all your nodes configured the same to do MPI or such? [20:27] kur1j, salt 'minions' pkg.upgrade dist_upgrade=True [20:29] Where 'minions' = a pattern that you want to target. possibly '*' for all machines [20:30] Yes, but I am takling about 'plug and play'. Yo just connect a new computer to the system and it automatically installs the OS. Not the BOX OS but a image customized by you to contain the tools you want. [20:32] Younder, MaaS will do that, I think [20:33] kur1j, #salt channel exists here too, I'm also on there [20:35] lordcirth: thanks. is salt all cli, programming based like Ansible or does it use some type of UI? [20:35] kur1j, there's an enterprise webui, which I've never seen a need for. I use it purely cli. It's pretty easy to learn [20:36] I just have a toy system here consisting of 20 PI's 10 PI 2 and 10 PI 3 B. they are controlled by a dell server. A Poweredge T110 II. THey run the usuual tuff. MDI cerberus, ldap, NFS, apt-cache, clock, slurm etc [20:36] Younder, fun [20:37] hi, i have ubuntu installed on an embedded system (like a raspberry pi but not an RPi specifically). unfortunately, i set up the user account a while ago and do not remember the password. i'm logged in because apparently i did not specify the system to require a password to log in. is there a way to recover / reset the account password without reinstalling? i tried holding shift during boot to get into grub but that didnt work so i'm guessing [20:37] there is no grub? [20:38] tnecniv: sudo su then passwd [20:38] leftyfb doesn't work. sudo requires a password. [20:39] it appears the account does have a password, i just have no clue what it is [20:39] I'm on mint(which has the same network manager as ubuntu) and have a small networking issue. I would like to set it so that both connections can be used but when a program doesn't care it uses the wifi interface rather than the ethernet. For example curl google.com would use my wifi interface and curl --interface eth0 google.com would use my ethernet interface [20:39] You can boot a live OS and chroot [20:39] tnecniv: this is Ubuntu running on the pi or Raspbian? [20:40] leftyfb ubuntu 16.04. the specific microcontroller i'm using is an up-board [20:40] tnecniv, so it's install on an SD card? [20:40] lordcirth: it's an image [20:40] no grub [20:41] yeah there's no grub when i boot. it's on whatever internal storage the board has [20:41] like, a dd'd iso? [20:41] lordcirth: dd of a disk image [20:41] i installed it from a live USB [20:41] That's how SBC's work [20:41] wait, really? [20:42] i could just reinstall it but i feel like there should be a simpler way to fix this [20:42] tnecniv: track down which boot loader it's using and edit the kernel line an add in init=/bin/sh [20:43] tnecniv: out of band [20:43] also, don't forget system passwords [20:49] Hey all [20:49] Is it possible to get an access to server via IPMI and user '' with password 'admin'? [20:49] V7, IPMI support seems offtopic here, unless I misunderstand your question [20:50] V7: I don't think the IPMI spec allows authentication without a user [20:50] leftyfb will do, thanks [20:50] lordcirth: I've been here for a while today and understood that public key for SSH on ubuntu is somehow wrong [20:50] Even if it was this would be a massive security breach to reveal it here. [20:51] So, there's IPMI access, but we can't find credentials to it [20:51] V7, using pubkeys for ssh is generally good - what's wrong with it? [20:51] Or do you mean you don't have the right key? [20:51] Don't know, yesterday all was working, but today it just says: Permission denied(publickey) [20:52] V7, try with -vvv [20:52] lordcirth: Already [20:52] V7: I would get local access to find out what's going on. [20:52] V7, so, did someone do an update, or something? [20:52] leftyfb: Server is over 500km from my spot [20:52] Yoes for a PKI infrastructure you need a public PKI server. If you use a locally generated key the browser will complain that that it is not a secure key. I suggest ignoring it. [20:53] yes [20:53] V7: ok, then if you don't have a plan to get in OOB, then you failed to plan appropriately. [20:53] V7, who is hosting the server for you? [20:54] V7 and VjdfMQ are friends [20:54] Yup we are [20:54] So, btw thank you: https://hastebin.com/vefeboquma.sql http://termbin.com/3gcn [20:54] These are permissions and -vvv output [20:55] Again cerberus and PAM is a pin to set up but wit will work. If you want to bigger Globus. Look up a book called 'PKI infrastructure' at amazon [20:55] lordcirth: This server is ours, no datacenter is used [20:55] pain [20:55] Younder: none of this helps their situation [20:55] VjdfMQ, so you have someone in the building, then? [20:55] Yup, far 500km from here [20:56] VjdfMQ: the permissions for the provate keys are wrong [20:56] Which should be? [20:56] that'll do it :) [20:56] they are 644, should be 600 [20:57] the public keys can be 644, no biggie there. But the private cannot [20:57] hggdh, ssh isn't complaining in -vvv output, though? And the dir is 700 [20:57] Now they are -rw------- 1 user user 3326 Nov 22 13:04 id_rsa [20:57] But this output is given with sudo [20:57] If without sudo it gives: ls: cannot access '.ssh/id_rsa': Permission denied [20:57] -????????? ? ? ? ? ? id_rsa [20:58] the client logs show that it sends the key properly [20:58] Gives Failed to add the host to the list of known hosts (/home/v7/.ssh/known_hosts). [20:58] lordcirth: sort of. ssh stated it cannot find the keys [20:58] debug2: key: /home/user/.ssh/id_rsa (0x17xba716b95) [20:58] debug1: Offering public key: RSA SHA256:i18HW71hWDicayDJWDi19wjDYDmcnxzMVkvnIWDD83o /home/user/.ssh/id_rsa [20:59] Its found the key and is trying it [20:59] And its denied. You need to look at sshd logs on the server. [20:59] Oh, sorry Failed to add the host to the list of known hosts (/home/user/.ssh/known_hosts). [20:59] lordcirth: We can't [20:59] No other way to get an remote access to it [21:00] VjdfMQ, I thought you said you have someone there? [21:00] 500km [21:00] Oh, you mean a guy [21:00] Nope, noone there, me nor V7 [21:00] Well, you dun goofed [21:00] Look sudo can look at everything, you have to give it read access first with chmod +r [21:01] Younder: fixed [21:02] VjdfMQ, I do notice that the client keys were edited on Mov 22 [21:02] November* [21:02] Yup, they were modified about week ago or so [21:02] That seems like a red flag [21:02] Do you know what the change was? Have backups? [21:03] Yesterday we were able to connect, but today - nope [21:03] ^ [21:03] logithack: the .ssh directory should br 700, the pub keys 644, and the private keys 600 [21:03] lordcirth: ^ logithack -- sorry [21:04] hggdh: fixed permissions. The same. [21:04] VjdfMQ, what are the OS versions on the client and server? [21:04] VjdfMQ: also, please show another -vvv in pastebin [21:04] VjdfMQ: why are you using sudo to do anything with ssh? [21:04] leftyfb: What do you mean? [21:04] lordcirth: Client KDE Neon [21:05] Server is Ubuntu 16.04 [21:05] VjdfMQ: If without sudo it gives: ls: cannot access '.ssh/id_rsa': Permission denied [21:05] VjdfMQ, kde neon is a Ubuntu respin. What version? [21:06] VjdfMQ, what user's homedir is .ssh in? [21:06] What usually messes me up with ssh is login in to another machine with a different user that I created the key for. NEVER allow root access. [21:07] Younder: keys do not care about users [21:07] leftyfb, I think he means trying to login as the wrong user [21:07] THey so do [21:07] not the client side user [21:07] yes [21:08] Younder: ssh keys do not care about users. It only matters which user's directory on the remote system the public key resides in [21:08] whats the difference in the "ubuntu-18.04.1-live-server-amd64.iso" image and the "ubuntu-18.04.1.0-live-server-amd64.iso"? [21:09] Younder: you can use and keypair for any user on any system, regardless of which user you create the key for/as, as long as you copy the key files to the appropriate directories [21:09] Is there a list of the most up to date version of the openssl FIPS packages? I know they're only available for download by Advantage Advanced subscribers, but I need to verify that the versions are keeping pace with non-fips versions. [21:09] lordcirth: KDE Neon 5.14 [21:09] Sort of, but that is resticted by the access restrictions of a user. [21:09] Also, this is new -vvv [21:09] Example: https://launchpad.net/ubuntu/xenial/+source/openssl This is the openssl version listing for 16.04. I need to know the fips versions available. [21:09] https://hastebin.com/veqofujidu.makefile [21:09] lordcirth: /home/user/.ssh [21:09] hggdh: ^ [21:10] VjdfMQ, ok, so why is sudo involved here? [21:10] I don't know [21:10] What do you mean? [21:11] if you are logged in as 'user', and the keys are in 'user''s homedir, and owned by 'user', then you shouldn't need sudo [21:11] lordcirth: habit [21:12] Bad habit. Don't use sudo unless you need it and know why [21:12] It causes problems [21:12] You're 100% right [21:12] BAD habit [21:12] valuable lesson [21:12] Because there's no FIPS-specific security advisories, I was assuming that the FIPS versions would keep pace with the non-FIPS versions. But someone's trying to tell me that 1.0.2g-1ubuntu4.fips.4.6.3 is the latest FIPS version and honestly I don't believe them. [21:12] Btw, we have an access to IPMI on 623 port [21:12] Also we know login '' and password 'admin' [21:13] SKaplan: try #ubuntu-security [21:13] VjdfMQ, so can you get a terminal? [21:13] But IPMI tells "insufficient privileges" [21:13] Oh thanks. I hadn't seen that one on the list. [21:13] lordcirth: Nope, without correct password for ADMIN user [21:13] VjdfMQ: you need a user [21:14] Seems to me the blaze attitude twoards security is what caused this problem. Don't just copy a key from one user to another create a new key. [21:14] We just can't find it on out credentials server. Someone just forgot to copy them to notes [21:14] VjdfMQ: your server is offering s DSA key. You do not have a DSA key === jeffreylevesque_ is now known as jeffreylevesque [21:14] What do you mean [21:14] Is it possible to create DSA key using RSA? [21:15] as far as I know, no [21:15] hggdh, what line are you referring to? [21:15] And no [21:15] DSA and RSA are created with the same tool genkey [21:15] lordcirth: line 71 -- debug1: Server host key: ecdsa-sha2-nistp256 SHA256:Aj1Cwu219ajDHawju18wdjAWDu281ADkvoiNJDu28HJ [21:15] hggdh, that's ecdsa, not dsa [21:15] Also it's being accepted [21:16] OK [21:16] It is probably some simple problem - on the server end [21:16] VjdfMQ: why can't you get someone local to the server to help you out? [21:16] someone removed the key, or mangled the permissions, or broke /etc/sshd_config [21:16] leftyfb: Only tomorrow [21:17] We though that we could recover it tonight [21:17] VjdfMQ, wait till tomorrow, and come here for help if you have trouble once you have access === capella is now known as capella|away [21:17] So, we don't have any idea how to get an access? [21:18] w/o local access [21:18] VjdfMQ: we do, but you didn't manage things properly so we cannot help you in doing so [21:18] We have RSA key and IPMI running on 623 [21:18] VjdfMQ: If you have the credentials for IPMI, then you can use that [21:19] Here we have another problem. Only one user with name '' and password 'admin', but IPMI says 'insufficient privileges' [21:19] VjdfMQ: you do not have the proper credentials for ipmi [21:20] Seems so [21:20] VjdfMQ: That is a problem of mismanagement of credentials and beyond the scope of this support channel since it has nothing to do with OS [21:20] Yup. [21:20] We'll talk with manager tomorrow [21:20] VjdfMQ, SSH is quite good at keeping out people who aren't allowed in, and you goofed :P [21:20] Thank you btw guys, really. [21:21] So, all in all, the problem: public key is invalid? [21:21] or private? or both? [21:21] VjdfMQ: At this point, get local access and create a new keypair [21:21] Roger [21:22] VjdfMQ, the server is rejecting it. Which could be for any number of reasons. odds are 'systemctl status ssh' on the server will tell you exactly why [21:22] Thank you [21:22] And fix IPMI and document the credentials :P [21:22] VjdfMQ: if you're having this much trouble getting into your own server (assuming multiple people involved), there's no telling who has access or where those keys live. [21:22] this is a fair point [21:22] hello, is there a way to disable all the fixes/workarounds for spectre/meltdown/etc ? [21:22] dorei, what for? [21:23] there's no need to patch my home pc for spectre/meltdown/etc [21:23] is there? [21:23] dorei: why? [21:23] This is small firm company, but ... this is unbelievable ... [21:23] dorei, does it have internet access? [21:23] leftyfb: i think the reason we have spectre/meltdown is the crazy unsecure optimizations that intel made [21:24] lordcirth: yeap, but no other users can login to my home pc [21:24] dorei: ok, why do you want to disable the patches? [21:24] I mean, this disrespect of credentials and staff [21:24] dorei, if it has internet access, let alone a browser, you need the patches [21:24] to make it faster, no? [21:24] dorei: wait. So you intend to solve the crazy insecure optimisations by allowing them to be exploited? [21:24] dorei: you won't notice a difference [21:24] dorei, 1) without those patches a website could root your PC 2) the performance impact is quite small [21:24] hggdh: i think that only on cloud environment they can be exploited, no? [21:25] trying to install openvpn, but getting errors, how do I figure out what's going on here? https://gist.github.com/aaronmu/0268fa07940bacde660b20441b1f97e3 [21:25] lordcirth: isnt a specially compiled firefox enough? [21:25] I believe both are needed [21:25] beepbeep_: which version of ubuntu? [21:25] dorei: no. Many of the exploits can be given to you, some even in Javascript (erm, via a web site) [21:25] hggdh: then i have a bigger security issue, my browser, no? [21:26] is anyone else having trouble getting chrome to run in the system tray on ubuntu? [21:26] leftyfb: 18.10 [21:26] beepbeep_: what's the output of: sudo apt-cache policy openvpn # use pastebin [21:26] dorei: not the browser as much as the lack of all other fixes, partial or not, so far implemented [21:26] ash_worksi, never heard of it til now [21:27] dorei, short version: Linus would not have put the patches into all kernels if it were not needed by basically everybody [21:27] at least if they had performance impact [21:27] ash_worksi, It is probaly not what you think [21:28] I think maybe it has something to do with appindicators? [21:28] dorei: also, most of the browsers implemented a few workarounds for some of the spectre exploit collection [21:28] hggdh: if these workarounds are enough, then there's no need to patch the kernel [21:29] dorei: You are good. No need to do anything [21:29] dorei: if they solved everything, they would not be workarounds, they would be solutions. [21:30] i don't collocate servers in my home pc, i only need a browser that's immmune to spectre/meltdown [21:30] at one point I was trying to move notifications from the top of the screen to the bottom right [21:30] I don't remember how I was advised to do that [21:30] dorei: You do not need to change anything. You will not notice any changes in performance if you remove security parches from your kernel [21:31] can anyone help me out with that? [21:32] ash_worksi, Yes that makes more sense. You ysed the GUI in a way not anticipated and something broke. [21:32] always stay on the happy path xD [21:33] Younder: so if you wanted to move the notification bubble to how it used to be on 17, what would you recommend someone do? (I am trying to figure out what I attempted) [21:33] brb [21:34] back [21:34] ash_worksi, Let's see. At firsr it worked. [21:35] ash_worksi, How about the steps that lead to it not working. (For me I never remeber every detail sharply enough, but maybe you are better) [21:35] I don't remember when it broke, but yes, when I opened chrome and set 'allow processes to run background' under chrome's settings, it would show an icon in the system tray; this would also allow hangouts to run in the system tray [21:35] Younder: I can't remember; I was told to install x or y from a recommended webpage [21:36] notify-osd [21:36] apparently [21:36] ash_worksi: hangouts will run in the tray without the "run in background" setting enabled. When chrome is opened. [21:36] leftyfb: well, it's not working :\ [21:36] ash_worksi: which part isn't working? [21:37] the part where it shows an icon in the try [21:37] leftfb is still here? [21:37] da fook? haha [21:37] * EvilAIM == tekky [21:38] tray*; actually, I really don't care if chrom(ium) shows in the tray, but I've never seen the hangouts extension properly put an icon there without seeing the chrom(ium) icon there as well [21:38] EvilAIM: from EFnet [21:38] ash_worksi: I'm not sure how it works with chromium === daniel is now known as Guest79902 [21:38] I can't remember which webpage I was referred to to get notify-osd [21:39] leftyfb: what I mean is, it _was_ the same situation between chrome and chromium [21:39] ash_worksi: https://askubuntu.com/a/1056242 maybe? [21:39] ash_worksi: is it just chrome apps or anything that runs in the notification tray? [21:40] yea [21:40] been a while haha === EvilAIM is now known as evilaim [21:40] erm, I specifically mean the little icon where the speaker and battery icons are on 18.04 [21:40] *yawn* [21:40] anybody has a recent guide on how to run docker in lxd? [21:41] Well, I'm back at it so I'll help out again. [21:41] been a bit lazy in my days haha [21:41] I did put security.nesting = true on a container but it still refuses to run (docker-ce) [21:42] isn't lxd a container software in-of-itself? [21:42] ash_worksi, Did you do a 'remove from favorites' when you right clicked the icon? [21:42] evilaim: 2013/02/23 [21:42] Younder: not that bar [21:42] ash_worksi: in a different way [21:42] ash_worksi, In what bar? [21:43] bjonnh: why do you want to run docker in lxd? [21:43] bjonnh: you might try #docker or #lxcontainers [21:43] bjonnh: dunno if I can help, but curious [21:44] oh nevermind it was an issue with the fact that I use a ubuntu minimized [21:44] ash_worksi: to test deployment systems [21:44] I use lxd to make sorts of mini-vms without the overload associated with vms [21:44] (because I have less needs for isolation here) [21:44] wowsa. 5 years ago :) [21:44] whats that url shortener called again? [21:45] but I'm working on projects that require docker so I would like to leverage my servers running lxd [21:45] I think I've been doing ubuntu since 7.04... [21:45] I feel so old :( [21:45] evilaim: same here [21:45] 10.04 [21:45] I don't know much about lxd, so I guess I can't appreciate that statement [21:46] lxd works great [21:46] but I'll look into it I guess. [21:46] bjonnh, What is lxd? [21:46] you really handle the lxc' containers like real machines [21:46] erm... omg... I used it so many times... [21:46] leftfb, you run this place or? [21:46] Younder: a canonical front-end thingy (in go) for lxc [21:46] can I shove a bot here for testing? No spam/channel interaction [21:47] evilaim: no. Lets take it to #ubuntu-discuss [21:47] hmm, nm, I'll just run my own network [21:47] bitly! [21:48] Younder: https://bit.ly/2ABn0xD -- where the volume and battery icons are in the top right [21:49] Look I nibble at my part, you nibble at yours. This system is to big for any of us to see the whole picture. [21:49] ash_worksi: I deploy my services fully on my servers directly from ansible [21:49] bjonnh: :) [21:49] ash_worksi: "I want a 'machine' with 4 cpus and 64GB ram then install all of that on it" [21:49] bjonnh: is lxd like rancher? [21:49] and I can have server loads or client loads (using x2go) [21:49] bjonnh: or maybe rancher would be better for you? [21:49] so I can have clean and reproducible Xorg environments for my users [21:50] ash_worksi: what would it bring? [21:50] is that some kubernetes stuff? no too convoluted [21:50] bjonnh: you should take the server discussion over to #ubuntu-server [21:51] well, I don't that much about linux environments, but xorg sounds like you have people spawn desktop environments? [21:51] yeah probably [21:51] leftyfb: but it is also about ubuntu clients ;) [21:51] bjonnh, No, If it was kubernetes ti would probaly be familiar to me [21:51] bjonnh: eh. Regardless, we try to keep the chatter down here. Keep it to support questions/help only [21:51] ash_worksi: yes I have full fledged xorg environments my users connect to using x2go (so they can have a Ubuntu desktop from any OS) [21:51] leftyfb: sure I'll move [21:52] bjonnh: there's also #ubuntu-offtopic [21:52] well it is not offtopic [21:52] We need a new group #ubuntu-cryptic ;) [21:53] for kubernetes discussions? absolutely [21:53] bjonnh: if it's not looking for or helping with support, it's offtopic [21:53] bjonnh: #ubuntu-offtopic is for discussions or all sorts. This channel is for support only [21:54] yes, sort of.. [21:54] But there are so many packages, not all in the standard distributien discussed here [21:55] Younder: packages not available in the official repos are not supported here [21:55] leftyfb: lxd is standard repo… [21:56] leftyfb, I was talking about packages not installed by the installer. [21:56] anyway for those that were curious: https://help.ubuntu.com/lts/serverguide/lxd.html [21:56] and lets move that to #ubuntu-server [21:56] bjonnh: There's also #lxcontainers [21:57] true [21:58] how do you trigger the touch keyboard? having to type "onboard" without a physical one seems like an oversight [22:00] anyway! [22:00] so any help in the system tray department? [22:01] ash_worksi: tried the link I gave you? [22:01] maybe I missed it? [22:03] this one ?https://askubuntu.com/a/1056242 === Skybot is now known as OhPie [22:03] leftyfb: oh I see it; I'll look [22:03] whats Budgie? [22:04] A desktop environment [22:04] oh; what DE is 18.04 use by default again... gnome? [22:04] gnome3, yes [22:05] Budgie shares some stuff with gnome, not really familiar, though [22:07] found the site i was referenced: fossmint.com ... is that a popular site to link to in here? [22:08] 2018 Nov 16 13:52:29 ash_worksi, gnome-tweak does not, but this page gives that time tweak https://www.fossmint.com/notifyosd-desktop-notification-bubbles-in-ubuntu/ [22:08] yup [22:08] that's the one [22:09] I followed both those actually (even though I think it's an OR thing) [22:09] should I reverse those steps? [22:11] Holy cow. Anybody else think libreoffice is a pos? [22:12] jmgb4: depends what you are using it for [22:14] jmgb4: please take rants to #ubuntu-offtopic. This is a support channel === patr0clus is now known as s3nd1v0g1us [22:15] * Younder reaches for his crutch ;) [22:15] no advice on the fossmint thing? [22:16] ash_worksi: I'm about to head out so I will not be able to dive deep enough to help you with i [22:16] it* [22:16] :\ [22:16] mint is fine, not ubuntu [22:16] oh yeah I forgot its not alright unless the higher uppers do it leftyfb [22:16] thanks for the help so far leftyfb :) [22:17] Younder: that's just then name of the site... it's a ubuntu 18 system we're talking about [22:18] something weird happening with ubuntu, I created a new user and in the home directory there's a folder called Template but it is in chinese characters, what is going on? There seems to be 2 word docs in it, 2 powerpoint files, and 2 excel files. All with chinese characters in the filename. wierd [22:18] everytime i create a new user, this pops up [22:18] note, i don't speak or read chinese [22:19] 18.10 k3b not recognizing any media i put in :/ [22:19] sheep dip :( [22:19] fx9 not suprising as there is no language called Chinese. There are a 130 languages in china.. [22:20] Younder: There are 2 that start with "chinese" [22:21] people had this before? [22:21] probably Canji [22:23] hi :) [22:24] So I need to know why my internet connection keeps going down, requiring me to reboot my system [22:24] Younder, https://pastebin.com/rBHQL9Q6 [22:24] which seems to resolve the issue until it goes down again [22:24] any ideas? [22:25] Additionally, the WiFi Indicator at the top, has a question mark on it [22:26] fx9, lol, have you accidentally set the language [22:26] Younder, how did that happen? [22:27] fx9, I have no idea. But you should be able to check it [22:28] https://websiteforstudents.com/how-to-change-to-your-native-language-on-ubuntu-17-10/ [22:30] any help on my indicators problem? [22:30] Younder, chinese isn't even installed [22:30] its on english [22:31] how can I check what repos have been added to apt? [22:31] or rather, if a specific one exists? [22:33] so, a friend has just asked me about an error to do with 'ejecting' a hard disk in a hot swap bay.. Nautilus reports that the commend failed with a non-zero exit status.. [22:34] I am assuming this is because his 'hot swap bay' probably reports itself as 'removable', so nautilus puts an 'eject' icon next to the drive. [22:34] clicking that icon does in fact unmount the drive, but then tries to 'eject' the drive, which fails. [22:35] is there a way to tell nautilus to not display the eject icon for that item? [22:38] is it safe to remove libnotify-bin? [22:39] it says it will remove ubuntu-desktop [22:39] does it sound fine to you? [22:40] not really, but I had to install it as part of the process earlier on fossmint [22:42] ash_worksi: "ubuntu-desktop" is just a virtual package that depends on everything required to make the desktop - removing it won't actually remove anything important [22:42] soory, not virtual, 'meta' [22:43] TJ-: that felt like a contradiction [22:43] I'm not saying it is, but how is something that is required to "make the desktop" not important? [22:44] ash_worksi: there's nothing useful in ubuntu-desktop, it just has a control file that has a Depends: on every package required to make the DE [22:44] I seems to me mint and gnome share a lot of the same libraries, so it is probably used by both. [22:44] so... I can remove it? [22:44] no [22:44] Younder, thanks, I solved it, there was that weird chinese directory in /etc/skel which was probably put there by wps-office, removing that dir in /etc/skel solved it [22:44] I mean, I just had it removed, then got cold feet and reinstalled it [22:44] ash_worksi: it's not recommended but yes, it won't stop the desktop working [22:44] (and haven't logged off yet) [22:45] what about libnotify-bin? [22:45] fx9: glad you resoved it. That was a really wierd error [22:45] ash_worksi: I thought you said that trying to remove that wants to remove ubuntu-desktop? in which case that suggests only those two will be removed [22:45] when I removed that, it removed ubuntu-desktop, but when I installed it, it didnt' say it was installing ubuntu-desktop, so I assume I can install ubuntu-desktop without libnotify-bin [22:46] ash_worksi: that package only contains the notify-send CLI tool [22:47] ash_worksi: ubuntu-desktop declares libnotify-bin as a dependency, so if you install u-d it'll pull in libnotify-bin. That's why u-d has to be removed to allow libnotify-bin to be removed [22:48] Yes, but why was it so important to remove it that you risked compromizing your entire system? [22:49] TJ-: okay, so let me start from the beginning, I wanted to revert the 18.04 notification bubble to the way it looked on 17; I was directed to fossmint where it told me to install libnotify-bin (to no avail); where I then continued to try notifyosdconfig which I don't even know if worked because I didn't see anything I could actually configure... in any event, since then (approximately) chrome doesn't [22:49] show up in the system tray and I can't pin just hangouts because it comes up as a chrome window [22:50] TJ-: okay, but then what did `sudo apt install libnotify-bin` do for me in the first place? nothing? (I assume that 18.04 ships with u-d, right?) [22:50] ash_worksi: that sounds like you removed libnotify itself [22:51] back sorry [22:51] TJ-: I just followed this: https://www.fossmint.com/notifyosd-desktop-notification-bubbles-in-ubuntu/ === JFox762_ is now known as JFox762 [22:51] hi [22:51] So I need to know why my internet connection keeps going down, requiring me to reboot my system [22:51] which seems to resolve the issue until it goes down again [22:51] Additionally, the WiFi Indicator at the top, has a question mark on it [22:51] ash_worksi: according to the package descriptions for libnotify-bin and libnotify4, the notification-daemon package does the actual work of displaying [22:52] ji D: [22:52] TJ-: so... I assume the leolik package of libnotify-bin was just an unncessary step for me perhaps? [22:53] ash_worksi: that article does say "I’m not sure of Ubuntu 16.10 and later but the following instructions work with Ubuntu 14.04 up to 16.04" [22:53] ¬.¬ [22:53] * ash_worksi "wah wah waaah" noise plays [22:53] ash_worksi: when you plan on making changes to core functionality it does help to read the entire article and consider what downsides there might be [22:54] eh heh ^ ^; [22:54] so uh, recommendations? [22:54] ash_worksi: as well as testing in a virtual machine before messing up your primary host :D [22:56] How can I make sure my graphics card is being used instead of integrated graphics right now? [22:56] k, I guess I'm gonna try restarting... I removed those ppas and notifyosdconfig and reinstalled libnotify-bin [22:56] so I hope it all works [22:56] brb [22:57] doesn't appear to have had any effect... should I completely restart my machine? [22:59] hello? [22:59] maybe there was some other configuration change I made that's causing this issue? [22:59] any ideas on where I should start looking? === [1]MrMobius is now known as MrMobius [23:10] still no dice [23:13] anyone know what might be causing my connection issue? [23:13] It seems like my Lenovo keeps losing connection with the hotspot [23:15] some hotspots give a lease of 5 minutes, or 10, 15 [23:17] What does the d-i stand for and/or do in a preseed file? [23:17] sgen: "debian-installer" [23:18] sgen: the preseed files sets answers to questions the installer would otherwise ask the user [23:18] TJ-: Cool thanks! [23:25] umm === Richard is now known as Guest21610