[00:02] <arooni> how come with  "${distro_id}:${distro_codename}-updates"; ;; enabled in unattended-upgrades it doesnt seem to auto install them for me
[00:02] <arooni> on 18.04 desktop
[00:04] <leftyfb> arooni: https://help.ubuntu.com/lts/serverguide/automatic-updates.html.en # first result on google for "ubuntu unattended-upgrades"
[00:35] <plongshot> I want to install this:  https://syncope.apache.org/  on a remote ubuntu 18.04 server (it's a linode).  Will I be able to use the gui interface shown in the screenshots out of the box with only installing syncope?  Or are there other dependencies? If so, what / how?
[00:35] <plongshot> Thanks in advance for any help. It is much appreciated.
[00:36] <leftyfb> plongshot: you'll have to Apache for support with syncope.
[00:36] <leftyfb> contact ^
[01:30] <virmaha> hello. I've weird issue with 'tar'ing a file. I do tar -czf /some/path/a.tar.gz /some/dir . When I cd to /home/ and run tar xzf /path/to/a.tar.gz, I expect /home/some/dir but instead I get /home/abs/path/of/some/dir
[01:31] <virmaha> how can i fix that?
[01:37] <OerHeks> virmaha, where 'abs' is the user?
[01:42] <virmaha> OerHeks: abs as ain absolute path of some/dir
[01:57] <deker> hi all, anyone around to assist with setting up a guest account in 18.04.2
[01:57] <deker> ?
[02:02] <OerHeks> very short manual: http://wiki.groovix.org/index.php?title=Enable_Guest_Logins_in_Ubuntu_18.04_Bionic
[02:02] <OerHeks> but beware of the riscs
[02:02] <deker> thanks oerheks
[02:03] <deker> so i did manage to get guest login working
[02:03] <deker> my next step is to try and apply some modifications
[02:03] <deker> for example, i want guest login to default to the flashback gnome window manager
[02:04] <virmaha> OerHeks: any help?
[02:30] <Anja> 18.10, some of my apps have white menus with light grey words. what setting should I be looking for to increase the contrast?
[02:42] <samba35>  i have problem with efi boot loader on macbook pro ,which was working fine but i try to install systemd-boot and it seems problem start there
[02:42] <samba35>  how do i fix it ,i can see efi boot loader bar
[02:42] <samba35> --> kamdard (~kamdard@2605:6000:1526:4ca5:679a:5f1e:aa08:d3d2) has joined #ubuntu-unregged
 is there any combination of key to press to boot or what?
[02:42] <samba35>  when i press v i can see systemd-boot 237 (x64),UEFI spec 1.10 vendor apple 1,10
[02:43] <samba35> when i press d i can see default boot entry select
[02:44] <guiverc2> Anja, i don't know, but the types of apps may dictate where to look.  snaps vs deb packaged, gtk+ vs qt; more detail may help people to advise or at least be a clue on where to look
[02:48] <Anja> guiverc2: fair call
[02:48] <Anja> hexchat & gimp are two that come to mind straight off the bat
[02:48] <Anja> smplayer
[02:49] <Anja> aptana studio
[02:49] <guiverc2> gimp is gtk+, hexchat I believe too, smplayer is Qt
[03:09] <Frankfurtsoup> Hi everybody
[03:25] <friendlyGoat> heya, i was wondering, ever since updating to 19.04 i've been seeing a quick little error before the login prompt appears, it says  iwlwifi: BIOS contains WGDS but no WRDS but i have no idea how to go about getting that to go away
[03:53] <friendlyGoat> if anyone has an answer please ping me so i can see when i wakr up
[05:39] <tomreyn> friendlyGoat: looks more like a warning or informational message to me. are there any issues with your wireless?
[05:49] <tomreyn> friendlyGoat: downloading the latest iwlwifi-9*.ucode3 files from https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/linux-firmware.git/tree/ to your /lib/firmware/ may remove this message.
[05:58] <tomreyn> actually, it is an error, reporting that geographic adjustment of radio emission levels is not possible, because only the adjustment level but not the base level is known / could be determined. it is certainly not critical, though, and should also not impact proper operation.
[07:18] <nikolam> Hi, does this means I have a bad RAM? Virtualbox VM is keeps halting on that machine... RAM is Unbuffered ECC..  https://bpaste.net/show/dbc764687680
[07:20] <SwedeMike> nikolam: yes, that seems to report Uncorrectable Errors
[07:22] <nikolam> SwedeMike, yet, testing RAM passes..
[07:23] <nikolam> thanks SwedeMike , I were trying to find any info before on those messages meaning and didn't find anything precise
[07:24] <nikolam> I have 4 DIMMs , maybe to see to remove one by one and see if one of them is affected.
[08:58] <yan0s> Hello all!
[08:59] <yan0s> is there an opensource option for live patching in Ubuntu?
[08:59] <lotuspsychje> !livepatch | yan0s
[09:01] <yan0s> is livepatch open source?
[09:02] <afidegnum> anyone familiar with copyq, now do i assigned pinned clipboared to specific keys?
[09:12] <Jackneill> hi
[09:12] <Jackneill> tomreyn, https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1829402/
[09:12] <Jackneill> GRUB_CMDLINE_LINUX_DEFAULT="dis_ucode_ldr" seems to be the solution, never hang
[09:13] <lotuspsychje> Jackneill: tomreyn might be afk for now, i will pass the message to him, thank you for the feedback
[09:13] <Jackneill> lotuspsychje, thank you
[09:13] <Jackneill> shit intel microcode
[09:28] <tbrox> Does the archive.ubuntu.com mirrors implement some kind of rate-limiting? Some of the mirrors give me a consistent 16KB/sec transfer speed. It also seems to depend on what OS I make the request with, and from what network.
[09:29] <lotuspsychje> tbrox: you could talk to the #ubuntu-mirrors channel for known issues
[09:29] <tbrox> Thanks lotuspsychje
[09:30] <Jackneill> what is the preferred / *lazy* way of changing username? and why do i cant give a username with a . in the installler?
[09:34] <lotuspsychje> Jackneill: are you looking for systemsettings/details/users ?
[09:35] <Jackneill> lotuspsychje, i meant not the associated the name but the username
[09:35] <banisterfiend> why is it that " sudo echo -e "$newIndex\txvpnrt" >> /etc/iproute2/rt_tables" fails with a permission error, but if i actually change to root and do it, it works?
[09:35] <Jackneill> lotuspsychje, thats in passwd, and the group with it
[09:36] <lotuspsychje> Jackneill: are you reffering to the hostname?
[09:38] <Jackneill> lotuspsychje, no. to simply put user@host:~$ here the user part.
[09:38] <Jackneill> thats not the same as that users setting, just a fancy variant of it.
[09:38] <Jackneill> the username is the name of the hopme directtory
[09:38] <Jackneill> i want to change that username.
[09:38] <Jackneill> because the ubuntu installer didnt allow a period (.) in it..
[09:47] <Triffid_Hunter> banisterfiend: because sudo only applies to echo, not >>. try echo blah | sudo tee -a /etc/iproute2/rt_tables instead
[10:03] <banisterfiend> Triffid_Hunter thx bb
[10:08] <wholesum> rebooted my box today and I am getting "operating system not found"
[10:09] <cfhowlett> what was your last system change/operation wholesum
[10:09] <cfhowlett> and are you the only user
[10:09] <wholesum> been working fine for 1+ year
[10:09] <wholesum> i am the only user
[10:09] <cfhowlett> what was the last system change?  i.e. updates, installs, etc
[10:09] <wholesum> gparted was showing the entire disk as unallocated
[10:10] <wholesum> a while ago
[10:10] <wholesum> so i used gdisk and repaired it, so now i can see the partitions
[10:10] <wholesum> i already tried boot-repair
[10:10] <wholesum> cfhowlett: "a while ago" is to answer your question
[10:11] <wholesum> not when the disk showed up as unallocated
[10:11] <wholesum> that was after i couldn't boot
[10:11] <wholesum> now i am booted into the live cd and i don't know what else to try
[10:12] <cfhowlett> systems don't  generally "unpartition" themselves.  however, as you stated the partitions were recovered?
[10:13] <wholesum> yes, i was able to recover but cant boot to this disk.
[10:13] <wholesum> to make it harder, it is Secure Boot (UEFI)
[10:13] <jeremy31> wholesum: check disks program, SMART and see if it shows any failures
[10:13] <wholesum> already tried disabling secure boot.
[10:13] <cfhowlett> more and more good news:)
[10:13] <wholesum> it's an SSD
[10:13] <cfhowlett> ssd's do fail
[10:15] <wholesum> i'm going to try smartctl. but the TBW on this disk is less than 80, and it's an Samsung 970 PRO (enterprise version)
[10:16] <cfhowlett> all sounds good if you ignore that even major OEM's have had components fail out of the box.  test to be sure.
[10:17] <wholesum> smart shows no errors
[10:17] <wholesum> https://i.imgur.com/RwrbK3E.png
[10:18] <cfhowlett> can you mount the ssd from your live session and see if the data exists.  I suggest at this juncture you concentrate on data backup
[10:19] <wholesum> yes, that's what i did, and the data is there
[10:19] <wholesum> here's the partition setup https://pastebin.com/akZ5WT7S
[10:19] <cfhowlett> and did you backup your important data?
[10:20] <wholesum> i have a backup of most of the data, and I can backup the rest via the live cd
[10:21] <wholesum> i also had a fully system backup that failed today, so i deleted it to redo the backup. left for an appointment and when i got back rebooted the machine to restart Acronis (and see if it would back up properly)
[10:21] <wholesum> so ofc the system fails on the one day i don't have a full system backup!
[10:22] <wholesum> in other words my data is safe, but i would like to try to recover this machine
[10:22] <cfhowlett> then I think you might be looking at reinstalling to the existing partition table.  set the previous user/passwd credentials.  assuming a dedicated /home, you should be good.
[10:22] <wholesum> thank you.how would i go about doing that?
[10:23] <wholesum> i have all the users and their pwords
[10:23] <cfhowlett> on the installation page, select the manual partition options "do something else" IIRC.
[10:23] <wholesum> so reboot the live USB and chose install ubuntu instead of try it?
[10:23] <cfhowlett> add your admin account, then set the other accounts.  that should do it.
[10:24] <cfhowlett> NOTE however, this has worked great for the past 8 years.  but when I did it again for ubuntu 18.04, I couldn't see my data.  it existed in the partitions but I missed a step somewhere cuz I couldn't get at it.
[10:25] <wholesum> do you mean following these steps here? this is the only thing i haven't tried yet: https://www.tecmint.com/rescue-repair-and-reinstall-grub-boot-loader-in-ubuntu/
[10:27] <EriC^^> wholesum: what exactly did you back up?
[10:27] <cfhowlett> Let me be emphatic: reinstall is your FINAL option if all else fails
[10:28] <wholesum> gotcha.
[10:28] <jeremy31> wholesum: Reinstalling grub on EFI might be different than what that webpage says
[10:28] <wholesum> Eric^^: i have data backedup. the system partition (with the home directory) is not backed up. the system backup was lost, today ofc
[10:29] <wholesum> jeremy31: hmm, the live USB is using EFI.
[10:32] <jeremy31> wholesum: you can do this from live https://askubuntu.com/a/831241
[10:33] <wholesum> jeremy31: will try that!
[10:34] <wholesum> note: the system in question is server 18.04.1
[10:38] <wholesum> the one partition i can't mount in the live USB is the EFI System Partition. is this partition supposed to be mountable?
[10:59] <yan0s> Is Landscape On-Premises free of charge?
[11:09] <datasmurf> Anybody here running atop successfully on 18.04.2 LTS? Having troubles to get the atop deamon started. Installed package and systemd faild to start service ERR MSG says something about kernel "bug", fetched newest release from github 2.4.0 ( dpkg version is 2.3.0 ). Daemon atopacctd runs but the atop.daily scripts fails to run exit code 7. Hints welcome :-)
[11:11] <Fixion_> how to fix BScreen::BScreen: an error occured while querying the X server.
[11:11] <Fixion_>         another window manager already running on display :0
[11:11] <Fixion_> Error: Couldn't find screens to manage.
[11:11] <Fixion_> Make sure you don't have another window manager running.
[11:11] <Fixion_> (trying to run fluxbox)
[11:13] <bazhang> Fixion_, did you select flux at the login window
[11:14] <Fixion_> uh no ... uh its ok just ignore it for now...
[11:15] <bazhang> Fixion_, if you want to run kde (kubuntu) lubuntu (lxde) or whatver, you need to select at login
[11:16] <Fixion_> I'm kinda barking up the wrong tree ... (WSL doesn't have a login window/DM tat way)
[11:16] <Fixion_> I'm in #ubuntu-on-windows but nobody replying right now
[11:16] <bazhang> Fixion_, you should mention that right away
[11:17] <Fixion_> I'm not sure how to determine if its a problem with another manager running, just an administrator lock in /var somehwere or if theres a problem with the vcxsrv not connecting with the internal linux x-server
[11:18] <bazhang> Fixion_, is wsl ever allowed to use flux even
[11:19] <Fixion_> I'm using it on my fedora remix and I've used it on ubuntu before
[11:19] <Fixion_> I can get mine working
[11:19] <Fixion_> but my friend is getting that error for his and I don't know enough to help him as to what is causing this particular problem
[11:20] <Fixion_> it should be working I think...
[11:20] <Fixion_> for the most part you really just need to export DISPLAY=:0
[11:20] <Fixion_> and have a running X-Server application (on the windows side)
[11:21] <leftyfb> datasmurf: atop from the official repo works fine for me on Ubuntu 18.04. Sorry, but we can only support the packaged version from the official repos. Not compiled from github.
[11:21] <Fixion_> I talked him through that, but he started X with the command X at first, I told him to exit everything, he did that, I'm wondering if the X he ran that way is still running or something ...
[11:21] <leftyfb> Fixion_: sorry, but we do not support WSL here.
[11:24] <jeremy31> wholesum:  you should be able to mount the efi system partition
[11:24] <BluesKaj> Hi folks
[11:39] <Fixion_> apparently he got it working...
[11:39] <Fixion_> thx anyway
[11:40] <Ublx> Hello, how can I change a color in a pdf? Could be in command line. I have a PDF and I want to change some colors in it but because there are many pages it should be automated. Is there a tool?
[11:42] <bazhang> Ublx, is the pdf protected
[11:42] <mouses> Ublx: hi, you might /msg ALIS help and search for a channel a bit more suited to that - your question really has nothing to do with ubuntu
[11:43] <bazhang> it's fine mouses
[11:44] <bazhang> Ublx, you probably wont get  a cli tool, but apt-cache search pdf should offer some options
[11:44] <fefa2k> Hello, I'm connecting to a remote server behind NAT with a reverse SSH, to do so I just run 'ssh administrador@localhost -p 19999' that works fine, but I'm trying to do now is use it as proxyjump but I get "Permission denied" every time, the command I'm using is 'ssh -J administrador@localhost -p1999 manager@192.168.81.202'
[11:44] <bazhang> pdftk being one
[11:45] <fefa2k> *first but = what
[11:48] <Ublx> Thanks, bazhang and mouses.
[11:48] <bazhang> welcome
[11:48] <mouses> Ublx: best of luck :-)
[12:23] <b1ack0p> hi. how can i install TeX::Hyphen module in ubuntu?
[12:26] <guiverc2> b1ack0p, https://help.ubuntu.com/community/LaTeX contains some info on add-on package install options
[12:26] <b1ack0p> installing cpan now
[12:27] <b1ack0p> it seems easy
[12:31] <Synaptic> my bluetooth adapter rt3290 work only if im running this command sudo modprobe rtbth
[12:31] <Synaptic> possible to make it automatic_ i mean...
[12:31] <Synaptic> that it works alone
[12:32] <mgedmin> should be!
[12:33] <mgedmin> it would be best to file a bug (detailing the exact vendor/device IDs from lspci or lsusb, whichever is appropriate) so in a future update it'll Just Work
[12:33] <mgedmin> meanwhile I think you'd need to create a udev config file that tells it to load this driver for that device
[12:33] <mgedmin> let me google a bit
[12:34] <b1ack0p> finally
[12:34] <b1ack0p> cpan was extremely easy than i guessed
[12:42] <mgedmin> oh, Synaptic has quit without waiting for an answer, which is https://www.linuxquestions.org/questions/linux-hardware-18/forcing-a-specific-driver-for-a-device-through-udev-836411/
[12:55] <uzee> Hi, can I use a local machine to serve as an installation source for preseed?
[12:55] <VasTappendage> just managed to f u. dist.upgrade, crashed during install.. any ideas smarter than me people?
[12:55] <tomodachi> VasTappendage: usually you can just run it again to continue
[12:56] <uzee> I've copied the entire contents of the ubuntu 18 server iso to a local machine and am trying to use that machine as the installation source in preseed, for e.g:
[12:56] <uzee> d-i mirror/http/hostname string http://myhost.mydomain
[12:56] <uzee> d-i mirror/http/directory string /ubuntu/ubuntu18
[12:56] <uzee> but the installer gives me an error saying "no kernel modules found" is having the iso copied locally not enough?
[12:56] <tomodachi> shouldn't be a problem
[12:56] <mgedmin> VasTappendage: I fixed many snafus by repeating sudo dpkg --configure -a && sudo apt dist-upgrade --fix-missing until it told me there was nothing left to upgrade/configure
[12:57] <mgedmin> uzee: iirc the new ubuntu server ISOs contain a live system instead of a bunch of .deb packages, so you cannot use it as an apt repository
[12:57] <mgedmin> there should be alternative server ISOs that contain debs and those might work, maybe?
[12:58] <uzee> mgedmin: thanks very much, I'll try and look for other images
[13:00] <VasTappendage> tomodachi ...it just hangs at the boot login screen
[13:00] <tomodachi> VasTappendage: boot? or login?
[13:01] <VasTappendage> boot
[13:01] <tomodachi> so before grub? or after grub?
[13:03] <EriC^^_> !ping
[13:04] <VasTappendage> i think it is after grub, starts to load the fails at utmp update
[13:05] <mgedmin> VasTappendage: can you get to the grub menu (hold down shift, iirc) and select the rescue option to get a root shell?
[13:06] <VasTappendage> hang on le tme try
[13:09] <VasTappendage> yes have got it into recov mode
[13:13] <VasTappendage> ok got it into recov mode. now what
[13:26] <samba35> i am using ubuntu 18.04.2 on macbook pro
[13:27] <samba35> but i am not able to save any firefox or chrome page (page save as option)
[13:28] <cfhowlett> error messages?
[13:30] <VasTappendage> what happened to the smarter than me people?
[13:31] <lotuspsychje> !patience | VasTappendage
[13:32] <lotuspsychje> VasTappendage: another alternative is re-ask your question to the channel, with all details
[13:32] <samba35>  gvfsd[3076]: mkdir failed on directory /var/cache/samba: Permission denied
[13:33] <samba35>  brcmsmac bcma0:1: START: tid 1 is not agg'able
[13:33] <samba35> i have installed b43 package but i have no idea how to configure it
[13:34] <samba35> b43-fwcutter
[13:34] <samba35> firmware-b43-installer
[13:37] <mattfly> anyone using nvidia proprietary graphics driver has some serious issues when returning to the desktop after suspension?
[13:37] <mgedmin> VasTappendage: sorry!  have you already tried dpkg --configure -a && apt dist-upgrade -f ?
[13:37] <mattfly> like this https://ibb.co/3FB5JZc
[13:37] <lotuspsychje> samba35: you asking 3 questions at the same time?
[13:37] <mattfly> https://ibb.co/cTgbWHj
[13:37] <mattfly> is there any workaround for those glitches, is that something known?
[13:38] <lotuspsychje> mattfly: ubuntu version? graphics card chipset? driver version?
[13:39] <mattfly> ubuntu 18.04.2, nvidia-430, gtx 1050 ti
[13:39] <lotuspsychje> mattfly: stock driver or from a ppa?
[13:39] <samba35> i thought they are co related as b43 driver might has issuse
[13:40] <mattfly> propretary driver
[13:41] <lotuspsychje> mattfly: for GTX we reccomend the ubuntu graphics ppa, try out a few drivers from there as a test
[13:41] <lotuspsychje> !nvidia | mattfly
[13:41] <mattfly> https://ibb.co/7WYsk2H
[13:42] <mattfly> I think im already using that
[13:42] <lotuspsychje> mattfly: try a switch to 418 or 415 perhaps? see if you can reproduce
[13:42] <mattfly> okay
[13:45] <lotuspsychje> samba35: did you not say you installed ubuntu server on your mac?
[13:45] <VasTappendage> thanks .. you know how it is. mgedmin.. yes managed to get it into rootshell ran apt -dist-upgrade from there it is chugging away installing freebsd  or something ..will see when the bell goes off.
[13:47] <mgedmin> hopefully when the distro is fully upgraded, all the problems will be gone!
[13:47] <mgedmin> (I had many broken ubuntu upgrades that were successfully finished with a command-line apt-get dist-upgrade -f)
[13:49] <datasmurf> leftyfb: Ok, i got it "working".
[13:55] <Assid> i have a SSD drive of an installed version of ubuntu.. is there a  way i can make it into a VM
[13:58] <SwedeMike> Assid: yes, you can either put the SSD into another computer or just block copy the entire drive to a file somewhere else and boot off of it in a VM.
[14:01] <Assid> SwedeMike: once i put it into another pc.. not sure what to do exactly
[14:02] <Steveccc> I want to download ubuntu to try out in a VM on a mac so 64bit. Does it make any difference choosing the 32bit image over the 64 when running in a VM
[14:02] <SwedeMike> Steveccc: there is no reason not to choose 64bit.
[14:03] <Steveccc> swedemike: thanks - i used to run ubuntu years ago and considering it for a production laptop for development would you recommend the LTS over the other or maybe server version to run the web server on the same machine that i am developing on?
[14:04] <samba35> i was trying to find more information about mac its mac book pro 6 2
[14:04] <Assid> SwedeMike: knmow a good resource i can use?
[14:04] <cfhowlett> Steveccc= LTS all the way!
[14:05] <rypervenche> Steveccc: If you're going to be using it as a desktop, then normal Ubuntu is fine and you can still run a web server on it. I'd use an LTS if you want stability. If you want newer packages, then maybe go with a newer release. Depends on your needs.
[14:05] <SwedeMike> Steveccc: depends on your requirements, what you're developing for and their requirements.
[14:05] <mgedmin> do you enjoy upgrading your OS every 6 months?  then use the latest; otherwise use the LTS
[14:05] <lotuspsychje> samba35: did you not say you installed ubuntu server on your mac few days ago?
[14:05] <mgedmin> running a web server on your dev laptop is a very normal thing to do
[14:05] <SwedeMike> Assid: what kind of computer are you putting it in, and what OS does it run and what do you need to do with it?
[14:05] <Assid> Steveccc: if youre asking about it.. i think LTS would be best for you
[14:05] <mgedmin> the only difference between server and desktop versions are what packages get installed by default
[14:05] <Steveccc> its web development - php and mysql mainly so just want stability, speed etc but also good driver support for the machine
[14:06] <samba35> yes
[14:06] <rypervenche> Steveccc: Then go with LTS.
[14:06] <Steveccc> ok thanks
[14:06] <lotuspsychje> samba35: ask your issues in #ubuntu-server then please
[14:06] <samba35> i give up that idea that did not work
[14:06] <Assid> SwedeMike: sorry mate.. windows desktop pc , probably either hyperv or virtualbox for the hypervisor
[14:06] <lotuspsychje> samba35: so you have desktop installed now?
[14:06] <leftyfb> mgedmin: 9 months
[14:06] <samba35> yes
[14:06] <lotuspsychje> samba35: ok..
[14:07] <SwedeMike> Assid: ok, I don't know how those work on windows.
[14:07] <samba35> i have two hardisk swap hardisk
[14:07] <Assid> SwedeMike: have a guide for ubuntu ?
[14:07] <leftyfb> Assid: it's less work to just install ubuntu from scratch and reinstall your applications and restore the settings you need from the SSD
[14:08] <SwedeMike> Assid: I can google it, https://askubuntu.com/questions/308897/convert-ubuntu-physical-machine-to-virtual-machine seems like one guide that touches this. You can probably find more.
[14:08] <Assid> leftyfb: lots of data on it .. i dont wanna re-create the database structure etc.
[14:08] <Assid> SwedeMike: yeah thats what i was seeing.. which pointed me here : https://askubuntu.com/questions/34802/convert-my-physical-operating-system-to-a-virtualbox-disk
[14:08] <leftyfb> Assid: depending on the DB you're using, that's literally just a dump and restore of the DB. A single command line
[14:09] <Assid> leftyfb: yes im aware.. but i was trying to avoid the whole re-creation of the box. since this was pretyt much configured
[14:09] <samba35> ok let me try server
[14:09] <samba35> bye for now
[14:09] <leftyfb> Assid: I'm just telling you, it's less work and cleaner to just reinstall and restore. Not to mention, you get better and the process.
[14:10] <AvidWolf43> hi everyone
[14:10] <Assid> alrite.. let me see if i can pull down any machine here.. it was a test dev box .. which was recycled BACK  to a windows desktop when the project was cancelled
[14:16] <lotuspsychje> welcome AvidWolf43 how can we help you?
[14:17] <AvidWolf43> anyone have experience with grafana / prometheus / metrics in general?
[14:17] <lotuspsychje> AvidWolf43: we try to focus on specific ubuntu problems in this channel, is it related?
[14:18] <AvidWolf43> lotuspsychje: ubuntu is the host but aside from that no
[14:18] <OerHeks> Grafana is not available in the Ubuntu repos
[14:18] <OerHeks> neither prometheus..
[14:18] <lotuspsychje> AvidWolf43: you can try #ubuntu-offtopic then or !alis to find a relevant channel
[14:19] <AvidWolf43> thx
[14:19] <BluesKaj> !prometheus
[14:19] <BluesKaj> !info prometheus
[14:20] <BluesKaj> odd
[14:21] <lotuspsychje> BluesKaj: apt-cache shows it here on bionic
[14:21] <OerHeks> it is an AWS thingy .. https://www.howtoforge.com/tutorial/monitor-ubuntu-server-with-prometheus/
[14:21] <BluesKaj> guess they reinstated prometheus in 19.10
[14:22] <BluesKaj> I'm testing 19.10 and it's in the repos
[14:22] <lotuspsychje> AvidWolf43: if it can help, you can also use the manpage for prometheus
[14:23] <AvidWolf43> lotuspsychje: I guess im just confused by all these tools and which would fit my use case
[14:23] <BluesKaj> lotuspsychje, guess the apt bot isn't up to date
[14:23] <AvidWolf43> for instance when I search for grafana I usually get it combined with prometheus, influxdb, elasticsearch, graylog
[14:23] <lotuspsychje> AvidWolf43: come to #ubuntu-discuss please
[14:24] <OerHeks> combined? as it is not in our repos, you follow a wild guide
[14:24] <BluesKaj> AvidWolf43, grafana is not in the repos
[14:24] <webmind> hi
[14:25] <webmind> can someone explain me how to prevent snap related tasks from being executed on boot?
[14:25] <leftyfb> webmind: what is your issue?
[14:25] <lordcirth> webmind, disable snapd?
[14:25] <lordcirth> However, yes, why?
[14:25] <webmind> I've got a nextcloud instance, but I don't want anything touching that snap before I do something as a user
[14:26] <webmind> there's an encrypted partition that needs mounting first
[14:26] <webmind> so I only want it to manually be started
[14:26] <leftyfb> webmind: disable the nextcloud service
[14:26] <webmind> I tried
[14:27] <webmind> but there is no one nextcloud service in systemd it seems
[14:27] <leftyfb> webmind: what exactly did you try? (what command) and why do you think it didn't work?
[14:27] <webmind> I tried looking for a service to disable
[14:27] <webmind> and I disabled snap-nextcloud-13453.mount, but that gets run anyway
[14:30] <webmind> now I did systemctl  disable snapd
[14:30] <webmind> but after a reboot, snapd is still running
[14:30] <mgedmin> webmind: I think you can systemctl edit snapd.service and add an After=your-special-filesystem.mount or something
[14:30] <webmind> or again
[14:31] <mgedmin> to the override file it creates
[14:31] <friendlyGoat> pardon i was asleep, tomreyn. im gonna look at that link of yours. i tried installing the new version of the driver but i dunno if i loaded it properly as i've never loaded a new driver before. gonna try looking into it. fortunately i can use my computer normally, it just bothers me seeing that little message before i load it up. i can use wifi too.
[14:32] <webmind> mgedmin: oh, how does that work?
[14:32] <mgedmin> webmind: 'systemctl disable' is tricky; it only prevents the service from starting if nothing else requires it
[14:32] <mgedmin> there's systemctl mask, but let's not go there
[14:32] <mgedmin> man systemd.unit describes what After=, Require= etc do
[14:32] <lotuspsychje> friendlyGoat: tomreyn might be afk for now, maybe idle in #ubuntu-discuss until he's back?
[14:32] <mgedmin> *Requires=
[14:33] <mgedmin> I'm fuzzy on the details myself, tbh
[14:34] <webmind> but that will cause systemd to try and mount it no?
[14:35] <webmind> I can't have an automated system try and mount the mountpoint
[14:36] <mgedmin> that's the difference between After and Requires, AFAIU
[14:36] <mgedmin> one waits until the thing becomes available, the other tries to make the thing available
[14:37] <webmind> Requires = Configures requirement dependencies on other units. If this unit gets activated, the units listed here will be activated as well.
[14:37] <webmind> so it isn't requires
[14:39] <webmind> So I guess I need to use after, but I can't make out what I need to fill in after After=, I guess some systemd module? How do I write that?
[14:40] <mgedmin> webmind: mount the thing you need mounted, then try systemctl list-units -t mount
[14:40] <mgedmin> see if you see your mount point there
[14:40] <mgedmin> it should be mount path with slashes converted to dashes, ending in .mount
[14:41] <mgedmin> afaiu systemd creates these units dynamically when it sees new filesystems getting mounted
[14:42] <mgedmin> yeah, the systemd.mount man page confirms that
[14:42] <cfhowlett> snap command to remove deprecated snaps??
[14:43] <webmind>  var-snap.mount it seems, so I do After=var-snap.mount
[14:43] <mgedmin> there's a risk maybe that if you do what I suggest and make snapd.service wait until the special mountpoint that it will time out after several minutes after boot and just fail instead of patiently waiting
[14:43] <mgedmin> but eh, at least it won't start with /var/snap not mounted
[14:43] <webmind> I'll find out now :)
[14:44] <webmind> can the timeout be lowered?
[14:44] <mgedmin> you mean, will it block the boot process?  I hope not, that would suck
[14:44] <webmind> hmm, I see startup messages for nextcloud being started
[14:46] <webmind> snapd and nextcloud still got started
[14:46] <mgedmin> ugh :/
[14:46] <webmind> maybe I should mask it, and manually unmaks it...
[14:49] <plongshot> I've tried 2 different screencasting software and both only record a black screen. One gives an error message and fails when run from the command line.  I need to create a short video for my boss before 9 am (less than an hour 15 min. CAn anyon pleas help?
[14:50] <webmind> which ones did you try?
[14:50] <webmind> masking snapd doesn't help either
[14:52] <plongshot> webmind: OBS (which I think was working on 16.04 before a dist upgrade to 18.04 - but iirc though).  And just installed Kazam to try it.  Both will record my voice from the mic but not any video.
[14:53] <plongshot> I did crate a sceen and add sources with obs (should have bee setup correctly)
[14:54] <benishor> plongshot: try simplescreenrecorder
[15:01] <tomreyn> friendlyGoat: so things work and it's just a nuisance, i see. well, you could try the newer firmwares. or try to learn to ignore it. another option would by to try and change the iwlwifi drivers' verbosity, but this bears the chance of loosing other messages you'd want to see.
[15:01] <friendlyGoat> how would i go about trying newer firmwares?
[15:02] <mgedmin> webmind: maybe ask on askubuntu.com?  this might be too complicated for irc
[15:05] <webmind> mgedmin: ok
[15:06] <lotuspsychje> webmind: or, re-ask here with all details, or talk to the #snappy channel
[15:09] <plongshot> webmind: I installed simplescreenrecorder and attempt to use but it has some kind of but where you select the output file. It does not allow to select the file from the nautilus pop up dialogue. The button in the upper right corner to seelct is inactivated always
[15:09] <plongshot> During setup process
[15:10] <friendlyGoat> tomreyn, i decided to remove verbosity from the driver, im looking into how to do it now. when i was in my drivers however i found some odd ones i dont know the purpose of. what are intel_telemetry_pltdrv and intel_telemetry_core
[15:10] <friendlyGoat> also im going to check to make sure my wifi is in fact working
[15:11] <friendlyGoat> wifi is in fact working, theres just that annoying error on boot
[15:13] <plongshot> Do you have to create an empty file for it or somethign?  aack!
[15:19] <plongshot> Thyis is what I get for output when trying to "start recording" with simplescreenrecorder:   ""
[15:19] <plongshot> [PageRecord::StartPage] Starting page ...
[15:19] <plongshot> [PageRecord::StartPage] Started page.
[15:19] <plongshot> [PageRecord::StartOutput] Starting output ...
[15:19] <plongshot> [Muxer::Init] Using format mp4 (MP4 (MPEG-4 Part 14)).
[15:19] <plongshot> [Muxer::AddStream] Using codec libx264 (libx264 H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10).
[15:19] <OerHeks> plongshot,  you know, don't paste like that
[15:20] <plongshot> good god
[15:20] <OerHeks> !paste
[15:20] <plongshot> I'm so sorry
[15:20] <plongshot> I'm horrified!
[15:21] <OerHeks> the only thing i can imagine, is that you run a wayland session?
[15:21] <plongshot> I'm so screwed. No screen capture software will work for me and I hav 39 min to get somethign done with it
[15:21] <plongshot> oh my god!
[15:21] <OerHeks> simplescreenrecorder works fine
[15:21] <plongshot> not for me
[15:21] <plongshot> and I don't have the time to look how to usee the dam thing
[15:22] <lordcirth> plongshot, try recordmydesktop? Optionally -gtk
[15:22] <popey_> plongshot: do you want help or not?
[15:28] <tomreyn> friendlyGoat: i don't know about intel_telemetry_*, where do you see those?
[15:28] <plongshot> yes yes
[15:28] <friendlyGoat> i did lsmod but its fine now i unplugged them
[15:28] <plongshot> I'm sorry I got freaked out and was off trying to make the bass* work
[15:29] <plongshot> I have till 9 am start time and could make a vid in a few min even just to show my proposal
[15:29] <plongshot> lordcirth:   popey_
[15:29] <plongshot> ?
[15:30] <plongshot> simple
[15:30] <mgedmin> plongshot: gnome-shell has a builtin screen recorder; start recording with ctrl+alt+r, stop recording with the same key, find the recorded .ogv file in ~/Videos after the encoding is done in the background (might take a few minutes)
[15:30] <popey_> plongshot: open a terminal, type "echo $XDG_SESSION_TYPE" and tell me what it says (without the quotes)
[15:30] <mgedmin> a red dot appears in the top bar when the recorder is active, so you can see if it works
[15:30] <plongshot> popey_: one moment..
[15:31] <plongshot> popey_: says "wayland"
[15:31] <popey_> there's your problem
[15:31] <mgedmin> many old X11-based screen recording tools do not work in wayland sessions
[15:31] <urgodfather> hey everyone. im having an environment issue. using bash and perlbrew. made .bash_profile to source perl5 but its not getting read or loaded or something. what am i overlooking. 18.04. thanks in advance
[15:31] <popey_> logout, log back in and choose xorg from the drop down session plongshot and then simple screen recorder will work
[15:32] <plongshot> I will try thank you
[15:32] <mgedmin> urgodfather: .bash_profile is getting read for login shells only; did you log out and log back in?  I don't think opening a new terminal tab counts as a login on ubuntu
[15:32] <urgodfather> mgedmin of course
[15:38] <urgodfather> i know this isn't best practice but i added a few lines to .profile to check for perl5 and if so, add it to PATH. maybe this will work. running compile right now
[15:39] <mgedmin> I do the same thing, except in my .bashrc
[15:39] <mgedmin> because .profile is loaded too rarely
[15:39] <urgodfather> maybe it was .bashrc
[15:39] <mgedmin> oh, the tricks of setting up PATH for your session, it's terrible
[15:40] <mgedmin> I'm still not getting the right $PATH when my browser launches my mail client
[15:40] <urgodfather> it was .profile not .bashrc
[15:42] <mgedmin> (my .profile sources my .bashrc, if the shell is bash, otherwise it sources my ~/.bashrc.path, which is dash-compatible)
[15:42] <urgodfather> personally since i only have to do this for me to compile on one project id rather just add it to it temp instead of perm. editing either requires a logout which is unfavorable
[15:42] <mgedmin> a logout is not necessary if you source your updated .profile/.bashrc in the existing terminal bash session
[15:43] <mgedmin> or if you do PATH manipulations in .bashrc and open a new terminal tab, which is why I do PATH manipulations in my .bashrc
[15:43] <mgedmin> of course then you have to be careful and check if the PATH is already modified so subshells don't keep adding the same directories to PATH over and over
[15:43] <mgedmin> computers are terrible
[15:45] <urgodfather> grr.... still getting perl error
[15:46] <urgodfather> mgedmin can i see your .bashrc to reference?
[15:47] <mgedmin> https://github.com/mgedmin/dotfiles; you probably want profile, bashrc, and bashrc.path
[15:52] <TJ-> urgodfather: mgedmin this is how I deal with adding to PATH in .profile: http://paste.ubuntu.com/p/GvPQ8t6fjs/
[15:53] <urgodfather> so perlbrew includes its own bashrc file. does that cause a conflict?
[15:53] <mgedmin> TJ-: does it work?  there seems to be an extra i in the P="i$HOME/bin" line
[15:53] <TJ-> mgedmin: haha, typo there, nicely spotted
[15:54] <mgedmin> urgodfather: noidea, I'd expect it to save that bashrc somewhere and modify your .bashrc to source it?  noidea; it's why I hate these curl|bash self-installers, you can never know how they modify your dotfiles
[15:54] <mgedmin> ... unless you keep your dotfiles under version control, which I do B)
[15:55] <mgedmin> TJ-: also it seems your script leaves the PATH with a trailing : -- doesn't that actually add the current working directory to the end of the $PATH?
[15:55] <mgedmin> yup, it does!
[15:58] <TudorMan> I'm trying to compile some software that depends on another piece of software that I've compiled. However whenever I run ./configure it says that it can't find the dependency (because I haven't installed it as a package, just self-compiled). How can I trick it into using my compiled version instead?
[15:58] <tomreyn> friendlyGoat: i looked into the module parameters exposed by iwlwifi (modinfo -p iwlwifi) but there don't seem to be any regarding verbosity. so your only option seems to be https://superuser.com/questions/351387/how-to-stop-kernel-messages-from-flooding-my-console - which applies to all modules, so you'd need to suppress all errors, which is maybe not something you want to do.
[15:59] <lotuspsychje> TudorMan: we dont really support own compiling on ubuntu, instead use the packages from the official repos
[16:00] <tomreyn> friendlyGoat: so i'd rather suggest you try the firmware update suggestion if you need a counter measure.
[16:00] <TudorMan> lotuspsychje: the packages on the repos aren't up-to-date enough
[16:00] <lotuspsychje> !latest | TudorMan
[16:00] <friendlyGoat> tomreyn, how would i go about doing that?
[16:01] <urgodfather> im banging my head, i added it to .bashrc and logout/login try env | grep PERL  and nothing
[16:03] <mgedmin> TudorMan: ./configure never looks at the package database, there must be some other reason
[16:03] <urgodfather> ah but if i $PATH | grep PERL i see it
[16:03] <urgodfather> $PATH | grep PERL
[16:03] <urgodfather> bash: /home/ov//perl5/perlbrew/etc/bashrc:/home/ov//perl5/perlbrew/etc/bashrc:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin: No such file or directory
[16:03] <urgodfather> except it has 2 /
[16:03] <urgodfather> why?
[16:03] <mgedmin> eh, extra / are harmless
[16:03] <TudorMan> mgedmin: if I install the dependency it requests, it goes through. But I want it to use the version I've built myself, rather than the version in repos
[16:03] <urgodfather> ::thumbsup::
[16:03] <mgedmin> wow do not add /home/ov//perl5/perlbrew/etc/bashrc to $PATH, it's a script, not a directory!
[16:04] <urgodfather> maybe it will complete this time?
[16:04] <mgedmin> rather, inspect it for safety and source ~/perl5/perlbrew/etc/bashrc from your own .bashrc
[16:05] <mgedmin> TudorMan: can you be specific about the names of the software and the depencency?  it's hard for me to be abstract at this hour
[16:05] <urgodfather> thats what i just did :D
[16:05] <TudorMan> mgedmin: specifically I'm trying to build ifuse using a self-compiled version of libimobiledevice
[16:06] <mgedmin> how are you installing your self-compiled libimobiledevice?
[16:07] <mgedmin> most importantly, are you passing a --prefix option to ./autogen.sh when you follow the instructions at https://github.com/libimobiledevice/libimobiledevice?
[16:08] <TudorMan> mgedmin: I haven't installed it (I generally try to avoid make install, or at least use checkinstall), and no I haven't passed a prefix
[16:08] <mgedmin> ah, there's your problem
[16:09] <anikras> hillyu, I have ubuntu 18.04 installed and I don't know if I can encrypt my hard drive or my partition
[16:11] <mgedmin> I would probably try to install (without sudo!) into ~/opt/ifuse or something
[16:11] <mgedmin> so for libimobiledevice, ./autogen.sh --prefix=~/opt/ifuse && make && make install
[16:12] <TudorMan> mgedmin: okay, running that now
[16:12] <mgedmin> and then inside ifuse dir, PKG_CONFIG_PATH=~/opt/ifuse/lib/pkgconfig ./autogen.sh --prefix=~/opt/ifuse && make && make install
[16:13] <mgedmin> I'm not 100% sure about the PKG_CONFIG_PATH, it might be share/pkgconfig instead of lib/pkgconfig
[16:13] <mgedmin> look inside ~/opt/ifuse and find where all the *.pc files are
[16:13] <mgedmin> and then to run it you'd have to add ~/opt/ifuse/bin to your $PATH
[16:13] <mgedmin> and possibly set LD_LIBRARY_PATH?  not 100% sure about that
[16:14] <TudorMan> mgedmin: okay sounds good, I'll give it a go
[16:14] <mgedmin> there's a reason installing software from scratch is not exactly recommended; there's bazillions of details that packages take care of
[16:16] <urgodfather> mgedmin just saw your post about not setting to PATH good catch!! so you are saying to use the second one not the first, right https://paste.ubuntu.com/p/jbBS8QDPfX/
[16:17] <mgedmin> both are wrong
[16:17] <mgedmin> no!
[16:17] <urgodfather> yeah i need to clean it up
[16:17] <mgedmin> I cannot read at this hour, sorry
[16:17] <mgedmin> second is right
[16:17] <mgedmin> no!
[16:17] <mgedmin> I cannot read at this hour, sorry!!
[16:17] <mgedmin> second is half-right -- change it to source "$HOME/perl5/perlbrew/etc/bashrc"
[16:17] <mgedmin> without the :$PATH bit
[16:17] <lotuspsychje> easy on the enter button mgedmin
[16:17] <urgodfather> thats the part im removing as we speak
[16:18] <urgodfather> lol at lotuspsychje
[16:20] <VasTappendage> mgedmin.. thanks for nfo. will keep banging away at it till it resolves. its always something.
[16:21]  * urgodfather loves it when vendors publish only half-baked source because they have to for GPL, but never the pubilsh the fixed stuff
[16:26] <TudorMan> mgedmin: awesome, thanks for your help, that's got it working, and hopefully that info will be useful for me in future.
[16:27] <urgodfather> with the exception of the mk .bash_profile line, does the perl install seem right?
[16:27] <urgodfather> https://paste.ubuntu.com/p/fSNtqDXqkr/
[16:40] <TudorMan> mgedmin: instead of adding ~/opt/ifuse/bin to my $PATH, can I just cd to there and run them that way?
[16:46] <webmind> is there a good way to run a nextcloud snap on an encrypted storage that needs ssh-user-interaction to unlock?
[16:48] <pragmaticenigma> webmind: simple solution is to not allow access to the webserver except through 127.0.0.1... then force users to login via SSH with a tunnel setup to link them to the webserver
[16:48] <pragmaticenigma> (assuming the SSH server and webserver are on the same machine)
[16:49] <webmind> ah no
[16:50] <webmind> I meant that nextcloud runs off an encrypted partition
[16:50] <webmind> and a ssh-user needs to unlock that partition
[16:51] <pragmaticenigma> webmind: That isn't a recommended setup. Nextcloud will want to run on boot, if the drive/partition is encrypted it will fail to start... you would have to unlock the partition and startup the next cloud instance each time
[16:52] <leftyfb> webmind: why not just make a script that you run via ssh that mounts your encrypted partition and then restartes the nextcloud services?
[16:52] <pragmaticenigma> kind of defeats the purpose of an encrypted drive if the password is on the same machine as the encrypted volume leftyfb
[16:53] <leftyfb> pragmaticenigma: I didn't say anything about saving the password anywhere
[16:54] <webmind> leftyfb: because on boot, nextcloud starts and creates a new nextcloud setup and gets confused it seems
[16:54] <pragmaticenigma> leftyfb: I assumed "script" => save password in script file
[16:54] <webmind> pragmaticenigma: starting the nextcloud instance manually is fine, problem I have is preventing it starting automaticly
[16:55] <leftyfb> webmind: start it automatically. Who cares if you're just going to login manually and mount the encrypted drive and then restart the services?
[16:55] <webmind> nextcloud seems to care, it breaks
[16:57] <webmind> I'm mountint /var/snap or /var/snap/nextcloud, the latter I haven't got working yet
[16:57] <webmind> *mounting
[16:57] <webmind> but there is also data in /snap
[16:58] <leftyfb> webmind: yes, it will be broken until you login and mount/restart
[16:58] <leftyfb> wait
[16:58] <leftyfb> why are you mounting anything in /var/snap?
[16:58] <webmind> no it is broken -after- mount & restart
[16:58] <webmind> because that's where the data of nextcloud is
[16:58] <leftyfb> so change that
[16:59] <webmind> how?
[16:59] <leftyfb> webmind: read the nextcloud documentation
[16:59] <webmind> and with data, I mean all the data
[16:59] <webmind> inc the database
[17:00] <webmind> leftyfb: I did, and unless I missed it, there was no easy option for changing that
[17:01] <webmind> oh wait
[17:01] <webmind> misread a bit
[17:01] <webmind> perhaps
[17:03] <webmind> how do you change $SNAP_DATA ?
[17:03] <leftyfb> webmind: I mean just the storage files. You really shouldn't care about the DB all that much.
[17:03] <TJ-> webmind: is snap an essential part of what you're doing? are you using it because it includes a web server and php ?
[17:04] <webmind> TJ-: it seemed the easiest route for running nextcloud
[17:04] <webmind> leftyfb: it's kindoff a requirement, partly due to not knowing what is in the data and I don't want to leak any meta-data
[17:05] <webmind> *in the database
[17:05] <webmind> I'm guessing filenames might end up there
[17:06] <popey_> webmind: you said "there's data in /snap" - there isn't. /snap is read only, contains the app
[17:06] <leftyfb> webmind: maybe start with just reconfiguring the data store and see if that works for you. Keeping the DB in the /var/snap and unencrypted. If it works, then tackle the DB stuff after
[17:06] <leftyfb> popey_: /var/snap
[17:06] <popey_> leftyfb: "< webmind> but there is also data in /snap"
[17:07] <webmind> popey_: I meant that app as data, I think it gets confused because there is an app there, and no data in /var/snap/ and then created a new app
[17:07] <leftyfb> oh. Yeah, that should be ignore is is probably what's causing all these issues if he's trying to mount/encrypt /snap
[17:07] <popey_> yeah, you can't modify /snap it's just a mount point of the .snap file buried in /var/lib/snapd/snaps
[17:07] <webmind> I haven't tried mounting /snap yet
[17:08] <webmind> but something causes confusion when nextcloud gets started with an empty /var/snap
[17:08] <popey_> it gets mounted when the snap is installed
[17:09] <webmind> what does?
[17:09] <popey_>  the /snap/nextcloud mount point
[17:09] <webmind> popey_: oh sure
[17:10] <popey_> The snap might not work for your use case.
[17:10] <popey_> I would be interested to see it documented if you get it working
[17:10] <webmind> leftyfb: it quite likely works to have just file-data encrypted, but it's also very likely that doesn't work for this usecase
[17:11] <TJ-> webmind: presumably your issue is the snap assumes the data is available unencrypted under /home/$USER/ ?
[17:11] <webmind> the request I got is to have all nextcloud data encrypted and unlocked via ssh
[17:11] <TJ-> webmind: so if $HOME for $USER isn't unlocked, then it doesn't find the data, and creates a new empty database thinking its the first time it ran
[17:11] <webmind> TJ-: it created new data in /var/snap after the next reboot, a next revision?
[17:11] <webmind> and that revision nr is in /snap
[17:12] <webmind> I don't think it touched $HOME
[17:12] <TJ-> webmind: hmm, in which case I canot see how encrypted $HOME is affecting it, unless the snap tries to read something from there
[17:12] <webmind> if I someone could prevent nextcloud from starting on boot, I think this should work
[17:12] <popey_> it will be running daemons which will probably look /root/snap/nextcloud
[17:12] <webmind> TJ-: this is not about an encrypted $HOME
[17:12] <popey_> (as the $HOME of the user the daemons run as)
[17:13] <Brumble> SchrodingersScat, fixed! :D https://streamable.com/ncebm
[17:13] <webmind> the encrypted partition in mounted on /var/snap
[17:13] <Brumble> I also got it's ac adaptor today.
[17:13] <webmind> or well, now I tried /var/snap/nextcloud, but then nextcloud won't install
[17:13] <TJ-> webmind: oh! sorry I misread earlier
[17:17] <TJ-> webmind: if you were using a regular install then you could use systemd mounts/ordering to manage it cleanly
[17:18] <webmind> TJ-: yeah, I tried that with this, but the 'After=' thing didn't actually seem to work
[17:19] <webmind> systemd ran the processes anyway
[17:19] <TJ-> webmind: After= on its own won't it also needs a Wants= or better still a Requires=
[17:19] <webmind> Requires= isn't an option, because systemd can't mount the mountpoint
[17:20] <TJ-> webmind: indeed, if you're using snap, unless you add an override to the snap so it isn't wanted by multi-user.target
[17:20] <TJ-> webmind: to snapd/snapd-socket/whichever!
[17:20] <webmind> no in general, systemd will never be able to mount this mountpoint
[17:23] <webmind> so I'm not sure how systemd will manage to deal with this
[17:27] <TJ-> webmind: if you have a systemd.mount for the encrypted device that when it mounts has a Triggers=snapd.service (or whatever!) linked to a systemd-cryptsetup@var_snap.service which the user activates from SSH, that would work
[17:29] <webmind> TJ-: but what about services that ignore After= ?
[17:29] <webmind> they still got run
[17:30] <TJ-> webmind: you have other services with After=snapd.service ?
[17:30] <webmind> I meant for instance snapd.service having After=<mountpoint>.mount but ignoring that
[17:31] <TJ-> webmind: as I said, After= isn't sufficient, that jsut controlls *ordering*. You want snapd.service to be delayed until other units are ready, it needs a Requires= or Wants=
[17:32] <webmind> ok, that doesn't seem to match with how I read the manual, but dinnertime.
[17:32] <TJ-> webmind: as long as snapd.service isn't a wants of multi-user.service or some other service that m-u wants, it'll work. You can drop an override file into /etc/systemd/... to unlink snapd*.* units from m-u
[17:34] <TJ-> webmind: what I mean is, you need to cancel/remove the /etc/systemd/system/multi-user.target.wants/snapd.service -> /lib/systemd/system/snapd.service so snapd not starting doesn't hang the boot
[17:37] <adac> Guys my firefox does not load any page anymore
[17:38] <TJ-> webmind: you'd do that simply with dropping /etc/systemd/system/snapd.service.d/override.conf with "[Install]\nWantedBy=" (I think - not sure if Install section gets re-evaluated at runtime)
[17:38] <adac> other browser work just fine
[17:38] <TJ-> adac: did Firefox recently upgrade?
[17:39] <adac> TJ-, yes it said that it cannot load pages anymore and I would need to restart it
[17:39] <adac> which I did
[17:39] <adac> but still no pages are loaded
[17:39] <TJ-> adac: there are a few suggestions here: https://support.mozilla.org/en-US/kb/firefox-cant-load-websites-other-browsers-can
[17:40] <TJ-> adac: we're had a few similar reports recently
[17:41] <adac> TJ-, ok I see. Are there workarounds? Do you perhaps know?
[17:41] <sarnold> is this it? https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/1830096
[17:41] <TJ-> adac: read the article
[17:41] <adac> cannot open it^^
[17:42] <TJ-> adac: open a terminal do "w3m https://support.mozilla.org/en-US/kb/firefox-cant-load-websites-other-browsers-can"
[17:43] <adac> hehe :)
[17:43] <adac> Opened it with chromium for now
[17:44] <TJ-> Terminal is king
[17:44] <adac> true :)
[17:50] <adac> sarnold, TJ-  I  don't get a message like "Using an older version of Firefox"  as in this link
[17:50] <adac> So I guess it is another issue?
[17:50] <ioria> adac, a new profile or Refresh Firefox
[17:51] <adac> ioria, how to refresh Firefox?
[17:51] <ioria> adac, about:support   (would be resetted: warning !)
[17:53] <adac> ioria, there is also safe mode
[17:53] <adac> I tried this now
[17:53] <adac> that works
[18:04] <urgodfather> whats the right way to remove something from path that i added via bashrc? remarking and reboot does not correct.
[18:04] <sarnold> what does it mean to "remark" something?
[18:06] <urgodfather> ##
[18:06] <sarnold> aha. that should work
[18:06] <urgodfather> didnt
[18:06] <urgodfather> idk y
[18:10] <lordcirth> urgodfather, probably you have it set somewhere else
[18:11] <urgodfather> thanks lordcirth looking into
[18:27] <CoolerZ> help
[18:27] <CoolerZ> I have 2 cursors on screen
[18:27] <CoolerZ> it happened when i logged out and logged back in
[18:27] <lordcirth> CoolerZ, do they move together?
[18:27] <CoolerZ> the other cursor is stuck
[18:27] <OerHeks> restart gdm perhaps ?
[18:28] <OerHeks> unless you have a fancy desktop/wm
[18:28] <CoolerZ> how
[18:28] <OerHeks> how
[18:28] <OerHeks> ??
[18:28] <lordcirth> CoolerZ, check Settings -> Displays; is there an "Unknown monitor"
[18:28] <OerHeks> tell us your ubuntu version, perhaps?
[18:28] <CoolerZ> 18.04
[18:29] <CoolerZ> lordcirth, nope
[18:29] <CoolerZ> just the built in display of the laptop
[18:29] <OerHeks>  sudo systemctl restart gdm3
[18:34] <CoolerZ> OerHeks, nope that was a bad ide
[18:34] <CoolerZ> a
[18:34] <CoolerZ> it went into a boot loop
[18:34] <OerHeks> then your system had an issue, restarting gdm3 should be oke
[18:35] <OerHeks> some zombieprocess or the like
[18:35] <CoolerZ> yeah there are lots of issues
[18:36] <CoolerZ> Everytime i restart the laptop i get the "System encountered an error" message
[18:36] <CoolerZ> and i click "Report problem"
[18:55] <tomreyn> CoolerZ: which graphics card do you have there?
[18:56] <tomreyn> and which driver?   lspci -knn | grep -A3 VGA | nc termbin.com 9999
[19:30] <eel_breath> When installing programs with python dependencies, how do I prevent my server (18.04) from installing python2.7 (instead of using 3.67/3.7) that are already installed?
[19:32] <lordcirth> eel_breath, programs can depend on either python 2 or 3.
[19:32] <lordcirth> If it needs python2, then 2.7 will be installed.
[19:34] <OerHeks> there is no override option to use 3.x
[19:34] <eel_breath> lordcirth I get that, but when I install flexget (3.3-3.6 supported), 2.7 shows up as a dependency
[19:36] <sarnold> python 2 and python 3 are different languages.
[19:36] <sarnold> it would have been better to name python3 "blackadder" instead.
[19:38] <lordcirth> eel_breath, what package is that?
[19:38] <ioria> eel_breath, are you sure flexget is an ubuntu pkg ?
[19:38] <leftyfb> it's not
[19:38] <eel_breath> it's not
[19:39] <ioria> right
[19:39] <leftyfb> eel_breath: you'll need to contact flexget for support
[19:40] <OerHeks> eel_breath, there is a #flexnet channel here on #freenode ?
[19:40] <OerHeks> i cannot find good info about 3.7, only the changelog
[19:41] <OerHeks> FlexGet version: 2.20.16 supports Python version: 3.6.7
[19:41] <srji> i did install i3-wm, but i am not able to find a option at the login screen to set i3-wm as window manager for the session
[19:41] <ioria> srji, did you reboot ?
[19:42] <eel_breath> thanks, I'll check it out - but it's not just flexget, there are at least 2 others that I've installed (apt-get) and the dependencies that show up are for 2.7 when they are compatible with 3.x
[19:42] <srji> i did
[19:42] <ioria> srji, how did you install it ?
[19:42] <srji> via apt
[19:43] <ioria> srji, what cmd exactly ?
[19:43] <OerHeks> what guide did you follow?
[19:43] <srji> sudo apt install i3-wm
[19:43] <ioria> srji, i'd say i3 not i3-wm
[19:44] <ioria> srji,   sudo apt install --reinstall i3
[19:44] <lordcirth> I thought the same, but I don't see anything in that metapackage that is relevant?
[19:49] <srji> still no option, sry
[19:49] <ioria> srji,   what vrsion of ubuntu and what dm ?
[19:50] <srji> oh wait
[19:50] <srji> i found the option
[19:51] <srji> its a small cog wheel close to the login button
[19:51] <srji> doubt
[19:53] <srji> ty
[20:15] <urgodfather> finally!!!!
[20:15]  * urgodfather doing happy dance
[20:17] <coconut> which irc dance this time, urgodfather?
[20:18] <urgodfather> the ubuntu rules happy dance
[20:18] <urgodfather> :D
[20:20] <coconut> yeah, that one rocks, i know :)
[20:23] <urgodfather> coconut especially when a vendor tells you to use EOL version and wont support and then ubuntu says use newer version to get support with ubuntu
[20:26] <b33m> So I disabled my external monitor in 18.04 (still using the usb-c cable for power, just not video), but now that I want to use it again ubuntu doesnt seem to detect the external monitor @ all in display settings
[20:26] <b33m> Is there any way to get it to manually detect the monitor?
[20:26] <coconut> opensource can be(is) great, i agree
[20:28] <Mdlpe> hi, I do not remember how to send the result of the find command to make a tar archive
[20:42] <scientes> how do i get ssh -X to work with xwayland?
[20:42] <scientes> oh damn it just worked!
[20:43] <scientes> probably going to be uuuuuuuuuunusable however
[20:44] <leftyfb> I didn't think -X was supposed to work with wayland
[20:44] <leftyfb> oh, xwayland
[20:45] <scientes> yeah its too slow with this much latency
[20:55] <blackflow> scientes: try x2go
[21:00] <JonHanDin> Hello peeps, have any of you used ElectronMail on Ubuntu? I'm concerned that with it being 3rd party it may expose / degrade my security. Am i just being paranoid?
[21:10] <gvvg_> Hi - I made a huge mistake and tried to enable php7.3 on my hosting server - now apache won't start -I'm trying to downgrade to 7.0 which is what it was using before but I am getting this error
[21:10] <jadax> if it ain't broken, don't fix it
[21:10] <gvvg_> https://pastebin.com/JyQf5852
[21:10] <jadax> how much money are you losing every hour, gvvg_?
[21:11] <gvvg_> jadax: good advice - I feel very stupid
[21:11] <gvvg_> any suggestions? jadax
[21:11] <jadax> how did you install php7.3?
[21:11] <gvvg_> apt-get
[21:11] <jadax> did you use apt-get?
[21:12] <tomreyn> also, which ubuntu release is this
[21:12] <gvvg_> actually it was installed already I just tried to enable it
[21:12] <gvvg_> 16.04 LTS
[21:12] <tomreyn> how was 7.3 installed already?
[21:12] <jadax> any chance you can update OS to 18.04 LTS? that's the distribution that's currently supported everywhere
[21:12] <gvvg_> I'm not sure I only do apt upgrade
[21:13] <tomreyn> 16.04 offers php 7.0.33 and nothing else
[21:13] <gvvg_> It must have been my old sysadmi
[21:13] <gvvg_> admin
[21:13] <jadax> 18.04 offers php 7.3
[21:13] <gvvg_> I don't know
[21:13] <tomreyn> jadax: 16.04 LTS is supported, too
[21:13] <gvvg_> I see it in the apt-cache search
[21:13] <jadax> well, I figured he's got the system messed up currently, might update now
[21:14] <jadax> as he would have to update at some point in future anyways
[21:14] <gvvg_> any ideas where to start?
[21:14] <tomreyn> gvvg_: please post this:  sudo /bin/true && cat &>/tmp/aptlog < <(sudo apt-get -qqy update 2>&1; apt-cache policy 2>&1; sudo apt-get -syV full-upgrade 2>&1;); nc termbin.com 9999 </tmp/aptlog; rm /tmp/aptlog
[21:14] <tomreyn> gvvg_: also this: nc termbin.com 9999 < <(lsb_release -ds;cat /proc/{version,cmdline};echo "$XDG_CURRENT_DESKTOP ($XDG_SESSION_TYPE)";)
[21:15] <tomreyn> gvvg_: and this:  dpkg -l php* | nc termbin.com 9999
[21:15] <sarnold> gvvg_: do you have an application that can't run on 7.3? if so be sure to file a bug report with them while you're at it
[21:15] <gvvg_> https://termbin.com/s4g8
[21:16] <gvvg_> https://termbin.com/w3j1
[21:16] <gvvg_> https://termbin.com/4jx9
[21:16] <gvvg_> no 7.3 is not a requirement right now
[21:16] <tomreyn> you have this PPA configured which offers co.installable PHP versions http://ppa.launchpad.net/ondrej/php/ubuntu
[21:17] <gvvg_> ok cool
[21:17] <gvvg_> any idea why apache2 won't starty?
[21:19] <tomreyn> gvvg_: no, you haven't shown any logs showing appache httpd not starting, ye3t
[21:19] <gvvg_> sorry
[21:19] <gvvg_> https://pastebin.com/JyQf5852
[21:19] <gvvg_> tomreyn: I did share them before - here they are again
[21:19] <tomreyn> those logs don't show that
[21:20] <gvvg_> sorry this is the apache2 error log which one would be helpful?
[21:20] <tomreyn> there is "caught SIGTERM, shutting down", but this can be a normal exit
[21:20] <tomreyn> do you actually know that it fails?
[21:21] <tomreyn> how do you tell?
[21:21] <gvvg_> https://pastebin.com/GCPPsXky
[21:22] <gvvg_> May 28 17:06:43 web2 systemd[1]: apache2.service: Control process exited, code=exited status=1
[21:22] <gvvg_> May 28 17:06:43 web2 systemd[1]: Failed to start LSB: Apache2 web server.
[21:22] <jadax> do you have enough memory on the system?
[21:22] <jadax> free -h would show
[21:22] <gvvg_> plenty 16GB
[21:23] <gvvg_>        total        used        free      shared  buff/cache   available
[21:23] <gvvg_> Mem:            15G        482M         13G         31M        1.4G         14G
[21:23] <gvvg_> the issue started when tried to enable php7.3 from php7.0
[21:23] <gvvg_> I must have missed a step
[21:23] <tomreyn> so apparently module php7_module is loaded multiple times. apache2 already fails while trying to parse the configuration file. this suggests not only a misconfiguration but rather a mix of incompatible modules or something similar.
[21:23] <gvvg_> right
[21:24] <gvvg_> how do I find the offending conf lines?
[21:24] <sarnold> gvvg_: if you want to use ondrej's php packages it's probably a good idea to remove all the ubuntu-supplied php packages
[21:24] <tomreyn> have a look at https://termbin.com/4jx9, maybe grep ^i
[21:25] <tomreyn> you have php packages installed using different php sapi versions
[21:25] <sarnold> sapi?
[21:25] <gvvg_> so I look for lines starting with i
[21:25] <gvvg_> and make sure the versions correspond to php7.0?
[21:26] <gvvg_> is there anyway to see which config line it stops on?
[21:26] <tomreyn> sarnold: the versioned php api, isn't it called SAPI?
[21:26] <sarnold> tomreyn: I don't know :) heh
[21:26] <sarnold> tomreyn: I was wondering if there was something in this dpkg -l output that I had overlooked re a sapi verison
[21:27] <tomreyn> no, no. "Server Application Programming Interface"
[21:27] <tomreyn> basically different versions of the mod_php API
[21:27] <gvvg_> I think it's php-common
[21:27] <gvvg_> not sure how I can install php-common for php-7.0
[21:28] <tomreyn> gvvg_: so php 7.0 is what you want, and only that?
[21:28] <gvvg_> yes
[21:28] <sarnold> gvvg_: probably the ppa-purge package would be a good way to get there
[21:28] <gvvg_> I'm not sure I understand?
[21:29] <tomreyn> gvvg_: "note down" which pacages are now installed, so you can ensure you'll re-install those you need (php modules mostly) later:   dpkg -l php* | grep ^i > php_packages_which_were_installed.txtr
[21:29] <tomreyn> gvvg_: "note down" which pacages are now installed, so you can ensure you'll re-install those you need (php modules mostly) later:   dpkg -l php* | grep ^i > php_packages_which_were_installed.txt
[21:30] <sarnold> you could install the ppa-purge package and then use the ppa-purge command to remove ondrej's PPAs
[21:30] <gvvg_> is there a way to install php-common for php7.0?
[21:31] <tomreyn> you need to downgrade this and sevceral other php mpackages, ppa-purge can do so for you.
[21:31] <tomreyn> you need to downgrade this and several other php packages, ppa-purge can do so for you.
[21:31] <tomreyn> sorry about the typos
[21:32] <gvvg_> I'm quite confused
[21:32] <gvvg_> if I try and install php7.0 wouldn't that work?
[21:33] <tomreyn> maybe this would work, but it would not suffice, there's more to be done
[21:33] <sarnold> I think you'd be better served by removing the mess that you've got onw
[21:33] <tomreyn> and these commands can do it for you:   sudo apt update; sudo apt install ppa-purge; sudo ppa-purge ppa:ondrej/php
[21:33] <sarnold> and *then* install the packages you want
[21:34] <tomreyn> i agree
[21:35] <tomreyn> the above commands would remove the ondrej ppa and its packages. you would then be able to install ubuntu's php7 version and modules
[21:35] <causative> I have a loud fan.  If I physically stop this fan, my system shuts down after a couple minutes, despite all cpu temperatures being low.  pwmconfig detects no pwm-capable sensors that I could use to control the fan.  What can I do to disable this fan?
[21:35] <gvvg_> ok I'm trying that
[21:36] <causative> physically disabling it would be fine if I could stop my system from shutting down as a result
[21:38] <gvvg_> all working again!
[21:38] <gvvg_> amazing
[21:38] <gvvg_> thank you very much
[21:39] <compdoc> causative, unplug it, or better yet, replace it
[21:39] <ryuguns> Hey
[21:39] <tomreyn> gvvg_: you're welcome. it's common to have a development system with (almost, ip addresses and a few other settings excluded) identical configuration for running such tests as you just did.
[21:40] <jadax> causative can you configure fan speeds/temperatures in BIOS/UEFI?
[21:40] <gvvg_> Right
[21:40] <gvvg_> Good advice
[21:40] <gvvg_> I'll be more careful in the future
[21:40] <gvvg_> Thank you again
[21:41] <causative> there is a problem getting into the BIOS because it is a laptop, the screen is broken but used automatically for the BIOS, and the external monitor does not turn on until I'm already past the BIOS
[21:42] <jadax> just remember what's the combination to get there
[21:42] <jadax> and keep pressing keys as you boot up
[21:42] <jadax> you will enter the BIOS and then screen comes up
[21:42] <causative> well but even once I'm there if it's still using the laptop screen, I won't be able to do anything
[21:42] <jadax> I see
[21:42] <tomreyn> gvvg_: you could configure the apache httpd virtualhosts on this webserver to accept requests for both maindomain.tld and dev.maindomain.tld, then just clone this system and have dev.maindomain.tld point to its (different) IP address, enabling you to easily  before doing run such tests.
[21:43] <compdoc> you wont be able to stop the fan making noise in the bios
[21:43] <jadax> compdoc often times you can lower the RPM
[21:43] <jadax> low RPM = no noise
[21:45] <tomreyn> gvvg_: if you created the php_packages_which_were_installed.txt file earlier (as i suggested then) you can check which modules you had installed - it is possible that you will need to manually reinstall those now, since they may be named differently on ubuntu's package archive than they are named on ondrej's PPA.
[21:47] <tomreyn> gvvg_: dpkg -l php* | grep -E '^(r|.c)' are PHP packages which are no longer installed but for which configurations may still be in place - you may want to "sudo apt purge PACKAGE" those (after ensuring you no longer need those configurations).
[21:49] <gvvg_> Excellent thanks
[21:49] <jadax> he's good
[21:56] <tomreyn> causative: you'd need to physically unplug the LCD - if that's possible without breaking the system (in the worst case this can mean disassembling the entire laptop - then plug in the external screen and hope for the bios output to show up there. it's also more of a question for ##hardware
[22:02] <Jon30> hey guys, i've been reading on Best Practices for SSH server, and none of them asked me to instal SSL on SSH server. Is installing SSL on SSH not a common practice? Is there a reason for that?
[22:03] <sarnold> Jon30: ssh provides for its own authentication and secrecy
[22:05] <OerHeks> libssl is a dependency https://launchpad.net/ubuntu/bionic/arm64/openssh-server/1:7.6p1-4
[22:05] <OerHeks> and my best guess, openssl is standard?
[22:05] <OerHeks> !info openssl
[22:06] <blackflow> yes and yes
[22:07] <OerHeks> so, it won  be pulled in, as dependencie, it is already there
[22:29] <TJ-> SSH doesn't use SSL, it uses the libcrypto part of libssl (which is libssl + libcrypto)
[22:31] <jayjo> I want to forward all ssh connections with the git user upstream, can I do this with an easy `Match User ForceCommand` ? I am only planning on matching and forwarding one user, the git user
[22:35] <tds> jayjo: is this for some git server appliance that exposes its own sshd? most I've seen are also able to generate an authorized_keys file to be used by the systems's sshd, and forcedcommands for each of those back to their own scripts
[22:38] <jayjo> that seems to sort of the recommended approach by the gitlab folks: https://blog.xiaket.org/2017/exposing.ssh.port.in.dockerized.gitlab-ce.html + https://github.com/sameersbn/docker-gitlab/pull/1731 (although I know this second repo is not official, the same approach seems to be taken from gitlab.com). I am also working with docker. Is there a reason to prefer the authorized_keys file way, vs fowarding all
[22:38] <jayjo> git user requests from sshd to gitlab-shell ?
[22:39] <TJ-> jayjo: probably be better using a ProxyCommand
[22:39] <gehn> how do I search for packages that statisfy the dependency of another package
[22:40] <TJ-> jayjo: if you use ForceCommand it makes it difficult for the user to update their key on the bastion host and the target
[22:40] <jeremy31> apt depends packagename
[22:40] <gehn> for example if I have a package that depends on libgl-dev how do find packages which provide libgl-dev
[22:41] <gehn> on Ubuntu 18.0.4 there is not a package named simply libgl-dev
[22:41] <tomreyn> then you won't be able to satisfy this dependency.
[22:42] <tomreyn> looks like you'Re working with incompatible packages. don't do that.
[22:43] <tomreyn> oh libgl-dev is a virtual package, ok
[22:43] <OerHeks> !find libgl-dev
[22:43] <TJ-> gehn:  awk '/^Package:/{P=$2} /^Provides: .*libgl-dev/{print P; P=""}' /var/lib/dpkg/status
[22:44] <nikolam> Why Ubuntu's ZFS does not have support for: com.delphix:spacemap_v2 (Space maps representing large segments are more efficient.)
[22:45] <nikolam> That is why I can't mount read/write OpenZFS dataset created on illumos/Openindiana
[22:46] <TJ-> nikolam: that's likelt because ZoL doesn't support it, but you'd need to check with the kernel devs who package it
[22:49] <TJ-> nikolam: which version of ZoL are you using?
[22:52] <nikolam> TJ-, it's currently XUbuntu 18.10
[22:53] <nikolam> 0.7.9-3ubuntu6
[22:53] <TJ-> nikolam: I *think* the features you're asking about weren't added to ZoL until 0.8, or later
[22:54] <Randolf> How can I get JavaFX working on Ubuntu 18.04.1?  Both Java and JavaFX are installed via apt, but only Java works.  I do see "javafx" files under the /usr/share/java/ directory.  Thanks.
[22:55] <nikolam> I'll be on 19.04 but I have ICH7 sound problem, with sound cracking all the time
[22:57] <gehn> how do releases work wrt to LTS, it seems that 19.04 is not yet LTS for example, and that the original 18.04 was not LTS until 18.04.2, if I upgrade to 19.04 now will the standard software-upgrade program get me onto LTS eventually or would I need to do a dist-upgrade to get that?
[22:57] <jayjo> TJ-: in this particular case, users manage their keys in GitLab, and GitLab would check via public keys. Does what you're saying about bastion hosts and targets apply still when GitLab is the target?
[22:57] <jayjo> re: using ForceCommand vs ProxyCommand
[22:58] <gehn> also, how viable is it to dist-upgrade from 18.04 to 19.04? I usually stay away from dist upgrades like that and tend to prefer a fresh install to avoid potential problems
[22:58] <TJ-> jayjo: I'm not entirely sure, its an unusual case. Best to set up tests systems and try both approaches
[22:58] <causative> I "solved" my fan problem:  by unscrewing 3/4 screws on the plate that covers the fan, and stretching a shoelace underneath that plate so that it presses against the fan hub, stabilizes the fan sufficiently that it can spin silently without wobbling/clattering
[22:59] <jadax> nice, you earned the DIY badge
[22:59] <TJ-> !lts | gehn
[22:59] <jadax> look how many problems could be solved with a shoelace
[22:59] <TJ-> gehn: LTS releases are every 2 years, so the sucessor to 180.4 will be 20.04
[23:00] <gehn> ohh, right, even numbers are LTS
[23:00] <gehn> ok, but as to the question of whether software-upgrade will get me on to the "real" LTS version, i.e. in context of the original 18.04 not being LTS until 18.04.2
[23:00] <tomreyn> !HWE | gehn: 18.04 was LTS from the start, you're mixing this up with
[23:01] <Bashing-om> gehn: The releas cycle is every 6 months. Those released in even 2 years in April are LTS. To upgrade from 18.04 to 19.04 - the path is through 18.10. The next LTS release is to be 20.04.
[23:01] <gehn> Bashing-om, tomreyn, ok thanks
[23:02] <TJ-> gehn: generally, upgrades *to* an LTS aren't automatically offered by the system until the first point release (so 18.04.1)
[23:02] <jadax> is there any reason why installing from scratch would be better idea than updating 16.04 to 18.04?
[23:02] <Randolf> jadax: No left-over baggage?
[23:02] <Randolf> jadax: Otherwise you should be fine.
[23:02] <TJ-> jadax: clearing cruft; changing architectures
[23:02] <jadax> I'm using AWS, so from time perspective it could take the same amount either way
[23:03] <TJ-> jadax: depends on how much configuration you need to carry over
[23:03] <jadax> what is left-over baggage? is there any in 2019?
[23:03] <Randolf> jadax: If starting fresh is just as easy, then I'd go for starting fresh.
[23:03] <gehn> jadax, historically with most distros I've regularly (or at least occasionally enough to be frustrating) had problems with doing an upgrade
[23:03] <gehn> well, starting fresh (at the moment) is not as easy for me
[23:04] <jadax> interesting, now that upgrades are so common and there are designed tools and procedures
[23:04] <causative> also solved the BIOS problem by unplugging the cable for the laptop screen from the motherboard, however, now it seems to take much longer to boot.  Perhaps it's spending time searching for the laptop screen?
[23:04] <TJ-> jadax: This last weekend I do-release-upgraded from 12.04 ESM through 14.04, 16.04, to 18.04, without too many problems, and that server started out with 5.04 (2005!)
[23:04] <gehn> well to be fair, I've been installing fresh like, forever now
[23:04] <Randolf> gehn: Start with a full backup first.  If you can test in a separate system, do so first.
[23:04] <gehn> I haven't tried doing upgrades in a long time
[23:04] <gehn> I do have some automated backup tools I wrote already, it's enough, but a fresh wipe/upgrade still tends to take a while overall
[23:05] <gehn> TJ-, well that's reassuring
[23:05] <brimonk> Hey, I installed some nvidia cuda packages and now my steam won't start. I get the error 'libGL error: No matching fbConfigs or visuals found' and 'libGL error: failed to load driver: swrast'. Any ideas on what I need to check?
[23:05] <TJ-> jadax: I had a lot of configuration changes to deal with due to various packages changing their syntax, but I dealt with those methodically
[23:05] <gehn> TJ-, but also there's a small thing I'd note "without *too many* problems" :)
[23:05] <gehn> sooo... there were some problems
[23:05] <jadax> TJ- what if everybody used POSIX
[23:06] <jadax> imagine that
[23:06] <TJ-> jadax: I was dealing with primarily with BIND9, postfix, dovecot, apache2, mysql, postgresql and several other server services
[23:07] <jadax> apache2 changed syntax too? you mean configuration files? vhosts etc?
[23:07] <TJ-> gehn: d-r-u-s through 3 release you've got to expect a few glitches, and be prepared
[23:08] <sarnold> the entire apache authorization and authentication mechanism was redesigned; only the simplest apache configs don't need changing over the last N years
[23:09] <Randolf> jadax: As long as you don't rely on any suEXEC stuff, the changes in the newest Apache 2 configuration files are pretty straight-forward.  Mostly to do with Require directives.
[23:09] <TJ-> jadax: yes, e.g. ReWriteLogLevel and RewriteFile retired for mod_rewrite in favour of LogLevel rewrite:trace2 and the primary log file
[23:09] <Randolf> sarnold: Yeah, that tripped me up with my ModPerl2 coding when I upgraded a while back.  I think it is better now though.
[23:10] <TJ-> jadax: and allow/deny replaced with "Require all granted" and so on
[23:10] <jadax> I usually keep a git repo for all apache related configs and symlink from my repo to /etc/apache
[23:10] <jadax> that somehow makes updates/transitions easier
[23:10] <jadax> do you do any tricks like that too?
[23:11] <TJ-> I've pushed on to apache2 2.4.39 so I can use mod_md (automatic letsencrypt handling)
[23:11] <TJ-> jadax: I have /etc/ as a git repo; there's an inotify hook that does a commit after a file is changed
[23:11] <jadax> if I have to move all my stuff from one machine to another it's quite painful otherwise
[23:12] <jadax> unless one did docker, maybe
[23:12] <jadax> but then docker comes with set of other challenges
[23:12] <jadax> that's smart, TJ-
[23:14] <gehn> is there somewhere that comprehensively describes the OpenGL packages in the standard Ubuntu package repos and how they're organized
[23:14] <TJ-> jadax: the git repo itself is on a separate file-system, that is snapshot-ed and cloned to a bacup system
[23:15] <gehn> it appears the core OpenGL lib and functionality is provided by mesa, but some of the details of what packages exist and why and what depends on what and why doesn't seem very clear. I also haven't found anything even close to an explanation of these packages on any of the community sites like help.ubuntu.org or wiki.ubuntu.org or the community forums etc
[23:15] <jadax> that by design also help with having secondary test infrastructure
[23:15] <jadax> not much effort to maintain it, right?
[23:16] <TJ-> it took some time to plan, test, and get right and be confident of it :)
[23:16] <jadax> would you consider doing docker instead?
[23:16] <TJ-> never
[23:16] <jadax> why not?
[23:16] <TJ-> if anything I'd use lxc
[23:19] <jadax> I've been playing recently with infrastructure that somebody else did 100% based on docker containers (on Ubu 18.04)
[23:19] <jadax> it was interesting
[23:19] <jadax> but docker needs resources, especially memory
[23:24] <jayjo> is there a way to submit easy changes to pages like https://help.ubuntu.com/community/SSH ? https://help.ubuntu.com/14.04/serverguide/openssh-server.html is broken link should be https://help.ubuntu.com/lts/serverguide/openssh-server.html.en
[23:26] <xamithan> Read here: https://wiki.ubuntu.com/DocumentationTeam
[23:28] <TJ-> jadax: if you want a rip-roaring read about Why Not Docker read to the end of https://thehftguy.com/2016/11/01/docker-in-production-an-history-of-failure/  :)
[23:29] <wholesum> been stuck for 22 hours already... grub boot partition got corrupted on an server 18.04 install (using secure boot)
[23:29] <wholesum> i am now on desktop 18.04 and installed refind
[23:30] <sarnold> jayjo: if you've got the desire, you could follow the "You can contribute to this wiki" link near the bottom of the page
[23:30] <wholesum> did refind install to /dev/nvme0n1p1 (~513MB, preceeded by 2M unallocated space)
[23:31] <Ryvius> Hello, is Firefox 68 not available for 18.04 yet?
[23:43] <Bashing-om> !firefox bionic | Ryvius
[23:43] <Bashing-om> !info firefox bionic | Ryvius
[23:45] <OerHeks> still in beta testing https://launchpad.net/~mozillateam/+archive/ubuntu/firefox-next?field.series_filter=bionic
[23:46] <jayjo> what does it mean to use `ssh -t` - keeping a pseudo terminal. Does this just keep the session interactive ?
[23:46] <jayjo> especially in the context of ForceCommand https://askubuntu.com/a/773903
[23:47] <Ryvius> Alright thanks
[23:48] <OerHeks> .. force tty allocation, even if SSH has no local tty ..
[23:48] <OerHeks> -t' Force pseudo-tty allocation. This can be used to execute arbitrary screen-based programs on a remote machine, which can be very useful, e.g. when implementing menu services. Multiple -t options force tty allocation, even if ssh has no local tty.
[23:48] <sarnold> jayjo: some programs change their behaviour if they are hooked up to a terminal vs not hooked up to a terminal
[23:53] <jayjo> sarnold: thanks, makes sense
[23:54] <TJ-> jayjo: -t is useful for doing things like "ssh -t user@remote sudo mkdir /mnt/test"  to get the "[sudo] password for user: prompt
[23:56] <Elw3> I am on 14.04 atm, does someone know if i have to replace the sources list entries with the "old-releases" lines already?
[23:57] <sarnold> Elw3: trusty is still on the archive mirrors http://archive.ubuntu.com/ubuntu/dists/ even though it is no longer receiving standard support
[23:58] <Bashing-om> !14.04 | Elw3 Trusty is no longer supported with out going !ESM:
[23:58] <tomreyn> Elw3: i assume you can still run do-release-upgrade
[23:59] <tomreyn> Elw3: so editing your apt sources should not be needed to upgrade just yet
[23:59] <Elw3> I dont want to upgrade....