[00:00] <jeremy31> qwebirc64497 all I have is https://wiki.debian.org/BuildingTutorial and I know I found a better one somewhere
[00:15] <SuperLag> RHEL has kickstart configurator for giving you at least a template to work with for a kickstart file. Is there an equivalent for Ubuntu preseed files, to generate them?
[00:31] <gbellinoz> If my system only knows about my ISP's nameservers, how does dig +trace work?
[00:36] <katamo> SuperLag https://help.ubuntu.com/lts/installation-guide/example-preseed.txt
[00:37] <katamo> SuperLag I spent 3 weeks in that documentation recently for a project on my desk. It goes significantly deeper than tha
[00:44] <texla> Ubuntu 18.04 ubuntu/gnome de When I boot num lk pad and light are active thru login them light turns off and back on but numlk pad buttons(keys) are dead..Have to use keyboard to enter any numbers..Switch to cinnamon and all systems work properly
[01:23] <satrioirc> if my syslog is full of records like: Creating SSL connection to host // SSL connection using ECDHE_RSA_CHACHA20_POLY1305 // Sent mail for my@email (221 2.0.0 closing connection i125-v6sm5077864wmd.23 - gsmtp) uid=0 username=root outbytes=5877
[01:24] <satrioirc> that means sthing bad, right?
[01:29] <satrioirc> i spammed my logs to full the other day when i launched a sparta script test against my ssh, but now im receiving messages when the system is idle. how to get harden my defenses against this?
[01:31] <gbellinoz> Same messages now as from when sparta was running?
[03:26] <hays_> is there a way a user can get into a group other then /etc/groups
[03:31] <hays_> i seem to be in the users group, but i don't see it in /etc/passwd
[03:33] <realies> doing a touch Tiësto in an ubuntu:bionic docker container generates a 'Ti'$'\303\253''sto' file
[03:33] <realies> this doesn't make it go away http://www.iasptk.com/ubuntu-fix-locale-issue/
[03:33] <hays_> i should say /etc/group
[03:34] <realies> how to fix?
[03:35] <realies> none of the locales are registered still
[03:36] <realies> setting them manually also doesn't register them
[03:36] <realies> what
[04:25] <fullstack> my hi
[04:29] <fullstack> My underscore _ is mysteriously missing in my terminal, I can't see it. It shows up as a space and it is really annoying. I had this problem before, any reason?
[04:29] <fullstack> _ <--- I see nothing
[04:29] <lotuspsychje> fullstack: did you check terminal settings?
[04:31] <lotuspsychje> fullstack: try change to blinking cursor
[04:31] <laudecay> hi so my python install is hopelessly screwed because i was stupid earlier. what do i need to do to completely remove it from my system and start fresh?
[04:31] <lotuspsychje> laudecay: can you explain what you did exactly?
[04:32] <laudecay> lotuspsychje: well, i ran pip as sudo, somehow ended up with python3 and python3.6, and also somehow pip, pip2, pip3, and pip3.6 all point to pip3.6.
[04:32] <lotuspsychje> laudecay: your ubuntu version please?
[04:32] <laudecay> i am not entirely sure how it got this way, may not have been entirely me too
[04:33] <fullstack> Its system wide not just my terminal, except for graphical X11 application
[04:33] <laudecay> is there a reason lsb_release isn't working
[04:33] <laudecay> wtf
[04:34] <lotuspsychje> plz no cursing here laudecay
[04:34] <laudecay> okie
[04:34] <lotuspsychje> laudecay: lsb_release -a
[04:34] <laudecay> lsb-release: command not found <- i just installed lsb and lsb-release
[04:35] <laudecay> something is horrendously broken
[04:35] <laudecay> ok im stupid
[04:35] <lotuspsychje> laudecay: your forgot the -a
[04:35] <laudecay> never mind that was an underscore help
[04:35] <laudecay> Ubuntu 16.04.5 LTS
[04:36] <laudecay> nah i had a dash not an underscore
[04:36] <lotuspsychje> laudecay: ok, so did youa dd external ppa's to your system?
[04:36] <lotuspsychje> add
[04:36] <laudecay> i did, i have one that gets me python3.6
[04:36] <lotuspsychje> we dont support external ppa's here laudecay
[04:36] <laudecay> what should i do to get python3.6 then?
[04:36] <lotuspsychje> laudecay: we advice to go back to vanilla sources.list to unscramble your system
[04:37] <lotuspsychje> !ppapurge | laudecay
[04:37] <lotuspsychje> laudecay: we also advice to stick to package versions, meant for your ubuntu version
[04:37] <laudecay> how do i list ppas? sorry im an arch user idk much about aptitude
[04:38] <fullstack> What in Ubuntu would make the underscore invisible? This is completely bewildering me and causing lots of bugs and headaches
[04:39] <fullstack> I have to be able to type, using my keyboard, and see the key, on the display. Its a fundamental concept in computers
[04:39] <laudecay> fullstack: font bug
[04:39] <laudecay> definitely a font bug
[04:40] <laudecay> if it was a parsing thing your system would definitely be imploding a lot more than it is right now
[04:40] <fullstack> I've been googing this for hours and not found a thing. Just tried "font bug" with all my queries... nothing useful
[04:41] <laudecay> https://github.com/Microsoft/vscode/issues/38133
[04:41] <laudecay> https://github.com/SWI-Prolog/issues/issues/42
[04:41] <laudecay> i googled "font rendering bugs ubuntu underscore"
[04:41] <laudecay> lots and lots of results! go play
[04:42] <fullstack> I read those already
[04:42] <fullstack> I am not using vs code. This is Xterm or even this chat terminal Hexchat. Everything except chrome/firefox
[04:42] <laudecay> chrome def has weird font rendering going on
[04:42] <laudecay> i remember when i had a rendering bug it didn't affect chrome
[04:43] <laudecay> hm this second one thinks you should mess with resolution settings
[04:43] <laudecay> i think that that's a likely culprit, make sure you're using the right resolution for your monitor.
[04:43] <laudecay> or try another default font
[04:45] <fullstack> ok yeah if I change the font on xterm, and change it back, it displays it
[04:45] <fullstack> now I can see _
[04:45] <fullstack> _ <-- don't see it in hexchat though
[04:46] <laudecay> seriously have you looked at your resolution
[04:51] <fullstack> I can't change my resolution. It is fixed at 4k
[04:53] <laudecay> huh idk then
[05:29] <neildugan> I am trying to get simple-scan to work properly.... if I run simple-scan as root (i.e. using sudo) it work fiine... but when run as a user it fails to find the scanner ... how can I fix this?
[05:43] <energizer> fullstack: i've had that issue before. i just changed my fonts to something else and it worked.
[05:46] <lotuspsychje> neildugan: is your system up to date to latest?
[05:57] <Triffid_Hunter> neildugan: probably need to add yourself to a group, plugdev maybe, or fiddle with udev rules so the thing's device node gets appropriate permissions
[06:04] <neildugan> lotuspsychje, yes
[06:06] <lotuspsychje> neildugan: i had an issue before like that, then simple-scan offered me to find the driver for the scanner button
[06:07] <neildugan> Triffid_Hunter, adding myself to the plugdev group had no effect
[06:07] <neildugan> lotuspsychje, I don't think it a driver problem... it works as root
[06:08] <lotuspsychje> neildugan: is it a combo printer/scanner?
[06:10] <Triffid_Hunter> neildugan: if it works as root then it's a permissions problem. strace can help with those
[06:10] <neildugan> lotuspsychje, yes it is
[06:11] <neildugan> Triffid_Hunter, I tried strace but it output 100's of lines most of which I don't know how to read
[06:12] <lotuspsychje> neildugan: can you open simple-scan from terminal? see what it says?
[06:12] <talx> hello guys,
[06:12] <neildugan> lotuspsychje, nothing
[06:12] <talx> did anyone ever used
[06:12] <talx> https://help.ubuntu.com/community/RemoveOldKernels#Configure_Unattended_Upgrades_to_Remove_Unneeded_Kernels_Automatically
[06:12] <lotuspsychje> talx: welcome, what can we do for you?
[06:12] <talx> ?
[06:12] <lotuspsychje> talx: unattended upgrades has now auto kernel cleanup
[06:12] <talx> I'm having an issue with 30-40 virtual machines
[06:13] <talx> they have /boot of 500mb
[06:13] <lotuspsychje> talx: wich ubuntu version please?
[06:13] <talx> and old kernels taking all the space
[06:13] <talx> 16.4
[06:13] <talx> 16.04
[06:13] <lotuspsychje> talx: up to date to latest?
[06:14] <lotuspsychje> neildugan: could try on another user perhaps?
[06:14] <lotuspsychje> neildugan: might be indeed a permission thing as Triffid_Hunter mentioned
[06:14] <talx> lotuspsychje: I'm not sure but I think not
[06:15] <lotuspsychje> talx: update your systems to latest 16.04.5 and try to enable unattended-upgrades
[06:15] <talx> what if there is not enough space in boot
[06:15] <lotuspsychje> !kernelcleanup | talx for manual
[06:15] <talx> I know how to remove actually
[06:15] <lotuspsychje> read the link talx
[06:16] <talx> would you think its a better option then making a new /boot partiton ?
[06:16] <lotuspsychje> no
[06:17] <talx> had to do it yesterday
[06:17] <talx> but what if I can't update those machines ?
[06:17] <lotuspsychje> talx: its not safe to not-update
[06:17] <talx> yeah but I meant
[06:17] <talx> to unattended upgrades
[06:18] <talx> if its not latest and I can't do an update?
[06:18] <talx> its says 16.04.04
[06:18] <talx> atm
[06:18] <lotuspsychje> talx: sudo apt update && sudo apt full-upgrade please
[06:19] <lotuspsychje> talx: always keep your systems to latest
[06:19] <talx> won't I face new bugs?
[06:19] <talx> heh :P
[06:20] <talx> I love to upgrade but its always abit scary
[06:20] <cim209> how is the performance of ubuntu mate 18 on a 2007 machine?
[06:20] <lotuspsychje> talx: you need to change the thinking, its when you leave behind on older versions that security flaws can occur
[06:21] <lotuspsychje> cim209: what kind of specs mate?
[06:21] <talx> okay I'm making the upgrade
[06:21] <talx> only because its a test machine
[06:21] <talx> heh
[06:21] <cim209> 2.5gb ram intel core 2 duo
[06:21] <lotuspsychje> cim209: think that will run fine, 18.04 has also a minimal option in setup you could choose
[06:22] <cim209> hmm, i'm running 16.04 right now
[06:23] <cim209> what would be the benefit of upgrading to 18?
[06:23] <AHemlocksLie> I'm trying to install the AMD drivers for my RX 580 graphics card, and the installer fails to install a package, libdrm-amdgpu-common. It recommends I run apt --fix-broken install to fix it, but that doesn't work. Anyone familiar with this issue?
[06:23] <lotuspsychje> cim209: its the users choice
[06:23] <lotuspsychje> cim209: 16.04 has still time of support so, you choose whats important to yourself
[06:24] <cim209> is there any performance gain in 18?
[06:24] <lotuspsychje> cim209: 18.04 has newer kernel, newer packages, for speed ask in #ubuntu-mate please
[06:24] <cim209> ok
[06:25] <lotuspsychje> cim209: i would think mate is already lightweight..speed differences will be close
[06:25] <lotuspsychje> AHemlocksLie: wich ubuntu version please?
[06:26] <AHemlocksLie> lotuspsychje, 18.04
[06:26] <lotuspsychje> AHemlocksLie: https://linuxconfig.org/how-to-install-the-latest-amd-radeon-drivers-on-ubuntu-18-04-bionic-beaver-linux
[06:28] <AHemlocksLie> lotuspsychje, that's pretty much what I did, but it failed to install one of its packages properly
[06:29] <lotuspsychje> AHemlocksLie: is it possible to hastebin the output?
[06:29] <cim209> lotuspsychje, what's the command to clean unused packages?
[06:29] <cim209> was it autoremove?
[06:29] <lotuspsychje> cim209: is apt autoremove what you need?
[06:30] <AHemlocksLie> lotuspsychje, just a sec, I'll get the initial output
[06:30] <cim209> lotuspsychje, what does that command do exactly?
[06:30] <lotuspsychje> cim209: it can help clean/sort apt problems
[06:30] <talx> lotuspsychje: an update is not an option right now but I will check about the unattended, though regardless to security updates, the unattended removal tool should work
[06:30] <talx> right?
[06:30] <lotuspsychje> cim209: its safe to do
[06:31] <cim209> ok
[06:31] <cim209> i'll run it this one time cause i was installing some themes and stuff
[06:31] <AHemlocksLie> lotuspsychje, https://paste.ubuntu.com/p/QjyF7Trtft/
[06:31] <talx> You can remove old kernels with a simple autoremove command in a terminal:
[06:31] <talx> sudo apt-get autoremove --purge
[06:31] <talx> thats from the lotuspsychje shared
[06:31] <talx> from the link*
[06:32] <lotuspsychje> talx: im not sure if it will work on a non-updated system, i would strongly advice to update your systems..
[06:32] <cim209> one more thing, why run apt-get, sudo apt <command> runs fine
[06:32] <AHemlocksLie> Hmmm looks like it's conflicting with ids-amdgpu...
[06:33] <lotuspsychje> AHemlocksLie: i think its fighting with versions: Unpacking amdgpu-dkms (18.30-641594) over (18.20-606296) ...
[06:33] <AHemlocksLie> ids-amdgpu is listed in the packages that were automatically installed and no longer required, so I think I can just rip it out and continue...
[06:35] <AHemlocksLie> ...How do I remove a package while I'm having dependency issues? When I try, it throws a fit about unmet dependencies, which I could fix if it'd let me REMOVE THE CONFLICTING PACKAGE
[06:35] <lotuspsychje> cim209: another way to test, is trying a 18.04.1 mate liveusb, and see for yourself if you like? like=>install
[06:35] <cim209> indeed
[06:35] <lotuspsychje> AHemlocksLie: let me ask first, did you add external ppa's of any kind?
[06:35] <cim209> i'll wait until after my backups are done. i'm running 2 shell scripts to ssh into my servers and download the backups
[06:35] <cim209> running on cron
[06:35] <AHemlocksLie> lotuspsychje, not as a separate step in this install, but the logs suggest the install script sets up something like a ppa for its own use kinda
[06:38] <AHemlocksLie> lotuspsychje, although it looks like I might have another PPA, but I don't think it should be serving any of the relevant packages
[06:38] <lotuspsychje> AHemlocksLie: did you do the update & upgrade part in the link i provided?
[06:40] <lotuspsychje> AHemlocksLie: and a sudo apt autoremove could do some magic too
[06:40] <AHemlocksLie> lotuspsychje, that's not part of installing the proprietary drivers, but I did do and update to at least refresh the package list
[06:40] <AHemlocksLie> lotuspsychje, and autoremove is ALSO broken until I can fix this dependency issue
[06:41] <AHemlocksLie> apt is basically refusing to work until I fix this, but... I need apt to work to fix it.
[06:41] <AHemlocksLie> I'm 99% sure I know what I need to remove, but I can't touch it. It'd probably be removed by autoremove, yeah, but I can't do that or specifically remove the problem package
[06:41] <lotuspsychje> AHemlocksLie: could it be your kernel is still at .29?
[06:42] <AHemlocksLie> lotuspsychje, yeah, 4.15.0-29-generic
[06:42] <AHemlocksLie> according to uname -r
[06:43] <lotuspsychje> AHemlocksLie: always update your system to latest before installing packages like this, to avoid dependecie nightmares
[06:43] <cim209> would be wise to upgrade from 16 to 18?
[06:43] <lotuspsychje> cim209: you already asked that..
[06:43] <cim209> i mean is it safe?
[06:44] <lotuspsychje> cim209: that depends what you mean by safe?
[06:44] <AHemlocksLie> lotuspsychje, I'll definitely keep that in mind next time
[06:44] <AHemlocksLie> In the meantime, does anyone know how to unscrew apt?
[06:44] <cim209> it won't brick my system?
[06:44] <lotuspsychje> AHemlocksLie: can you rollback, perhaps try a purge, then update system and try again?
[06:45] <lotuspsychje> cim209: make a good backup before you upgrade..
[06:47] <ducasse> AHemlocksLie: can you pastebin the full output of 'apt install -f'?
[06:47] <cim209> ok
[06:49] <AHemlocksLie> Okay, pretty sure I fixed it. Found something that says apt is totally broke in such a situation, and you have to use dpkg, which I had COMPLETELY forgotten about.
[06:50] <ducasse> AHemlocksLie: dpkg can often fix messes apt can't resolve
[06:51] <AHemlocksLie> ducasse, I didn't run that specifically, but I'm guessing it gets the same error message as literally every other conceivable apt command got, immediately nagging about the broken dependencies
[06:53] <lotuspsychje> AHemlocksLie: amd web says amdgpu-pro-uninstall to uninstall
[06:53] <talx> lotuspsychje I want to try the unattended kernels removal
[06:53] <ducasse> AHemlocksLie: it will try to resolve the problem, but if it can't it will usually tell you which package(s) to purge with dpkg
[06:53] <talx> did you say its already enabled ?
[06:53] <lotuspsychje> talx: check in software&sources and enable it, after that kernels will auto remove
[06:54] <lotuspsychje> talx: but again, i dont know if it works on non-updated systems..
[06:54] <talx> hmm
[06:54] <AHemlocksLie> lotuspsychje, which would have helped, I bet, if it had come with it. For whatever reason, it did not.
[06:54] <talx> Note: This way will not remove all automatically installed old kernel providing packages as fallback versions are kept; the list of kept kernels is maintained and automatically updated in the file /etc/apt/apt.conf.d/01autoremove-kernels as a list of matching regular expressions.
[06:54] <talx> I don't really get this part
[06:55] <AHemlocksLie> ducasse, it just told me to try apt --fix-broken install, which led to the exact same error
[06:55] <AHemlocksLie> But it's installed! So I really appreciate you guys trying to help, lotuspsychje and ducasse
[06:55] <lotuspsychje> !yay | AHemlocksLie
[06:56] <lotuspsychje> AHemlocksLie: now quick, update your system :p
[06:57] <AHemlocksLie> lotuspsychje, lol I'm all over that
[07:06] <lotuspsychje> talx: just make it easy for yourself, update system to latest .5 perhaps move to the hwe kernel and enable unattended upgrades
[07:17] <talx> lotuspsychje: what is hwe kernel
[07:17] <talx> hardware enablement
[07:17] <talx> :p
[07:17] <ducasse> !hwe
[07:18] <talx> something is not clear for me though
[07:19] <talx> if the machine is not connected to the internet how does it reproduce new kernels ?
[07:21] <fullstack> energizer, do you remember how you changed your font?
[07:22] <energizer> fullstack: there's a default font menu somewhere in the gnome settings
[07:22] <energizer> or unity or whatever
[07:22] <fullstack> i'm using fluxbox
[07:22] <fullstack> it also happens when I don't use x11, like screen terminal
[07:22] <energizer> i dont know what that is, but probably there's a 'set default font' menu somewhere
[07:36] <talx> lotuspsychje: are you still here?
[07:37] <lotus|NUC> talx: yes
[07:37] <lotus|NUC> talx: whats up?
[07:39] <talx> I found something else
[07:39] <talx> installing byobu
[07:39] <talx> and the purge-old-kernels
[07:39] <talx> http://ubuntuhandbook.org/wp-content/uploads/2016/05/purge-old-kernel-script.jpg
[07:40] <talx> what do you think of it ?
[07:40] <lotus|NUC> talx: thats old
[07:40] <talx> to stick with unattended then ?
[07:42] <lotus|NUC> talx: i told you few times what i think mate, update your systems, enable unattended upgrades, and auto kernel remove will be active
[07:42] <talx> okay
[07:42] <talx> sorry, ty.
[07:44] <lotus|NUC> !hwe | talx consider also
[08:40] <xrandr_mac> Hi there. I am having an issue connecting a gluster node on ubuntu to a gluster server running CentOS
[08:40] <xrandr_mac> I keep getting PEER_CONNECT and PEER_DISCONNECT events in the logs
[08:48] <ducasse> xrandr_mac: try #ubuntu-server
[08:48] <xrandr_mac> Ok, thanks
[08:51] <blackpope> hey peeps!
[08:51] <blackpope> anyone there?
[08:54] <zaggynl> blackpope: welcome to IRC, most people are idle and don't monitor the chat, if you have a question fire away, someone may respond at any time
[08:55] <blackpope> Thanks for that zaggynl! Just a quick dobut, i am new to IRC, i am using a term based client! Are there any cool things you might wanna share with me regarding IRC?
[08:56] <ryuo> Welcome to IRC. Time for the traditional greeting.
[08:56]  * ryuo slaps blackpope around for a bit with a large trout.
[08:57] <blackpope> Also, Can you please guide me how can i make a private channel where I can privately talk to my friend?
[08:57] <blackpope> haha ryuo thanks!
[08:57] <ryuo> > /msg
[08:57] <ducasse> blackpope: like what? for irc support, ask in #freenode, this channel is for ubuntu support
[08:57] <ryuo> for 1 to 1. if you need a group, make a channel.
[08:59] <blackpope> so how do i make a channel? /CONNECT -add channel? like this, idk what i have written xD
[09:00] <ryuo> > /join
[09:00] <blackpope> or wait do i just /join <myrandomchannelname> ?
[09:00] <ryuo> It may or may not work.
[09:01] <ryuo> Freenode restricts channel names.
[09:01] <blackpope> and then /msg chanserv REGISTER #channelname password description
[09:01] <ryuo> or w/e. I don't have any channels registered.
[09:01] <blackpope> alright, are there any cool things on this channel?
[09:02] <ryuo> Uh... it's a support channel. What do you expect?
[09:03] <blackpope> idk, I am a bash enthusiast, and wanted to share my github work! Was expecting some feedback, idk why but yeah thats why i joined
[09:03] <blackpope> lol
[09:03] <zaggynl> There might be some bash specific channels
[09:03] <blackpope> alright i'll lookup online, nice to meet you guys, will come back later with some ubuntu specific doubts!
[09:03] <zaggynl> You might find more feed back at other social media like reddit
[09:03] <blackpope> cheers
[09:04] <blackpope> signing off!
[09:04] <zaggynl> bye
[09:09] <paul98> does any one use sendemail? i'm trying to send a email from command line but the password has ;!/@@} in the password and when i try to send it i get event not found with half the password from ! I tried using incasing it with " / ` and ' but then i get auth fail
[09:10] <ikonia> ubuntu dosn't use sendmail by default
[09:10] <ikonia> have you removed postfix
[09:11] <paul98> i haven't removed postfix
[09:11] <paul98> i know it doesn't by default
[09:12] <ikonia> postfix can't exist with sendmail
[09:12] <ikonia> so if you've not removed postfix how are you using sendmail
[09:13] <paul98> I said sendemail
[09:13] <paul98> not sendmail
[09:14] <ikonia> oops, sorry
[09:14] <paul98> no worries
[09:14] <paul98> basically I want to send a email based on someone logging on, so was going to use sendemail
[09:15] <paul98> but if something more default ?
[09:15] <lesshaste> is there any way to access onedrive ? I thought "online accounts" through gnome might do it and it has a long list but not onedrive it seems
[09:16] <ryuo> lesshaste: browser. otherwise I don't know of one.
[09:16] <lesshaste> ryuo, can you drag and drop files via the browser?
[09:17] <ryuo> lesshaste: No idea. You can give it a try.
[09:17] <ryuo> lesshaste: I recall drag & drop being a feature of modern browsers, but it's still up to the websites whether it works.
[09:17] <ducasse> lesshaste: there's a software called duplicati that can use it for storage of backups, but that's all i know about
[09:17] <lesshaste> ryuo, I will try that
[09:18] <lesshaste> ducasse, interesting.. given the list that online accounts gives you, I can't see why onedrive isn't there
[09:18] <TJ-> paul98: "!" is interpreted by bash - enclose the password in single-quotes
[09:19] <lesshaste> ryuo, seems drag and drop works!
[09:21] <ryuo> lesshaste: all about support. these are proprietary cloud systems, so native clients may not always exist.
[09:27] <paul98> single quotes being ' TJ- ?
[09:32] <paul98> still gives me auth failed, although I 100% know it's correct and i'm using this emaila ccount for other email relays etc and they are working
[09:36] <satrioirc> gbellinoz, you still up?
[09:37] <satrioirc> i fell asleep :D
[09:43] <TJ-> paul98: you can try escaping the ! using \! instead
[09:45] <paul98> it gives me auth failed if in ' ' if i remove it it tells me file not found
[09:47] <geirha> escaping the ! with \ won't work
[09:48] <geirha> the only way to escape ! from history expansion is with single quotes
[09:48] <geirha> or by simply disabling history expansion, of course. set +H
[09:48] <Triffid_Hunter> msmtp allows me to use a credentials file, no need to stuff around with shells eating stuff and putting your password in ps output
[09:49] <paul98> i used an account with a normal password and that come up with auth failed aswell
[09:49] <paul98> so i don't think password is the issue here
[09:50] <jost> Hi! Is there a way to increase the space of a disk temporarily, without unmounting that disk? Maybe by attaching some RAM to it? Situation is that a long-running process has almost exhausted the disks space, and I really don't want it to error because of disk full.
[09:50] <jost> There is nothing on that disk I could delete
[09:50] <Triffid_Hunter> paul98: perhaps your smtp server doesn't like your passwords?
[09:51] <paul98> it's gmail
[09:51] <Triffid_Hunter> jost: what filesystem? btrfs? you can add more storage
[09:51] <paul98> we use this account else where so i know it works
[09:51] <Triffid_Hunter> paul98: yeah you need to set up a separate app password and enable low security apps or something
[09:51] <[twisti]> i somehow messed up my network setup. any attempts to ping outside ips (as in not lan) result in 'network unreachable'
[09:52] <[twisti]> any advice on how to debug ? i can ping lan ips and i can resolve names to ips fine
[09:52] <pumba> trying to use an ethernet usb adapter on 18 bionic getting [  225.834205] usb 1-2: device descriptor read/64, error -71
[09:52] <pumba> can someone help
[09:52] <Triffid_Hunter> [twisti]: you have no default gateway
[09:52] <pumba> [  225.054133] usb 1-2: device not accepting address 10, error -71
[09:52] <[twisti]> Triffid_Hunter: how would i go about repairing that ?
[09:52] <TxRaspPi> gmail only likes ssl passwords
[09:52] <Triffid_Hunter> pumba: that usually means that either the usb cable, the device, or the usb port is damaged
[09:53] <[twisti]> (ideally on a permanent base, rebooting has not helped)
[09:53] <jost> Triffid_Hunter: ext4
[09:53] <Triffid_Hunter> [twisti]: add one.. sudo ip route add default via (router ip)
[09:53] <Triffid_Hunter> [twisti]: permanent? fix your dhcp server
[09:53] <Triffid_Hunter> jost: no can do then
[09:53] <[twisti]> it is incredibly unlikely that this is a problem by my dhcp server
[09:53] <jost> Triffid_Hunter: D'oh
[09:54] <[twisti]> i was just messing with my network settings when it broke, what are the odds that the dhcp server randomly broke at the same time ?
[09:54] <gloomy> lotus|NUC: Hey there :) Sorry, I didn't see your last messages drom yesterday
[09:54] <Triffid_Hunter> [twisti]: wireshark will tell you for sure.. if your dhcp responses have gateway but your dhcp client isn't using them, then something very weird is happening with your networking client setup
[09:54] <gloomy> What was that laptop-tools thing you mentioned?
[09:54] <Triffid_Hunter> [twisti]: heh you were messing with network settings? set something to manual perhaps?
[09:55] <[twisti]> Triffid_Hunter: i dont know, it was a docker image that messed up somehow, im not entirely sure what it did
[09:55] <[twisti]> but thats of course entirely possible
[09:58] <[twisti]> the dhcp lease includes a gateway, so somehow the docker container must have overridden it. can i somehow say 'go back to using the dhcp settings' ?
[09:59] <Triffid_Hunter> [twisti]: depends what network setup you're using.. personally I'd just dhcpcd -n enp3s0 and I'd be set, but that's 'cos I'm using openrc+dhcpcd here
[09:59] <Triffid_Hunter> with systemd/networkmanager I have absolutely zero clue
[10:02] <ikonia>  /join #theforeman
[10:02] <ikonia> oops
[10:04] <[twisti]> Triffid_Hunter: thank you, that worked (after manually adding the gateway in order to download dhcpcd and the nrunning the command with my interface name (enp0s8)
[10:05] <gloomy> I'm trying to fix an issue where closing the lid of my laptop doesn't trigger 'suspend'
[10:05] <paul98> Triffid_Hunter: but it works from all other services like gitlab, teamcity our own website etc.
[10:06] <Triffid_Hunter> paul98: well perhaps sendemail is broken? try msmtp
[10:07] <gloomy> I'm quite new to ubuntu's architecture so I'm not quite sure where to look exactly. I read that I should modify /etc/systemd/logind.conf, so I set the lines `HandleLidSwitch=suspend` and `HandleLidSwitchDocked=suspend
[10:07] <gloomy> `
[10:07] <gloomy> But it didn't solve the issue
[10:08] <gloomy> running acpi_listen shows `button/lid LID close` and `button/lid LID open` upon closing/opening the lid, so the issue isn't there
[10:09] <gloomy> (I'm running budgie(bionic) on a macbook pro btw)
[10:09] <MacroMan> libtirpc1 updated to version 0.2.5 today and now access to my NFS share is slow, in the region of 12mb/s. It was easily in the region of 60-80mb/s before this update. What could be happening?
[10:09] <gloomy> Ideas?
[10:15] <paul98> Triffid_Hunter: must be as msmtp works :D
[10:15] <paul98> thank you i been banging my head for hours lol
[10:15] <gloomy> On a side note: suggestions for a nice, modern terminal emulator?
[10:16] <gloomy> I'm used to iTerm3 on macOS, something similar would be nice
[10:18] <SimonNL> https://github.com/systemd/systemd/issues/7137     gloomy any useful info there ?
[10:18] <zamba> i have a running system that i'm able to access remotely using ssh.. i already have a running X session there.. how can i connect to this session remotely?
[10:20] <gloomy> SimonNL: Going through it, it may take a while as I'm not familiar with most of the utilities ;)
[10:20] <SimonNL> me neither :)
[10:22] <gloomy> Hmm, this entry is suspicious:
[10:22] <gloomy>      Who: luca (UID 1000/luca, PID 1426/gsd-power)
[10:22] <gloomy>     What: handle-lid-switch
[10:22] <gloomy>      Why: Multiple displays attached
[10:22] <gloomy>     Mode: block
[10:22] <gloomy> (hope I'm not above the paste limit :p)
[10:23] <gloomy> Ah, but that's one of the possible blocks, it doesn't mean it's currently active
[10:24] <blackflow> gloomy: anything more than one line should go to paste
[10:24] <blackflow> *bin
[10:24] <gloomy> Duly noted, sorry :-)
[10:33] <gloomy> SimonNL: Yes, seems to be the same issue
[10:33] <gloomy> So now I know the responsible is `gsd-power`
[10:34] <gloomy> *the mystery unfolds*
[10:34] <SimonNL> gloomy: use ubuntu options to handle lidclose/open
[10:34] <SimonNL> power management or screen saver I think
[10:34] <gloomy> My flavor (budgie) doesn't seem to offer those settings in the GUI
[10:35] <arndff> Hello! I am new to Ubuntu and have a problem with "sudo apt-get update". Didn't find a solution myself, so would like you to help me.
[10:35] <d2r2> HI
[10:35] <arndff> Here you can see the last few lines which cause the problem: http://snippi.com/s/kkg4lgx
[10:35] <arndff> Thank you in advance, lads. Cheers!
[10:36] <d2r2> I am trying to install specific version of sqlite what should be the command?
[10:36] <d2r2> I want sqlite 3.7.8 as the book uses that one and recommends it
[10:37] <d2r2> I am using learn sql the hard way book so please help me with that if u have any idea
[10:54] <d2r2> no body? :(
[10:55] <enzotib> d2r2: you cannot install from repos the version you want, ubuntu doesn't work this way
[10:56] <arndff> @enzotib: could you answer my question as well?
[10:56] <d2r2> enzotib: so can I only have the latest version? TY for reply :)
[10:57] <enzotib> d2r2, I don't think it will be much of a difference
[10:57] <enzotib> arndff, you have a PPA configured, why? Remove it
[10:57] <d2r2> arndff: Hi arndff What are you working on?
[10:57] <d2r2> enzotib: ok TYSM have a great day
[10:58] <arndff> Have a problem with sudo apt-get update and that's the last few lines which cause the problem: http://snippi.com/s/kkg4lgx
[10:58] <arndff> @enzotib, so if I remove that PPA and later on want to update my themes, what shall I do?
[10:58] <arndff> Sorry for the noob questions but as I said I am very new to Linux and Ubuntu :)
[10:59] <d2r2> arndff: Sorry I'm a noob as well lol. I hope someone helps you
[11:00] <arndff> Cheers
[11:00] <arndff> No worries, will wait to see what enzotib can say
[11:00] <enzotib> arndff, I can help to remove the PPA, that seems to be broken, cannot say how to update
[11:00] <arndff> I succeded in removing it
[11:01] <arndff> Opened 'Software and Updates' -> 'Other Software'
[11:01] <arndff> Then removed that PPA and sudo apt-get update goes well now
[11:01] <arndff> Ok, thanks
[11:02] <arndff> And one last Q -- just seen that GNOME 3.30 has been released. Is it available to download or not yet?
[11:04] <blackflow> arndff: package versions are tied to ubuntu releases with some minor exceptions, like firefox for example. so you can't just randomly install any version of any package. doesn't work like that. if you must, then you need to use containers
[11:04] <blackflow> (which is probably not going to work for something so core like a desktop environment)
[11:05] <arndff> Let's see whether I got that right. When Ubuntu decides to release a new version, most prob it'll use GNOME 3.30, right?
[11:06] <arndff> Until that, I have to stick with Ubuntu 18.04 and Gnome 3.28
[11:08] <blackflow> arndff: no idea. right, gnome in the next release of Ubuntu, which is 18.10 pending this October is 3.28. Which version will be released with 19.04 in April 2019? no guarantees it'll be 3.30.
[11:08] <blackflow> *right now
[11:09] <arndff> I see. Thanks for your explanations!
[11:10] <blackflow> arndff: note there are "snaps", which are containerized applications, designed so that vendors can release at their own pace, regardless of what's supported by official Ubuntu maintainers, via apt.
[11:10] <blackflow> !snaps | arndff
[11:11] <blackflow> meaning, who knows, gnome might become a snap and you'll be able to enjoy latest release as it comes out upstream. though personally I doubt that, a DE is very deep in the system and one can't just randomly use different versions on demand
[11:16] <BluesKaj> Hey folks
[11:19] <d2r2> I learn so much randomly just reading here haha
[11:19] <d2r2> and it's not boring at all
[11:20] <d2r2> who knew there firefox is a snap
[11:21] <d2r2> are all the software that I can update in the ubuntu software center, snaps?
[11:22] <blackflow> d2r2: no, not all are snaps
[11:23] <d2r2> blackflow: ah! ok, so I do I find out which one are?
[11:24] <blackflow> d2r2: through software center? I have no idea, never used the thing. through command line, there's 'snap' command and its functions like list, search, etc...  or use snapcraft.io/store
[11:26] <d2r2> so i did snap list but it doesn't show firefox. why is that? It shows VLC and OBS
[11:27] <blackflow> d2r2: I see "Source: Snap Store" might be the clue, in he software center, when you click to view an item.
[11:27] <albech> trying to get 18.04 running on my old mac mini. I managed to get the installation working with 'nomodeset' boot option. This resulted in the system not being able to detect multiple monitors. Then i installed lightgdm and is not able to get the login manager running without the nomodeset, but i cannot get x to start after login. Any suggestions?
[11:27] <ducasse> d2r2: because the firefox snap is not installed. try 'snap find firefox'
[11:27] <blackflow> d2r2: so you don't have FF installed from snaps, but from regular package
[11:28] <blackflow> Not sure what will happen if you install Firefox from snaps in parallel with Firefox from packages.... sounds... fishy :)
[11:29] <albech> the system is running on a geforce 320M, but using nouveau drivers
[11:31] <d2r2> blackflow: yes "snap store" found out but it's not always necessarily true right? like it shows blender is from ubuntu-bionic-universe  but we can upgrade it mannually too
[11:34] <d2r2> yeah I installed firefox from cmd line. Would it's source be snap store if installed from software center?
[11:35] <hateball> d2r2: I think that is the default yes, same with Steam
[11:35] <hateball> and you probably dont want that, since the snap doesnt create proper udev rules like the apt package does
[11:36] <blackflow> d2r2: there's an overlap in that some applications exist both in snaps and in regular packages via apt. that's why I dislike the software center and prefer being explicit in what I want to install and how, on the command line.  snap vs apt.
[11:38] <d2r2> blackflow: yeah right. good to know though. I never discriminated apps like that. Now I know :)
[11:39] <blackflow> d2r2: that's a big problem, IMHO. because with snaps, one should be VERY careful which snaps they install, as anyone can upload a snap to the store. we already had an issue wiht.... "malware" being present in one.
[11:40] <blackflow> snaps are excellent idea, but they're removed from official ubuntu maintainership so one must do extra checks and trust the vendor.
[11:42] <d2r2> But if we compare them with android or ios store aren't all mobile apps some kind of snaps? These stores seem to handle them very well I guess. So does chrome web store right
[11:44] <blackflow> d2r2: snaps are containers, like flatpaks. that means they carry all the required dependencies and operating system parts (minus the kernel) that they need to run. there are some exceptions to that with "runtime" snaps providing common facilities, but the general gist is that they're containers, and thus isolated and independed filesystems on their own, from the main OS.
[11:45] <blackflow> I don't know how android or ios software delivery compares to that.
[11:45] <ilias_gr> Hi all. I just cleanly install on a notebook with an Intel(R) Celeron(R) CPU  N2840  @ 2.16GHz with 4 gb ram the new 18.04.1 distro and I realized that in idle position the system needs 570 mb ram without any application open. On the other hand on an another notebook with exactly the same hardware's configuration (an Intel(R) Celeron(R) CPU  N2840  @ 2.16GHz with 4 gb ram) with 16.04.5 on idle position needs only 270 ram approximately. I ch
[11:45] <ilias_gr>  ecked 'Session and Start up' and both systems have the same autostart applications. Does new distro need more ram to run or do I have to make any other configuration to reduce the ram the new system needs ?
[11:46] <blackflow> ilias_gr: yeah. xenial is lightdm+unity, bionic is gdm+gnome. full gnome. gnome is a RAM eater. gnome loooooooves RAMs.
[11:46] <gloomy> Hi again :-) I'm investigating an issue where closing my laptop lid doesn't trigger sleep
[11:47] <gloomy> And delving into how power management works in ubuntu at the same time (Installed it 3 days ago)
[11:48] <gloomy> Where can I see logs of stuff like "suspended", "hibernated", etc.?
[11:48] <blackflow> gloomy: journalctl
[11:48] <gloomy> I found that `acpi_listen` displays physical actions like closing the lid and pressing the power button
[11:48] <blackflow> gloomy: you mentioned earlier you fiddled with logind.conf? I've heard/read somewhere that changes there might be required in such situations
[11:48] <gloomy> blackflow: Thanks, will look
[11:49] <ilias_gr> blackflow: thanks. any idea to bypass this and reduce memory's need recourses using xubuntu bionic distro?
[11:49] <blackflow> ilias_gr: xubuntu is xfce, that should be using much less RAM on start up
[11:50] <gloomy> Yes, but the issue that Simonious (I think) pointed to seemed to imply that those settings were overridden by gsd-power
[11:50] <blackflow> y'all should be more precise in your opening questions. ubuntu is gnome now. xubuntu is, always was, xfce.
[11:50] <ilias_gr> blackflow: sure it is and uses 570 on start up but this the double on comarison with xenial
[11:50] <blackflow> ilias_gr: so you're having more startup memory consumed, on xfce, between bionic and xenial?
[11:51] <gloomy> blackflow: so what do I find with journalctl as opposed to /var/logs?
[11:51] <blackflow> gloomy: that I don't know, sorry.
[11:51] <ilias_gr> blackflow: that;s right! that's the point !
[11:51] <blackflow> gloomy: I don't know about gsd-power, I mean
[11:51] <d2r2> blackflow: Ok so does that mean you need to add extra stuff in snaps to communicate with os than the apt? Does that makes the size larger?
[11:53] <blackflow> ilias_gr: dunno. could you pastebin   ps --sort -rss axu | head -n 20
[11:53] <blackflow> d2r2: apt is a tool that manages .deb packages. snap manages whole containers which are full OS trees of all the software required, MINUS teh kernel and minus common stuff installed with the "core" snap
[11:54] <blackflow> d2r2: so a .deb package only carries the files for the software in question, and maintainers scrips, configs, ....    a container is full OS tree.
[11:54] <gloomy> What's the go-to wiki/website for researching stuff like journalctl/systemd/etc ? most of the stuff I find on google dates many years back
[11:54] <gloomy> And manpages are a bit too beginner-unfriendly :p
[11:55] <ilias_gr> on both systems? when both are idle on start up?
[11:55] <blackflow> gloomy: google with filter to show last year's content only :)     ain't kiddin'   that's what I do
[11:55] <gloomy> huh
[11:55] <blackflow> ilias_gr: primarily on bionic. wanted to see what the main consumers of memory were
[11:55] <gloomy> ok :]
[11:55] <d2r2> blackflow: TY :)
[11:55] <ducasse> gloomy: i like the arch wiki, depending on what you're looking for
[11:55] <blackflow> gloomy: meanwhile the manpages should be up to date
[11:56] <gloomy> Manpages are quite terse for generic information like this :)
[11:56] <d2r2> You are awesome. have a great Dah
[11:56] <ilias_gr> blackflow: in a moment please as I am on xenial system right now and I have to boot the other system with bionic
[11:56] <blackflow> gloomy: and what information are you looking for?
[11:57] <blackflow> ilias_gr: in that case do both, so we can compare...
[11:57] <gloomy> anything that will help me learn how everything fits together :-)
[11:58] <blackflow> gloomy: then start with any tutorial no matter its age. manpages are for specific up to date details. and what ducasse mentioned, arch wiki is a great repository of mostly up to date knowledge.
[11:58] <blackflow> but it's arch and many things don't fit in ubuntu on face value / verbatim.
[11:58] <ilias_gr> blackflow: ok , I have to reboot also xenial system
[11:58] <gloomy> Alright, will do
[11:59] <blackflow> gloomy: the principal systemd developer's blog is also a great source of knowledge, as he explains the concepts and things they do.   http://0pointer.net/blog/
[12:00] <gloomy> Thanks. I don't even really know what systemd is so I think I'll start there ;)
[12:02] <blackflow> gloomy: it's primarily a whole repository of tools and utilities for a middleware between the kernel and userland applications. it's second function is that it's the main process manager / PID1 for modern linux distros. So it's important to know what you're reading, when you're reading, is it about the process manager / PID 1,  or about the whole middleware repo of additional, not strictly required,
[12:02] <blackflow> tools.
[12:02] <blackflow> and ignore the hate. most hate comes from total misunderstanding of what system the PID1 does and how it does it. people think it's drop-in replacement for the old sysv init, and it's totally not.
[12:02] <gloomy> So every process in ubuntu is a child of systemd?
[12:03] <gloomy> *every user process
[12:03] <blackflow> directly or indirectly, yes
[12:03] <blackflow> of PID1, always, be it system'd, sysv, ...
[12:15] <neredsenvy> Is there a way to watch service status messages ?
[12:16] <neredsenvy> The service who's status I want to watch does not have a log file.
[12:16] <blackflow> neredsenvy: can you elaborate?    systemctl status some.service   not it?   or   journalctl -u some.service     even -f with that
[12:17] <neredsenvy> Ah that works
[12:17] <neredsenvy> Thanks
[12:17] <blackflow> yw
[12:18] <ilias_gr> blackflow: https://pastebin.com/vyxEKaTz
[12:18] <blackflow> neredsenvy: btw you can have a log file by adding a regex to rsyslog. by default journal forwards all messages to rsyslog
[12:20] <blackflow> ilias_gr: yeah it seems newer versions eat up a bit more ram and that cumulatively totals up.   also look at that xorg. weird it's not in top 20 on xenial
[12:20] <blackflow> oh wait, it is.... just not at the top
[12:22] <blackflow> well.... bloatware, increasing codebase and feature creep as time passes. the scourge of any softare, FOSS or proprietary.
[12:22] <ilias_gr> blackflow: so, any solution ?
[12:23] <blackflow> ilias_gr: none other than removing stuff you don't really need. consider if you need a full DE even. perhaps a (tiling) wm based minimalism is what you'd like?
[12:25] <blackflow> I mean.... that weather applet eats 32MB of friggin' ram. do you need it? system load 25MB. that's 50+MB just like that!  *snaps fingers*
[12:25] <ilias_gr> blackflow: i can live with minimalism but i like XFCE as I use it for years until now. I think I run only basics
[12:26] <ilias_gr> blackflow: i understand...
[12:27] <ilias_gr> blackflow: i can live without weather although is useful but load monitor is really more useful
[12:28] <ilias_gr> blackflow: two more questions
[12:29] <ilias_gr> blackflow: xenial will receive updates till 2021 isn't it? but it will stop xfce updates on 2019 ?
[12:30] <blackflow> no, bionic will until 2021 and xenial up to next year
[12:30] <blackflow> for xubuntu that is.
[12:31] <ilias_gr> blackflow: oh ok ! so update is no way to avoid it.
[12:31] <blackflow> avoid what exctly?
[12:32] <ilias_gr> blackflow: avoid an update to bionic which seems to consume more ram
[12:32] <blackflow> looks like it.
[12:35] <ilias_gr> blackflow: OK. I 'll see what to do in recent future. Do you know why in bionic there is no directory '.Private' and '.ecryptfs' although the hd is encrypted? In xenial both are under /home.
[12:36] <blackflow> ilias_gr: Bionic doesn't do ecryptfs by default any more. not sure what happens when you upgrade from xenial to bionic, I guess it does support that use case
[12:37] <blackflow> full disk encryption with LUKS is the way forward.   also, check the release notes for other changes.
[12:37] <ducasse> it does
[12:37] <ilias_gr> blackflow: i cleanly install bionic from scratch
[12:39] <blackflow> ilias_gr: yeah, "encryption" then only means FDE with LUKS and LVM
[12:39] <ilias_gr> blackflow: i just compare both two systems and on xenial exist both dirs but on bionic no dirs are present
[12:40] <blackflow> those dirs are created by ecryptfs. ecryptfs is useful when you only want to encrypt some directories.
[12:40] <blackflow> in previous releases that meant encrypting only your $HOME
[12:42] <narakrish> I've 2 PCs running Ubuntu. The PCs are from different vendors. They have IO port interface via RJ11 connector. They are used for accessing the cash drawer. On one PC, I'm able to access the cash drawer at 0x48C. But the same does not work on the other PC. I do not have the spec for both cash drawer and the PC.  I would like to know how can I get the IO port address to access the cash drawer. I've tried cat /proc/ioports, but not ab
[12:42] <ilias_gr> blackflow: oh yes ! you have right ! i remeber that user have to check home directory encryption too . At this case /home will be encrypted too as an extra option. But i don't remember if i saw his option during installation. That's it anyway.
[12:44] <blackflow> narakrish: #ubuntu is not the best place for that question. some things to check first: ubuntu versions, kernel modules / drivers, are they the same between the PCs. otherwise you'll need specific hardware manuals for programming for them that low
[12:45] <ilias_gr> blackflow: thank you for advises today !
[12:45] <blackflow> you're welcoome
[12:46] <qsl> Hey,
[12:47] <qsl> If I want to encrypt some local files on my pc would you suggest veracrypt, cryptomator or something different?
[12:50] <tomreyn> qsl: if it's just a few files, and you dont need to change them regularly, tar + gpg2
[12:50] <tomreyn> qsl: if it's many files or you'll edit them regularly, a crypto container such created by dmcrypt-luks
[12:52] <blackflow> tomreyn: even a loopback mounted image file would suffice for the container, right?
[12:52] <tomreyn> sure
[12:52] <blackflow> I mean dmcrypt-luks works with block devices only afaik, or does it use other node types?
[12:53] <tomreyn> not that i know of, only block devices
[12:53] <blackflow> yeah. qsl, so if there's no room to spare for a dedicated partition, a loopback could be used
[12:53] <blackflow> I'll have to look up why ecryptfs was dropped and isn't recommended any more
[12:54] <tomreyn> but creating an image file is easy, and if you'll use nautilus or any other gvfs-backed graphical file browser, you'll be able to use the GUI to mount it
[12:56] <blackflow> oh hey, ext4 can now do encryption built-in  too
[12:56] <qsl> I only need a few files once in a while that i want to encrypt,
[12:56] <qsl> I only have a small 223gb ssd on this laptop so I dont have that much space
[12:56] <tomreyn> blackflow: i dont think this is stable, yet, or is it?
[12:56] <qsl> whops why did i double send this msg
[12:56] <qsl> oh nvm
[12:57] <blackflow> tomreyn: dunno. it's been years. i'll have to look it up
[12:57] <aglaia> hi
[12:57] <blackflow> qsl: mounting an image file via loopback with LUKS would be my preferred way. you can copy/carry/backup   one file, that's a filesystem internall, very easily.
[12:58] <blackflow> s/loopback with/loopback, and with/
[13:00] <aglaia> I am running an AWS EC2 instance with ubuntu 18 with a few simple process on it and very simple tasks. About once per day, randomly, the CPU goes to 100% without reason. It looks like it is kswapd that hits the CPU, during about 20 minutes, then all comes back to normal. I cannot figure out with there is this CPU spike.
[13:00] <blackflow> tomreyn: looks like "not yet"   https://askubuntu.com/questions/1029249/how-to-encrypt-home-on-ubuntu-18-04/1031509#1031509
[13:01] <blackflow> aglaia: does it correspond to apt-daily timers triggering?
[13:02] <blackflow> aglaia: systemctl list-timers   will show when individual timers have hit last. apt-daily by default is all over the place with large randomized delay
[13:02] <tomreyn> right, probably soonish
[13:02] <aglaia> blackflow i don't know, what is apt-daily timers ? this issue occurs randomly, never at the same hour.
[13:03] <blackflow> aglaia: right, see what I just posted
[13:03] <aglaia> ok, trying
[13:04] <blackflow> anyway, if kswapd is running during those spikes, then something is having a memory spike and swapd is on the call.
[13:04] <qsl> blackflow: i just read about loopback and LUKS a bit
[13:04] <qsl> but tbh, atm Im still learning to get used to working with the terminal ^^
[13:04] <qsl> do you maybe have a tutorial or some help site you could link me?
[13:04] <aglaia> blackflow looks like it doesn't correspond with daily timers. This morning this issue happend at 5:00 am, and systemctl list-timers shows nothing at this time
[13:05] <aglaia> blackflow indeed, but i have tried to log everything, i have all logs about CPU and memory at the moment the issue appears, but i can't know which process do it
[13:05] <aglaia> blackflow all seems ok
[13:06] <aglaia> i have been working on this bug 3 last days... getting desperate
[13:08] <blackflow> aglaia: what "all logs about CPU and memory"?  are you using sar?
[13:08] <aglaia> blackflow: yes, using sar, top and atop
[13:08] <aglaia> blackflow: i will show you the sar log
[13:08] <blackflow> yah that has 5 minute resolution. not sure it'll help
[13:11] <aglaia> blackflow: that is the sar log of yesterday morning, the issue started at 6:45 till 7:20 : https://pastebin.com/C5s9uST9
[13:11] <blackflow> qsl: I don't, you'll have to hit up google.   Use `truncate` to create an image file. Use `losetup` to create the loopback. Use cryptsetup to create a LUKS container atop of the /dev/loopX you created with losetup. use mkfs.ext4 or whatever fs you want, with /dev/mapper/...  device you opened for the LUKS container.    these are all hints to look up specific commands and how to use them.
[13:12] <aglaia> blackflow: it starts with %iowait during 15 minutes, then %system during 20 minutes
[13:12] <blackflow> aglaia: it's been a while since I used sar, does it not also store snapshots of process activity too?
[13:13] <aglaia> blackflow: no but i have logs of process with top, i will show you
[13:13] <blackflow> aglaia: oh wait... sudden large iowait could indicate storage layer issues. not uncommon with aws
[13:13] <aib> anyone able to set up certificate auto-renewal in 18.04 with letsencrypt and nginx?
[13:13] <aglaia> blackflow: ah yes ?
[13:14] <blackflow> in other words, something clogs the storage layer outside of your VM and then every little I/O becomes hugely delayed, everything grinds to a halt, errors start cropping up if it's taking longer than 120s. did you check dmesg for corresponding kernel errors about I/O being stalled?
[13:15] <blackflow> aib: yes, with dehydrated, my favorite "light" acme client
[13:15] <pragmaticenigma> aib: Yes, I was able to with the instructions provided by the letencrypt/certbot
[13:16] <aglaia> blackflow: no, i don't know about dmesg. I will try that. (and, if needed, this is the top log at the time of the issue : https://pastebin.com/pFWfEiKG )
[13:16] <aib> blackflow: pragmaticenigma: sorry, I forgot to add "without installing external packages"
[13:17] <aib> blackflow: I will check out dehydrated, though, thanks!
[13:17] <blackflow> aib: ehh no go. you need an acme client or write your own.
[13:17] <aib> pragmaticenigma: hmm, did you have to install certbot?
[13:17] <na2s> On 18.04, the "sleep" button on my USB keyboard puts the system in suspend mode, and I want to disable this.  Changing dconf settings like "button-sleep", "button-suspend", "button-hibernate" in org.gnome.settings-daemon.plugins.power has no effect.  Ideas?
[13:17] <pragmaticenigma> aib: What are you trying to do exactly? There is no way around installing external packages... certbot is a Python based script, so it leverages Python which is already installed on your computer anyways
[13:18] <na2s> I can't even figure out what daemon is causing it to suspend!
[13:18] <aib> blackflow: letsencrypt (ubuntu package now) does an okay job on its own. the webroot plugin seems to be able to renew automatically, too. I just need an nginx restart. and a mechanism to schedule all this
[13:18] <blackflow> aib: funny thing, kswapd0 grinding the CPU but there's no swap
[13:18] <blackflow> aglaia: ^^^^
[13:18] <huck> Hello. I'm getting the following error when trying to run CUDA application: CUDA ERROR: Memory alloc (eventsData bufer): CUDA driver version is insufficient for CUDA runtime version. I have CUDA 9.2 and the nvidia driver 396.44. Any idea?
[13:18] <aib> pragmaticenigma: I mean without bothering with PPAs and/or repositories. `letsencrypt` looks like a dumbed-down certbot and works okay for obtaining and renewing certificates. it's an ubuntu package
[13:19] <aglaia> blackflow: yes, that is why i'm getting crazy... it looks so strange. And the top command with MEMORY doesn't show anything relevant. Just common memory consume with php (about 20 % of memory) and mysql (20 % too)
[13:19] <pragmaticenigma> aib: The best way to install is through the PPA... that way when there are security updates, you will receive them automatically
[13:19] <darkn33ss> hi
[13:19] <darkn33ss> anyone on?
[13:20] <pragmaticenigma> !ask | darkn33ss
[13:20] <aib> pragmaticenigma: yeah but there's also a way without PPA (the letsencrypt package) I'm just wondering if anyone has managed auto-renewal with it yet
[13:20] <blackflow> aib: if it works for you...... personally I don't like certbot, it's bloated, opinionated behemoth that wants permissions to (re)start services and I ain't giving it that. dehydrated works fine for me, and I can easily add shell script hooks for various stages, so it's all automated, with nice privilege separation. I just need to build an apparmor profile for it
[13:20] <blackflow> aglaia: I have no idea what would kswapd do like that with no swap in use.
[13:21] <pragmaticenigma> aib: In all the documentation I have read, the recommeded way is to use the packages from the PPA for Ubuntu. They are modified specifically for the setup and environmnet of Ubuntu. Going your own way, you will have to figure out the differences between what the Developers computer environment was, and how yours is actually setup
[13:21] <blackflow> aglaia: here's an idea. since oyu don't have swap, see if setting vm.swappiness = 0   sysctl helps. that should
[13:21] <blackflow> aglaia: s/that should//
[13:21] <aglaia> blackflow: ok. thanks.  i will try that
[13:22] <pragmaticenigma> aib: The whole point of a package management system is to avoid that. letting the Ubuntu developers take care of that for you, so your not left running in circles trying to figure out why this config file is placed here instead of over there.
[13:22] <grumbel> Can anybody recommend a font viewer? Looking for something that takes a .ttf (without installing it) and a custom string and shows me how that string would be rendered.
[13:23] <pragmaticenigma> aib: Also, this channel will be better equiped to support your future questions, should you have any. We only support the packages provided through the PPAs because they're documented according to Ubuntu's environment.
[13:23] <aib> pragmaticenigma: sorry, are you talking about PPAs as opposed to official ubuntu packages?
[13:24] <pragmaticenigma> aib: For the moment, I'm treating them one in the same. Offical ubuntu packages is what I implied
[13:24] <blackflow> pragmaticenigma: I thought PPAs weren't really supported here? as in, who knows what their maintainers are doing....
[13:24] <pragmaticenigma> blackflow: Technically the Ubuntu main repos are PPAs... 3rd party ppas are not supported
[13:24] <blackflow> (other than how to add/purge a PPA, obvs)
[13:24] <aib> pragmaticenigma: I am talking about the official ubuntu `letsencrypt` package. I'm trying to about PPAs and other custom solutions. I'm wondering if there's any way to use the official `letsencrypt` package to automate certificate renewal without writing custom scripts
[13:25] <blackflow> pragmaticenigma: not quite, PPA = Personal Package Archive. the official ubuntu repos are not the first P ;)
[13:25] <pragmaticenigma> blackflow: fair enough
[13:25] <blackflow> that's teh important difference, as who knows what individual PPA maintainers are doing
[13:25] <aib> sorry, brainfart there. Ignore the weird sentence
[13:25] <pragmaticenigma> aib: I'll simplify... this channel is not able to support custom solutions or the official letsencrypt package.
[13:26] <pragmaticenigma> aib: for assistance with those, you will have to seek out their support forums
[13:26] <blackflow> aib: not going against what pragmaticenigma just wrote, but I'd really recommend you to look into dehydrated. you only need two lines of shell code, in appropriate hook files (its' all documented) to copy the cert files to wherver you want, and to reload nginx
[13:27] <aib> pragmaticenigma: https://paste.ubuntu.com/p/gCRbqpMy2g/
[13:28] <aib> pragmaticenigma: I don't understand your distinction between packages in the official ubuntu repo and "official ubuntu packages"
[13:29] <leftyfb> aib: https://www.digitalocean.com/community/tutorials/how-to-secure-apache-with-let-s-encrypt-on-ubuntu-16-04#step-3-%E2%80%94-verifying-certbot-auto-renewal
[13:30] <leftyfb> aib: that should work with the official ubuntu packages
[13:30] <leftyfb> aib: no need for the ppa or the python package
[13:31] <pragmaticenigma> aib: Official Ubuntu packages are provided through the official Ubuntu repositories... I don't see what's difficult in understanding that
[13:32] <pragmaticenigma> leftyfb: The instructions you linked to are for the certbot renewal package... it's written in Python
[13:33] <leftyfb> pragmaticenigma: the cron will work with the packaged version of "certbot"
[13:33] <leftyfb> though to be honest, I've been using the certbot-auto from the official github page for ages
[13:33] <pragmaticenigma> leftyfb: I know... I'm saying the application "certbot" is written in python...
[13:33] <leftyfb> ok? What difference does that make?
[13:34] <leftyfb> oh!
[13:34] <leftyfb> by python package I meant the python-certbot-apache from the link
[13:34] <leftyfb> that shouldn't be needed
[13:35] <aib> pragmaticenigma: what's difficult is that I'm trying to tell you that there is an official ubuntu package called `letsencrypt` and you keep talking about 3rd party solutions and PPAs
[13:35] <pragmaticenigma> aib: Because you're trying to ask us about custom installations and using software that isn't provided in the Ubuntu software repos
[13:35] <aib> I don't want to install custom packages or PPAs is precisely why I'm asking if there is a recommended solution using the official `letsencrypt` package in the official ubuntu repositories
[13:36] <leftyfb> aib: use the official package. Stick the certbot renew command in a cron. Done
[13:37] <pragmaticenigma> leftyfb: I assume aib contraint is they're not allowed to use anything that isn't provided in the default Ubuntu repos
[13:37] <leftyfb> again
[13:37] <leftyfb> use the official package. Stick the certbot renew command in a cron. Done
[13:37] <leftyfb> as in, the ubuntu repo package
[13:37] <leftyfb> I don't see what's so difficult about this
[13:38] <pragmaticenigma> leftyfb: because at the moment you are not quantifying what you mean by official... Ubuntu Offical meaning from Ubuntu maintained software channels... The let's encrypt offical client provided through the certbot ppa... etc...
[13:38] <darkn33ss> hey
[13:38] <pragmaticenigma> !chat darkn33ss
[13:39] <pragmaticenigma> !chat | darkn33ss
[13:39] <darkn33ss> i need help
[13:39] <pragmaticenigma> !ask | darkn33ss
[13:39] <darkn33ss> need to hide my ip on freenode
[13:40] <leftyfb> sudo apt install letsencrypt -y && echo "0 5 * * 1 certbot renew" | sudo tee -a /var/spool/cron/crontabs/root
[13:40] <pragmaticenigma> darkn33ss: that question belongs in #freenode
[13:40] <leftyfb> aib: ^
[13:40] <darkn33ss> sorry my bad
[13:40] <huck> Hello. I'm getting the following error when trying to run CUDA application: CUDA ERROR: Memory alloc (eventsData bufer): CUDA driver version is insufficient for CUDA runtime version. I have CUDA 9.2 and the nvidia driver 396.44. Any idea?
[13:41] <pragmaticenigma> !details | huck: what application?
[13:41] <m_i_k_e> aib: https://certbot.eff.org/
[13:42] <aib> leftyfb: the default config doesn't play well with nginx. the nginx plugin doesn't seem to be available. webroot works, but needs a manual nginx restart afterwards. updating everything to see if any new versions are better
[13:43] <leftyfb> !info python-certbot-nginx
[13:44] <leftyfb> aib: that plugin most certainly exits
[13:46] <huck> ubottu: Ubuntu 18.04. The application is mine and it was working fine. Today I typed: "apt install cuda-9.0" (just because I'm getting toubles installing TensorFlow - it searches for libraries from CUDA 9.0. Actually I thought that is will install in /usr/local/cuda9.1). After this I got that message on the cuda driver (not sufficient for the runtime that I have). I installed again cuda 9.2 but still I getting this error: CUDA ERROR: M
[13:46] <blackflow> huck: you're talking to a bot.
[13:47] <leftyfb> huck: ubuntu packages will not install to /usr/local ... that would be if you were compiling application
[13:47] <huck> blackflow, what I'm missing here?
[13:48] <huck> Thanks leftyfb.
[13:50] <aib> leftyfb: ah, installing package python-cerbot-nginx installs a plugin for package certbot? that makes my life much easier
[13:52] <leftyfb> aib: you should really learn how to read documentation and research things on google
[13:54] <aib> leftyfb: please tell me where the idiosyncrasies in the ubuntu package naming system is documented
[13:55] <leftyfb> aib: if you google for "ubuntu letsencrypt nginx" the very first link helps you
[13:55] <aib> leftyfb: and where certbot says it scans my system for installed python packages and configures itself accordingly
[13:56] <leftyfb> aib: ignoring the PPA since the package in bionic is adequate
[13:57] <aib> never read past the ppa part
[13:57] <aib> or are you telling me doing part of what a webpage says is good practice?
[13:58] <leftyfb> aib: https://certbot.eff.org/lets-encrypt/ubuntuartful-nginx
[13:58] <leftyfb> follow except for the ppa
[13:59] <leftyfb> please go read, research and understand
[13:59] <leftyfb> aib: good luck
[14:01] <pragmaticenigma> aib: what leftyfb is telling you is to read the entire page, and then do a little research to see what best applies to your situation
[14:03] <aglaia> blackflow: Hi again. Trying dmesg i got something interesting
[14:04] <aglaia> blackflow: at the time of the issue : [Thu Sep  6 05:08:12 2018] Out of memory: Kill process 6463 (php-fpm7.2) score 294 or sacrifice child [Thu Sep  6 05:08:12 2018] Killed process 6463 (php-fpm7.2) total-vm:840376kB, anon-rss:267396kB, file-rss:2580kB, shmem-rss:25620kB
[14:05] <blackflow> aglaia: yeah, something wants RAM and it gets kswapd worked up, but you have no swap
[14:05] <aglaia> blackflow: do you think i should focus on this php-fpm7.2 process ?
[14:06] <aglaia> blackflow: sure, as you advised, i will set the swapiness to 0. But, beside that, if there is an issue with this php-fpm process, i could try to fix too
[14:06] <blackflow> aglaia: not sure, that needn't be the process that caused overcommit
[14:06] <aglaia> ah... too bad
[14:06] <aglaia> i hoped
[14:07] <pragmaticenigma> aglaia: The message that is logged was what was deemed sacrificial due to the requirement for more ram. Another way to put it, low hanging fruit. What triggered the out of memory would have been on the other end of the list
[14:08] <pragmaticenigma> aglaia: You mentioned that it's once per day? correct?
[14:09] <aglaia> pragmaticenigma: yes, but the the server is in use only between about midnight till i wake up. I suppose that if it was running 24/24, the issue could occur more than once a day. (i don't know)
[14:09] <aib> leftyfb: and you be nicer to people who did not grow up with software with inadequate manual pages that warrant online browsing and software which scan your system for libraries dynamically, okay?
[14:10] <aib> leftyfb: I spent 10 minutes trying to explain pragmaticenigma that there is an official `letsencrypt` package in the official ubuntu repository
[14:10] <aib> actually... never mind
[14:10] <blackflow> aglaia: what I'd do.... create and enable some swap. then when the same situation happens, hopefully swap would help and the system won't OOM anything. then you can inspect which process had increase in RSS during the time that swap increased. sar should have all the info for that
[14:10] <very_sneaky> hi all not sure if this is the right channel for this; I'm trying to modify the ps2 mouse driver such that the axes are inverted. I *think* I've narrowed it down to psmouse-base.c and the psmouse_process_byte function, but I'm strugglingt o interpret the functions and what everything does. Is there comprehensive documentation on each of the functions anywhere? I haven't been able to find anything
[14:11] <pragmaticenigma> aglaia: Do you have auto-updates enabled? I know that I get processor spikes when apt goes and does its daily update
[14:11] <compdoc> dynamically scanning libraries?  sounds fascinating
[14:11] <leftyfb> aib: That's fine if you're asking about installing Firefox or something. But you're trying to install a package that manages SSL certificates on a public server hosting webpages. There's an assumption of competency when managing such an environment.
[14:11] <pragmaticenigma> aglaia: and dpkg goes and does its optomize
[14:11] <blackflow> pragmaticenigma: random times, so not cron based, and apparently not correlating with recorded systemd timer triggers
[14:11] <aglaia> pragmaticenigma: i don't know if i have auto-updates unabled
[14:12] <pragmaticenigma> very_sneaky: This is not the best forum for those questions. Programming and software development should be in #programming
[14:12] <pragmaticenigma> very_sneaky: Or try #kernel
[14:12] <very_sneaky> pragmaticenigma: sweet as, thanks mate
[14:13] <aib> leftyfb: do you seriously expect me to read articles that start with "install this 3rd party program" to the end to learn how to configure this official package with a different name?
[14:14] <leftyfb> aib: yes. It's called research and learning.
[14:14] <aib> leftyfb: and it's a painful process thanks to people like you
[14:14] <pragmaticenigma> aib: Please stay on topic. If you have an issue with another individual in this channel you are best to either ignore them or leave. This isn't a place for that
[14:14] <leftyfb> aib: you're welcome.
[14:14] <aib> nowhere does it even say that letsencrypt si the same thing as certbot...
[14:16] <blackflow> aib: certbot messed things up there, it used to be called "letsencrypt" then a few trademark issues later, it renamed to "certbot"
[14:16] <blackflow> oh they left.
[14:17] <blackflow> and if they listened and used dehydrated, they'd be up and running long time ago.    *shrug*
[14:18] <aglaia> blackflow: when i do less /var/log/apt/history.log i don't see any update at the time of issues
[14:19] <blackflow> aglaia: your best bet is finding the process that increases RSS at the same time swaping does. for that you'll need a swap(file).  maybe you could run without it, and rely on OOM killer to free up space, but the key is to observe rss of processes and see what's going up with the OOM slaughter starts.
[14:20] <blackflow> *when the
[14:20] <aglaia> blackflow: if i understand you well, if i enable a swap file, then the sar and top commands should then give me more relevant informations at the time the memory grows ?
[14:22] <blackflow> aglaia: it's more about not causing pathological states with OOM killing, because OOM could kill exactly the process trying to eat more RAM, and you wouldn't know then. with swap, you ease the situation so the culprit can gobble up RSS in peace.
[14:22] <blackflow> aglaia: but yes, in short, when that happens, when swap starts going up, you must find which process' RSS starts to grow. that correlation is indication of your memory eater.
[14:23] <aglaia> blackflow: i agree i definitely need to know what the process causes this OOM
[14:24] <blackflow> right. ideally you should not overcommit, especially not without swap. that means every service must be limited so that total sum of limits isn't bigger than available RAM.
[14:24] <blackflow> aglaia: also... that's a good idea. limit processes' RSS in their service files. the one wanting more ram will fail, hopefully loudly in the logs, so that's another way to find it.
[14:25] <aglaia> blackflow: so far, top command only show "kswapd". If i add a swap file, will it show the real process eating memory ?
[14:25] <gloomy> Hi again :-)
[14:26] <gloomy> I want an easy way to share data between macOS and ubuntu (dual booting)
[14:26] <gloomy> Should a make a 'shared' partition formatted as exFAT?
[14:26] <blackflow> aglaia: no, the swap file would only be to help the memory pressure so the process eating memory can do so in peace. you catch it by observing the rss changes when swapping increases
[14:26] <gloomy> Other (better) ways?
[14:26] <gloomy> macOS partition currently has apfs, and ubuntu ext4
[14:26] <blackflow> aglaia: otherwise you can't see which process is being swapped out, it doesn't work like that. the kernel is paging out based in usage, not process ownership.
[14:27] <blackflow> (you could after the fact check how much of individual process is in swap, via procfs status for the pid)
[14:27] <blackflow> but the process eating ram won't necessarily be the one swapped out. swapping is about _unused_ pages, so it's actually the the opposite, the one using active ram _won't_ be swapped out
[14:28] <gloomy> Also, is it better to add/resize partitions from Ubuntu (gparted) or from macos (diskutil)?
[14:29] <aglaia> blackflow: ok ok... (thinking, it's so hard and new for me !)
[14:29] <gloomy> Or, maybe install some ext driver in macos and put the shared data in the linux partition
[14:29] <aglaia> blackflow: thanks for all your explanations, i was far to know and understand all that
[14:29] <gloomy> Linux definitely can't read apfs
[14:30] <pragmaticenigma> gloomy: my intial search of the web says to do any disk partition within macos... and mac can read ext4 if you install osxfuse... for both, they're out of scope for this channel. You might want to find a channel specializing in Linux on Mac
[14:30] <aglaia> blackflow: last thing : if i enable a swap file, i guess i shouldn't set the swapiness to 0 as you told at the begining ?
[14:31] <blackflow> aglaia: correct, not 0. you could leave it at default 60. or even increase it so the kernel swaps out more aggressively, that really depends on the use case. if you're not sure, leave default (don't set any swappiness via sysctl)
[14:32] <aglaia> blackflow: okie
[14:32] <gloomy> pragmaticenigma: Well, if I ask on #osx they'll tell me to ask on #ubuntu, so I gotta try somewhere ;)
[14:32] <gloomy> Simonious: How do you manage that? :-)
[14:33] <gloomy> Also #ubuntu people are generally more friendly than #osx people (*subtle manipulation attempt*)
[14:33] <aglaia> blackflow: i will try and see how it goes next time. I hope it will provide me more information
[14:33] <blackflow> aglaia: maybe you can already do it based on existing sar data
[14:34] <aglaia> blackflow: ah yes ? how would i proceed ?
[14:34] <blackflow> aglaia: compare "before" and "during" process snapshots, if anything has increased RSS
[14:34] <blackflow> been a while since I used sar, but iirc it stores ps tree snapshots
[14:35] <blackflow> rss, or memory%, anything that would hint at increase in memory usage
[14:35] <aglaia> blackflow: i did it, i pasted the result : https://pastebin.com/Xgu7XipT
[14:35] <pragmaticenigma> gloomy: we can help to a degree... this channel attempts to stay within scope of Ubuntu specific topics. Per what I've read, I would create the partition in MacOS side. Pick a filesystem that is readily supported by both OSs, MacOS is more restrictive, which is another reason to choose there. You already know that Linux/Ubuntu can support nearly any filesystem except apfs.
[14:35] <intrepidsilence> does anyone know if there is a cli interface to add/delete entries in the ad_access_filter field in sssd.conf - yes I know one can edit it with vi
[14:35] <blackflow> aglaia: no, you need ps snapshots.  like output of "top"
[14:36] <aglaia> blackflow: i also have outputs of top, that i log each 2 minutes (CPU and Memory)
[14:36] <gloomy> Fair enough.
[14:36] <blackflow> aglaia: and that's exactly what you need
[14:37] <aglaia> blackflow: ok, let me show you
[14:37] <blackflow> aglaia: check if a process increases memory% or rss when kswapd starts to hog the cpu
[14:37] <pragmaticenigma> intrepidsilence: I doubt it... unless the developer provided a utiltity, it's rare to see another project. Most configuration files are meant to be edited with a text editor.
[14:38] <intrepidsilence> pragmaticenigma: yep, i get it - just looking for a way to enable this for our lesser capable folks - when using authentication method of simple then you can use realm permit and deny
[14:39] <intrepidsilence> but it does not appear that they have added the same kind of permit/deny commands for auth ad
[14:39] <blackflow> aglaia: I'm sure you don't expect me to put those numbers in a table and see what's increasing over time :)
[14:39] <blackflow> that's the work _you_ have to do. look for any process that's increasing memory% or rss during that period.
[14:39] <pragmaticenigma> intrepidsilence: I'm not too familiar with ad_access_filter... Does the config file support an automatic "look in directory for more configs" ability?
[14:40] <blackflow> a.k.a   the conf.d  pattern
[14:40] <aglaia> blackflow: so this is the ouput of top command looged at the time of the CPU spike : https://pastebin.com/kEQ9xnwh and this is the output of top at the same time, with memory option : https://pastebin.com/MmL1ZnrZ
[14:40] <pragmaticenigma> intrepidsilence: Like Apache or DNSMasq... where you have the primary config file, and then can define something like a apach.conf.d directory to put additional configs within?
[14:41] <intrepidsilence> pragmaticenigma: no, this is a field in the /etc/sssd/sssd.conf file for specifying what AD users and groups can login
[14:41] <blackflow> aglaia: right, and this other, sorted by memory, you need to compare _before_ and _after_ the kswapd spike
[14:42] <blackflow> or before and _during_ more precisely
[14:42] <pragmaticenigma> intrepidsilence: So you're looking for a quick/safe way to add remove users/groups from that config
[14:43] <intrepidsilence> pragmaticenigma: right, without having to use vi - the format of the field is cryptic and error prone and i was hoping to find an easier way for our level 1 and 2 folks to use
[14:43] <aglaia> blackflow: ok, this is 2 hours before, when all is perfectly quiete and peaceful : https://pastebin.com/Vcfpf1vZ and this is during the kswapd spike (same as i posted above) : https://pastebin.com/MmL1ZnrZ I don't see a big difference !
[14:44] <pragmaticenigma> intrepidsilence: I'm sure something exists, short and sweet would be to build a bash script that accepts a user or group as a parameter and does the heavy lifting
[14:45] <blackflow> aglaia: right, so at a first glance, pid 6462 went from 146M to 222M
[14:46] <ExoUNX> morning
[14:46] <ExoUNX> is there a way to enable a "testing" repo for Ubuntu 18.04
[14:47] <aglaia> blackflow: yes... this php-fpm process receives one image per minute. For each image, it applies a simple transformation on it with image-magick. That could be the cause, althought that is a very very simple and common image transformation (just a resize)
[14:47] <pragmaticenigma> ExoUNX: Can you elaborate further on what you are trying to do and what you mean by "testing" repo?
[14:47] <intrepidsilence> pragmaticenigma: that is not nearly as short and sweet as you might think - it would have to take a short user or group name, then do a lookup in AD and find the object path, convert the path to ldap format, then check against what is already in the file, and finally add the info if not already there
[14:47] <blackflow> aglaia: but then 6463 was at 163M, disappeared later. that's the problem with transient peaks like that, pids can come and go, and are very hard to catch
[14:47] <intrepidsilence> pragmaticenigma: certainly it can be done, but not short and sweet - probably why it does not exist yet
[14:47] <blackflow> aglaia: "just a resize" has to unpack the image into memory, which takes at least 3 bytes for each pixel in it. so how big are those.
[14:47] <ExoUNX> pragmaticenigma, I'm hoping to install restic 0.9.2 but the universe repo only has 0.8.3.
[14:47] <pragmaticenigma> !latest | ExoUNX
[14:48] <pragmaticenigma> ExoUNX: Any reason for attempting to get the latest?
[14:48] <ExoUNX> I understand, it's only for restic
[14:48] <aglaia> blackflow: images are average 50 k size
[14:48] <blackflow> aglaia: anyway, this is getting beyond the scope of the channel. you know what you have to do: log/grap/trace memory usage before and _during_ the incident to see what's causing it.  consider using swap to have less pathological states with OOM. definitely don't overcommit and limit service max memory usage.
[14:48] <blackflow> s/grap/graph/
[14:48] <pragmaticenigma> ExoUNX: If restic provides a PPA (not supported here) you can use that. Otherwise, you'd have to compile it yourself (also not supported here)
[14:49] <ExoUNX> I use CentOS, I have to fight to get bleeding-edge sometimes
[14:49] <pragmaticenigma> ExoUNX: You realize this is an Ubuntu support channel? not CentOS
[14:50] <blackflow> aglaia: with a picture worth a tousand words, really consider creating a graph from those numbers. spikes will be immediately obvious.
[14:50] <ExoUNX> I know this, I was just explaining why I'm well aware of stability,backports,etc
[14:51] <blackflow> ExoUNX: there are other distros suited better for bleeding edge. with ubuntu, it's either DIY or maybe there's a snap/flatpak straight from teh vendor.
[14:51] <pragmaticenigma> ExoUNX: Okay, still, the latest version isn't always the best. If there is a specific feature you require, that makes sense. But to have the latest just for having the latest isn't always a great idea in Linux/Unix world
[14:52] <pragmaticenigma> ExoUNX: Most of us here are going to advocate using what is readily available. As blackflow stated, there are distros better suited for obtaining bleeding edge
[14:52] <aglaia> blackflow: ok... looks like i'm far to figure out, with an issue i cannot reproduce and that happens randomly each night
[14:53] <ducasse> ExoUNX: see https://github.com/restic/restic/issues/355
[14:53] <blackflow> aglaia: quick question, how is FPM configured wrt max process count and max memory per php process? are you overcommitting with php?
[14:54] <ExoUNX> ducasse, yah I was just on that
[14:54] <aglaia> blackflow: i don't know at all...
[14:54] <aglaia> blackflow: i will check
[14:55] <ducasse> ExoUNX: a comment there says the debian package is installable on ubuntu as-is, at your own risk, ymmv etc
[14:55] <aglaia> blackflow: actualy my first suspicion is not about fpm, but the single php7.2 process. That is a thread that deals with websockets. Only one websocket, but pinging all the time each 10 second
[14:56] <blackflow> aglaia: yeah fpm the daemon itself is probably not the culprit but worker processes. still, it's in the domain of fpm config, process count, and php.ini for memory limits
[14:56] <ExoUNX> ducasse, as far as I can tell 0.3.3 is the debian version, which is much older
[14:56] <pragmaticenigma> intrepidsilence: Have you seen this? https://ldap.com/ldap-tools/
[14:56] <ExoUNX> but I'm looking at this - https://tracker.debian.org/pkg/restic
[14:57] <aglaia> blackflow:  memory_limit 128 M for fpm
[14:57] <blackflow> aglaia: you mean php.ini?
[14:58] <aglaia> blackflow: yes the php.ini of php-fpm
[14:58] <blackflow> and the process count?
[14:58] <blackflow> I'm guessing default on-demand fpm mode, so what's the max process count and does that number, times 128MB fit into available RAM
[15:00] <hggdh> ExoUNX: for for the record, restic 0.9.2 is being built for Cosmic (future 18.10). Currently Cosmic has restic 0.9.1. I would not expect 0.9.2 to appear in any previous versions, unless from a PPA
[15:00] <aglaia> blackflow: pm.max_children = 5
[15:02] <aglaia> blackflow: and there is both php and php-fpm running. php for a long time running script that manages a websocket, and php-fpm that receives and manages images.
[15:02] <blackflow> aglaia: I see. well, you know what to do. the only way to find the culprit is to allow it to take the ram and catch it. ideally graph the rss requirements of top 10 processes and you'll find it easily. I have to leave now, good luck.
[15:03] <aglaia> blackflow: ok, thanks for your help
[15:07] <satrioirc> if im running a plain ubuntu 18.04 LTS server w some security (which were recommended on many sites, such as hardening the transport layer, RSA authentication for SSH, etc) for educational purposes, and using openssh, ssmtp, ufw, fail2ban, and psad, what can spam my syslog with tons of "Creating SSL connection to host // SSL connection using ECDHE_RSA_CHACHA20_POLY1305 // Sent mail for my@email (221
[15:07] <satrioirc> 2.0.0 closing connection z11-v6sm7725756wrm.94 - gsmtp) uid=0 username=root outbytes=4435" like messages even when the system is idle? and if its a hack in progress, how can i prevent it?
[15:09] <satrioirc> no ports forwarded to the host other than 22
[15:09] <pragmaticenigma> satrioirc: The only options are to furlough the message to another process that will also indicate that there is an active attempt to access the machine.
[15:09] <TJ-> satrioirc: is z11-v6...94 a hostname of that server?
[15:09] <pragmaticenigma> satrioirc: Also, in all those articles, should have been the recommendation to not run your SSH server on a default port
[15:10] <subvhome> I'm running VLC with a unix socket interface. the socket is ~/vlc.sock
[15:10] <pragmaticenigma> !ask | subvhome
[15:10] <pragmaticenigma> !details | subvhome
[15:11] <subvhome> problem is when i send a command like: nc -U ~/vlc.sock <<< "pause"   .. it works but im stuck in the nc command and needs me to CTRL-C to drop back to shell.. is there any other way i can write to the socket without staying trapped in netcat?
[15:12] <pragmaticenigma> subvhome: suggest reading up on netcat... see "man nc"
[15:12] <subvhome> got it.
[15:12] <subvhome> doing that now
[15:15] <qwebirc39514> I did follow the Ubuntu's recommendations to protect my machine against L1TF by updating the Kernel, but I am still vulnerable :-(
[15:15] <qwebirc39514> $ cat /sys/devices/system/cpu/vulnerabilities/l1tf Vulnerable
[15:16] <subvhome> used socat
[15:16] <subvhome> works like a charm
[15:16] <pragmaticenigma> subvhome: or nc with "-q 0"
[15:16] <satrioirc> TJ: nope, it has a differenct hostname.
[15:16] <pragmaticenigma> subvhome: but best to use with the right tool ;-)
[15:17] <subvhome> still reading the man pages for nc.. but thanks... yes socat will work :)
[15:17] <subvhome> appreciate the push
[15:17] <satrioirc> pragmaticenigma: indeed. should i switch to a non.default port then? and what process you mean?
[15:18] <tomreyn> qwebirc39514: which recommendation areyou referring to specifically, which actions have you taken to mitiagte the vulnerability?
[15:18] <pragmaticenigma> satrioirc: By process I mean, you could create a firewall rule to block the IP of the incoming conenction, but then the firewall will alert that it is blocking connections to that IP.
[15:18] <tomreyn> qwebirc39514: and which ubuntu release and kernel version were and are you running now?
[15:19] <pragmaticenigma> satrioirc: Any port above 2000 is unreserved and can be used.
[15:19] <tomreyn> qwebirc39514: also the version of intel-microcode would matter
[15:20] <satrioirc> pragmaticenigma: ye i wanted to, my fail2ban is set to permaban auth retries, but the thing is: syslog doesnt show any ip for those messages, so there is no attacker to get rid of o_O
[15:23] <pragmaticenigma> satrioirc: I don't know the origins of the message... it almost appears interanlly originated
[15:23] <qwebirc39514> tomreyn: Ubuntu recommends to update the kernel so i tried different kernel on Ubuntu Xenial 4.4.154, and Bionic 4.15
[15:24] <qwebirc39514> tomreyn: my CPU is Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz
[15:24] <pragmaticenigma> qwebirc39514: Where did you find this recommendation?
[15:25] <qwebirc39514> pragmaticenigma: I found it there https://wiki.ubuntu.com/SecurityTeam/KnowledgeBase/L1TF
[15:25] <tomreyn> qwebirc39514: so which kernel versions did you have after applying the patches?
[15:26] <qwebirc39514> tomreyn: I tried 4.4.154 and 4.15
[15:27] <pragmaticenigma> qwebirc39514: Also, why are you manually updating these kernels? Also noted, did you read the entire article? Specifically: In some cases, which are believed to be rare, there may be no mitigation available due to the PTE inversion technique being incompatible with the number of physical address bits used by the processor and the amount of memory installed in the system. In these cases, the file will indicate the
[15:27] <pragmaticenigma> no mitigations are in place and that the system is vulnerable
[15:27] <tomreyn> qwebirc39514: please provide the output of "dpkg -l linux-image\* intel-microcode" and "cat /proc/version" on systems you have patched
[15:29] <satrioirc> pragmaticenigma: thanks anyways. port change: will do. if the messages will keep to show up, that will be a proof of something internal jobs generating them, but still, what the heck? :D
[15:29] <pragmaticenigma> satrioirc: can't hurt... easy enough to switch back
[15:29] <crazyisaac>  Hey i have been trying to autorun Guake when i startx but it doesn´t work.I tried to use crontab and     sudo cp /usr/share/applications/guake.desktop /etc/xdg/autostart/ too, but nothing works. Can someone help me?
[15:29] <iateadonut> snap packages: they can't see any files.  can i fix this?  what are the alternatives?
[15:29] <pragmaticenigma> !details | iateadonut
[15:30] <qwebirc39514> qwebirc39514: here it is : https://pastebin.com/ySPxwwwL with last kernel for Xenial LTS
[15:30] <pragmaticenigma> crazyisaac: Set it up per the documentation. crontab is not the correct location
[15:31] <iateadonut> snap packages, like pdftk, or telegram, cannot find files that have permissions from my user.
[15:31] <qwebirc39514> pragmaticenigma: I did read this but my CPU is very common
[15:31] <pragmaticenigma> crazyisaac: guake requires an active DE to run, and all DEs have a settings area for specifying autolaunch applications
[15:31] <iateadonut> so if i try to drag a file to telegram, telegram cannot accept it.
[15:31] <pragmaticenigma> iateadonut: Are you certain it's not telegram preventing drag-and-drop?
[15:31] <iateadonut> or if i try to merge several pdf's with pdftk, that won't work either - it simply doesn't see files, as if the user running the executable from the snap package does not own the files.
[15:32] <iateadonut> pragmaticenigma, yes, i'm sure, because it happens on pdftk, too, and because of this thread: https://askubuntu.com/questions/964377/snap-application-doesnt-see-files-from-another-partition
[15:33] <pragmaticenigma> iateadonut: did you read the solution? because the answer is right there
[15:33] <pragmaticenigma> iateadonut: it may not be exactly the same solution, but that should set you in the right direction
[15:35] <iateadonut> no, it didn't work for me.
[15:37] <tomreyn> qwebirc39514: can you show the exact version of intel-microcode you have installed there, again, it was cut off
[15:37] <tomreyn> qwebirc39514: dpkg -l intel-microcode | cat
[15:38] <qwebirc39514> tomreyn: ii  intel-microcode 3.20180807a.0ubuntu0.16.04.1 amd64        Processor microcode firmware for Intel CPUs
[15:38] <pragmaticenigma> iateadonut: You will need to reach out to the developer/maintainer of the SNAP application you are using for support than.
[15:38] <tomreyn> qwebirc39514: thanks. so i guess this suggests that this is one of the few CPUs which cannot be fixed (but i'll try to verify this).
[15:39] <iateadonut> prgamaticenigma - it did lead me in the right direction.  i'll have to take it further.
[15:39] <qwebirc39514> tomreyn: thanks
[15:40] <iateadonut> i have all my stuff mounted on /mnt but will have to change it to /media instead
[15:40] <iateadonut> that's about a 10 year old habit, though.  time to get with the times, i guess.
[15:40] <qwebirc39514> tomreyn: how can you keep me updated ? are you chatting here often?
[15:41] <iateadonut> anyway, thanks.
[15:42] <tomreyn> qwebirc39514: yes, i'm here regularly. also if you can stay around for another 60 minutes i *may* be able to give a better reply already. i'lltry to chekc this with someone from the security team.
[15:42] <qwebirc39514> tomreyn: thanks, I stay here
[15:46] <tomreyn> qwebirc39514: here's the question i asked, can you confirm that i'm stating the facts correctly?
[15:46] <tomreyn> A user is reporting that they have a xenial system with intel-microcode 3.20180807a.0ubuntu0.16.04.1 and linux-image-4.4.0-134-generic 4.4.0-134.160 (booted into, confirmed via /proc/version), and an Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz, and that "cat /sys/devices/system/cpu/vulnerabilities/l1tf" still reports "Vulnerable".
[15:47] <Sven_vB> what's a good program to draw equalizer curves in a narrow frequency range? I tried audacity's equalizer but when I make the window so huge that the interesting range is large enough to edit stuff, the editor becomes so slow it's unusable.
[15:48] <qwebirc39514> tomreyn: yes perfect and accurate
[15:48] <tomreyn> qwebirc39514: thanks. would you also like to review these details for the other system you said is also affected?
[15:49] <qwebirc39514> tomreyn: I only have this problem with this machine/CPU
[15:51] <tomreyn> qwebirc39514: you stated that a system running ubnntu bioniuc, kernel 4.15, was also affected. is this the same system then?
[15:52] <qwebirc39514> tomreyn: yes it was on the same system
[15:55] <tomreyn> qwebirc39514: i see. so i've not gotten a response, yet, which is to be expected at this time of day (we'Re beyond UK business hours). would you like to file a bug on this to make sure you don't miss a reply?
[15:55] <qwebirc39514> tomreyn: yes, with pleasure
[15:56] <tomreyn> qwebirc39514: you'd just need to run "ubuntu-bug linux" on the affected system. it will then print a URL (after some seconds or a minute), which you can then access form a web browser.
[15:56] <tomreyn> ... to complete filing the bug
[15:56] <tomreyn> you'll need to register with lauchpad.net in the process (unless you already have an account)
[15:57] <qwebirc39514> tomreyn: do I need to install a specific package to run ubuntu-bug ?
[15:58] <tomreyn> qwebirc39514: apport, but it should already be installed
[15:59] <Lope> how can I modify nvidia's UDEV rules to only load the driver if the device is present? currently it pegs the CPU at 100% when I boot up in a diff machine. I don't want to change the udev rule files every time, It should be possible to probe for the device before loading udev? http://codepad.org/AYRxqsGK
[16:02] <qwebirc39514> tomreyn: ok I am doing it
[16:03] <tomreyn> qwebirc39514: cool, (if you don't mind) please let me know the url of the bug report after you have finished the submission from your webbrowser
[16:04] <Sia-> Hi, my issue is i have a m.b with 2 SSD sda and sdb, if i install any **ubuntu last setup failed because trying to install grub on sda. and sda is hackintosh. Why can't self chose the bootloader disk as we have in Debian or Solus and manjaro?
[16:04] <Sia-> is possible to let the installer install the bootloader on sdb and not the firs disk?
[16:05] <tomreyn> qwebirc39514: if this is not a production system and you can afford to reboot it to bionic again, it'd be great to add the version details (of the versioned kernel image package and the intel-microcode package, combined with the outputs of "cat /proc/version" and finally "cat /sys/devices/system/cpu/vulnerabilities/l1tf", to the same bug report, too.
[16:08] <tomreyn> Sia-: normally (if more than one device is detected) you can choose the target device for the boot loader installation. which ubuntu version and variant were you installing there?
[16:09] <Sia-> tomreyn 18.4.1
[16:09] <qwebirc39514> tomreyn: I did it, i pm you the link
[16:09] <Sia-> i've tried with kubuntu, ubuntu ..etc same issue as last setup. Doesn't let me to chose the bootloader disk
[16:10] <tomreyn> Sia-: so you're installing using desktop installers, right?
[16:10] <Sia-> yes
[16:11] <Sia-> it let me to chose the installation disk, but the bootloader is sda by default can't change or don't have this option
[16:12] <tomreyn> Sia-: did you try manual partitioning?
[16:12] <Sia-> no
[16:12] <Sia-> autos
[16:13] <tomreyn> Sia-: on the manual partitioning install screen, you can choose the target device for boot loader installation on the bottom (drop down box):  https://i.stack.imgur.com/3DBJC.png
[16:13] <Sia-> aha thanks
[16:16] <tomreyn> Sia-: i assume that in automated mode it will just try to install to the first disk, which is usually the first bios disk, which will usually be the one the bios tries to boot (first, or only).
[16:16] <tomreyn> it's a bit different for uefi, but the same assumptions may apply there.
[16:43] <Hanumaan> I have Multi-OS (Fedora, Ubuntu and Windows) With Fedora managing the Grub now trying to upgrade Ubuntu from 16.04.5 to 18.04.1 .. however I get this error : https://paste.ubuntu.com/p/FcvkZz8SqP/
[16:51] <Hanumaan> EFI is managed by Fedora and so it is not in Ubuntu how do skip this error and upgrade?
[16:54] <NoImNotNineVolt> is ubuntu-for-windows offtopic here?
[16:54] <lotus|NUC> !ubuwin | NoImNotNineVolt
[16:54] <NoImNotNineVolt> thanks.
[16:54] <TJ-> Hanumaan: on that 16.04 is grub-efi or efibootmgr installed?
[16:55] <Hanumaan> TJ-: I suppose not installed or at least it is not getting used, how do I confirm that?
[16:58] <Hanumaan> TJ-: When I tried to install I always used to get this error and was not needed probably as it is managed by fedora: http://dpaste.com/03WGYK0
[16:58] <TJ-> Hanumaan: I'e looked at the d-r-u source; it checks for this list of packages. if any are installed you'll see that message: "bootloaders = ["shim-signed", "grub-efi-amd64", "grub-efi-ia32", "grub-efi-arm", "grub-efi-arm64", "sicherboot"]"
[16:59] <TJ-> Hanumaan: ensure those are not installed on 16.04 and it should go ahead
[17:03] <texla> ubuntu 18.04  ubuntu/gnome My numLK pad works thru login ..then the light blinks off/on and the numLK pad number no longer work..If I switch to cinnamon all keypad numbers work correctly
[17:08] <tomreyn> texla: try: gsettings set org.gnome.settings-daemon.peripherals.keyboard remember-numlock-state true
[17:11] <texla> tomreyn, tried that gsettings with no positive results
[17:12] <lotus|NUC> texla: did you reboot after?
[17:12] <tomreyn> texla: so you ran the command and rebooted since, and numlock still turned off?
[17:13] <TJ-> texla: what does this report? "gsettings get org.gnome.settings-daemon.peripherals.keyboard remember-numlock-state"
[17:13] <texla> tomreyn, that is correct tried it yesterday and have rebooted numerous times numLK still turns off after login
[17:15] <texla> tomreyn, gsettings get org.gnome.settings-daemon.peripherals.keyboard remember-numlock-state
[17:15] <texla> true
[17:18] <ioria> texla, universal access -> point & click -> mouse keys   (is On or Off  ?)
[17:21] <texla> ioria, point & click is on
[17:21] <ioria> texla, set to Off
[17:23] <texla> ioria, Switched to off 123 see the numbers from numLK
[17:24] <ioria> texla, meaning ?
[17:25] <texla> Ioria tomreyn TJ thanks to every one will reboot and see if it still working after boot and login
[17:25] <ioria> ok
[17:25] <skr> Anyone could explain what am I doing wrong with squidguard? conf is here https://pastebin.com/73agZihx
[17:26] <mohnish> Is it possible to install lxqt on xubuntu?
[17:27] <ioria> mohnish, never tried on xfce , but i think you can
[17:27] <mohnish> ioria: Okay, so it's possible, but does it break your system?
[17:28] <ioria> mohnish, nope... why ?
[17:28] <mohnish> Just asking ioria
[17:28] <mohnish> I'm gonna be installing lxqt in mine
[17:28] <ioria> mohnish, http://xubuntuhowto.blogspot.com/2016/11/trying-lxqt-desktop-on-xubuntu-1604.html
[17:29] <mohnish> ioria: What is openbox by the way?
[17:29] <ioria> mohnish, a wm
[17:29] <ioria> !info openbox
[17:30] <texla> ioria, Back on line worked at login...these numbers from numLK pad 5460 all working again Thanks for the help
[17:30] <ioria> texla, good, no prob
[17:30] <mohnish> ioria: What is i3 used for?
[17:30] <Lope> what's the min disk space for a minimal ubuntu server install?
[17:30] <ioria> mohnish, it's a good tiling wm
[17:31] <mohnish> ioria: Can I run applications like Firefox in it?
[17:31] <ioria> mohnish, sure ... no overlapping windows
[17:32] <mohnish> ioria: What do you mean no overlapping windows?
[17:32] <ioria> mohnish,  https://en.wikipedia.org/wiki/Tiling_window_manager
[17:33] <ioria> mohnish,  overlying if you prefer the word
[17:33] <mohnish> okay
[17:34] <mohnish> If I run my games in it, will they run faster? ioria
[17:34] <ioria> mohnish,  idk for sure, but you'll  need fewer resources
[17:35] <mohnish> ioria: Fewer resources, that's a good thing
[17:35] <ioria> sure
[17:35] <mohnish> I'll be sure to check it out
[17:35] <pragmaticenigma> mohnish: i3wm isn't really designed for games.
[17:35] <mohnish> Oh, okay
[17:36] <pragmaticenigma> mohnish: While your game may run, your experience may not be what you expect
[17:36] <mohnish> Thank you for your time though ioria
[17:36] <ioria> mohnish, ok
[17:36] <mohnish> Okay, I'd rather use lqxt then
[17:36] <mohnish> Does lxqt require openbox??
[17:37] <ioria> mohnish, yep... let me check
[17:37] <mohnish> okay
[17:37] <mohnish> !info arch
[17:37] <ioria> mohnish, yes
[17:38] <mohnish> okay
[17:38] <luka_33> I use arch btw
[17:38] <mohnish> So I need openbox for running lxqt?
[17:38] <pragmaticenigma> mohnish, ioria : Yes, LXQT uses openbox as its default window manager
[17:38] <ioria> yep
[17:39] <mohnish> Okay
[17:53] <RabidWeezle> Stupid question... you think a 3ghz pentium 4 with hyperthreading could run ubuntu server?
[17:53] <RabidWeezle> with like 2gb of ram, 32-bit
[17:54] <pragmaticenigma> RabidWeezle: Yes, I've run one with the same specifications. As well as a Celeron 2Ghz Laptop with 1GB of RAM in a laptop running server
[17:55] <RabidWeezle> ahh nice. I didn't know if server was still shipped in the 32-bit variety :)
[17:55] <subvhome> I have a bash script with if else conditions and statements.. I know the conditions work since one of the statements is "touch ~/test1" ... however the following statement doesn't work for some reason:  socat UNIX-CONNECT:/home/user/vlc.sock - <<< "add /mnt/sdcard/Videos/Playlist/AM.m3u"
[17:55] <subvhome> am i supposed to use certain single or double quotes somewhere?
[17:55] <ioria> RabidWeezle, not 18.04
[17:56] <RabidWeezle> 16.04?
[17:56] <ioria> RabidWeezle, yup
[17:56] <RabidWeezle> nice
[17:56] <RabidWeezle> I use that on my company mail server, so that should be a breeze
[17:56] <subvhome> RabidWeezle: server hardware runs on practically anything. im running 18.04.1 on 1gb ram Intel Compute Stick...
[17:56] <ioria> i think so
[17:56] <subvhome> but 64bit
[17:56] <jehorn> I use an old atom netbook as a file / git server at home. So yeah you can use older hw.
[17:57] <RabidWeezle> I use an atom zbox for our mail server, dual core with hyper-threading with 4gb of ram :)
[17:58] <RabidWeezle> but I didn't know about 32-bit support, so guess it's 16.04 for that
[17:58] <subvhome> RabidWeezle: product: Intel(R) Atom(TM) CPU  Z3735F @ 1.33GHz
[17:58] <RabidWeezle> subvhome: nice
[17:58] <RabidWeezle> Atom + linux = happiness
[17:59] <RabidWeezle> nothing like a 16 watt server
[17:59] <leftyfb> subvhome: you should try #bash for your question
[18:00] <RabidWeezle> Right now I'm running 2 servers in VM at home, but I'd rather not leave my 8700k running 24/7 with a custom loop that could potentially spring a leak while I'm at work :/
[18:01] <mohnish> What is the desktop environment the takes the least resources??
[18:02] <tarzeau> wmaker or amiwm
[18:02] <tarzeau> but it's probably not a desktop environment
[18:02] <mohnish> Oh
[18:02] <RabidWeezle> lubuntu uses lxde, it's pretty light
[18:02] <mohnish> How about lxqt
[18:02]  * tarzeau runs them on his raspberry pi or other elder hardware
[18:02] <RabidWeezle> last I checked
[18:02] <jehorn> LXDE, LXqt for desktop, or something like blackbox or such for a window manager only.
[18:02] <mohnish> Okay okay
[18:02] <Anthaas_> Hi guys - I have 2 computers in my house - this one (running Linux) has stopped connecting to the internet via powerline adapter.
[18:02] <RabidWeezle> blackbox is super light, aka fluxbox
[18:03] <mohnish> I'm installing lxqt
[18:03] <jk0ne> Anthaas_: have you reset your powerline adapters (both sides?)
[18:03] <compdoc> Anthaas_, what ip address does the non-working computer have? do you use dhcp?
[18:04] <mohnish> exit
[18:04] <mohnish> Oops, I thought I was in a terminat'
[18:04] <compdoc> you can never leave, mohnish
[18:04] <mohnish> terminal
[18:04] <mohnish> compdoc: LOL why?
[18:04] <compdoc> we are in control now
[18:04] <mohnish> I thought my terminal was open
[18:05] <Anthaas_> Uhhh, I have reset.
[18:05] <mohnish> compdoc: Oh my god! I can't even shut down my pc and leave oh no! LOL
[18:05] <Anthaas_> DHCP? I think my router does this?
[18:05] <compdoc> one of those  powerline adapters have probably died
[18:06] <Anthaas_> aaaand, the IP address? Haha - I'll be honest, I have 2 from `ip address show` and Im not sure which it is.
[18:06] <mohnish> I'm gonna get back here once I'm logged in to my lxqt desktop
[18:06] <mohnish> Bye!
[18:06] <Anthaas_> compdoc: It does connect for like 5 minutes after turning the machine on, but after that it disconnects and doesn't reconnect...
[18:07] <compdoc> and it worked fine when you first got them?
[18:07] <Anthaas_> Uhh, it worked fine until I connected the 2nd PC up with a powerline adapter it seems.
[18:07] <Anthaas_> It _seems_ that only one computer can be connected via powerline at a time...
[18:07] <compdoc> oh. how about using another power outlet?
[18:08] <compdoc> thats not good
[18:09] <Anthaas_> I'll try that in a bit - my birthday, about to go out for dinner, but I'm not hopeful on that one.
[18:09] <Anthaas_> For sure if that is the case, then it'll be the Windows machine that needs to be swapped.
[18:09] <Anthaas_> The weird thing is, the windows machine connects absolutely fine. This one is the one that has stopped holding a steady connection.
[18:09] <Anthaas_> I am currently tethering from my mobile phone.
[18:09] <pragmaticenigma> 18.04 is availble for 32 bit... You won't be able to use the regular installers. You can install via the mini.iso though
[18:10] <pragmaticenigma> ioria: 32-bit is still supported, the installer is only available in the mini.iso thought
[18:10] <albech> Gave up on getting 18.04 running on my old Mac Mini. Ubuntu Mate installed with no issues and ran out of the box. Cheers to those who tried to fix the issues.
[18:10] <pragmaticenigma> RabidWeezle: You can install 32-bit 18.04... You will have to install via the mini.iso is all.
[18:17] <Hanumaan> TJ-: great it worked and I was able to upgrade
[18:18] <pragmaticenigma> Hanumaan: TJ- has departed... glad to hear you're up and running
[18:22] <seanrdev> So I am getting an error when installing anything on my ubuntu server machine. It isn't allowed to remove older linux-header files. To be specific dpkg: error processing package linux-headers-4.15.0-29 (--remove):
[18:22] <nacc> seanrdev: there will be more lines than that, use a pastebin and paste the whole command and output
[18:24] <fedorafan> hey I got System program problem detected how to fix it?
[18:24] <pragmaticenigma> !details | fedorafan
[18:24] <fedorafan> I dont know it pops up and can cancel or report problem
[18:24] <cim209> !pastebin
[18:25] <nacc> fedorafan: typically that means something crashed, either in the current or prior boot.
[18:25] <nacc> fedorafan: and it's just trying to ask you to report the bug. There's nothing to 'fix' without knowing what the issue is.
[18:25] <fedorafan> ok
[18:26] <seanrdev> nacc: https://pastebin.com/HgT8gRbb
[18:26] <nacc> seanrdev: can you stat that path?
[18:27] <nacc> seanrdev: on line 18
[18:27] <fedorafan> how to upgrade kernel?
[18:27] <fedorafan> !kernelupgrade
[18:28] <nacc> fedorafan: ... what do you mean? the kernel is just a package, use apt/apt-get like normal.
[18:28] <pragmaticenigma> fedorafan: Kernel updates are provided through regular updates. Unless you disabled features of system update, you will receive a notification when updates are avaialble to be installed.
[18:28] <fedorafan> nacc I didnt get it before with newer kernel but I had problems with ethernet so I deleted it by a package manager
[18:28] <fedorafan> now I would like to upgrade again manually
[18:29] <nacc> fedorafan: i don't follow? what was the "it" you deleted?
[18:29] <seanrdev> nacc: https://pastebin.com/uWBWGYeP
[18:29] <fedorafan> by synaptic I chose all kernels expect the one I use now
[18:29] <fedorafan> 4.13.0-45-generic
[18:30] <nacc> fedorafan: what version of ubuntu are you on?
[18:30] <fedorafan> I guess 16.04
[18:30] <nacc> fedorafan: 4.13 is not a supported version of any ubuntu currently
[18:30] <fedorafan> ok
[18:30] <nacc> seanrdev: hrm; as root can you delete that file manually? (`sudo rm /usr/src/linux-headers-4.15.0-29/net/bridge`)? Did you see any errors in `dmesg` from when the apt-get ran.
[18:33] <fedorafan> yeah it is 16.04
[18:34] <nacc> fedorafan: well, only 4.4 and 4.15 are supported on 16.04
[18:34] <fedorafan> ok
[18:34] <nacc> fedorafan: the appropriate ubuntu packages of each
[18:34] <fedorafan> how to install them
[18:35] <nacc> fedorafan: either linux-image-generic or for 4.15 (hwe): https://wiki.ubuntu.com/Kernel/LTSEnablementStack
[18:35] <nacc> fedorafan: and stop removing metapackages if you don't understand what they are for.
[18:35] <fedorafan> I followed instructions just wanted to get rid of them
[18:36] <nacc> fedorafan: ... yes, don't do that. It easily makes your system unsupported.
[18:36] <pragmaticenigma> fedorafan: followed by you coming here with a broken system or system issues and expecting us to fix them. You need to stop doing that
[18:36] <nacc> fedorafan: and means you have not had any security or bugfix updates in some time.
[18:36] <fedorafan> and what are you doing if you got a big collection of kernels?
[18:36] <nacc> fedorafan: that's why you use the metapackage and follow the autoremove advice that apt always spits out.
[18:37] <fedorafan> I didnt expect anything just asked what to do
[18:37] <pragmaticenigma> fedorafan: There is already a documented process for removing older kernels
[18:37] <seanrdev> nacc: I get an input/output error...  rm: cannot remove '/usr/src/linux-headers-4.15.0-29/net/bridge/Kconfig': Input/output error
[18:37] <fedorafan> so with apt ok
[18:37] <nacc> seanrdev: when you do that, do you get an error or any logs in `dmesg` ?
[18:38] <pragmaticenigma> fedorafan: all software changes, including kernels, are handled through package management. When new kernels are installed, older kernels can be removed with "sudo apt autoremove" which will remove all unneeded packages and/or unused kernels
[18:38] <fedorafan> got it
[18:38] <pragmaticenigma> fedorafan: if the kernel isn't listed for autoremove, then you shouldn't force it.
[18:38] <fedorafan> ok
[18:40] <seanrdev> nacc: I tried doing a simple 'ls' and got an IO Error. I grabbed the dmesg during that command.. https://pastebin.com/yMKKJ6Bd
[18:40] <nacc> seanrdev: it would appear your disk is failing.
[18:40] <nacc> seanrdev: and your filesystem may have errors.
[18:40] <nacc> seanrdev: you should reboot and force a fsck
[18:40] <nacc> seanrdev: well, actually, that may not fix it alone, if your disk is failing
[18:42] <seanrdev> nacc: Damn.... Thanks a lot!! I appreciate your help. It's a 15 year old laptop I'm using as a log server so I guess that makes since.
[18:42] <pragmaticenigma> That's a huge set of log files!
[18:43] <fedorafan> pragmaticenigma what is enigma in your case
[18:43] <pragmaticenigma> !chat | fedorafan
[18:43] <fedorafan> so not here
[18:43] <pragmaticenigma> no
[18:44] <nacc> seanrdev: yeah, that isn't great, sorry :(
[18:44] <ShaunR> I have a weird issue that I cant seam to figure out.  I'm usint the apache2 build from ubuntu 18.04 btw.  I have confiugred apache for 3 virtualhosts, domain1.com, domain2.com, and sub1.domain.com.  Those sites load as expected. Whats weird is when i try to create sub2.domain.com. When i do that all of a sudden domain1.com starts showing content from sub1.domain.com  and I cant seam to figure
[18:44] <ShaunR> out why.
[18:45] <pragmaticenigma> ShaunR: Possibly a cut and paste issue?
[18:46] <fedorafan> sorry for asking :)
[18:47] <pragmaticenigma> ShaunR: It's hard to determine from your example domains... is sub2.domain.com a sub domain of either domain1.com or domain2.com?
[18:47] <ShaunR> pragmaticenigma: I dont see how.  I've even copied the sub1.domain.com to sub2.domain.com and did a search and replace for sub1 to sub2
[18:47] <m_i_k_e> ShaunR: same ip? same directory path? issue pragmaticenigma asked?
[18:47] <ShaunR> same ip
[18:47] <ShaunR> directory different
[18:47] <m_i_k_e> are you viewing by the domain name or ip?
[18:48] <m_i_k_e> in the browser
[18:48] <albech> in ubuntu 16.04 i would disable dnsmasq to get dns servers configured right when connecting to an openvpn server, but it seems like this was changed in 18.04. any hints?
[18:49] <pragmaticenigma> albech: dnsmasq is used as a local caching dns server in 18.04... disabling it will break all connectivity
[18:50] <pragmaticenigma> albech: not sure why you were disabling it in 16.04 ... that should not have been needed
[18:50] <pragmaticenigma> ShaunR: did you see my earlier question about the sub domains?
[18:50] <m_i_k_e> ShaunR: do you have "ServerName" in the configuration file?
[18:52] <albech> pragmaticenigma: i really just want to use the DNS servers my DHCP and VPN server is handing out or at lease use them as upstream DNS servers
[18:52] <nacc> albech: that doesn't require disabling dnsmasq (afaik)
[18:53] <albech> nacc: in 16.04 it was the only ay i could reliably make openvpn connections use a DNS server in the data center for internal name resolving.
[18:53] <ioria> albech, 18.04 uses systemd-resolved service if this can help you
[18:54] <pragmaticenigma> albech: So your system connecting to the VPN needs to have access to local machine on the local network as well?
[18:54] <pragmaticenigma> albech: So your system connecting to the VPN needs to have access to local machines on the local network as well?
[18:55] <nacc> albech: i think that might mean you misconfigured your VPN to perhaps handle all traffic or something. Were you using NM to configure the VPN?
[18:55] <albech> nacc: no it only routes traffic destined to the VPN networks through the VPN
[18:56] <pragmaticenigma> albech: so the system was acting as a gateway for traffic bound to go over the VPN?
[18:58] <albech> pragmaticenigma: 2 sec. locating the documentation
[19:00] <pragmaticenigma> albech: I'll return in about an hour... have a meeting to attend
[19:00] <fedorafan> https://docs.oracle.com/cd/E19182-01/820-7851/inst_cli_jdk_javahome_t/
[19:01] <fedorafan> im using the bourne shell but got problems with the line JAVA_HOME = jdk-install-dir
[19:01] <fedorafan> command not known
[19:02] <fedorafan> its bash not bourn sorry
[19:02] <subvhome> leftyfb: thanks will do!
[19:03] <fedorafan> export not found
[19:04] <nacc> fedorafan: no spaces. That help is formatted terribly.
[19:04] <nacc> fedorafan: and obviously not literrlly 'jdk-install-dir'
[19:04] <nacc> *literally.
[19:04] <fedorafan> ok
[19:04] <fedorafan> so within ''
[19:05] <nacc> albech: right, but did the vpn connection override the DNS maybe?
[19:05] <fedorafan> not a valid identifier
[19:05] <nacc> fedorafan: ... what?
[19:05] <fedorafan> what do I need to type
[19:05] <ShaunR> Here are the apache confs
[19:05] <ShaunR> https://pastebin.com/vL0J2HNK
[19:05] <fedorafan> export JAVA_HOME=bla there no spaces?
[19:05] <albech> nacc: it did.. and thats what i wanted.
[19:06] <fedorafan> doesnt work
[19:06] <nacc> fedorafan: use a pastebin, stop just saying it doesn't work.
[19:06] <fedorafan> Im on a laptop and doing stuff on desktop machine
[19:06] <nacc> fedorafan: why does that matter?
[19:06] <fedorafan> it works ok
[19:07] <fedorafan> it doesnt matter
[19:07] <fedorafan> youre right the format isnt that good
[19:07] <albech> acc: i wish to use other DNS servers (internal) when on the VPN..
[19:10] <albech> nacc: see lots of other people having same issue
[19:10] <fedorafan> java cannot be found
[19:11] <fedorafan> I followed the lines in the guide
[19:12] <fedorafan> pastebin.com/vmKsrtbc
[19:12] <leftyfb> fedorafan: https://www.digitalocean.com/community/tutorials/how-to-install-java-with-apt-get-on-ubuntu-16-04
[19:12] <leftyfb> fedorafan: use that to install java on Ubuntu 16.04
[19:13] <fedorafan> thanks leftyfb
[19:14] <albech> nacc: https://github.com/systemd/systemd/issues/6076
[19:16] <fedorafan> I would like to get it worked by the instructions given by the officials
[19:17] <fedorafan> !export
[19:17] <leftyfb> fedorafan: If you want it to work correctly on your system, use the link I gave you. Otherwise, seek support from Oracle with it
[19:17] <fedorafan> no I think thats a good way now im just curious to get it run by the link I already posted here
[19:18] <leftyfb> fedorafan: good luck. You're on your own with that.
[19:19] <fedorafan> stubborn sorry but you have helped me a lot I appreciate
[19:19] <fedorafan> :)
[19:25] <fedorafan> what does export do?
[19:26] <fedorafan> its a nick dont want to highlight you
[19:27] <nacc> fedorafan: you want #bash or `man bash`
[19:27] <RabidWeezle> no joke, #bash is super helpful
[19:28] <fedorafan> I did echo $SHELL and it says /bin/bash so its bash right
[19:28] <leftyfb> fedorafan: go to #bash for help with BASH
[19:28] <RabidWeezle> I love those guys
[19:28] <fedorafan> im going there :)
[19:29] <madLyfe> this is odd and not the first time its happened but i cant power off my pc. it turns off and i can hear the power(fans/drives spin down) go off then the computer boots back up again.
[19:33] <albech> disappointing to see that resolved does not honor DNS servers handed to it by openvpn and the hacks suggested in this thread is not feasible to the ordinary user. I am guessing this will affect LOTS of users: https://github.com/systemd/systemd/issues/6076
[19:33] <RabidWeezle> madLyfe: do you have the setting on in uefi that turns on as soon as it detects power hitting the system?
[19:34] <RabidWeezle> I use that on my server
[19:34] <RabidWeezle> in case of total outage/ups failure
[19:34] <pragmaticenigma> ShaunR: I think the issue comes from load order of the config files in apache... try enumerating the config file names with a numerical prefix like 0001_dev1.domain.com 0002_dev.domain.com
[19:34] <madLyfe> RabidWeezle: nah. this doesnt happen all the time. it functions properly, until it doesnt.
[19:35] <madLyfe> i saw this but i dont think: https://www.linuxquestions.org/questions/ubuntu-63/lubuntu-won%27t-shutdown-reboots-it%27s-the-microcode-firmware-4175635589/
[19:35] <pragmaticenigma> ShaunR: I'm not 100% certain as I've never had wildcards with subdomains before
[19:35] <RabidWeezle> madLyfe: a short in the switch?
[19:35] <RabidWeezle> wires pressing against an internal power button on some motherboards?
[19:35] <madLyfe> im not using the switch, im using shutdown on the desktop.
[19:35] <RabidWeezle> o_O
[19:35] <RabidWeezle> weird
[19:39] <Mokresh> I'm having some troubles booting Ubuntu 18 after fresh install. First Reboot after completing install will dump me to a black screen with no GRUB menu etc
[19:39] <Mokresh> http://pastebin.ubuntu.com/p/Zw3mTsHt3S/
[19:40] <pragmaticenigma> madLyfe: remove all devices from USB and other ports. Except for your monitor, keyboard and mouse. TRy then... if it stays off, then one of your USB or periphial devices is trigger the boot up
[19:42] <madLyfe> pragmaticenigma: could it be that a USB device doesn't trigger it every time but sometimes?
[19:43] <pragmaticenigma> madLyfe: could be a lot of things... bad power supply, stuck power button, residual power in a peripheral device giving feedback, static electricity in the air around the computer, power surge in the mainline power, etc.
[19:44] <xp0int3r> what is the name of the screenshooter application on ubuntu?
[19:46] <madLyfe> pragmaticenigma: well it worked the first time with the usb drive removed. still doing it now.
[19:47] <xp0int3r> nvm. it is gnome-screenshot.
[19:48] <pragmaticenigma> madLyfe: Could also be a sign that the Power Supply Unit is failing. Hardware isn't really something we can troubleshoot over the internet. You would need to find someone with experiencing in electronics that might be able to help diagnose that issue.
[19:50] <madLyfe> ok thanks for the help. though if I halt the system manually and then hold the power button down till it shuts off it doesn't turn back on.
[19:51] <madLyfe> this is only when I use the software shutdown button on desktop.
[19:51] <madLyfe> haven't tried the command shutdown
[19:53] <pragmaticenigma> madLyfe: menu command issues the same "shutdown" command
[19:54] <somenew> hi
[19:54] <somenew> anyone here
[19:54] <pragmaticenigma> !ask | somenew
[19:54] <pragmaticenigma> !chat | somenew
[19:55] <Lope> when I run `nbd-server -C /dev/null 0.0.0.0@1234 /dev/sdb1` it forks and when a client connects, nbd-server segfaults. If I run it with -d appended meaning "don't fork" it works perfectly, I've tested many clients with local and remote clients, same version client and diff version. Any ideas?
[19:55] <somenew> I have a PC with 8gb ram and an ssd, and whenever I use up too much ram, my system freezes and completely. I have 6gb of swap but it doesnt even look like its really being used, what can i do?
[19:56] <pragmaticenigma> somenew: What are you doing at the time the computer freezes?
[19:56] <Lope> somenew, your freeze might be unrelated to ram usage.
[19:56] <Lope> somenew, lots of things to check. CPU temp, memtest.
[19:58] <somenew> it is always different
[19:58] <somenew> sometimes i run a game, sometimes i run virtualbox
[19:58] <somenew> or an IDE
[19:58] <somenew> something that consumes a lot of ram
[20:00] <pragmaticenigma> somenew: That's making an assumption that it's something consuming ram.. all those items sound like stressing the machine and like Lope mentioned... possible the computer is overheating or other issues
[20:01] <nacc> Lope: https://github.com/NetworkBlockDevice/nbd/issues/77 ?
[20:02] <somenew> pragmaticenigma: i have an i5 -8250 and its below 30% usage most of the time
[20:06] <madLyfe> pragmaticenigma: i had power on by lan set in the bios, i turned that off and was able to shutdown. not sure why it would have been shutting down fine until now? or if i used the power button it would shutdown fine.
[20:09] <madLyfe> why would an internal(non OS) drive all of the sudden only be read only?
[20:10] <madLyfe> all other internal drives/network drives are RW.
[20:11] <leftyfb> madLyfe: either a config change or the drive has a physical issue
[20:11] <madLyfe> actually thats a lie. my other internal drive, win 10 drive, is also now only readable.
[20:12] <madLyfe> its a dual boot system with 3 internal drives. nix/win10/storage.
[20:12] <madLyfe> nix/win10 are ssds.
[20:12] <EriC^> look at dmesg and smart tests
[20:12] <madLyfe> win10/storage are read only now.
[20:18] <madLyfe> i see these errors: https://paste.ubuntu.com/p/gBjKMkt8r3/
[20:19] <pragmaticenigma> madLyfe: With the forced shutdowns, the drives may have been flagged with unclean shutdown and are in write mode to protect from corruption
[20:20] <madLyfe> this is the full dmesg: https://paste.ubuntu.com/p/7rtjP9KxMg/
[20:22] <pragmaticenigma> madLyfe: Boot windows, see if asks for a checkdisk
[20:22] <madLyfe> well the drive issue was first. then the not being able to shutdown started. but that had happened before.
[20:24] <leftyfb> madLyfe: sounds like a hardware problem to me
[20:25] <madLyfe> nah I can boot windows fine and read/write to the other drive that was read only under nix.
[20:27] <nacc> madLyfe: what filesystem is the win10 one?
[20:27] <madLyfe> NTFS I believe
[20:27] <nacc> madLyfe: and did you respond to what pragmaticenigma said about the forced shutdowns?
[20:28] <madLyfe> I don't know how to respond to it. the drives are working fine outside of nix. I didn't have any force shutdowns when the drives issue started.
[20:29] <nacc> madLyfe: which disk is the one that's readonly?
[20:29] <leftyfb> typically if NTFS is "dirty", it'll complain when you're trying to mount it and won't mount it at all. I don't think this has to do with the NTFS filesystem being left in a dirty state
[20:30] <madLyfe> 2 internal drives that are not the nix drive. one is a win10 drive and one is just a storage drive.
[20:30] <madLyfe> both read only drives are mounted and i can view contents and copy from them.
[20:30] <leftyfb> madLyfe: to help with details, "nix", "win10" and "storage" are not drive or filesystem types.
[20:30] <nacc> madLyfe: ... i meant which disk in the system by path.
[20:31] <nacc> madLyfe: we can't parse the log to know which disk is which without that information
[20:31] <nacc> madLyfe: is it simply mounted readonly?
[20:33] <madLyfe> the drives are /dev/sdb and /dev/sdc
[20:33] <nacc> madLyfe: which are which!
[20:33] <madLyfe> it could be, how do i check that?
[20:34] <madLyfe> win 10 is sdb and storage is sdc
[20:34] <nacc> madLyfe: i know what drives are on your system, they are clearly listed in `dmesg`. I do not know what filesystem should be on which, etc.
[20:34] <nacc> madLyfe: ok
[20:34] <nacc> madLyfe: look the output of `mount` ?
[20:34] <madLyfe> both are ntfs
[20:35] <madLyfe> https://paste.ubuntu.com/p/Y7pxfFgmkd/
[20:35] <nacc> madLyfe: /dev/sdb4 and /dev/sdc1 are mounted readonly.
[20:36] <madLyfe> they were RW until today for some reason.
[20:36] <nacc> does ntfs use fuse? I thought it used the ntfs module
[20:36] <nacc> leftyfb: --^ do you know?
[20:37] <nacc> madLyfe: yes, i understand. YOu don't need to repeat that. Have you tried remounting them read-write?
[20:37] <nacc> madLyfe: if the fs has errors, it will complain about them
[20:37] <leftyfb> yes
[20:37] <madLyfe> i havent tried that.
[20:37] <leftyfb> NTFS support is done with ntfs-3g which uses FUSE
[20:37] <nacc> leftyfb: ah! thanks
[20:40] <madLyfe> do i just want to remount that sdb4 and sdc1 rather than sdb and sdc?
[20:40] <leftyfb> yes
[20:41] <nacc> madLyfe: yes, you mount partitions, not disks.
[20:41] <leftyfb> sdb4 is the 4th partition on the sdb drive
[20:41] <leftyfb> if you're mounting an entire drive, you're doing it wrong :)
[20:41] <madLyfe> `sudo mount -o remount,rw /partition/identifier /mount/point` ?
[20:43] <madLyfe> `sudo mount -o remount,rw /dev/sdc1 /media/kinghat/Staging Storage`?
[20:43] <Iarla> Anybody know how to burn a CD from a .cue file? The .cue references a .iso and several .wav files. Normally I just burn .iso files to CD but the .wavs are required.
[20:44] <madLyfe> 'mount: bad usage'
[20:44] <Iarla> Brasero recognises .cue but asks me whether to burn the .iso as a file in the CD or something else. So I'm confused.
[20:44] <leftyfb> madLyfe: sudo mount -o remount,rw /mount/point
[20:45] <leftyfb> Iarla: where did this .cue file come from?
[20:45] <nacc> madLyfe: it's already mounted, so you don't need to tell it what to remount. Just the mountpoint to remount.
[20:45] <Iarla> It's part of an archive I downloaded. To be burned to CD for a games console.
[20:46] <madLyfe> hey that worked
[20:46] <Iarla> leftyfb: ^
[20:46] <leftyfb> Iarla: an archive of what?
[20:46] <madLyfe> does it need to be set to be always mounted as rw?
[20:46] <madLyfe> or will it be that from now on?
[20:46] <Iarla> leftyfb: a .cue, a .iso and several .wav files.
[20:46] <leftyfb> madLyfe: edit your fstab and change the ro's to rw
[20:47] <compdoc> if it were failing
[20:47] <compdoc> if it were failing
[20:47] <compdoc> ooops
[20:47] <madLyfe> i acutally looked there the first thing and they arent in there and from attaching my samba share via fstab i dont remember them ever being in there.
[20:47] <Iarla> leftyfb: I'm told the .cue is important because it tells the burning software what order t add the iso and .wav contents. But the others are using Nero and can't help with Brasero.
[20:48] <leftyfb> Iarla: What exactly is the iso of?
[20:48] <Iarla> leftyfb: the .iso is a bootable filesystem.
[20:49] <Iarla> For the Amiga cd-32.
[20:49] <Iarla> Brasero asks: Do you want to create a disc from the contents of the image or with the image file inside?
[20:49] <madLyfe> leftyfb: ya they are not in fstab.
[20:51] <leftyfb> Iarla: What you are doing is not a typical activity amongst Ubuntu users and not documented. It also has nothing to do with ubuntu and everything to do with your emulator. You should seek support from where you got the files
[20:55] <Iarla> leftyfb: it's not for an emulator. It's going in a CD tray. As I understand .cue files are relatively standard and burning software (even Brasero) knows how to use them. I just don't know. But maybe I will try a very approaches and test the resulting CDs when I am next at the lab.
[20:56] <Iarla> leftyfb: thanks for checking the docs, much appreciated!
[20:58] <fedorafan> leftyfb the channel wasnt that nice but now it runs
[20:58] <leftyfb> fedorafan: good luck in your future dealings. You will not be receiving help from me any longer
[20:58] <fedorafan> lol
[20:58] <fedorafan> why
[21:16] <Chris56> Hello there... Does anyone have experience using the cp command in the exec section of the find command?
[21:18] <hggdh> Chris56: it should be just -exec cp {} /some/where/else/{} \;
[21:18] <Chris56> yes!... but, when I use cp -n (no-clobber) it does nothing
[21:20] <Chris56> -exec cp --no-clobber {} /mnt/s3-drive/backup/hanab1/monthly/ \; => does nothing
[21:20] <Chris56> -exec cp --no-clobber {} /mnt/s3-drive/backup/hanab1/monthly/{} \; => cannot create regular file
[21:21] <hggdh> Chris56: indeed the second would fail because {} carries the fully-qualified file name. My fault
[21:21] <Seveas> Chris56: -exec cp -t /mnt/s3-drive/backup/hanab1/monthly/
[21:22] <Seveas> no need for {} tricks :)
[21:23] <Chris56> yeah, but I need to copy and not overwrite existing
[21:23] <Chris56> that's why the -n parameter
[21:23] <Seveas> you can still have the --no-clobber
[21:23] <Chris56> let me check
[21:23] <Seveas> -exec cp --no-clobber -t /mnt/s3-drive/backup/hanab1/monthly/
[21:24] <Chris56> Seveas => cp: missing file operand
[21:25] <Seveas> what's the full commandline?
[21:25] <Chris56> the original one: find /mnt/s3-drive/backup/hanab1/data -type f -newermt "2018-08-31 00:00:00" ! -newermt "2018-08-31 23:59:59" -exec cp -n {} /mnt/s3-drive/backup/hanab1/monthly/ \;
[21:26] <Chris56> if I remove the -n it copies the file, but that is a trouble
[21:28] <mobile_c> how do i force the installation of an already installed package
[21:29] <Chris56> I think it should work with: apt-get --reinstall install rsync
[21:30] <Seveas> Chris56: you may need to quote the {} in that one
[21:30] <Chris56> let me check
[21:31] <Chris56> Seveas same thing
[21:31] <mobile_c> how do i get past this
[21:31] <mobile_c> E: The repository 'http://ppa.launchpad.net/flosoft/cross-apple/ubuntu bionic Release' does not have a Release file.
[21:31] <mobile_c> N: Updating from such a repository can't be done securely, and is therefore disabled by default.
[21:32] <mobile_c> https://launchpad.net/~flosoft/+archive/ubuntu/cross-apple
[21:34] <ducasse> mobile_c: there are no packages for bionic in that ppa, only for really old ubuntu releases
[21:34] <Seveas> Chris56: ok, time to go for the axe. find /mnt/s3-drive/backup/hanab1/data -type f -newermt "2018-08-31 00:00:00" ! -newermt "2018-08-31 23:59:59" -print0 | xargs -0 -r cp -n -t /mnt/s3-drive/backup/hanab1/monthly/
[21:35] <Chris56> omg... let me check
[21:35] <mobile_c> how do i remove it but still be able to aquire packages from that ppa
[21:35] <Chris56> Seveas same thing D:
[21:36] <Seveas> mobile_c: that means that the ppa doesn't support bionic. Try using its artful or zesty variant.
[21:37] <Seveas> Chris56: which same thing? You've had various errors with the different variante
[21:37] <Chris56> it does nothing
[21:37] <Seveas> no errors? That means there may be nothing to do :D
[21:37] <mobile_c> can it be inctalled manually by adding these to sources?
[21:37] <mobile_c> deb http://ppa.launchpad.net/flosoft/cross-apple/ubuntu precise main
[21:38] <mobile_c> deb-src http://ppa.launchpad.net/flosoft/cross-apple/ubuntu precise main
[21:38] <hggdh> Chris56: replace -exec by -ok -- this will force find to ask you if it should execute the command. You can then see if it is doing something
[21:38] <Seveas> mobile_c: precise is *ancient*, packages built for precise are very unlikely to work.
[21:38] <mobile_c> or will apt stil default to attempting to aquire the bionic release from that ppa
[21:38] <Chris56> hggdh => cp ... /mnt/s3-drive/backup/hanab1/data/SCHEDULED_2018-08-31_0200.tar.gz > ?
[21:39] <hggdh> Chris56: there you go, it *is* finding the file and trying to -exec
[21:39] <Chris56> yes, but it does not copy the files
[21:40] <hggdh> Chris56: would the files, by any chance, have the same name?
[21:40] <Chris56> the path is empty :S
[21:40] <mobile_c> can i recompile them for bionic?
[21:41] <Chris56> if I remove tne -n it copies the file with no problems
[21:41] <mobile_c> by adding the src ppa
[21:41] <hggdh> Chris56: -ok will not show you the full command
[21:41] <mobile_c> if so how do i add 1024R/679C4EC6014ED9DF9F4E18A4730943C4D0611AA0 to the key thing
[21:42] <Seveas> mobile_c: you can try :) Add the source ppa for recise, apt-get source to get the sources; dpkg-buildpackage and pray :)
[21:42] <Seveas> apt-key adv --recv-keys 679C4EC6014ED9DF9F4E18A4730943C4D0611AA0
[21:43] <mobile_c> W: GPG error: http://ppa.launchpad.net/flosoft/cross-apple/ubuntu precise Release: The following signatures were invalid: 679C4EC6014ED9DF9F4E18A4730943C4D0611AA0
[21:43] <mobile_c> E: The repository 'http://ppa.launchpad.net/flosoft/cross-apple/ubuntu precise Release' is not signed.
[21:44] <Seveas> fun! Just dget the .dsc files
[21:44] <mobile_c> Executing: /tmp/apt-key-gpghome.4gZQr30I83/gpg.1.sh --recv-keys 679C4EC6014ED9DF9F4E18A4730943C4D0611AA0
[21:44] <mobile_c> gpg: key 730943C4D0611AA0: "Launchpad PPA for FloSoft" not changed
[21:44] <Chris56> strange... the file is found, the destination path is empty, but not copying
[21:45] <mobile_c> i still get W: GPG error: http://ppa.launchpad.net/flosoft/cross-apple/ubuntu precise Release: The following signatures were invalid: 679C4EC6014ED9DF9F4E18A4730943C4D0611AA0
[21:45] <Seveas> mobile_c: dget -u http://ppa.launchpad.net/flosoft/cross-apple/ubuntu/pool/main/a/apple-uni-boost/apple-uni-boost_1.55-0flosoft2.dsc
[21:45] <Seveas> repeat for the other packages
[21:46] <Chris56> well... gotta go now... will see it tomorrow... hope you're here... take care :)
[22:06] <jaziz> heya
[22:06] <jaziz> I'm trying to dual boot windows/ubuntu on two separate hard drives
[22:07] <jaziz> I've got this m2 ssd that I've installed it on, but it's not being detected on boot
[22:07] <jaziz> the windows hdd is being detected just fine, though
[22:07] <jaziz> is it a matter of installing the bootloader in the corret place?
[22:07] <jaziz> not sure where to go
[22:13] <jaziz> fuckit, I'll format some things again and see how it goes haha
[22:14] <jaziz> I got no data on there anyway
[22:15] <hggdh> jaziz: please mind your language
[22:15] <jaziz> hggdh, apologies
[22:20] <jaziz> Hm, I don't understand
[22:21] <jaziz> it's just not being detected
[22:22] <ryuo> jaziz: is it NVMe?
[22:22] <jaziz> ryuo, what's that
[22:23] <ryuo> jaziz: well... m.2 supports two main types of SSDs. SATA and NVMe. NVMe isn't supported by all BIOS as a boot device.
[22:23] <jaziz> oh, is this the same as m2?
[22:23] <jaziz> oh okay
[22:24] <ryuo> basically, NVMe is a storage device hooked directly into PCIE.
[22:24] <jaziz> it's the ssd that the computer came with, I think, so I wouldn't see why it wouldn't detect it
[22:24] <ryuo> I see.
[22:25] <jaziz> real strange
[22:25] <ryuo> Could always be some BIOS issue.
[22:25] <jaziz> my boot devices are listed at 1)windows boot manager 2)yes
[22:25] <jaziz> sad stuff
[22:26] <jaziz> what if I did something like install the bootloader on the hdd that is detected?
[22:26] <jaziz> but install the os on the ssd?
[22:26] <ryuo> Might work, but eh. What's the machine model?
[22:27] <jaziz> Acer E5-575
[22:28] <ryuo> Hm.
[22:29] <jaziz> what's up
[22:29] <jaziz> exact model here https://www.cnet.com/products/acer-aspire-e-15-e5-575-521w-15-6-core-i5-6200u-8-gb-ram-256-gb-ssd-us-international/specs/
[22:29] <jaziz> (I think)
[22:29] <ryuo> What's your BIOS version?
[22:30] <jaziz> InsydeH20
[22:30] <jaziz> or is that capital O
[22:30] <ryuo> No, the version.
[22:30] <jaziz> oh
[22:30] <ryuo> Insyde is a type of BIOS.
[22:30] <jaziz> whoopsies
[22:30] <jaziz> v1.04
[22:30] <ryuo> Wow.
[22:31] <ryuo> You really should update the BIOS.
[22:31] <ryuo> You're still using the first release of it.
[22:31] <ryuo> It's up to 1.45
[22:31] <ryuo> You can update it from Windows I believe.
[22:31] <zimablue> hey, does anyone know how/whether it's possible to both dual boot into ubuntu and run it inside hyper V? I can't find any discussion of how to do
[22:31] <ryuo> I can't be certain, but it could help with your issue too.
[22:32] <ryuo> jaziz: i've had BIOS updates fix various Linux issues before.
[22:32] <jaziz> oh dang
[22:32] <jaziz> well
[22:32] <jaziz> that I'll do then
[22:32] <ryuo> Like my Probook 455 G5 function keys finally work correctly after a recent update.
[22:32] <ryuo> BIOS update
[22:32] <jaziz> into windows we boot
[22:33] <ryuo> jaziz: well, you don't have to, but it's generally a good idea to keep them up to date.
[22:33] <ryuo> jaziz: i think this might be it, but you'll need to confirm it for yourself.
[22:33] <ryuo> jaziz: https://www.acer.com/ac/en/US/content/support-product/6752?b=1
[22:35] <jaziz1> alrighty
[22:35] <jaziz1> let's see what we got here
[22:39] <hans_> the entries in `contab -e`, where do their default stdout/stderr go?  /dev/null ?
[22:40] <nacc> hans_: yes, not logged by default, afaik.
[22:40] <hans_> kk
[22:40] <nacc> hans_: if you set up local mail delivery, they will be mailed to the local user
[22:40] <hans_> huh, ok
[22:41] <nacc> hans_: i should say, mail delivery at all -- the easiest is local
[22:41] <hans_> nothing wrong with giving them a screen session?
[22:41] <nacc> hans_: not sure i follow
[22:43] <hans_> hold on a min
[22:44] <ryuo> jaziz1: it worked ok?
[22:46] <hans_> nacc, well http://paste.debian.net/plain/1041067
[22:49] <nacc> hans_: oh i see what you are saying. yes, that would be fine, to do. It may make more sense to make it a proper service rather than a cron entry if you're trying to make it long-lived
[22:51] <hans_> hmm, i don't know how to write a init.d script, and i suck at bash scripting
[22:52] <roadkill> add your commands to the end of the file.
[22:52] <roadkill> commands you would otherwise type in shell.
[22:53] <hans_> roadkill, do you mean `to the end of the crontab -e file` ?
[22:53] <nacc> hans_: well, i was not going to sugget an init.d script, as that's sysv and old
[22:53] <nacc> hans_: you want systemd units
[22:53] <hans_> don't know how to write those either, any getting started tutorials?
[22:54] <nacc> hans_: `man systemd.unit`
[22:54] <hans_> right, thanks
[22:55] <nacc> hans_: there are a bunch of such manpages
[23:03] <fedorafan> nacc
[23:04] <lotuspsychje> ask your question in the channel fedorafan
[23:05] <fedorafan> no didnt want to send it
[23:05] <lotuspsychje> oh ok
[23:07] <nacc> fedorafan: did you need something?
[23:07] <fedorafan> scroll up please
[23:19] <jfra> hello
[23:21] <tomreyn> hi there
[23:21] <lotuspsychje> what can we do for you jfra
[23:21] <jfra> ho, i juste come for idle, and if possible, grab some tips xD
[23:25] <hans_> how do you modify this https://i.imgur.com/HOYU7vK.png and this https://i.imgur.com/RYuoxed.png  post-installation ?
[23:26] <hans_> apparently the answer to the first 1 is `sudo hostname <new_name>`
[23:27] <nacc> hans_: that's not accurate any longer, afaik
[23:27] <nacc> i believe you are supposed to use hostnamectl
[23:28] <nacc> the latter, i think, could be controlled by `domainname`, and i think only affects the fqdn expressed locally and then /etc/resolv.conf
[23:28] <roadkill> anyone in here have experience dealing with intel bluetooth/wifi combo chipset issues? Ubuntu recognizes the bluetooth part, wifi part is completely missing from lsusb. I know one fix is to disable bluetooth but when I manually do that in modprobe, it doesn't seem to enable wifi.
[23:31] <lotuspsychje> roadkill: your ubuntu version and chipset please?
[23:31] <hans_> roadkill, does `lspci` list your wifi perhaps?
[23:31] <HellFire> roadkill: the wifi part is usually PCI(e)
[23:31] <roadkill> lotuspsychje: 18.04, 17.10, 16.04, Fedora 28. Same issue across all. Gemini Lake chipset from Intel. Pentiun N4100. Intel Wireless-AC 3165
[23:32] <HellFire> with a built-in USB hub for the bluetooth
[23:32] <roadkill> it's a USB device.
[23:32] <HellFire> what model?
[23:32] <HellFire> oh, you said
[23:32] <roadkill> Device ID is 8086-3165 in Win10.
[23:32] <roadkill> When listing in lsusb, bluetooth's device ID shows, wifi does not.
[23:32] <roadkill> This specific chipset seems to have been fully supported since kernel 4.4
[23:32] <lotuspsychje> roadkill: is your system up to date to latest? aka .33 kernel?
[23:33] <hans_> roadkill, `uname -a` ?
[23:33] <roadkill> I did the latest kernel last night, yes, but with no changes in wifi. I had to wipe the install to reinstall Win10 to get the usb device ids so I'm currently reflashing the USB to reinstall with 18.04
[23:34] <nacc> HellFire: also may be useful to provide the dmesg when you unplug/plug the device in
[23:34] <roadkill> it's built into the motherboard.
[23:34] <nacc> err, roadkill sorry
[23:35] <nacc> https://codeyarns.com/2017/02/04/how-to-make-intel-wireless-ac-3165-work-in-ubuntu/
[23:35] <roadkill> I followed that guide. No dice.
[23:35] <nacc> roadkill: without logs, it's hard to help
[23:35] <roadkill> I dumped the latest intel firmwares into /lib/firmware without any results.
[23:36] <roadkill> And I'm aware, which is why I'm reinstalling now.
[23:36] <nacc> it's also certified in all these systems: https://certification.ubuntu.com/catalog/component/pci/8086%3A3166/
[23:36] <roadkill> been at this for 3 days, btw. heh.
[23:36] <nacc> sorry, https://certification.ubuntu.com/catalog/component/pci/8086%3A3165/
[23:37] <roadkill> So glad I kept that USB Ethernet dongle...
[23:37] <nacc> this doesn't happen to be you? https://ubuntuforums.org/showthread.php?t=2382682
[23:37] <roadkill> Nope.
[23:38] <roadkill> I have a Teclast F5,
[23:38] <nacc> check their results
[23:39] <nacc> roadkill: also, based upon that user's logs, the wireless is definitely pci attached
[23:40] <nacc> roadkill: s/results/diagnostic steps/
[23:40] <lotuspsychje> roadkill: try that tail -f /var/log/syslog and plugout/plug in to paste the errors as nacc suggested
[23:40] <nacc> yeah, that would be the easiest first step
[23:40] <nacc> don't do any remediation steps without first establishing the baseline
[23:42] <HellFire> roadkill: i'm guessing this is a laptop? what model?
[23:42] <HellFire> https://forums.linuxmint.com/viewtopic.php?t=241071 could be your issue if it's a Lenovo laptop
[23:43] <nacc> HellFire: they said it's a "Teclast F5"
[23:44] <HellFire> ah, missed that
[23:44] <nacc> although one review i read said that ubuntu wasn't supported on it at all, so there is that
[23:44] <nacc> oh but a german one says 18.04 worked
[23:45] <roadkill> It boots, almost everything works. The german guy didn't seem to try to get online with it in his review.
[23:45] <roadkill> Touchpad works after a kernel update.
[23:45] <roadkill> Don't care about touchscreen.
[23:47] <roadkill> Installed. Restarting so I can start the logging.
[23:52] <roadkill> dmesg | grep iwl = nothing
[23:53] <nacc> roadkill: can you pastebin the dmesg, `lsusb`, and `lspci` output?
[23:53] <nacc> roadkill: oh and `uname -a` ?
[23:53] <roadkill> I'm ircing from a different system. give me a moment to switch over
[23:53] <nacc> roadkill: and i suppose the output of the commands from that AU article
[23:54] <tomreyn> xonsider ssh'ing instead
[23:54] <roadkill> well my irc is on a rpi so I SSH in anyway
[23:55] <nacc> roadkill: i mean, to pastebin the commands you can use pastebinit
[23:55] <nacc> the commands' output, i mean
[23:57] <roadkill> https://pastebin.com/HwK8gQre
[23:59] <roadkill> whew. 1920x1080 with xfce at 11" is kinda hard to read. heh