[02:08] so I did apt-get install ubuntu-server on my 18.04 install and then in the console it shows this https://apaste.info/zU9t what do I pic (i'm USA english normal inputkeyboard) [15:01] help please! i have ubuntu studio, bionic beaver. this morning, system asked me to do upgrade, so i did. then it told me to reboot to finish upgrade, so i did. when i went back in it won't start xfce. if i try startx from command line it tells me fatal server error: unrecognized option: vt1 -keeptty -auth /tmp/serverauth.1V9dTn8Dws [15:15] after today's dist-upgrade, x won't start. help please? [15:15] check the logs [15:15] see why it won't start [15:16] ikonia, if i try startx from terminal it's giving me 'fatal server error: unrecognized option: vt1 -keeptty -auth /tmp/serverauth.blah [15:17] don't run start x [15:17] view the logs from the session manager and xorg service starting up [15:17] you mean /var/log/syslog? or which? [15:20] thats the syslog [15:20] ikonia, in /var/log/Xorg.0.log I see things like 'falling back to old probe method for fbdev', 'falling back to old probe method for vesa', 'no devices detected', 'no screens found' [15:22] looks for errors, not warning [15:23] ikonia, right. well, the ones marked EE are "failed to load module 'blah'" and right at the end, "no screens found" [15:23] and "no devices detected. [15:23] failed to load module.... [15:24] the modules failing to load are: "nvidia", "modesetting", "nvidia" [15:24] so there is your problem [15:26] ikonia, ok, so what do i do? (feeling like an idiot asking, but haven't got a clue... at least that computer is connected to wifi. apt-get something?) [15:27] the nvidia module isn't loading [15:27] right, and? [15:27] find out why [15:28] that same log says "Failed to load module "nvidia" (module does not exist, 0)" [15:29] so the nvidia log is missing [15:30] so i should apt-get install nvidia? [15:30] no [15:30] why are you using a developer pre-release [15:30] you seem to be looking for a standard user experience [15:31] ikonia, sorry. it's worked great in the past to use the cutting-edge stuff [15:31] but you're probably right [15:32] it's not cutting edge, it's development, it WILL break [15:32] if you're not confident fixing it, you really shouldn't be using it [15:33] ikonia, understood. well, perhaps you can help me fix it for today and then i'll look into rolling back to the latest stable version? [15:33] I think I'd rather (personally) you just rolled back [15:33] (or re-installed to be accurate) [15:33] sigh [16:38] do you folks know how I can get past this message https://apaste.info/Ndel (I'm on 18.04 fwiw) [16:59] here is what I get when I try dpkg fix - https://apaste.info/AB54 [16:59] :( [17:07] donofrio: on my 18.04,, /etc/resolv.conf is a symlink to ../run/resolvconf/resolv.conf (e.g. /run/resolvconf/resolv.conf) [17:07] donofrio: i'm guessing that maybe you have a slightly altered config? [17:09] nope [17:09] donofrio: nope as in it's the same symlink for you? [17:09] it seems to be upset with systemd package, I'm just wondering how to skip this message and allow it to contiune finishing package things [17:09] donofrio: did you read what i just wrote? [17:10] donofrio: you can't "skip" the message [17:10] donofrio: on your system now, for instance, `ls -ahl /etc/resolv.conf` [17:10] oh yah I see that now....dunno why it's upset with resolve [17:10] donofrio: see what? [17:11] just noticed that "cp: '/etc/resolv.conf' and '/run/systemd/resolve/stub-resolv.conf' are the same file" was upset as well as systemd [17:11] donofrio: i'm sorry, i don't understand what your point is [17:11] systemd is not 'upset' [17:12] the systemd package failed to install because it was not able to run one of its scripts [17:12] because your /etc/resolv.conf is already the same (possibly via symlink) as /run/systemd/resolve/stub-resolv.conf [17:12] donofrio: hence why i asked what the status is o your system [17:12] donofrio: which you have not yet provided [17:14] status like? it's up and working or? [17:14] donofrio: really, I asked what is `ls -ahl /etc/resolv.conf` [17:14] lrwxrwxrwx 1 root root 39 Feb 15 03:35 /etc/resolv.conf -> ../run/systemd/resolve/stub-resolv.conf [17:16] donofrio: right, which, afaict, is not the default [17:16] donofrio: but that's the issue for systemd, which is about to do a `cp /etc/resolv.conf /run/systemd/resolve/stub-resolv.conf` or v.v. and it can't if they are symlinked to each other [17:16] it might be a bug [17:17] What should I do to make it like it should be......I mean /etc/resolve.conf is standard so I'm guessing its the nested one that is not default [17:17] what nested one? [17:17] donofrio: your /etc/resolv.conf is pointing to somewhere that systemd (package) does not expect it to [17:18] so I should break the symlink? [17:19] donofrio: how did your system get to 18.04? is it a fresh install? upgrade? [17:20] donofrio: you can 'fix' the systemd package issue by probably pointing /etc/resolv.conf to the file i mentioned it is on my system (which should also exist on yours), temporarily and thenputting it back [17:21] fresh install from https://partner-images.canonical.com/core/bionic/current/ [17:21] from https://partner-images.canonical.com/core/bionic/current/ubuntu-bionic-core-cloudimg-amd64-root.tar.gz [17:22] I've been apt-getting the system up to what I need 'ubuntu-desktop' and the like....and it's tripping over this....sdo not know if the 1.2k packages have had there template tags processed and the like? [17:23] here is full output of what I see - https://apaste.info/vSjK [17:23] starts at line 65 [17:24] and at line 568 is where it says it's upset with location of resolv.conf [17:24] donofrio: wait this is an ubuntu core image? [17:24] yep [17:25] no, that's not ubuntu core, it's something else [17:26] ubuntu core is at http://cdimage.ubuntu.com/ubuntu-core/16/stable/current/ubuntu-core-16-amd64.img.xz [17:27] donofrio: is this more WSL stuff? [17:28] of course it is :) [17:29] I couldn't update from 17.04 to 17.10 things broke so I reinstalled with 1709 of win10 (got lots of new fixes for WSL) and installed lxrunoffline that I used to install from the tar file to what I have now [17:39] here is me installing in (along with 16.04 on left (on right its 18.04)) https://1drv.ms/f/s!AsBlZbBf72iNoL8tiTMnKYmVWmeM9A - ignore unneeded onedrive login [17:40] screen2gif way cool [18:00] I mv'ed the /run stub resolve file to my homedir then I get https://apaste.info/GBpR should I mv theese files as well? wish I could 'blacklist' a package allowing it to stay installed just do not attempt up upgrade it....probably not the correct thought just a thought [18:04] so the stub one is needed for apt it seems.... [18:06] dang I removed systemd and it added it right back - lol [18:07] output of my "attempt" at a workaround - https://apaste.info/d2Gq [18:31] donofrio: what? no one said mv anyting [18:31] donofrio: also whtever image you used is absolutely the wrong thing [18:31] that's for docker [18:32] (i just checked with the image folks) [18:33] nacc, that image/tar file is what is needed to install when the "ms store" is not installed all details listed in this ticket https://github.com/DDoSolitary/LxRunOffline/issues/21#issuecomment-365856589 [18:34] donofrio: ok, so using a 3rd party tool to install an unsupported thing [18:34] great choices all around :) [18:34] and then asking ubuntu to solve your issues [18:34] only choice you mean [18:34] uh, no, just run ubuntu [18:35] no in corp setup like I use its the only option as there is no vmware (mdm agent) airwatch client fr uairwatch for ubuntu/linux [18:36] here is my setup (when I was running 17.04 - http://www.tinyurl.com/donofrioworkdesk) [18:36] i have no idea what any of that is [18:36] and i don't particularly care :) [18:36] I couldn't do-release-upgrade so I reinstalled windows 10 corp image that is mandated to use this on corp network [18:36] so don't use ubuntu [18:36] problem solved? [18:37] no not at all.... why the os bigotry can we all not coesist and enjoy the projects that make our daily lives better? [18:37] donofrio: it's not bigotry [18:37] you consistently ignore what is supported where [18:38] this is a 18.04 channel [18:38] donofrio: for *ubuntu* [18:38] not wsl [18:38] you know that [18:38] because i tell you that every week [18:38] I'm here with 18.04 and asking questions before is "all finished" in hopes that I cnaa help make this a better project [18:38] donofrio: installed in a way that is unsupported, as well [18:38] what do you mean? [18:38] so again, think about it from our (volunteer) perspective [18:38] you used a docker image [18:39] I start with the core tar file - doesn't get more standard than that [18:39] to do something not docker [18:39] that file is *not* what you think it is [18:39] I'm not using docker at all [18:39] donofrio: right, then you should not be using those files [18:39] did you review that github link? [18:39] i don't care about that link [18:39] since that's another not ubuntu thing [18:41] is is an ubuntu thing the whole thread has been about ubuntu, hence why I said "why os bigetory" I get the feeling like your upset because "no kernel" is running so why help, well userspace is what I need to get my job done and I'd love to not only help anyone else that is trying this but also my co-workers that will not use 17.04 because it's "three years old" [18:42] mistype "will not use 16.04" I meant [18:42] 'lxrunoffline' does not exist in ubuntu [18:42] so it's not an ubuntu thing [18:42] running some random program that installs ubuntu in windows does not make you supported [18:42] lxrunoffline is only the guide to get ubuntu installed [18:43] but it uses the presses ubuntu core tar file that is created by ubuntu folks [18:43] *for* docker [18:43] seriously. [18:44] so I do not see why it's a go away feeling I'm getting from you, I have 62 updates and relies from the author of that lxrunoffline project as he helped get this working and I'm only here because this resolve issue....and systems issue so please "keep an open mind" please and help [18:44] "You simply download the tar files built for Docker" [18:44] not using docker at all.. [18:44] so then seek support from that tool's authro [18:44] doing things that are not supported using their own tools [18:44] that you've thumbed your nose and not read any of the effort yet from that githib linkk :( [18:46] donofrio: you refuse to listen, and you've done this ... for months now? [18:46] i have given you clear information on how to fix your issue [18:46] and informed you that you are doing unsupported things [18:47] you have to realize while I'm an early adopter of this technology many others will be coming in the next two years so WSL and the hatred towards WSL and its uses will not help anyone out...just saying... [18:47] i don't hate WSL [18:47] it's explicitly not supported [18:47] which again, you ignore all the time [18:48] nacc, what "clear inforation on how to fix" all I heard was, install ubuntu bare metal, and I cannot get on my corp network without windows [18:49] donofrio: So is the issue that you can't use the regular WSL install mechanism because you have no network access? [18:49] donofrio: you need to fix the symlink that is broken [18:49] why is it "not supported?" It's a released product built into released version of windows that is Not insider just standard windows [18:49] Or am I misunderstanding the issue? [18:49] !ubuwin [18:49] Windows 10 has a feature called Windows Subsystem for Linux, which allows it to run Ubuntu (and other Linux distro) userspace programs without porting/recompliation. For discussion and support, see #ubuntu-on-windows or ##windows. For installation instructions, see https://msdn.microsoft.com/en-us/commandline/wsl/install_guide [18:49] donofrio: i have to send that faq to you *every* week [18:50] Odd_Bloke, what is the case is we have a NAC network here that requires the use of VMWare AirWatch product, and to date no linux client exists so its unable to get an dhcp from an ethernet hookup - hence why I'm installing though windows [18:50] donofrio: OK, but you can install WSL from Windows normally. [18:50] So why not just do that? [18:50] I work for a hospital that locked down all ethernet and wifi access [18:50] yep [18:52] Odd_Bloke, I am - tinyurl.com/donofrioworkdesk shows my full write up on how I used it wwhen I was w10 1703 w/17.04 apt-upgraded from 16.04 but cannot go from 17.04 to 18 breaks everything so I rebuilt windows 10 corp build 1709 with wsl and lxrunoffline beacuse our corp build doesn't have "the ms store" so the is the only way to install - using tar files and lxrunoffline [18:53] keep in mind this is corp desktop on corp network with corp lockdowns but I still gotta have my ubuntu....it's a godsend to keep me happy [18:53] and, how, exactly, is this relevant to this channel? [18:54] donofrio: Please move this conversation to #ubuntu-on-windows. [18:54] I'm running 18.04 and asking about what do I need to do to work around my current issues of https://apaste.info/d2Gq error [18:55] no, you are not. You are running Windows, and trying to install something in an unsupported way [18:56] I understand it would be nice to be able to do it, but this is NOT #ubuntu+1 [18:56] hggdh, did you see my apaste link? (I'm talking in ubuntu-on-windows now) https://apaste.info/d2Gq [18:56] nvidia-384 is complaining, e.g.: "Depends: libc6-dev but it is not going to be installed" [18:57] it wants libc-i386 and i would guess some other 32 bit libraries, is there a way to install those in bionic? [18:57] i don't see any ia32 packages [18:57] hggdh, when I type cat /etc/*release* it returns "18.04" so I'm still wondering why people say this is not an 18.04 topic [18:57] BLZbubba: i think this is a packaging bug [18:57] BLZbubba: the version in bionic is behind the one in artful [18:57] BLZbubba: i pinged about it yesterday, let me look again [18:59] BLZbubba: i'm asking the respective uploaders now to find out [18:59] BLZbubba: i think you can make it work by forcing it to use the version in bionic (which is a downgrade) [19:00] BLZbubba: i can try and walk you through it [19:00] i'm using bionic [19:00] BLZbubba: i know [19:00] BLZbubba: the problem is a security fix went into artful [19:00] and not bionic [19:00] gotcha [19:00] so the bionic version is less than artful [19:00] security is overrated [19:00] :) [19:00] for what i'm doing anyway :P [19:01] i think it might have raced with bionic opening, but i'm not sure [19:01] ok what should i try [19:01] BLZbubba: can you paste the exact command you are runing and the output? pastebin it, i mean [19:01] is it going to skip the 32 bit libs or is there something needed to get them to install? [19:01] BLZbubba: yeah we have to help apt out here to use the one fromm the archive (which is 'older') which apt won't do on its own [19:02] http://uberdl.com/bionic.txt [19:04] apt-get install nvidia-384 libc6-dev=2.26-0ubuntu2 libc6-i386=2.26-0ubuntu2 lib32gcc1 [19:04] refresh the link [19:05] add libc6=2.26-0ubuntu2 [19:05] ok its loading, thanks! [19:05] BLZbubba: yw [19:06] BLZbubba: basically for your own knowlede, we resolved the issue for apt, which by default refuses to downgrade packages to resolve installations [19:06] that is the right thing to do by default for sure [19:07] yeah [19:07] it's an 'unsafe' operation [19:07] but bionic is 'unsafe' sort of definitionally ;) [19:08] haha for sure [19:17] though i'm using it on a couple of desktops and a ceph cluster and it has been working great [19:18] yeah i've had very few issues here [19:21] is did ubuntu get sold? is it not canonical.com anymore? [19:22] I only ask because my tar file that I built with is from https://partner-images.canonical.com/core/bionic/current/ not 3rd party [19:23] donofrio: i'm not sure what the one question has to do with the other, but no, ubuntu did not get 'sold' [19:23] nacc, it's just a reply to you when you said that was from docker, so I relooked and it is from canonical.com and no one else [19:23] donofrio: i never said it was from docker [19:23] donofrio: i said it was *for* docker [19:25] only thing I'm having issues with now is these four lines "[/usr/lib/tmpfiles.d/tmp.conf:15] Failed to replace specifiers: /tmp/systemd-private-%b-*" (three more varants but same /tmp/systemd-private* varants) === Smedles_ is now known as Smedles [21:44] donofrio: As discussed, you aren't going to get support here; if you want to do something this unsupported, you need to be able to dig in to the problems and solve them yourself. [21:54] Odd_Bloke, I'm just trying to find help for something that should work, I am willing to help out just need to know where to do what? [21:58] donofrio: You'll need to dig in to the problem; look around for other similar error reports and work out if they apply to your situation, maybe dig in to the systemd source code to understand what the errors mean. [22:05] tbh, it seems like you're working from a buggy snapshot [22:05] i.e., one with systemd already there