[00:00] sarnold: _Sym_ it's done from ./Kbuild [00:02] <_Sym_> I can't see to figure out what changed to cause this [00:02] sarnold: _Sym_ it's probably failing on 'clean' due to not having run mrproper or prepare, may also be due to doing an out-of-tree build [00:02] _Sym_: are you missing ./Kbuild's "no-clean-files := $(bounds-file) $(offsets-file) $(timeconst-file)" [00:03] <_Sym_> TJ, yes [00:03] _Sym_: I'd suspect your git repo isn't virgin, or you're using some non-standard build tooling [00:03] <_Sym_> oh wait, no [00:03] <_Sym_> sorry [00:03] <_Sym_> its there [00:04] <_Sym_> this happens on a clean repo with no changes [00:04] <_Sym_> I have a legacy kernel project that I know has not changed and its also happening on that one too [00:05] _Sym_: then you've got some implicit/explicit make over-rides going on somewhere [00:06] (fun fact, downloading this .zip was the thing that filled /home :) go figure.) [00:07] <_Sym_> TJ, This is happening on my 4.19.63 kernel for my laptop and also my 3.10.108 kernel for my android device [00:08] <_Sym_> I have not changed anything in the kernel [00:08] <_Sym_> in either kernel [00:08] _Sym_: my best guess is you've got a 'no-clean-files' assignment as "-" [00:08] <_Sym_> it must be some update from the last day or so [00:09] <_Sym_> TJ, its also happens with "make mrproper" command [00:09] <_Sym_> but just "make" does not produce the error [00:10] _Sym_: create a new user account, use it to make, see if it still happens. That'll tell you if it is a per-user or system-wide issue [00:15] <_Sym_> TJ, I figured it out.. :-) it was because of a kernel submodule [00:15] <_Sym_> thanks for helping me [00:16] <_Sym_> thank you too, sarnold [00:19] _Sym_: woo :) [00:19] Hello, I was wondering if anyone knew who to make a shortcut for terminal. I've seen other people do it where they right-click terminal and a bunch of boxes pop up where they can click on a shortcut [00:19] google isn't helping very much [00:20] Kevin199, where do you want this shortcut to show up? [00:20] I want a desktop shortcut for irssi(for example) [00:21] Kevin199, are you using gnome? [00:21] Yes [00:21] Kevin199, ok, I dont use gnome, but, have you tried dragging the exectable to the deskto? Not sure that works with gnome [00:22] I'll try [00:22] Kevin199, or right click desktop, crate shortcut, type in gnome-terminal [00:23] What are you using instead of Gnome, I can switch real quick [00:23] Kevin199, I use Mate [00:23] coz_, let me install that real quick brb [00:24] the shortcut you create on mate may not work the same way on gnome. [00:25] I didn't realize Ubuntu Mate was its own Distro [00:27] Kevin199, I believe you can install mate-desktop then log into that, however, I generally don't suggest that. I prefer a cleaner system, if you can define it that way [00:29] coz_, I'm looking into it right now. I think i might stick to Gnome tho [00:30] my laptop is only 4GB of ram and apparently Ubtunu Mate runs a minimum of 1GB of ram [00:33] if I have automount working for my usb sticks.. and i pull one out (does it dismount automagically as well??) and put in a new one.. will they all use the same /dev/ "name" [00:34] like usb stick 1 gets identified as sdb, and sdb1 gets mounted into /media/movies.. if I pull the usb stick, it should umount /media/movies.. and if I plug in usb stick 2.. will it be a the new sdb, or will it take sdc? [00:35] I can make em all mount correctly I think using fstab.. but it'd be easier if I could use /dev/sdb1 vs. device ID and a bunch of lines [00:35] scythefwd2: you should not PULL it out without unmounting/ejecting it first via software [00:35] If you can suggest a way to do that via a kodi interface using a joystick.. I'll be very glad to hear it [00:35] scythefwd2: removable media devices are handled by udisks, and it mounts them under /media/$USER/ with either the file-system label or the UUID [00:36] well thats not gonna work.. [00:36] scythefwd2: that's something to ask the Kodi project about but if you do pull the device expect file-system damage [00:37] scythefwd2: it sounds like something else might be responsible if they're being mounted directly under /media/ [00:37] they're read only as it is.. so I'm not particularly worried about that (nor the data on them.. they'll be copies of movies) [00:37] they're not, atm.. I was thinking I could make that happen though.. I'll gladly accept I'm wrong though [00:37] scythefwd2: despite read-only the Fs metadata is updated [00:38] but I need them to all mount to same dir for my idea to work.. I'll see if there is a way to do it having it check /media/$user/ as a parent and scan sub dirs [00:41] scythefwd2: possibly you can add a service unit that creates/removes a symlink when udisks mounts a device [00:42] possible.. that shouldnt enough extra load to it to be problematic [00:42] scythefwd2: you might be able to use the udisks HintName and/or the udev UDISKS_NAME [00:43] Yeah, you just went right past my head there lol [00:43] I'm new, very new to linux [00:43] scythefwd2: or you could just ensure all your USB devices have the same file-system label [00:43] that should be easy enough [00:43] heh, I like that one :) [00:44] wonder if its possible to alias a button press to a umount command (remember.. only a game pad hooked up here) [00:45] scythefwd2: I don't use kodi so no idea but the principle sounds good [00:45] I'm kinda hoping it lets be do it at the os level lol. kodi is more of a wm [00:46] reboot ... [00:46] but I'm running on a pi zero.. so gotta keep it TRIM [00:47] scythefwd2: maybe this will get you started: https://kodi.wiki/view/Built-in_scripting [00:47] scythefwd2: section 4 shows a gamepad example [00:47] thanks.. I was planning on bugging them next lol.. figured these questions were more os level [00:48] scythefwd2: from there I find RunScript(script[,args]*) on https://kodi.wiki/view/List_of_built-in_functions [00:49] scythefwd2: and further down that page is "System.Exec(exec)" [00:49] thanks.. looks like I get a bit of digging in there still todo before I start building [00:50] I came in here half expecting .. yeah, it doesnt work like that, and what you want to do isnt possible [00:50] so this was a much better result [00:50] thanks [01:24] hello guys, I have created a VM using ubuntu 18.04 server [01:25] I set up my proxies and all, but when i do sudo apt-get update, I am getting "temporary failure resolving " [01:25] stewie925: DNS issue [01:25] oh [01:26] qq, TJ, when i was in the ubuntu 18.04 installation process, there was a part where it gave an option to install OpenSSHServer (which I did click), but there is also a DNS Server option, should I have clicked it? [01:27] stewie925: no, 'server' means it answer requests. The issue you've there is the local resolver client cannot find the proxy name when it queries your network's DNS server (which is usually given out by DHCP) [01:29] hi where is my custom binds located. i mean if i run bind '"k":"apa"' [01:29] where do linux ssave that custom bind [01:30] kemisten: what tool did you use? [01:31] sarnold, "bind '"key":"string"' [01:31] sarnold, bind '"key":"string"' is the cxommand [01:33] long working windows kvm/qemu guest suddenly won't work as it used to. any ubuntu devs which know the qemu thing well which are helping here I could IM? [01:33] with more technical details and so on. [01:34] kemisten: I can't find a bind executable in ubuntu; what package provides it? [01:35] sarnold, just type bind '"k":"test"' [01:35] it will work [01:35] sounds like an option of a specific program [01:35] tested my self [01:36] kemisten: what does "cat /etc/issue" report? [01:36] kemisten: AHHHH! :D [01:36] kemisten: moment [01:36] testill -9 lolz [01:37] ok so no official help with upgrades of qemu/libvirt packages or sudden borks of people's libvirt machiens. yikes.. back to the old fashioned 9 hrs of dismantlig ubutnu automagic [01:38] kemisten: thanks for persevering :D it never crossed my mind that this would be a shell builtin. sorry. [01:38] Ubuntu 19.04 \n \l [01:38] kemisten: so you're on about the BASH "bind" built-in ? [01:38] kemisten: so -- if you just this, all on its own, it's not going to be saved anywhere; it's only going to affect the process that you ran it in, and it's not going to be saved automatically [01:39] ive not installed anyuy package for running bind kommaNDF IT CAMES WITH DEBIAN,RASPBIAN AND UBUNTU IT SEEMS [01:40] kemisten: see "man bash" and the " Readline Initialization" section, they are read from the readline config file ~/.inputrc [01:54] Hello, just installed Disco Dingo and tryng to run docker... I install the docker.io package, started a docker using "docker run ..." and I am unable to stop it with ctrl+c or with docker stop with the container id... any clue..? [01:54] ^\ may do it [01:54] if not, ps auxw to find the pid, kill -9 [01:56] I was just googling regarding the "temporary failure resolving " issue - it pointed me to updating /etc/resolv.conf , but for 18.04 it always revert to original after I reboot [01:57] thanks sarnold it worked, but it's kind of heavy... [01:58] stewie925: that is bad advice, don't mess with /etc/resolv.conf, fix the underlying issue which is the name not being found [01:58] metres: yeah, there *really* should be a better way to kill things. I'm not sure where to suggest to find out though.. it might be worth a bug report if you can't find something [01:59] TJ-: thank you , let me find that [02:01] stewie925: try "systemd-resolve " see what it reports, and if it fails, check the log file for clues with "journalctl -n 25 -u systemd-resolved" === [b0b] is now known as b0b [02:13] TJ-: thanks again - just got back, checking right noe [02:14] *now [02:14] the "system-resolve " seems to be lagging [02:14] oops systems-resolve i mean [02:15] systemd-resolve (typo) [02:15] just failed now with "query timed out " - checking the log [02:16] stewie925: check the NSSwitch config "grep 'hosts:' /etc/nsswitch.conf" [02:17] hosts: files dns [02:18] is what I see in the nsswitch.conf file [02:19] stewie925: so it first looks in /etc/hosts then it uses the local dns resolver stub, which will be systemd-resolved which is referenced in /etc/resolv.conf (which is a symbolic-link) and should contain "nameserver 127.0.0.53" [02:21] stewie925: if the PC is getting its IP address via DHCP then the nameserver provided by DHCP will be given to systemd-resolved to use, you can check what the upstream DNS server addresses are with "systemd-resolve --status" [02:28] so i did a sudo apt install libopencv3.2-java and now i don't have a .so file in lib [02:28] what did this actually do? [02:28] dpkg -L libopencv3.2-java showed a bunch of jar fles ut no .so file [02:31] iirc A libopencv_javaXXX.so will be in /usr/lib/jni, to where you have to symbolical link libopencv_java.so. [02:32] sudo ln -s /usr/lib/jni/libopencv_java320.so /usr/lib/libopencv_java.so as example from https://sikulix-2014.readthedocs.io/en/latest/newslinux.html [02:35] hrm. i wonder why it doesn't just put it in /usr/lib/ [02:41] maybe it just gives back control === guiverc2 is now known as guiverc [03:19] hi. im curious if 'ufw allow ssh' enables firewall only port 22 tcp or any another port we replaced for 22 too? [03:50] QR: how do you mean "any other port we replaced for 22"? [03:55] tomreyn: i changed ssh port from 22 to another number. [03:57] 'ufw allow ssh' will cover new ssh port number? or its designed for default port 22 only ? [04:00] QR: UFW has a concept of something like service / application descriptions / profiles (i forgot the proper name it uses). one of those is for the ssh service, defining that it listens on 22/TCP. you should be able to find those in /etc/ufw, i think [04:00] (and be able to edit them as needed) [04:01] that's all from memory, haven't used ufw in a while [04:02] question: Does anyone know if a recient change/update is now making a Grub popup menu prior to Ubuntu booting? ... I'm having trouble with a computer running Ubuntu ver 16.04 booting anymore :\ [04:04] I'll check back for any replys ... thanks. [04:06] tomreyn: yes but this command covers for default 22 port only? otherwise i'll specify the port without type ssh [04:09] yeah, /etc/ufw/applications.d/openssh-server [04:10] QR: yes, 22 by default. and deltab just pointed you to the file which defines it. [04:22] deltab: ok then 'ufw allow ssh' command only covers port 22/tcp. i thought it should cover other port after i changed 22 to something else too. [04:22] I'd expect so too [04:23] I don't know much about ufw [04:23] tomreyn: i tested now but firewall didn't stop me [04:23] i didn't add new port for lfw as rule after changed ssh port. and i could login without problem [04:25] QR: do you mean s/lfw/ufw/? [04:26] ufw yes. sorry :) [04:26] QR: /etc/ufw/applications.d/openssh-server won't magically change after you make openssh-server listen on a different (or additional) port. personally i'd not edit /etc/ufw/applications.d/* and just put the specific port that's being used in the ufw rules. [04:27] did you use ufw app update ? [04:27] you may also need to systemctl reload ufw if it's implemented as a systemd unit [04:28] first time using ufw and Ubuntu. trying to secure system. 2018 lts has systemctl ? [04:29] deltab: didn't do it [04:29] ubuntu 18.04 uses systemd and thus systemctl, yes. [04:30] i didn't edit /etc/ufw/applications.d/openssh-server file. it shows ssh 22/tcp still. but i use another ssh port to login and ufw didn't stop me. [04:32] deltab: 'ufw app update' command returned as invalid syntax. [04:33] it needs a name (or 'all') and is used when you've edited the file [04:34] sorry I misunderstood what you meant about changing 22 to something else [04:35] changed ssh port 22 to something else at /etc/ssh/sshd_config [04:37] i'm curious if 'ufw allow ssh' command will cover new port too? or only old port as ssh has predefined 22 port in ufw config. [04:48] "Users can specify rules using either a simple syntax or a full syntax. The simple syntax only specifies the port and optionally the protocol" [04:48] I think it looks up ssh in /etc/services [04:50] oh, there is also ufw allow to use an app [04:50] but I think the app's name is openssh-server, not ssh [04:57] Hi - It looks like by changing something in my main.cf file for postfix it won't start - not sure how to debug the issue - any suggestions - sorry for a basic question [05:00] deltab: in web i read it as ssh. not openssh-server [05:01] try ufw app info ssh [05:14] deltab: couldn't find profile ssh [05:14] sudo ufw allow ssh (Checks the /etc/services file on your system for the port that SSH requires and allows it. Many common services are listed in this file.) [05:15] i checked services file says 22. so i deleted ssh rule for ufw and added port manually [05:16] i see that predefined rules static and not changing when i changed defaults. so its better to specify everything by self instead of typing service names to be sure. [05:39] i reinstalled ubuntu-desktop, because upgrading 18.04 to 19.04 broke my startx (didnt get a display, no error in Xorg.log either) [05:39] now my desktop looks blerg!? https://pasteboard.co/IqW61b3.png what happened? [05:43] that looks like vanilla gnome3, gnome session, logout, click your username, and change to ubuntu [05:55] hello guys , i am trying to run VNC server on my desktop, but when i follow this "https://www.digitalocean.com/community/tutorials/how-to-install-and-configure-vnc-on-ubuntu-18-04" the VNC Server works fine but when i loggin it does not show the User's main desktop [05:56] instead it loads a default gnome desktop session and all the icons i added to desktop is missing [05:56] do you know how is it possible to load the User's main desktop when vnc connected ? instead of empty desktop [05:58] i have a folder that i use for a cifs mount /mnt/steam and it's showing up size 0 on ls -al. it's not mounted, i can't rm -rf it, is there a better way to nuke the folder/directory? [05:58] "cannot remove 'steam' : Device or resource busy" [05:59] OerHeks dankjewel!!!! [06:15] oh when i upgraded from 16.04 to 18.04 cifs-utils got deleted/purged somehow, but the fstab entry still automounted it [06:15] somehow? that's a bug, right? [06:15] so the directory was just "resource busy" [07:56] Is there anyone experiencing problem with right click context menu in 18.04? [07:57] Across many software, when I right click to open a context menu, it sometimes auto-click the menu item. This was really frustrating. [07:58] I often ended up opening a new window or open inspector when right-clicking under firefox. [07:58] I also experience this when using other apps. [08:10] I suppose this is due to wrong positioning of the context menu. [09:00] hi there! i am wondering if there is a option for cifs that allows ... "reconnect". for example: i mount smb-shares on my laptop when i am on wifi with vpn. when i come back to my office, i plug in a wire and would like to continue using the share, but via wire. [09:01] i just tried, it doesn`t work at the moment [09:01] plugged in the cable, made sure i received a dhcp-lease, then disconnected vpn and wifi. [09:02] when i now try to ls the mounted share... the console hangs [09:02] my current options are: noauto,users,nobrl,soft,vers=2.1 [09:04] is this just known broken? [09:41] If it doesn't respond to a signal like HUP, you'd need to either have your network manager, ifup script, or similar fire a mount -o remount (WITH password in a file somewhere safe and accessible, if required!) or unmount and remount the share, or use an automounter daemon [09:43] darkmeson: i just tried a mount -o remount ... that also hangs itself up [09:44] you'll probably have to umount -l ; mount then [09:45] and if that doesn't work, there's also fusesmb. FUSE-based filesystems tend to lose a decent bit of performance, but it doesn't matter nearly as much for slow network filesystems [09:45] yeah, the umount also hangs [09:45] what a pain [09:45] and, of course, you always have veto power since you can always kick the daemon out from under the mount with kill -9 and THEN unmount [09:49] what i dont understand: under windows the same procedure is working absolutely transparent out of the box. i doesnt even take a second. i plug in a network cable and it automatically switches to the wired network, even open files dont cause an issue. what kind of snake oil is that? [09:49] personally, I only use sshfs anymore. Windows has mingw/cygwin+openssh, and Android has the same stacks as GNU/Linux without too much more effort (than stock GNU/Linux) [09:49] sshfs is to slow for me [09:50] davfs2 then. windows has native support for that, and apache, nginx, etc, have native builds for Windows and about everything else [09:50] (can you saturate a gig with sshfs?) [09:51] lately, yes. they fixed that several years ago [09:51] not even thinking about 10g :) [09:52] your question is a bit loaded whether you realized it or not though (heavily hardware/configuration-dependent) [09:52] i didnt :) [09:53] i like smb, especially modern versions of it. smb multichannel, smb direct stuff like that [09:54] I think it actually has support for the 'fsc' mount option these days ("FSCache"), so than and cachefilesd might help if there's enough local storage for the hot data and the workload is mostly read-dominated [09:54] s/so than/so that/ [09:55] It always amuses me how people get sucked in by marketing terms like that [09:56] well, its really nice in real world to [09:57] they wade just far enough into the shallow end of psychology to get people believing their too smart by half solutions are easier and more straightforward, when the reality is anything but [09:57] i do have infiniband connections, i do have multiple 10g links to my servers [09:58] infiniband is nice, but that just moves the bottleneck to the slow disks unless they're in a dangerously large array with inadequate redundancy [09:58] ... or one uses nvme ;) [09:59] always loved the rdma mechanisms too, but that's just a multisystem exploit waiting to happen. Especially in this day and age [09:59] yup [09:59] hi buntus [10:00] if there was budget to go IB (even if only 10g) AND solid state, then why are you cheaping out on the fs? :P [10:01] ib is actually 40g, but the access layer is 10g ethernet [10:01] anyway, keep fusesmb in mind just in case ;) [10:01] ill try for sure [10:03] * darkmeson got dual port DDR pulls cheap enough years ago, but ~$35 per 3-6ft, inflexible cable was total bullshit [10:03] funny, got the same stuff laying around [10:03] then there's also the fact that the PCIE bus overhead effectively capped the theoretical max at about 16G [10:04] those were my first tests with ib [10:04] huh? i never noticed that, did you get the cards with x4/x8? [10:05] hey guys, whenever i turn on my computer and log in, i always get this error: https://imgur.com/YKIHwJ5 [10:05] the mellanox ones or? [10:05] how can i fix it? [10:05] like I said, it's PCIE bus overhead, so no getting around it no matter what [10:06] and yes, they're mellanox. who else? :) [10:06] rudeguy: can't load that (easily) atm. grub boot prompt, or? [10:06] well, if pcie bandwidth + overhead > nic speed: there you go. doesn`t matter [10:07] In practice, other components like slow SBs are going to matter more anyway [10:07] darkmeson: no, it's ubuntu's usual crash window. it says dbus-daemon crashed [10:10] what exactly is dbus, anyway? [10:10] Interesting. I'd been having issues like that too, but I figured it was some of the other unusual things I was doing [10:11] dbus is roughly the GNU/Linux equivalent of Android's intent system, only more horribly insecure and broken [10:11] most distros still run it as the user 'messagebus", which is a bit more telling [10:12] (because it started life as a generic message bus for generalized IPC, kind of) [10:14] technically, you can do without it IF you avoid the major DEs like Gnome/KDE, but systemctl will throw tantrums and some parts of systemd won't work [10:16] (all the useless parts that have nothing to do with init responsibilities and contribute nothing anyway) [10:16] anyway, head over to freedesktop.org if you want actual specs and so forth [10:19] as far as dbus crashing, the only thing I can suggest there is to 'cat /var/lib/dbus/machine-id' and make sure it exists and has a string in it [10:20] i see... [10:21] it does have a string in it [10:21] no, I guess I can also suggest "ps aux |grep dbus-daemon" to get the exact dbus command invocation and running it manually if/when it fails and doesn't come back. understanding what you'll be seeing is a bit more involved there, though [10:25] rudeguy: you can see your systems' past reports by running: x-www-browser https://errors.ubuntu.com/user/$(sudo cat /var/lib/whoopsie/whoopsie-id) === jstein_ is now known as jstein === im0nde_ is now known as im0nde [11:28] anyone facing LAN connectivity issue with 19.04 [11:28] https://askubuntu.com/questions/1135412/updated-to-19-04-and-no-ethernet-now [11:38] hello [11:47] Howdy folks [11:53] weird. why does it keep rejoining channels? heh [12:32] Hello, I am having a weird issue... I have two screen, a VGA and a DVI. when both are connected on boot, only the DVI get a display after choosing the grub entry but if I only connect the VGA I can plug the DVI one once I am log in and everything works fine...Anyone got an idea? [12:34] metres, are you using gdm3 as display manager ? [12:36] ioria: sddm [12:36] Hello, I was wondering if anyone here knows how to make a Keyboard shortcut to open Terminal in Ubuntu Gnome [12:36] metres, no idea then [12:37] I go to Device, then Keyboad, then scroll down and clicked the + button. But I don't know what to put as a command [12:39] https://help.ubuntu.com/stable/ubuntu-help/keyboard-shortcuts-set.html.en [12:41] Kevin199, you know that it's already set with ctrl+alt+t ? [12:41] Kevin199: gnome-terminal ? [12:42] I'm very new to Linux [12:42] I don't 100% understand [12:42] ioria, ty [12:42] Kevin199, ok [12:48] metres, are you using proprietary video drivers ? i mean, do you have an nvidia card ? [12:50] I got it to work, ty for the link [12:54] ioria: no dedictated graphic cards, I got a amd APU [12:56] I wonder if I change the cable on one screen from vga to hdmi and then both screen will receive digital signal... [12:59] On another hand, apparmor seems to do not work well with docker... I can run container but it prevent me to stop them, [13:01] Hi! Could somebody please help me with fsck? I found a corrupt file on ext4 partition. The partition is mountable, still I want to repair it. The command fsck -p shows "inode has corrupt extent header". Is the next best step to "Clear inode" with answering yes? Thank you in advance for any hint! [13:32] hi I'm on ubuntu server and installed the ubuntu desktop minimal (since my friend wants a DE) and when I boot I get to the display manager. How can I switch to the terminal? ctrl alt f1 doesnt do anything [13:34] kek_: the GUI is on tty1, use tty2 through tty6 with Ctrl+Alt+F2 through F6 [13:37] TJ-, ah it worked when I switched my keyboard ty [13:53] Hi, is there a command to reboot in a specific grub entry (and to list them all?). I only have bluetooth keyboard so I can't select an instance at boot [14:14] Hi! I have two issues 1) gurb wont install into a hard disk that has been RECYCLED from a buffalo terastation (I think this is a sector size issue (low level format) ) cant find any utility that takes care of this (not even sure if its still possible on 1TB disks its toshiba disks (4) of them 2) want to install grub from the installer live cd in the terminal to see what errors it generates...complains about a /cow when I issue [14:14] -install /dev/sdb , I don't think I have to jump over the moon to do this but pretty close!!! Anyone know how to solve either errors? [14:19] transhumanist: you need to mount the choort at /target correctly to do grub-install correctly from within the chroot [14:21] transhumanist: " for n in proc sys dev etc/resolv.conf; do sudo mount --rbind /$n /target/$n; done" then "sudo chroot /target grub-install -v /dev/sdb |& tee /tmp/grub-install.log" then "pastebinit /tmp/grub-install.log" [14:28] thanks TJ will do that and come back [14:41] Hi [14:43] I want to add a shortcut to my blender install in the launch bar. [14:44] I cannot drag and drop and there is no option on the right click menu related to shortcuts [14:46] test [14:46] d3vnu11: test elsewhere [14:47] :( [14:59] Am looking for a method to permanently bind caps lock to home without remapping the whole damn keyboard [15:18] parak0vsky, switch 66 to 110 : http://xahlee.info/linux/linux_capslock_do_home_key.html [15:28] I hate when this happens... everything was working fine... then I start my PC the other day and everything is messed up. My monitors are not being recognized property, I don't have full resolution... wtf? [15:28] I don't even know where to start [15:29] wondows: ubuntu version? kernel version? [15:40] lotuspsychje 18.04 [15:40] wondows: make sure you do the apport-collect while running the kernel version causing the problem, not the 4.18 version which worked well [15:41] tomreyn I don't know what that is [15:41] hi lotuspsychje [15:41] ducasse, hi [15:41] omg.. cant believe I'll have to spend hours trying to fix this...my day is ruined [15:41] wondows: start your system in recoverymode [15:42] wondows: and share your dmesg in a pastebin please [15:42] wondows: ok, i assumed you had filed a bug since one which sounds similar was filed around the same time you discussed your experiences here. [15:45] dmesg shows me plenty of these messages https://paste.ubuntu.com/p/ZCWHxHVZFS/ [15:45] not in recovery mode btw [15:51] wondows: whats your nvidia chipset please [15:52] wondows: read the ATTENTION paragraph at https://usn.ubuntu.com/4069-2/ [15:54] so what? [15:54] it's a RTX 2080ti [15:54] wondows: i'm guessing that you run 18.04 with HWE and got the 5.x HWE kernel [15:55] guess it updated itself [15:55] what can I do? [15:55] update to 19? [15:55] wondows: try the latest nvidia driver from the ubuntu graphics ppa, from !nomodeset [15:56] !nvidia | wondows [15:56] wondows: For nvidia and matrox graphics cards, see https://help.ubuntu.com/community/VideoDriverHowto . For AMD/ATI graphics cards, see « /msg ubottu ati » and « /msg ubottu fglrxmissing » For the latest Nvidia drivers see https://launchpad.net/~graphics-drivers/+archive/ubuntu/ppa [15:56] wondows: before you last rebooted, was this the problematic boot? [15:57] it just happened [15:58] wondows: actually you probably just need to upgrade nvidia drivers, yes. if you don't mind, post journalctl -b | nc termbin.com 9999 [16:00] hmm no this message suggests user space has nvidia driver version 430.26 but kernel has 418.56. so the nvidia driver was updated, but the kernel module needs to be upgraded, yet [16:01] which probably means you need to sudo update-initramfs -c -k 5.0.0-23-generic [16:01] uname -r prints 5.0.0-23-generic [16:02] so i guessed that version right [16:02] what does that do? [16:02] "update-initramfs: Generating /boot/initrd.img-5.0.0-23-generic" [16:03] regenerate the initial ramdisk. i'm not certain that's where the updated nvidia kernel module needs to be, but can't think of anything else right now. [16:04] so should I reboot? [16:04] yes, shouldnt get worse [16:07] tomreyn: the ATTENTION about 3rd party is for the DKMS (re)build of the module I'd expect [16:07] TJ-: but are nvidia modules rebuilt via dkms? despite lack of source? [16:07] have no idea how you guys figure this out [16:08] that worked [16:08] I'll never understand this [16:08] tomreyn: the shim is I think [16:08] thanks [16:09] tomreyn: not used nvidia drivers in a long time though so it may have changed [16:09] TJ-: apprently it still helped, so my suspicion is the module was already there but dkms occurred after update-initram (due to kernel image update) [16:09] !info nvidia-dkms-415 [16:09] ...when updates were installed [16:09] Package nvidia-dkms-415 does not exist in bionic [16:09] wondows: i was really guessing, [16:10] !info nvidia-dkms-430 [16:10] Package nvidia-dkms-430 does not exist in bionic [16:11] what's up with ubottu? that does exist! 430.26-0ubuntu0.18.04.2 -> 500 http://gb.archive.ubuntu.com/ubuntu bionic-proposed/multiverse amd64 Packages [16:11] there's only nvidia-dkms-390 [16:11] ...in bionic [16:12] oh in -proposed [16:13] https://packages.ubuntu.com/search?keywords=nvidia-dkms-430 [16:13] tomreyn: I'd guess that DKMS did (re)build but the new module wasn't added to the initiamramfs [16:14] grrr, typos galore today [16:14] Hello everyone, I am currently having issues with my Dell XPS and Dell D6000 dock. It was working for weeks until last night when it suddenly stopped. If I have it plugged in when booting I get dropped to the maintanance shell, but after pressing ctrl+d it boots fine. [16:15] TJ-: yes, that's what i meant by: "my suspicion is the module was already there but dkms occurred after update-initram (due to kernel image update) ...when updates were installed" [16:15] tomreyn: aha :) [16:15] you expressed it better :) [16:15] my eyes are suffering much blurring today I think I'm missing a lot! [16:17] get well soon. [16:17] I should menton that I was running 18;04 now upgraded to 19.04 but the issue still exists [16:21] rj1705: can you reboot twice, once while connected to the dock, once while not? and when you enter the maintenance shell, type echo $REASON and note down any output (if any), and then continue to boot? === jstein_ is now known as jstein [16:22] tomreyn sure, I'll go do that now [16:22] rj1705: and when you're done with those two boots, post journalctl -b -1 | nc termbin.com 9999; journalctl -b | nc termbin.com 9999 [16:23] ok :) [16:23] tomreyn do that once I get back into my system? [16:23] after those two boots, yes [16:24] i can remind you about those commands when you'll be back [16:31] if I echo $REASON nothing comes back [16:32] @tomreyn could you send me those commands again when you have a chance :) [16:33] rj170: journalctl -b -1 | nc termbin.com 9999; journalctl -b | nc termbin.com 9999 [16:33] so the difference is just "-1" [16:33] hey, I think ubuntu's virtualbox package broke with a kernel update recently. In the end it complained about number of arguments when compiling something related to dkms. [16:33] when I added oracle's repo and installed virtualbox 6, everything worked. [16:34] i wiped out and reinstalled my LAMP stack on my ubuntu server via tasksel, but now i can't ssh into the server anymore, keeps saying connection refused. Idk what I did [16:34] rj170: this posts your full kernel log from last but one (-b -1) and the latest / current (-b) boot to termbin.com, a !pastebin like site. [16:34] didnt think LAMP and ssh had anything to do with each other [16:35] abdulhakeem: theyy don't, ssh also works in the dark. [16:36] @tomreyn https://termbin.com/jfid [16:36] hnn [16:36] hmm** [16:36] abdulhakeem: is openssh-server installed then? [16:36] no matter what i do i always get this message with respin: https://pastebin.ubuntu.com/p/3dczBY4tqd/ [16:37] i would think so because it was working before [16:38] rj170: so https://termbin.com/jfid is the output of "journalctl -b -1 | nc termbin.com 9999", your last but one boot, with the docking station connected, where you were dropped to the maintenance shell and continued? [16:39] abdulhakeem: does dpkg -l openssh-server say 'ii' in the beginning? [16:42] @tomreyn Yup docking station connected drop to the maintenance shell, ran echo $REASON but nothing came back, then hit ctrl+d and the machine booted fine but monitors not working from the dock [16:43] rj170: okay, and the other log? [16:47] I think i ran into a bug, tried this: sudo apt install --reinstall xserver-xorg-video-amdgpu . And got https://termbin.com/yovx [16:49] Fresh install of Ubuntu 18.04 LTS [16:49] @tomreyn sorry I think I mistook your instructions, I ran both those commands as one with a ; between [16:50] rj170: okay, how often did you boot since? [16:50] I havent Lucky I have two laptops that I can use ha [16:50] rj170: "ha"? [16:51] oh that's an expression of relief, i assume. [16:51] @tomreyn sorry yes it is :) [16:51] Means I can stay on IRC on one whilst trying to fix the other [16:52] When I try running journalctl -b -1 I get 'Specifying boot ID or boot offset has no effect, no peersistent jounal was found' [16:52] rj170, have you tried to boot an older kernel ? 4.15 or 418 if you still have it ofc [16:52] rj170: so i don't knwo which of those boots is uggested you'd do the log you posted is of . but we can just sort this out later. as a reminder, here's what i had suggested you'd do: rj1705: can you reboot twice, once while connected to the dock, once while not? and when you enter the maintenance shell, type echo $REASON and note down any output (if any), and then continue to boot? [16:53] rj170: oh oyu have no persistent journal, then this wont work [16:54] rj170: that's strange since it should be default on 19.04 [16:55] I should also point out I have recently moved my root partition to LVM [16:55] One second just booting again [16:56] rj170: so my understanding is that what you posted at https://termbin.com/jfid must is the log from the current boot, which you did with the docking station connected. [16:56] No thats with it disconnected [16:56] this line from the log supports this: systemd-udevd[383]: Spawned process '/opt/displaylink/udev.sh /dev usb-004-003-DisplayLink_Dell_Universal_Dock_D6000_1711034143 /dev/bus/usb/004/003' [445] is taking longer than 59s to complete [16:56]