[00:05] I formatted a drive to have three partitions. One realdy has for windows. Other two are ntfs. Ubuntu's installer doesn't recognize any partition and freezes. What to do here? [00:10] backthen: What is your goal here, as ntfs is a Windows file system. [00:15] Bashing-om: Install Ubuntu on one of the other two partitions. I assumed I could format the partitions during Ubuntu's install process [00:15] you can't install on ntfs [00:16] if you want to format one of those partitions as something else such as ext4 and you suspect the installer is freezing because of ntfs you could try to format those partitions from windows as something else such as fat or just remove the partitions altogether [00:19] midow: previously one parittion was unallocated so had no file system. Ubuntu install still didn't recognize anything. And it freezes when I use the '+'/'-' icons to add/remove partitions [00:22] backthen: is it possible your machine has "RAID" enabled? (it's actually Windows version of software RAID which won't be supported by Ubuntu) [00:22] backthen: I would look in your BIOS to see if you can find anything about RAID [00:23] leftyfb: ok. It's a laptop btw [00:23] backthen: yep, they still do it. It's typically to use some Intel pcie chipset to improve boot speed [00:25] Can some kind soul who understands iSCSI in Ubuntu set me straight on some things? [00:30] leftyfb: no raid. The C drive has 'bitlocker encryption' however. Don't know if that's standard [00:30] For instance, if I use targetcli-fb to admin and manage my iSCSI setup, I shouldn't need to have open-iscsi even installed should I? It's just for client/initiator, correct? [00:39] ctrlbreak: iscsi is already established, you say vmware currently is using it [00:39] ctrlbreak: and if it's already established, what is the point of your endeavor? [00:40] ctrlbreak: with that having been said, is it establish or NOT? === Glassjoe is now known as funnycalzoneman [00:57] Hi, I want to install picom (the compositor, for i3) on my unbuntu 20.04. I couldn't figure it out. The source code doesn't compile and I didn't find the picom package with apt search [01:02] SO... I have a client for my fledgling consulting business who was on Windows 7, which is now EOL, and put her on Ubuntu 20.04. She can print to her Lexmark CS410-dn printer from Ubuntu, but complains that it is slower than when she used Windows. Testing this, I found that when firing off a job from Windows, it will start the printer immediately and begin to kick out the page(s). On Ubuntu it takes several seconds before the printer fires up. [01:02] Any idea why this is? So far as I can tell, the printer is a Postscript printer and CUPS on Linux natively supports Postscript. I have tried using a "CS410-dn" PPD file as opposed to a "CS410 Series" PPD file that Ubuntu/CUPS automatically sets up, with negligible difference in print speed. Any thoughts? [01:02] I will be visiting her at her home tomorrow, and hope to have some idea(s) to try as I am out of my own. Thanks in advance! :-) [01:04] sounds not like a bug, at all. === zbenjamin is now known as Guest19677 === zbenjamin_ is now known as zbenjamin === Guest4516 is now known as joaquinito01 [01:16] In my experience, "slow" anything can be difficult to troubleshoot. It works, but it takes upwards of 40-90 seconds to start the printer engine. The printer is connected via USB to the same computer in either case (dual boot). [01:16] * sirlurksalot is lurking but will be back to check for answers [01:17] must be a large job to print, 40-90 seconds.. sounds different than several seconds. [01:18] pavlos, hi [01:18] I have a particular directory, in which when I cd xx; ls -la -> ls returns the list, then waits 1 sec+ to terminate [01:18] any ideas what could be wrong? [01:18] and ls - or ls -a or ls -la -- all have the same behavior in this directory [01:20] @oerheks, it's the same job in either case; just a printer-friendly rendering of an email from GMail in Firefox. [01:20] The exact same email prints almost immediately when using Windows 7 [01:24] geosmile: is that dir remotely mounted, like an nfs link? [01:26] geosmile: for example, accessing /home/geosmile should have no delay at all but accessing some dir -> nfs:/data might have a delay [02:07] pavlos, it is a local directory [02:10] geosmile: does that local dir have sym.links inside (that could point to another fs)? [02:13] Guys, in Ubuntu when the official repositories doesn't have a package what you usually do? I'm newbie and have seen something like AUR in Archlinux, I want to know if there is a similar thing in Ubuntu? Is luanchpad the way to go? [02:28] pavlos, all files here are local - nvme drive [02:38] @JackSK5 I don't know if there is a good rule of thumb. It depends quite a bit on the people that are making the software. My preference is to install from the Ubuntu repo, but if the repos are not kept up to date and there is another repo I tend to do it that way. [02:41] If you use a repo you will get updates. Sometimes all you get is a .deb. At least you have a package designed for debian/ubuntu. Some software is in active development and you might need to essentially do a git install. I don't think there is a single consistent installation method. [02:43] quite often I install one way and discover that there are problems and end up uninstalling and installing some other way [02:49] geosmile: sudo apt install nvme-cli, then run "nvme smart-log /dev/nvme0" paste output with pastebin [02:55] pavlos, thanks [02:58] pavlos, https://paste.ubuntu.com/p/wP4K3nNX73/ [02:59] geosmile: all look good so I have no idea why "ls -al" has 1sec delay at the end [03:01] geosmile: nvme error-log /dev/nvme0 and pastebin, I do not like line 16: 98 errors [03:01] pavlos, all entries in error-log are zero [03:03] geosmile: ok [03:06] Hello, exist some way for revert a 'rm -r folder' ? [03:07] other_rick: restore from back up [03:20] JackSK5, the AUR is basically scripts to download and build the software. With Ubuntu, if it's not in the official Ubuntu repositories (package archives), then the first thing is, usually, to see if there is a PPA (personal package archive) being maintained by someone for the software. If you add the PPA to your system, you can install software from there using the same infrastructure as the normal Ubuntu software installs (apt & co.). You will [03:20] also get updates when upgrading via the normal package management system. The second place I would look is to see if there is a Snap package available. Failing that, you can look into Flatpack support and see if it's available as a flatpack. If not available as a snap or flatpack, then you can see if the authors (or someone) makes a .deb (Debian package archive; Ubuntu is derived from Debian) that you can download and install with a tool such as [03:20] GDebi (graphical DEB installer). Lastly, you can almost always download the source code, compile it, and install it according to the author's build instructions. AUR packages pretty much do this in a semi-automated way. You will have to be mindful of dependencies (needs this version of that library, etc.) if you build it yourself. The DEB packages (which are what the Ubuntu and PPA repos deliver) have dependency information in them so that the [03:21] APT packaging system can resolve and install the needed supporting software. The AUR package info files also list dependencies. If you build it yourself, it's up to you to be sure you have the dependencies met. Snap and Flatpack packages are different; they include their dependencies (more or less) and are "containerized". At least Snaps are. I'm not as familiar with Flatpacks. There are also Appimage files, but I have limited experience with [03:21] that system. [03:53] jonnj: Thanks dude [03:53] sirlurksalot: That's was totally enlightening! Thanks a lot for elaborated explanation. Now things are clearer for me. :) [03:54] JackSK5, oh you're so welcome. glad to be of some help [04:01] In Archlinux, both AUR and official went through the official install manager (pacman, or variant). Even the system python packages are install by one system (pacman). So kind of centeral management if i put it correctly. Here in ubunut I'm using apt install (for official and ppa). Snap and flatpack are containerized. But how to deal with .deb packages and buids in a standard or sustainable way [04:01] such a way it doesn't leed to a hell? I mean what is the good practice to manage diverse sources of packages? [04:07] JackSK5, No... in Arch, pacman installs packages from the repository. OTHER programs like yaourt and the like can _also_ automagically build packages from the AUR scripts, then submit the packages (which are built on your system) to pacman for installation. Pacman doesn't deal with the AUR directly... Either you manually compile the AUR packages with the Arch Build System or you use a tool like yaourt, etc.... the result is a local package file [04:08] that pacman can install. [04:09] https://wiki.archlinux.org/index.php/AUR_User_Guidelines explains it in more detail [04:10] In Ubuntu, you have the official repositories, which the APT system uses. You can also add in other repositories that APT can use, like the Canonical Partners repo, or PPA's (also repos). The repos in a Debian-like system contain pre-compiled binary versions of the software. The AUR does not. It only contains package definitions that make it a little easier to find, download, and compile the software from source code. [04:13] Snaps are a different thing (and also include binary software, not source code). Flatpacks are somewhat similar. Both Snap and Flatpack systems are meant to be cross-distro, as they include their dependencies. Software packaged for a particular distribution or distro family is comiled against the versions of libraries and stuff that the distributions package maintainers include with the system. For example, ubuntu comes with such-and-so version [04:13] of Python, this-or-that version of libinput, yadda yadda. The software packaged FOR the distro is made to work against those versions. It's really a ton of work that the distro package maintainers do, to be honest. [04:14] So if you use a Snap package, it may have included within it some different version of a supporting library than what the distro (such as Ubuntu) comes with by default. [04:14] The Ubuntu repos, and the PPA archives (also just repos), deliver DEB files for APT to install. [04:16] "The Arch Way" does not really encourage you to use "helper" tools to get your AUR packages. (packages in the sense of the AUR are really a misnomer ... they are simple text files telling where to get the stuff to build the files and CREATE a package, locally, than you then install with pacman). Arch promotes a do-it-yourself and learn-by-doing mentality and strongly recommends you learn how to use the AUR manually before blazing ahead with [04:16] "helper" tools, so that you understand exactly what is going on. [04:18] sirlurksalot: I meant either from official repo or AUR, this is the pacman system that put the binary in the OS filesystem. Even for python I seen guideline in their website that install system packages only from ropo/AUR and for other just use virtual reality. HOwever, in ubuntu I'm clear about APT (repo or PPA) and also snap and flatpack. But I'm not sure about how to manage .deb installs and [04:18] build from source installed (make install right?) in a way that do not mess the system. [04:18] *virtual environment [04:18] Because there is really nothing more than some "where to get it" info in the AUR packagebuild files, you will find that someone usually creates a packagebuild file for most software. That's why there's so much stuff there. But all the heavy lifting is really done locally, on your machine. Pacman is really a very simple package installer system compared to the likes of APT (Debian and deriviatives) or DNF in the Redhat/Fedora world [04:19] the APT system basically downloads DEB files from the "repos" or "archives" and installs them. DEB files are the same thing, delivered outside of a "repo" system. [04:20] sirlurksalot: I get it the AUR is basically a warpper around the build system which produce a .xyz package adn that later can be used by pacman. [04:20] yep [04:20] so if I install a .deb file (eg. with dpkg -i...) then it's safe. Right? [04:20] And DEB files are basically tar (think zip) files with the binary software and a manifest [04:21] yeah, more or less. It's doing the same thing that APT does behind the scenes. Thing is that a manually installed DEB is not going to see automatic updates, as the repos get new versions from time to time and APT will download the new (DEB) and install it. you'll have to download and manually install any software updates for DEB files you install with dpkg -i [04:22] I imagine this scenario that snap/flatpacks are separate beast. APT also another manager and now GDebi or manual build is another manager. There is no clash between APT and snap. But it might be some mess when I try to use GDebi for debian .deb files...that was my point on what is the best practice [04:22] (manually get the new DEB with the updates) [04:22] aha [04:23] I get it. So when I'm using .deb or build, the automatic update is on my own. No worry about messing this (clashes or conflicts with APT) [04:23] I should manually keep track of the packages, I mean and that's become the user's responsibility. [04:23] Manually installing a DEB file is a throwback to how it was before package managers like APT came along. On RedHat systems they have a very similar package format called .RPM (redhat package manager) files. You may have heard of "dependency hell" as you'd have to track down all the dependencies for a .rpm and get them and install them and then try (and try again). APT came along for Debian and handles all that for you. And then on RedHat, Yum [04:23] (and now DNF) do that same thing. [04:24] I just get worry that using GDebi mess my ubuntu bcz it was another's distro package... [04:24] I don't know off hand, but there should be a way of using the family of tools for the APT system to list the manually installed DEBs. pacman has a similar feature that you can list all the "foreign" packages that were manually installed (the AUR packages, mainly). [04:25] sirlurksalot: exactly, I had an experience with fedora and went through that dependency hell (I switched back to windows as result for years) [04:25] There is also a tool called alien that can convert an RPM file to a DEB [04:25] Great [04:25] Well the modern DNF is very good, I understand. I'm not a Fedora user; mostly Ubuntu and Arch (acutally Manjaro these days) [04:26] sirlurksalot: Arch was a bit intimidating for me but I might test it later [04:27] sirlurksalot: Thanks a lot for the explanations! [04:30] I went through it for the learning experience. Then once I satisfied myself with that, I stopped being a masochist and just used AntergOS (until they disbanded). Antergos was, mostly, an Arch installer that included a couple extra package repositories for stuff they provided, like theming and the graphical front end to the package management system. not a lot else; it basically gave you a close-to-native Arch system and was not difficult to [04:30] "convert" into plain Arch as it used the Arch package repositories for most of the software. [04:32] Later I started using Manjaro, which is like Arch on a delay. The new software releases are very quickly delivered by Arch proper -- thus the "bleeding edge", as you may get to experience some new bugs or regressions along the way. Manjaro holds back the packages for a short while so they can be "shaken down" by others first, then when they are a tad more stable, they release them. So it's not quite as bleeding-edge, but it's close. Manjaro uses [04:32] their own repositories, though, as that is how they control when the software is released to you. If they used the Arch repos, you'd get the new stuff slightly sooner and slightly "raw"-er [04:32] sirlurksalot: any alternative then? with same purpose? I found people talking about ArcoLinux which probably have the same spriti [04:34] I'm into deep learning stuff and had a co-worker that things managing stuff (CUDNN, python, nvidia, etc) on arch is way easier...I'm still on ubuntu [04:34] *thinks [04:35] If you want an Arch system without the Arch install process, look at https://endeavouros.com/ (EndeavourOS). It is pretty much what Antergos was... a nice installer for Arch with a very few custom packages; it uses the Arch repos and AUR for 99% of the system. It's Arch, but without the pain (or education, to be frank). === vlm_ is now known as vlm [04:37] Installing Arch the classic way is educational. Beyond that, there is Gentoo (where everything is built from source), or you can build Linux From Scratch (which is basically an instructional guide to build your own linux, from scratch, mainly so you can be educated about how a Linux system is put together and the sorts of choices that distributuion already make for you [04:38] Even if you don't ever plan to go through the Linux From Scratch process, you may want to download the "book" and read through it to learn (a lot!) about how Linux systems are put together. [04:40] Deep Learning you mean Machine Learning? Or do you mean you like to deep-dive and learn stuff with your fleshy brain [04:40] http://www.linuxfromscratch.org/ [04:43] So with regard to your co-worker's opinion... I guess you can say that Distribution Maintainers generally make a lot of choices for you and do a LOT of work making a cohesive system. Arch is much more do-it-yourself. It gives you a minmial system (lean and mean), and leaves you in charge of fleshing out the system as you see fit. [04:43] But you get power with that level of control [04:44] In the Arch world there is not so much hand-holding by the distro itself; but rather the community is known for putting lots of good and detailed information on the Arch Wiki so you can figure it all out. The Gentoo Wiki is also a good resource. [04:46] for people that just wanna use their computer... [04:46] may I suggest Ubuntu [04:46] oh cool look where we are [04:47] Yeah some of us are tinkerers. I rarely play games ON my computer; I play WITH my computer [04:48] LOL this Ubuntu channel is getting a lot of non-Ubuntu chatter, no? Well JackSK5 wanted to understand the differences, as so many do. Ubuntu is a good, solid system to just get your work done on. No doubt. [04:48] And for people I will be supporting, I generally go with an Ubuntu derivative because it is so ubiquitous [04:49] You can generally always find help on how to make something happen on an Ubuntu system. [04:50] sirlurksalot: Thanks a lot. [04:50] Yes. Indeed I'm a tinkerer and have learn a lot through it. [04:51] JackSK5, To be honest, you're probably going to be better off using Ubuntu or a derivative until you are more comfortable on Linux and have a bit deeper understanding. You can "play" with alternatives in a virtual machine or something until you get comfortable with those. [04:52] yeah, JackSK5, no problem [04:52] On Arch, I like the philosophy but it's a trade-off between how much time you want to put learning your tools (OS). Right now I'm OK with ubuntu but in future I might get fed up with those choices that others made for my system and make things harder. My co-worker point was about this. It says since Arch is too lean you're closer to what developer experinces... [04:52] Now if someone can enlighten me on slow CUPS printing like I asked an age ago in here :-P [04:53] JackSK5, seriously just try the others in a virtual machine environment. It's super easy. Then you can break them all day long and just revert them 'till you get the feel for what you really want to do [04:54] Ubuntu is a solid choice, though. I've been using Ubuntu since it came out in 2004. Before that it was Mandrake and a lot of hair-pulling (dependency hell, for sure). [04:55] Those days are behind us, though [04:55] For your learning, seriously consider going through Linux From Scratch and Beyond Linux From Scratch [04:56] I'm on hp spectre 360x, and I can't get my sound to work. It seems some people have had the same issue, but I don't see any real solution. [04:58] dr`venom, use the lspci command on the terminal (lists the PCI bus devices) to see what sound hardware you have [04:58] Linux uses the ALSA system for the low-level drivers, and typically PulseAudio runs on top of that as a sound-server architecture for routing sound between applications and devices [04:59] You will often have more than one audio interface; one on the motherboard and another in your "graphics card" hardware to send sound over HDMI [05:00] 00:1f.3 Multimedia audio controller: Intel Corporation Comet Lake PCH cAVS [05:00] Is it that? [05:01] On the settings sound GUI, under output device I have Speaker-sof-hda-dsp [05:01] Even when I put on something like youtube, I see sound on the meter and I even have sound through the headphones. [05:04] seems like that would be it [05:07] You can go into Settings, then Sound, and you should be able to see a list of applications using audio, and their volume levels, as well as selections for Output and Input devices (and a way to test the output devices) [05:10] hi. [05:12] dr`venom, try installing pavucontrol which is a graphical program that will show you all the applications sending audio and let you re-route where they output [05:12] and control volumes, independently [05:12] it's a much more granular tool and might help you [05:41] I open a terminal during Ubuntu installation. Then I do fdisk -l to show partitions but get permission denied for all listed partitions. is this right? [05:56] backthen: sudo in front === ace_me1 is now known as ace_me [06:20] bindi: I used GPart tool as well. It shows no partition other than the USB drive itself. It's as if the hard disk is not found [06:36] Hi, after a recent system update on ubuntu 20.04 I'm facing this error whenever I run few programs - "/usr/bin/ssh: error while loading shared libraries: libkeyutils.so.1: cannot open shared object file: No such file or directory". I've tried reinstalling the library package but it fails with the following error: [06:36] https://paste.ubuntu.com/p/w4hqr4pHzs/. I'm not able to find anything online to fix the issue so can someone help me out? [06:40] systemd-analyze time [06:40] Startup finished in 12.519s (firmware) + 3.024s (loader) + 4.264s (kernel) + 1min 24.412s (userspace) = 1min 44.221s ==\n== graphical.target reached after 55.584s in userspace [06:40] How to reduce userspace time? [06:53] leibniz[m]: are you on 20.04?