[00:02] I wish to apologize. I deserve crucifixion. I did not in fact install nheko with apt, in fact it was flatpak. In my defense, I've installed lots and lots of stuff in the past 24 hours and forgot some details, and I had no idea that a flatpak could depend on a native package, so didn't realize this. [00:03] But in fact, it seems, after closer inspection, yep, I installed the flatpak, and it depends on native package of gnome-keyring. Weird. [00:03] e25 no problem, glad you found out yourself [00:04] e25: remove the flatpak and install the apt. T [00:05] they can reside, no? [00:07] oerheks here is my nvidia-smi http://paste.debian.net/1274960 [00:08] chonkin, running xorg, not wayland? [00:09] Yes. But since you asked, my boot logs show strange errors about wayland === tds7 is now known as tds [00:17] errors, or warnings? not all messsages are errors [00:21] Ok guys I love you very much and really don't want to miss a moment, but I want to switch to the new version I just built and installed of this irc client. So if I come back, it worked. [00:21] j #srain [00:31] how am i supposed to use gnome if its designed for people who dont switch tabs often and i always switch tabs [00:32] Well Pelo, that perhaps may be why other desktop environments exist, and ubuntu supports several. [00:32] pelo: you might want to rephrase that with an actual support question. [00:32] what is a good desktop environment for people who switch tabs often? [00:32] But, I guess with gnome, you could map some keyboard shortcuts? [00:32] Or Use some extensions? [00:33] pelo: alt+tab works completely fine in gnome [00:33] i think alt tab is really annoying to use [00:33] pelo: though, are you referring to switching apps or tabs? If tabs, which app? [00:33] What is it you want, Peto? [00:33] apps [00:33] You want a mac like dock? You want traditional style taskbar? [00:34] pelo: you think alt+tab is annoying? You do realize it's been the standard way to switch between apps across all OS's for about 30 years right? [00:34] i just want something easy to use when i have 20 applications to cycle through [00:34] pelo: like what? [00:34] i cycle through discord spotify hexchat google terminal and plenty of other apps constantly [00:35] ok? [00:35] alt tab takes a while when you have 20 apps open [00:35] pelo: so maybe you want to look into a tiling window manager? [00:35] Use the taskbar? [00:35] pelo with how much mem? [00:35] pelo: look into i3m and dwm [00:35] 2 gb? [00:36] intel internal gpu? [00:36] use the activity manager and search by the app by name [00:36] pelo: maybe also notion wm [00:39] oerheks, doesnt really matter [00:40] yes, it does [00:40] not to me [00:40] some stupid sites give megabytes of javascript, and youtub take a chunk [00:40] ny times 200 mb,.... [00:43] oerheks this is from January. The machine is no longer getting wayland warnings today. https://paste.centos.org/view/raw/1d154c98 [00:43] however, a more disturbing trend has emerged [00:46] oerheks look at the timestamps https://paste.centos.org/view/raw/f1ffa6a9 [00:48] chonkin, google give no clue, bing says asia [00:48] lolx [00:49] oerheks: I'm going to try to firewall him [00:49] you are just scanned,. sid you enable a firewall? [00:49] sid/did [00:50] disable password authentication, move sshd to a different port [00:50] tomreyn, +1 [00:53] does it hurt anything leaving the deb-src line for devel? or should i remove it after im done using it [00:53] in my apt sources [00:54] no problem keeping them, sources just use diskspace [00:55] and can make "apt update" slower [00:55] i'd comment them out if not regularly using them [00:56] ah ok, so may as well at least comment it out so its easy to bring back if i ever try the backport thing again [00:56] thanks [01:05] Earlier this morning, thermald was held back when I tried to install it with apt. [01:06] e25: https://ubuntu-archive-team.ubuntu.com/phased-updates.html it's being phased [01:06] maybe after reboot you can [01:06] Now, gnome-software is happy to install it. [01:07] Ah ok. Phazed. Will read and learn. [01:07] (I vaguely recall hearing about this earlier) [01:09] e25: the idea is that machines will decide to install an update or not based on a random number on the machine vs the phasing percentage -- we slowly ramp up the percentage, so more and more machines install it over time; that way, huge problems can be spotted when something like 10 or 20% of users have the update, rather than when 80% of users have the update === ledeni_ is now known as ledeni [01:10] Wow. [01:10] So has that actually happened? [01:10] Some phased update was halted cuz of problems in the early percentage? [01:10] !phasedupdates [01:10] Since Ubuntu 21.04, APT may hold back some updates on some systems while they are being phased in. This is called "phased updates". See https://ubottu.com/y/phased and https://help.ubuntu.com/community/PhasedUpdates for more info. [01:10] No, we do not want a huge wave of issues, despite proposed testers [01:11] e25: when there's a problem, the phasing will be turned back down to 0% [01:11] still, sudo apt get install (those packages) will suffice [01:11] e25: when it's at 40%, it means there haven't been enough problem reports to justify holding it back [01:11] but for regular users, be patient, the update gui does not show them [01:12] Sounds a bit more like preventing too much load on mirrors. [01:12] server users should have a testing system before applying updates === sig is now known as bebop [01:15] e25: security updates aren't phased; new kernel day is a *huge* load on the archive servers -- they almost always include security fixes, and every single machine has a kernel, they're pretty big packages, and we've got dozens of flavours [01:28] alright I have to test the SSH from this laptop with a different user [01:28] bye for now [01:35] I installed Ubuntu Server 22.04 as encrypted LVM on a 2TB drive, df is telling me now that I only have 100GB of usable space? https://paste.debian.net/1274970/ - Is this normal for encrypted LVM groups? [01:40] ForeverNoob[m]: fdisk -l, lvdisplay, pvdisplay, vgdisplay === a is now known as Guest9436 [02:01] booting from usb I immediately get "choose your install" but I just need a shell to fix grub. How do I just run it, NOT install it? [02:03] rbox: https://paste.debian.net/1274974/ [02:03] you dont see the menu "try ubuntu" or "install ubuntu"? [02:04] rbox: I see the option "try or install ubuntu" and that leads me to installing it [02:04] kombi: yes, and the first thing is an option to pick "try ubuntu" or "install ubuntu" [02:05] rbox: negative. Three options: 1. try or install 2. run with HWR kernel 3. i forgot [02:05] kombi: which iso is that? server? [02:05] ForeverNoob[m]: yeah, looks like its got the rest of the PEs unallocated, so you can just add them to the lvm [02:05] server [02:06] kombi: you need the desktop one [02:06] kombi: is it the live-server? [02:06] can't tell, how do i tell? [02:06] should be in the iso's name [02:06] point taken [02:07] I guess I have to make a new one. [02:07] kombi: what's wrong with grub? [02:07] (just spent hours upgrading a machine that must be back and running in the morning...;) [02:07] oof [02:08] EriC: did a few dist-upgrades and now there is no fs found [02:08] so I'd need a shell to grub-install /dev/sda [02:08] kombi: where do you get that msg? [02:08] trying to boot [02:09] initramfs [02:09] kombi: you have a separate /boot? [02:09] how do you mean? [02:09] a separate /boot partition [02:10] i figured... not sure, box is 10 years old...;) [02:10] standard debian install usually does that though [02:10] even 10 years back [02:11] EriC^^: If I had one, what would you do? [02:12] well you could hold shift to get grub, press 'c' and at least list the partitions/fs and see if the rootfs is there or not [02:12] I see [02:13] if it is indeed there, maybe see the grub entry by pressing "e" if it has the right root=UUID= there, maybe that's off somehow and it's not finding it [02:13] I will try that [02:14] afk [02:14] (plugging the screen into the other box now;) [02:16] rbox: Is this the correct command I should use in order to do so? `sudo lvextend -r -l +100%FREE /dev/ubuntu-vg/ubuntu-lv` [02:17] seems about right === chris14_ is now known as chris14 [02:20] Nice, that seems like it worked. Thanks rbox, thanks ChatGPT. [02:58] If I am intentionally modifying a .desktop file in /usr/share/applications, how can I tell apt to *not* overwrite that file when the associated program/package gets updated? [02:59] Every time Google Chrome updates my "enable dark mode" mod get plowed over. [02:59] *gets [02:59] (yes, I realize Google Chrome is third-party, but this is a question about how to do something with apt, which is supported.) [03:00] you can place the file in ~/.local/share/applications [03:00] it should override it [03:00] It would be extra-handy if the file that would have overwritten the old one could be saved somewhere else where I can look at it. [03:00] Hmm, that's not a bad idea. [03:00] * arraybolt3 tries [03:02] rbox: It works, thanks! [03:09] arraybolt3: hah, I just (re(re(re)))learned the this other day :) < FileJanitor> sarnold , found it. you just put path-exclude=path in a file in /etc/dpkg/dpkg.cfg.d/file [03:09] arraybolt3: rbox's approach is way cleaner but works just for your account [03:22] when my 'apt update && apt upgrade' reports nothing to do, the software-properties-gtk keeps popping up and telling me lots of software to install, why is that [03:23] I wonder, does apt full-upgrade do anything different? [03:23] last time I clicked on GUI it installed oem kernel and messed up my nvidia driver with cuda stuff, took me a day to settle down [03:24] oh yeah, that was a bad day [03:24] now I decided to ignore the GUI upgrade, but it keeps popping up [03:26] http://ix.io/4rzQ is my sources.list [03:28] wow that's *so* tame [03:28] what do you mean [03:29] the gui does not let me to turn it off fully, at least it must show itself every two-weeks [03:29] maybe I just uninstall it, doing it now [03:34] yolo: so many people come in with complaints about upgrades not going well and they'll have two or three PPAs and mix a dozen third-party debian stable, debian unstable, ubuntu xenial, ubuntu bionic, ubuntu focal repos for different applications, etc [03:34] yolo: and I'll be stunned their computer worked at all, and they'll be upset that it doesn't upgrade perfectly every time :) [03:34] yolo: yours is just so *easy* and *simple* [03:38] to create a usb install thumbdrive do i need this rufu etcher program [03:38] or can i simply copy the iso file contents and paste to a FAT32 formatted usb stick? [03:39] johnjaye: if you're starting from a windows machine, rufus or balena thingy are the way to go [03:40] fedora has one for windows too [03:40] johnjaye: the idea is to *overwrite* the memory stick, so the formatted filesystem on it doesn't matter -- it won't exist once you're done [03:40] matrixy: ooh? [03:40] its on the website [03:41] johnjaye: I would highly recommend balenaEtcher - it works well for me, and verifies the USB drive after it writes it. [03:41] Rufus is complicated, tricky, and difficult IMO. === docmax is now known as Guest2996 === docmax_ is now known as docmax [03:43] ok. i vaguely recall some situations where simply copying the iso files worked [03:46] maybe if you setup all the partitions right [03:51] how do I grub-install on a raid member? [03:52] box has only one disk right now, booted with a live distro. mount won't let me mount a 'linux_raid_member' [03:53] kombi_: Not sure how to do a grub-install on that, but you might need to use mdadm to get the RAID "array" mapped to a mountable device first. [03:53] kombi_: Though you might just have a /dev/md* device already that you can mount. [03:53] arraybolt3: even when there is only one disk? [03:53] already looked for md but no avail [03:54] blkid only gives me sda1 - 6 [03:55] kombi_: Well if you have a one-disk RAID "array", then it probably has some extra stuff in there that mdadm uses to identify it as a RAID array. [03:55] In which case, yeah, you might need that even with only one disk. [03:55] now trying to figure out the syntax.. :) [03:57] I think you want to use an assemble command? [03:57] (Take anything I say about mdadm syntax with a big grain of salt, I have never used it before.) [03:58] kombi_: Also is there a /dev/md directory? (As opposed to a file) [03:58] I think you'll probably need a command to assemble it, but just in case it's been autodetected already it might be wise to check. [04:01] is there a /dev/md directory? negative [04:02] ubuntu live distro has neither mdadm (apt install it) nor ifconfig to connect (apt install it). Where do I start? [04:03] never mind, found a mouse, will delve into settings and set up eth0 [04:04] ifconfig has been deprectaed for like 20 years, ip replcaed it [04:10] rbox: I'm an old man!...:) [04:10] thx [04:11] so what's the equivalent to ifconfig etho 2.2.2.2 gw 3.3.3.3 up? [04:12] I look that up, sry [05:44] How do I clear DNS cache in ubuntu? [05:44] server ip has changed [05:44] ubuntu doesn't know and keeps giving old ip [05:47] hm, it's at the gateway [06:02] anybody here? [06:02] !ask [06:02] Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience [06:05] heya [06:06] so i had my pi4 booted with zfs on root ..however, now its stuck on it not being able to boot because modprobe cant find it [06:07] its been down for a while and i finally decided to do something about it.. i think i had done an update and it crashed right after [06:44] I'm trying to set up a firewall with iptables and deny traffic by default with `iptables -A INPUT -j DROP`. I have https open and am forwarding it to 3000 with some nat rules. I also have loopback traffic enabled. However, these rules don't work unless I also open the port 3000, but I want that closed to the public. Any ideas? [06:50] why use ubuntu when there is debian? [07:37] Aloha, I have an issue when moving files, where i need the contents of a subdirectory without the directory itself http://ix.io/4rAx [07:48] spent the night doing four dist-upgrades. All went well until the last reboot. Now grub just won't let me install it, claiming the drive doesn't exist [07:53] hi === realivanjx0 is now known as realivanjx [08:50] I have a LVM question. I find a lot of results in Google but they all going into the wrong direction [08:50] DF shows me 53% free. pvdisplay shows me 0 Free PE. The system behaves like there is no space anymore [08:51] But if I use du, just half of the space is actually used [08:51] Most google pages says the volume is not the size it should be. But it spanning two physical volumes [08:59] back === B1773rm4n5 is now known as B1773rm4n === ssserpent2 is now known as ssserpent1 [09:34] test [09:36] Got it working with this http://ix.io/4rB8 Thank you, if anyone looked into it. === realivanjx9 is now known as realivanjx === realivanjx4 is now known as realivanjx [11:36] Hi all! My work laptop runs on Ubuntu 20.04 LTS. On every apt upgrade, I get this message: https://dpaste.org/ZQ5QV - is Ubuntu trying to extort us money with the Pro-thing? because, come on!... do we really need to go Pro to upgrade packages that have been always available from their repos? [11:36] In this sense: are those of you running Ubuntu 22.04 LTS facing the same ? [11:36] thanks [11:39] !ubuntupro | SteelRose [11:39] SteelRose: Ubuntu Pro is a service offered by Canonical for expanded CVE patching, ten-years security maintenance and optional support. Anyone can use Ubuntu Pro for free for personal use on up to 5 machines. For details please see https://ubuntu.com/pro and https://discourse.ubuntu.com/t/31018 [11:41] ravage: I know that Ubuntu Pro is... my question was not that... [11:41] If know what it is and want to discuss it there is #ubuntu-discuss [11:41] OK [11:45] Hi, I'm trying to update my cuda apt-key in 22.04 and it gives me there is a deprecated CUDA GPG installed. So, run "sudo apt-key del 7fa2af80". I just ran the command and nothing happened, any solution? [11:49] Hi, anyone has experience with ubuntu AWS EC2 VMs? [11:50] Having trouble with adding a desktop to the default Ubuntu 22.04 VM image, & run it. [11:51] starting desktop reverses back cli, or rebooting image defaults to cli despite desktop being installed. [11:51] Need some assistance, setting up this .. [11:53] Someone write a set of straightforward commands , that will turn Amazon's cli VM into operating GUI system .. [11:56] Using Android to connect to VM. [11:57] I exhausted the online guides, still doesn't work .. [11:59] Usi g Android connec5s to the instance without ssh smoothly. [11:59] Seems the trouble is xserver. [11:59] Tried budgie & xcfe .. both do not resolve positively ........... [12:12] Guest46 -- 1. how did you add a desktop to the Ubuntu 22.04 server VM? 2. how are you connecting to it? Did you install a remote desktop server (like xrdp)? 3. what type of instance? does it have a graphics card? [12:16] this is the general guide I suggest: https://discourse.ubuntu.com/t/deploy-ubuntu-desktop-jammy-jellyfish-on-aws-ec2/30381 [12:16] it's the steps i've used in the past when doing some testing [12:19] MSG nishit_ hiii === B1773rm4n7 is now known as B1773rm4n === realivanjx5 is now known as realivanjx === polymorp- is now known as polymorphic [14:07] hi, I'm unable to boot into desktop, on terminal now, is there a way to paste my terminal directly to pastebin? === gacuxz1 is now known as gacuxz [14:11] No idea, but I guess you can dump terminal output to a file, start up a ftp server on your machine, and make another machine connect to your ftp server, download dump file and paste it to pastebin [14:12] oh, or just connect to your malfunctioning machine via telnet/ssh, and copy&paste on another machine [14:12] hao: if the machine has internet you can run this: ( do some commands ; do some other commands ) | nc termbin.com 9999 [14:15] nice approach, new method learned [14:24] fishbone000: thanks, I've got no additional pc at hand. I'll just describe my problem by typing them out... [14:24] hao: does the pc currently have internet? [14:25] leftyfb: yeah, I'm on it inside console (ctrl+alt+f3), using irssi [14:26] hao: ok, the use the method I posted for sharing output from commands in your terminal [14:26] the/then [14:26] leftyfb: oh, I missed your message, thanks [14:30] so I'll just pipe the dmesg to termbin, here: https://termbin.com/uniu [14:31] hao: ( uname -a ; cat /etc/os-release ; lspci ) | nc termbin.com 9999 [14:33] leftyfb: https://termbin.com/7tnp [14:33] Looks like the BDL errors are a known kernel regression for the sound card. [14:33] I just remember I did a upgrade using the upgrader [14:33] The gdm error is what's stopping the desktop from loading. [14:33] hao: ( uname -a ; cat /etc/os-release ; lspci ) | nc termbin.com 9999 # please type this exactly, including the parentheses [14:34] leftyfb: oh, sorry, wait [14:34] hao: You could try booting to the previous kernel if you still have it in the grub menu. [14:35] leftyfb: https://termbin.com/mxjh [14:35] jhutchins: I tried, but still can't boot into desktop, same errors [14:36] hao: you could try this, but it doesn't looks like you have an discrete GPU: sudo ubuntu-drivers autoinstall [14:36] hao: Interesting, as it was presumably working before. [14:36] also, when trying `startx`, [14:37] hao: Was it working before the upgrade? [14:37] it looks like there's lots of other people with 22.10 with this same issue [14:38] jhutchins: leftyfb: I actuall done that after googling, seems nothing happen, I'll do it again. [14:38] hao: apt list --installed linux-image* | nc termbin.com 9999 [14:39] leftyfb: it output "all the available drivers are already installed" and done nothing. === ssserpent2 is now known as ssserpent1 [14:40] leftyfb: https://termbin.com/f0vx [14:41] hao: try this: sudo apt install linux-image-5.19.0-35-generic [14:41] sorry [14:41] hao: try this: sudo apt install linux-image-5.19.0-32-generic [14:42] I'm on a old mac, it's been smashed several times, some hardware doesn't work (built-in display, bluetootch, etc..) but it never held me from booting into desktop... [14:43] leftyfb: "unable to located package ..." [14:43] try linux-image-5.19.0-31-generic [14:44] leftyfb: that can be found, installing now... [14:44] hao: you probably also want linux-modules-5.19.0-31-generic [14:45] hao: and linux-modules-extra-5.19.0-31-generic for good mEASURE [14:45] then reboot into 5.19.0-32 [14:46] leftyfb: the "linux-modules-5.19.0-31-generic" is already newest, will try modules-extra... [14:49] installing modules-extra now, with this network connection, will take like ten minutes. [14:49] leftyfb: can you tell me what's going wrong? [14:49] hao: potentially missing drivers in the latest kernels you have installed. That's my guess [14:50] the "too many bad BDL ..." error is even flooding irssi now, when switching between ttys. [14:51] leftyfb: so what you told me doing, is actually roll back to old version right? [14:51] yes [14:51] hao: sudo sysctl -w kernel.printk="2 4 1 7" # this should quiet the kernel messages [14:51] after this, will those newer broken kernel version got cleaned up? [14:51] I am trying to install vmware workstation player - after adding the missing components, and trying to install the modules with "sudo vmware-modconfig --console --install-all", I get a bunch of errors htat seem to be caused by "the compiler differs from the one used to build the kernel" [14:52] hao: this is a troubleshooting step [14:52] how do I take the current very nicely working system and build new headers from what already/actually exists? === polymorp- is now known as polymorphic [14:55] Hello again, I'm trying to install stuff from the `ubuntustudio-installer` (which itself installed just fine: https://paste.debian.net/plain/1275018/), but it's failing (not recognizing my passwords: https://paste.debian.net/plain/1275022/) - I've asked over at #ubuntu-studio:matrix.org and I got redirected here because it might be a policykit misconfiguration? [14:56] Should I take this "sudo apt update && sudo apt install -y --upgrade linux-headers-5.4.0-105 linux-headers-5.4.0-105-generic linux-image-5.4.0-105-generic linux-modules-5.4.0-105-generic", but replace the numbers with 12.2.0-3ubuntu1 [14:56] ending with """sudo apt update && sudo apt install -y --upgrade linux-headers-12.2.0-3ubuntu1 linux-headers-12.2.0-3ubuntu1-generic linux-image-12.2.0-3ubuntu1-generic linux-modules-12.2.0-3ubuntu1-generic" [14:56] leftyfb: finished installation, will reboot. [14:56] hao: make sure you force it to bo..... [14:57] darn - have to go now :( [14:57] weedmic: no, because there is no version 12 lonux [14:57] :/ [14:58] gettin old apparently, can't keep up with the youngin peacing out [14:59] Maybe the youngins should use KVM instead of VMware :p === polymorp- is now known as polymorphic [15:11] howdy, anyone know about this problem, ubuntu 22.04, xorg, gnome, all of sudden all my window decorations, themes are as if they're set to light *except* for firefox which obeys appearance and actually changes the theme whenever I switch between styles :? Icons also changed, I did a full reboot too but it didn't help. My desktop looks so hideous now [15:13] leftyfb: there is no kernel 5.19.0-32, so booted into 5.19.0-31, the "BDL" errors are gone, but still leave me in console, and my usb wifi card's connection is gone (its driver is built with dkms, so I guess this is expected). So I'm now here again inside 5.19.0-38. [15:13] any more thing I can do? === coXZist1 is now known as coXZist === csexecutive is now known as Guest5431 === linuxmodder is now known as csexecutive === coXZist1 is now known as coXZist [15:29] hao: Are the log messages the same? [15:29] Two gdm errors? [15:35] jhutchins: sorry I didn't checkout those gdm errors, only noticed that those "bad BDL entries" errors are gone, I can reboot inside *-31 and see [15:36] also because network connection broke in *-31, I didn't save them to termbin.com [15:49] hi, anyone knows how to understand why ubuntu pro don't work anymore? the command "pro security-status" seems stuck [15:51] max_0076: What did you try to do? How did you try to do it? What did you expect to happen? What happened instead? === tscott is now known as TC [15:53] i have aa free account that allow me to use 5 ubuntu pro license. And same day ago after an upgrade this command don't work anymore... but i don't know what i can check or make a kond of "dubug" === TC is now known as tscott [15:55] max_0076: the Pro License stuff will only apply currently to LTS releases, what command "isn't working" for you? [15:56] pro works fine for me on 22.04 [15:58] for example this command /snap/bin/canonical-livepatch that live in /etc/update-motd.d and write a message at ssh login take exacly 5min and 12 second to respond. I try also the command pro security-status but it keep running without any result [15:58] the server have 22.04 release [16:01] max_0076: any chance the server is low on resources or overtaxed or the infrastructure it's running on is slow-speed disks? [16:01] asking 'cause i see 'slowness' lke this on bogged down or slow-disk-IO systems === tscott is now known as TC [16:05] does ubuntu uses musl as default libc ? [16:07] ratakor: No. GNU libc is used by default. [16:07] I don't think you can safely swap it out either. [16:08] If you need musl libc for a particular program, you might be able to put it in the same directory as the executable before running it. [16:08] There's nothing that keeps you from having multiple libc's on a system :) [16:08] I see there are a number of disk snapshotting or "time machine" tools. Which one is the most popular or recommended? [16:08] @teward, could be. I try to understand the problem of ssh and login get stuck at "debug1: pledge: filesystem", i should try to check the HD. This HD have more than 10 years lol. I was thinking something about a netwrk problem or name resulution dealy [16:08] oh okay thank you arraybolt3 [16:09] max_0076: if debug shows the handshake happens and then it's waiting on a reply from $REMOTE after login/auth then the issue is the speed of the remote system (full swap/ram, slow disks, etc.) [16:09] calamari: feel free to ask for software opinions in #ubuntu-disuss or #ubuntu-offtopic. Otherwise, I recommend installing one at a time and seeing if it works for you personal needs [16:10] leftyfb: That's the kind of tool you only know if it meets your needs or not once a disaster happens. I'd rather get some user experience feedback. [16:11] calamari: it should be fairly easy to test with another PC, VM or container [16:13] leftyfb: I mean sure, but I was hoping to narrow it down to the best options first. [16:13] calamari: best is relative [16:13] Some people like an app that looks blue, some red [16:14] leftyfb: give me a break [16:17] @teward thanks :) the problem could be the HD, anyhow if i deactivate this message (99-livepatch-kernel-upgrade-required) the login proceed without any problem. This computer have also shorewall as firewall and i think that it is loading the HD more than i was imagine [16:24] calamari: It was just an example. For something more applicable to your situation, how much of the disk are you wanting snapshotted? Just the system files (Timeshift), or the whole whopping thing (btrfs snapshots)? How fast does it need to be? Instant (true snapshotting) or lengthy-but-safer (rsync)? You see where I'm going with this. There are many good tools with many use cases out there. We don't know your exact use case, therefore it's [16:24] hard for us to make a good recommendation. [16:24] (Also btrfs snapshots aren't the only way to snapshot a whole disk. Nor does it work on filesystems other than btrfs.) [16:25] calamari: for instance, I personally recommend backups via rsync and an ansible playbook to rebuild from scratch. I don't even recommend just blasting an outdated image onto a storage device as a proper backup [16:26] arraybolt3: entire filesystem, incremental if possible, offsite. Currently looking at duplicity, is there a better option? [16:26] * ogra whispers "zfs" into arraybolt3's ear [16:26] or btrfs [16:26] err, yes, that is why i said it [16:26] or clonezilla [16:27] or ghost [16:27] or just dd, but doesnt do incremental [16:27] tar ! [16:27] casync [16:27] calamari: see, none of this is productive [16:28] Does clonezilla do incremental? [16:28] nope [16:28] I kinda thought it was entire partitions or disks only. [16:28] leftyfb: I think it'd be productive for me to put you on ignore, you're just being an ass at this point [16:28] And timeshift only does system files? [16:28] I always thought it was a full backup, guess I'm glad I've never had to use it. [16:29] anyway, we'Re massively off topic for this channel, pleas take it to #ubuntu-discuss [16:29] calamari: Please be respectful and follow the Ubuntu Code of Conduct and IRC Terms of Service when using this room. Someone pointing out that there are many options to choose from and that listing them off here is unproductive is not them trying to be mean. [16:29] s/Terms of Service/Guidelines/ (got wrong terminology stuck in my head) [16:29] he's just naturally an ass? [16:30] I think he's trying, doing it for fun. But you say he's not trying. You're entitled to your opinion I guess. [16:31] e25, let it go please ... [16:31] :) [17:28] Is this how /etc/polkit-1/ is supposed to look like on a desktop-ish system? https://paste.debian.net/plain/1275044/ [17:29] I'm trying to set up a firewall with iptables and deny traffic by default with `iptables -A INPUT -j DROP`. I have https open and am forwarding it to 3000 with some nat rules. I also have loopback traffic enabled. However, these rules don't work unless I also open the port 3000, but I want that closed to the public. Any ideas? [17:31] ForeverNoob[m]: i also have nullbackend.conf.d/ on this 18.04 LTS system, the rest matches what you posted [17:34] irc_user: you want a closed port behind NAT to show as closed to the WAN? why? [17:35] tomreyn: Users are meant to access the application on https, not on 3000. It's supposed to forward their connection to 3000 internally, but they shouldn't be able to access `example.com:3000`, no? [17:36] oh i get the scenario now. i guess you need different rules then. [17:37] also note you may want to use nftables rather than iptables on current ubuntu releases [17:42] Yeah, I was just wondering if there was something obvious I'm missing. And thanks, I'll check that out. [17:45] Should I use a wrapper for this sort of setup or is using nftables directly the way to go? [17:46] iptables IS a wrapper for nft since 21.10 [17:47] oh lol [17:47] but then you didn't mention which ubuntu release you're asking about [17:47] 22.10 [17:47] So would you still switch to nftables or would that be pointless? [17:48] i would start using nftables directly wherever possible [17:48] Got it, thanks :) [17:48] or through a wrapper, but an nftables wrapper, not an iptables wrapper that then wraps it to nftables [17:50] this was a side note, though, it's probably not the root cause for your NAT rules not working as expected (however, i think i read the iptables -> nft wrpapper has some deficits) [17:52] There aren't that many rules, I'm happy to just convert it to nftables anyways and maybe I'll get lucky and it'll solve the issue :P === shokohsc0 is now known as shokohsc === bhechinger_ is now known as bhechinger === Rahoul18 is now known as Rahoul1 [18:13] irc_user: The ipmasq howto has some pretty good (if dated) info: https://tldp.org/HOWTO/IP-Masquerade-HOWTO/ipmasq-intro1.1.html [18:15] Thanks, will check it out :) [19:25] hey all, I'm trying to help a friend with something and I told her use ubuntu live usb and set a password, but when we try to instal openssh-server the command sudo isn't found. How can I continue helping her? [19:25] trying with ubuntu 22.04 btw === llalaland is now known as llalaland-test === llalaland-test is now known as llalaland [19:27] also appearently help.ubuntu.com is down. [19:27] down for me too [19:28] gartral[m]: Are you sure that it's "sudo' that isn't found? [19:28] And not something else? What's the exact error message, and the exact command you're using? [19:29] (Also try running "sudo su -" and that should give you a root shell if sudo is present and working right.) [19:42] Better yet, sudo -i [19:44] I ran out of time to help her today, I'll try that tomorrow [19:50] arraybolt3 jhutchins https://cdn.discordapp.com/attachments/654102641500094512/1088542262528573552/20230323_201747.jpg [19:51] gartral[m]: You're typing a capital "S" in "sudo". Commands are case-sensitive, the S should be lowercase. [19:51] So `sudo`, not `Sudo`. [19:52] (You or the person you're helping, I mean.) [19:53] sudo should be renamed to https://youtu.be/r0qBaBb1Y-U [19:53] do'h how did I miss they were doing that >.< [19:55] funny, but slightly ot, no? [19:55] no [20:03] what do you use in terminal to print a text file and mark a specific word? for example with grep I can see the word printed in the line, but I can't see the rest of the text [20:05] brkcore: so, you want to see an entire file, but have a word made bold everywhere it appears, or something like that? [20:05] sarnold, yep [20:07] brkcore: it seems obvious once you *hear* it :) try this: sed "s/sbin/$(tput bold)\0$(tput sgr0)/g" /etc/passwd [20:07] brkcore: that'll bold all the 'sbin' uses in /etc/passwd [20:08] :) [20:08] any other ideas [20:09] grep -C99999999 or something ridiculous :) [20:09] that'll put a bunch of other garbage on screen though [20:10] cat /etc/passwd | grep sbin .... anything that I can add to make all lines of the text file print, and still mark up the grepped sbin? [20:10] -C99999 what is that :)) [20:11] brkcore: try it: grep -C9999 sbin /etc/passwd [20:11] thats good sarnold [20:12] exactly what I needed [20:12] why is that, how does it work [20:12]