=== Morfeus is now known as _--_ === money is now known as Guest9844 === Guest9844 is now known as Gambino === money is now known as Guest8020 === Guest8020 is now known as money === Scotty_Trees2 is now known as Scotty_Trees === money is now known as Guest7223 === money is now known as Guest1297 === Guest1297 is now known as Gambino === Gambino is now known as Guest1931 === money is now known as Gambino [01:11] Is there a way to fix the firefox-update situation? [01:12] Im getting nags to close firefox so it can be updated , but if I close firefox it doesnt actually update [01:12] sudo snap refresh firefox [01:12] that solves it one time [01:12] it solves it every time [01:12] I want a permanent solution [01:12] that I can do once [01:13] there is no fix released yet. it will come with one of the next snapd updates [01:14] ah [01:14] https://forum.snapcraft.io/t/wip-refresh-app-awareness/10736/90 [01:17] so how will it work after the update I dont understand what they are saying [01:19] i dont know the exact plans. but probably actually performing the update when you click the notification [01:34] ah nice === sans is now known as Guest4769 [02:12] https://github.com/wanfuse123/zram-zswap-with-parallelized-compression-using-facebook-zlib/settings <<< test on your system at your own peril. Tested 5 times on fresh install Ubuntu 22.04 instances. seems ok. Feedback welcome! [02:15] oops drop settings off the link [02:17] transhumanist: this isn't the place for that. Try #ubuntu-offtopic [02:17] ok sorry [03:49] hi there - any news which kernel will be in tomorrows point update to 22.04.1? [03:49] Guest82: I think it's still 5.15. [03:50] If you need a newer kernel there's ways to do that. === LabMonkey is now known as Mechanismus [04:17] "fuzzymanboob98: maybe only worry..." <- hey not a bad idea [04:17] also sorry about dropping off, had to take an emergency work call just got off. Appreciate y'all helping out === greknod16 is now known as greknod1 === Probie9680 is now known as Probie968 === Probie9685 is now known as Probie968 === Loshki_ is now known as Loshki === Probie9687 is now known as Probie968 === Probie9687 is now known as Probie968 === Gambino is now known as Guest1221 [06:21] f [06:22] exit [06:28] is it possible to install ubuntu on a portable ssd and plug it in other pcs ? i thought we could do that but havent succeeded [06:29] que passa ! [06:31] hello ?? [06:32] Vegombrei: welcome to #ubuntu [06:32] Vegombrei: do STAY logged in to IRC, let people answer. [06:32] Vegombrei: what error(s) do you get with the portable-ssd installation? [06:44] enyc so my ssd is an m2 drive in an external enclosure so it shows up as a usb to pcie bridge drive, it works like a normal drive though but im guessing thats the reason [06:47] Vegombrei, it may depend on the devices you're wanting to boot the external media with; I have some that require me to press & hold a key down when the device is off, it'll turn itself on and boot external device ... ie. that box is a pain b/c of it's firmware... otherwise just require a key during boot process, but I'd expct it to work as long as devices can boot external drives [06:51] i did that https://www.youtube.com/watch?v=MNVnO2aPB_c ,and https://www.youtube.com/watch?v=P0nHvY6noXA [06:51] both didnt work [06:55] linux doesnt see the drive but windows does [06:57] hi all , is there still support for ubuntu 32 bit ? if yes , anyone can tell me which version is that ? as i have some old eeepc [06:57] just 1 gig ram and can't be upgraded [06:58] asddass, 19.04 was the last release with x86 or i386 32-bit support but it's EOL; 18.04 LTS the last still supported though flavor/universe packages only partially. 32bit is still supported on ARM; ie. armhf [06:59] hmmm [06:59] i am using x86 though [06:59] how is the performanc eguiverc [06:59] how is the performanc guiverc [06:59] ? [06:59] for 19.04 ? [06:59] 18.04 LTS still has support; particularly Ubuntu Server 18.04 LTS [07:00] I used asus eepc with n270 for QA-testing releases up to 19.04; Lubuntu 18.04, Xubuntu 18.04 would still run but they're EOL with security fixes only being provided for packages that were common with Ubuntu 18.04 LTS Desktop/Server or a five year support release (flavors only get 3 years) [07:01] i mean this laptop come with windows 7 and it run fast :( , not sure should i upgrade to windows 10 32bit or ubuntu , how long does 19.04 official support going to end ? [07:02] Ubuntu 19.04 reached EOL in Jan 2020; it was NOT a LTS release [07:02] mind to point me on ubuntu website which is the one that you refering guiverc ? [07:02] so i can read at least . i try to see but so many version not sure which one is which one for 32 bit [07:02] 19.04 was the 2019-April release; but not being a LTS release it reached EOL long before the older 18.04 LTS release [07:03] 32-bit is vague; armhf is the 32-bit ARM supported processor; Debian & Ubuntu refer to 32-bit x86 as i386 [07:05] your cpu is likely more than a intel atom n270; as windows 7 did not come on any of those devices... what makes a machine fast depends on more than just cpu [07:06] nah not sure what is this laptop made of [07:06] is old asus eeepc [07:06] nvm i guess i have no choice to upgrade it to windows 10 32 bit . heard it pretty slow to for ubuntu [07:06] thank you guiverc [07:07] intel atom n270 I mentioned was only sold with windows XP; but eepcs were available with a few CPU options, some were 64bit capable (amd64). Do note many 3rd party apps no longer provide 32-bit support & older versions can be a security risk [07:08] yeah [07:08] thank you guiverc [07:08] guess just downloading windows 10 now [07:08] see ya guiverc === Probie9686 is now known as Probie968 [07:55] I'm seeing this 'The following packages have been kept back: python3-distupgrade ubuntu-release-upgrader-core' on a number of machines on 22.04, I guess I can "fix" it by installing those packages manually, but is it expected to happen and is just installing them the intended way forward? === Gambino is now known as polo [08:02] hawk: what does 'apt-cache policy ' show? [08:03] zsh: parse error near `\n' [08:04] EriC^^: ah, I see "1:22.04.12 500 (phased 20%)" there, is that some kind of slow rollout thing? I wasn't even aware that was a thing [08:04] yeah [08:04] it'll work itself out hawk [08:04] i think [08:04] itsx handled [08:06] EriC^^: Right, it seems like it should, yes. I've just been wondering about that for a while now and didn't realize how I could find out what that was about [08:06] EriC^^: thanks [08:07] hawk: no problem [08:28] hawk, you could file a whishlist bug asking for a more descriptive message 😉 ... like: 'The following packages have been kept back because phasing is in progress: ...' [08:28] ogra: That would be very helpful, maybe I should [09:28] is there a packaging channel for ubuntu? [09:30] murmel: whats your purpose exactly? [09:31] lotuspsychje: want to build something (locally) have fakeroot installed, but dpkg-buildpackage complains about it, when using -d it errors out with can't find binary fakeroot (but it's definitely there) [09:35] murmel, do you use -rfakeroot as option to dpkg-buildpackage (to tell it to use fakeroot) ? [09:35] ogra: no [09:35] try that then 🙂 [09:36] dpkg-buildpackage -rfakeroot -b ... to build a binary ... dpkg-buildpackage -rfkaroot -S -sa ... to build a source package (with orig tarball attached) [09:37] *-rfakeroot indeed [09:37] ogra: fakeroot binary not found [09:38] hello [09:38] hello [09:40] murmel, does "which fakeroot" find it ? [09:41] ogra: yes [09:41] hmm, weird [09:44] ogra: if it helps, am on an up2date ubuntu server 22.04 install [09:44] any paticular reason to use -d (and did you try without it) ? [09:45] ogra: yeah it complains and errors out even earlier (because of fakeroot) [09:48] ogra: okay, weird. fakeroot was installed. but when I apt build-dep init-system-helpers,it reinstalled fakeroot and now it works [09:48] hah === money is now known as polo [09:59] hi polo === de-facto_ is now known as de-facto === guiverc2 is now known as guiverc [12:12] I'm having issues running Tiberian Sun. It installs but when I select single player it just flashes for a second then closes [12:12] Won't run multiplayer or skirmish either [12:13] I downloaded the snap version [12:15] shawn: is your graphics card driver installed properly? [12:17] shawn: if its an issue about the snap itself, best to contact the maintainer; contact: https://github.com/mmtrt/cnctsun/issues [12:17] shawn, see https://github.com/mmtrt/cnctsun/issues ... [12:17] ah ... *snap* [12:31] Hi all [12:32] I had a windows 10 vm running under qemu with a gvt-d gpu passed through on 20.04. [12:32] It was working friday, when I suspect my work pushed updates to the vm, my computer or both. [12:33] now it refuses to boot (goes into windows startup repair) [12:33] I've looked through the update logs and it doesn't appear anything changed other than updating the kernel itself. [12:35] it's possible the vm itself got corrupted [12:35] eg: unclean shutdown [12:36] ikonia, thats what I thought too, so i restored it from a backup qcow2 I had from May. [12:36] same behavior [12:36] why does windows say it's failing [12:36] it normally gives an error/problem of why it went into repair mode [12:36] eg: file X is missing, or could not do Y [12:37] windows may have used to do that, but windows 10 does not. [12:37] it does [12:38] I fixed a missing file on my own machine a few weeks ago, it refused to boot complaining a file was missing, then give me options, auto repair, drop to a shell, reboot, restore from backup etc etc [12:39] yep thats the same thing I've got, except without a complaint of missing file [12:39] it normally says it in the top left of the screen why [12:39] white text on a light blue background [12:39] well lets see what happens this boot [12:40] loops into "choose your keyboard layout" [12:40] "troubleshoot" "Turn off pc" [12:41] doesn't seem like it can find the hard drive from the command prompt. [12:41] https://www.linuxserver.io/blog/2017-11-07-how-to-fix-a-windows-10-qemu-guest-stuck-in-a-prepearing-automatic-repair [12:41] trying this procedure [12:42] and in so doing, my os drive pops up as E: [12:42] so something in the bootloader is extra mad [12:42] and isn't loading the viostor driver [13:16] Hello [13:16] aNYBODY HERE [13:16] welcome choko [13:16] what can we help you with? [13:17] I am new to Ubuntu Distro and I was checking everything [13:17] choko: this is the ubuntu support channel, you can ask questions here (if you have issues) [13:19] Okay thank you! === jontyms279 is now known as jontyms27 [14:14] firewall noob here, I try ufw status it shows a bunch of lines. I know I am listening for TCP connections on 8084 but it doesn't show up in the output. Why so? [14:34] hi there [14:35] hi [14:38] bittin, hexo: Hello! [14:39] Anything we can help with? [14:39] actally, I have a question. I've spun up VM with 20.04 [14:40] and it seems I can't use shift-pgup to scroll up in console [14:40] what can i do so that it works again? [14:40] kernel is 5.4.0 [14:40] that has nothing to do with the OS but with the VM ... typically VMs re-map these keys [14:41] Like, to scroll in a TTY? [14:41] Or in a terminal emulator? [14:41] yes, scroll a tty [14:41] no terminal emulator at all [14:41] (Also, knowing which hypervisor you're using may be helpful. KVM? VirtualBox? VMware? Hyper-V?) [14:41] KVM, virt-manager [14:42] hexo: Can you try clicking inside the VM windows and then try it? Sometimes I have to click all the way inside the VM in order to get keyboard stuff to be recognized in there. [14:42] clicking did not hepl [14:42] have you tried "shift+pgup" ? [14:42] yes [14:43] ogra: not sure of that. I have a physical machine, different distro, and Shift+PgUp no longer works. I think the kernel tty driver is changed. [14:43] that tends to work in plain qemu [14:43] rob0: kernel version? **** dropped supprt in 5.9 [14:43] It looks like it's gone. [14:43] but this is 5.4 [14:44] You could use screen or tmux. [14:44] to it should still work [14:44] i *hate* screen [14:44] tmux is even worse :( [14:44] hexo: Are you sure you're on 5.4? I thought you'd be on 5.8 with 20.04, and if you're using the HWE kernel you're on an even newer one. [14:44] uname sais 5.4.0-122-generic [14:45] no HWE at all [14:46] this is newly installed and updated 20.04, like an hour ago [14:46] ogra: rob0: kernel disabled it, because of security [14:46] Hi, I am running Ubuntu 21.10 impish but somehow my sources in sources.list are outdated. Maybe I missed some update period (as I am not on the LTS)? [14:46] mad ... [14:47] this is my sources.list https://paste.ubuntu.com/p/8tkbq35nVx/ [14:47] !eol [14:47] End-Of-Life is when security updates and support for an Ubuntu release stop. Make sure to update Ubuntu before it goes EOL so you get updates promptly for newly-discovered security vulnerabilities. See https://help.ubuntu.com/community/EOL and https://wiki.ubuntu.com/Releases for more info. Looking to upgrade from an EOL release? See https://help.ubuntu.com/community/EOLUpgrades [14:47] Does Ctrl+Shift+Up Arrow do it? [14:47] tokam: 21.10 is EOL. [14:47] arraybolt3[m]: still removed from kernel [14:47] tokam: Best solution would be to upgrade to 22.04 as per the guide I'll link to in a bit. [14:47] what to do about it? [14:47] ctrl+shift+up didnt work [14:47] thank you! [14:47] tokam: https://help.ubuntu.com/community/EOLUpgrades Back up all your data and then follow this guide. [14:48] hexo: oh I should have pinged you. the kernel removed that feature because of security [14:48] sad EOL happens so quick! [14:48] hexo: Argh, that's annoying. [14:48] tokam, use LTS 😉 [14:48] murmel: yes, in 5.9 [14:48] tokam: LOL it made me panic too. Thankfully 22.04 is an LTS release and won't do that for 5 years. [14:48] 21.10 is less than 1 year [14:48] murmel: or they backported it to 5.4? can i re-enable it? [14:48] yep [14:48] hexo: can't remember which version. but maybe it was backported [14:48] (Normal releases go kaput in 9 months, LTS get 5 years.) [14:48] (21.10 is a normal release, not LTS.) [14:49] tokam, it is 9 months for the non LTS releases [14:49] hexo: you would need to recompile the kernel [14:49] i can do that [14:49] arraybolt3[m]: so basically, I just use the old-releases path? [14:49] murmel: I bet you're right - if the change was a security fix, it's probably backported into Focal. [14:49] and that's all to consider? [14:49] yeah ... [14:49] then bring your system up to date and run do-release-update [14:49] (if you dont already get a graphical upgrade prompt for 22.04) [14:50] tokam: Yep, that should let you upgrade to 22.04. That way you won't be running outdated software anymore, keeping everything secure. [14:50] is there a way to recompile kernel and not lose all the nerves along the way? [14:50] some tools? [14:50] please, tell me there are :D [14:50] hexo: There's a guide, I'll find it... [14:50] make ... 🙂 [14:50] https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel [14:51] You can probably pull down Ubuntu's kernel sources and then add back TTY scrolling yourself, then build it and install it. [14:51] hexo: idk how canonical does it's backporting, but I assume it's in one of the patches [14:52] yea, thats basically what i wanted to do [14:53] thanks a lot for now <3 [14:53] i'll try to bake my own kernel [14:53] hexo: Glad we could help! === blackest_mamba_ is now known as blackest_mamba [15:17] so, it was removed in v5.4.66 [15:18] now i'm looking into how to revert the drop [15:19] yeah, I thought it must have been removed. [15:20] did you find a comment explaining why it was removed? [15:20] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=973c096f6a85e5b5f2a295126ba6928d9a6afd45 [15:20] Commit 973c096 in kernel/git/torvalds/linux.git "vgacon: remove software scrollback support" [15:20] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=50145474f6ef4a9c19205b173da6264a644c7489 [15:20] Commit 5014547 in kernel/git/torvalds/linux.git "fbcon: remove soft scrollback code" [15:21] "This (and the VGA soft scrollback) turns out to have various nasty small [15:21] special cases that nobody really is willing to fight. The soft [15:21] scrollback code was really useful a few decades ago when you typically [15:21] used the console interactively as the main way to interact with the [15:21] machine, but that just isn't the case any more." [15:21] which is obviously bull***t [15:23] i cannot recall when i used graphics "desktop" last time on my powerpc [15:23] so, yes, the code is still used [15:23] :( [15:23] I still use Linux console sometimes. [15:25] oh so it wasn't security related [15:25] more like laziness-related :( [15:28] hexo: sounds like you need to use a multiplexer [15:28] like tmux? [15:28] screen/tmux [15:28] a can't use it [15:28] and it usually interferes with how software works [15:29] :( [15:29] like? [15:29] * hexo doesnt recall [15:30] i use it only as serial terminal to interact with mcu [15:30] I mean the only thing I can (obviously) see is that sometimes the color profile is different [15:30] maybe, that was it? [15:30] also, escape sequences in screen are beyond me :D [15:30] i need simple shift-pgup [15:31] :D.... can I have it in some way? [15:31] yeah both don't do it ootb [15:32] i also kept forgetting to run it [15:32] and then needing to scroll output... :D [15:32] yes, i'm the worst in this [15:33] hexo: well that's why its getting autostarted when logging in for example [15:35] i thought that apt-get source would fetch the patches for me, but the files seems to be patched already [15:36] and seems like it does, but it's one large diff file and it gets applied automatically [15:36] so, seems like i have to go git route [15:38] hexo: oof [15:52] byobu? [15:53] arraybolt3[m]: something from a canonical employee [15:53] https://ubuntu.com/server/docs/tools-byobu [15:57] murmel: Yeah, I was thinking maybe it would help hexo. [15:58] yeah that's what I was saying also earlier [15:59] especially because of the ncurses interface [16:16] oh, thanks a lot [16:16] i was apparentnly blind [16:16] i'll try it out [16:16] <3 <3 [16:24] Yo [16:25] hi [16:36] cloning that kernel is an experience, i tell ya :D [16:36] stalls at couting objects: 8318020 [16:36] thats a lot of objects :D [16:39] hexo, uuuh ... use -depth 1 !! [16:39] else it will ake until the turn of the century (or close to that) [16:39] *take [16:40] :)) i'll try depth 1 [16:40] i wanted to see a quite a bit of history, tho [16:41] oooo there is shallow-since [16:41] this is gonna be more fun than i thought [16:41] FSVO fun ... [16:42] ls [16:42] ups [16:45] server does not support shallow-since, so, i'll have to take a 3000years-long-clone path [16:46] again, thanks for help, /me bbl [17:02] I have a question.                               I have two folders let's call them folder 1 and 2.                           Folder 2 is in. A separate partition and I want to sync the contents to 1 to 2 and vice versa [17:03] man that formatting :( [17:05] hello, i had problems with my gpt partition table but managed to restore them with testdisk, and now the partitions are numbered differently but the EFI and Linux ext4 are intact. [17:05] what should i check to make my 18.04 system work again? i have access to the rootfs via chroot, and i've already updated /etc/fstab [17:05] nthing in ubuntu uses partition numbering ... everything uses UUID ... [17:06] /etc/fstab is correct for now [17:06] so as long as you did not re-format the filesystems on these partitions, it all should still be fine [17:06] it isn't though unfortunately [17:06] whats the error [17:06] one error was waiting too long for suspend device [17:07] webchat48: so what's the question? [17:13] Is it possible to sync the contents of folder 1 to folder 2 and vice versa using symlinks [17:15] webchat48: you would need to setup a cronjob to do it on a regular schedule [17:16] Is it possible to use symlinks [17:16] And which software sound I use [17:16] no [17:16] rsync [17:17] The folders are on the same system [17:17] yes? [17:18] Are there any guides I can follow [17:20] webchat48: as rsync is quite complex, I wouldn't follow any guide. best is to read the manpage [17:20] i found teamviewer in system processes, I don't know if I ever install it. Could have forgotten it if so. How do I check when a package or app have been installed? [17:20] Is the a comand I can use [17:21] If I edit a file in say folder /directory 1 will it be synced with folder 2 if I use rsync [17:21] webchat48: unlikely, as every command is a bit different, and we don't know what you wnt [17:22] webchat48: if the timer/cronjob runs yes [17:22] Ok I will read the man page [17:22] so, I'm trying qo get old software (electrum 1.8) to run on old Ubuntu ( 12.04, freshly installed [17:22] and it's complaining that I don't have python-qt4 [17:22] it says to do apt-get install python-qt4 [17:22] but the package doesn't exist [17:22] even though the instructions are from the same timeframe... [17:23] murmel: Yes, if you symlink a file from one location to another, both files will be the same. You do not need to take any action to synchronize them. [17:23] webchat48: ^^ [17:23] jhutchins: not me, but their are different directories, not symlinks [17:23] arthur-_: did you already adjust the repos? [17:24] 👍 [17:24] and I really wonder why you want to use such old software [17:24] how do i see info about when and how an app was installed? [17:24] brkcore: logs [17:24] does vmlinuz, config, initrd have to be in the same location as the EFI directory? [17:25] nrb: depends on the bootloader [17:25] GRUB [17:25] then yes [17:25] murmel, I need to study the behavior of that software to help a friend with a problem that has its origin at that time. studying the modern software doesn't help. I'm not sure what you mean "adjust the repos". I just installed ubuntu 12.04.4 LTS as-is in a virtual machine, and it works, but it doesn't have python-qt4 [17:25] ah yeah, ok i will grep "teamviewer", any suggestions for a good way/command? [17:25] murmel, [17:26] arthur-_: the ubuntu 12.04 repos were moved, so if you want install something, you would need to adjust the repos [17:26] brkcore: grep :)? [17:27] yeah like showing the logs and then | grep "install" and "teamviewer" and it will show the installation package, and perhaps a date if ask for it, but dont know how, murmel [17:29] brkcore: you don't need to show the logs, just grep the file [17:30] murmel, thanks, that worked! [18:13] I got a lot of activity on teamviewer, but didnt use it and cant remember installing it. Any advise please? [18:13] brkcore: remove it if you don't recall installing it (you or someone with access to your computer did) [18:14] brkcore: teamviewer is not part of the ubuntu repositories so it would have had to have been done manually [18:14] leftyfb, that will probably remove the logs too [18:14] i have these files in its folder: brkcore@ Connections_incoming.txt install_teamviewerd.log signaturekey.log TeamViewer15_Logfile.log teamviewerd_FI_15.30.3_2022-06-05-171353.amd64.stack TVNetwork.log [18:14] brkcore: feel free to reach out to teamviewer for further support with their application [18:14] leftyfb, even on ubuntu 22.04 not in the repos? [18:15] brkcore: Who may have had access to your system? [18:15] brkcore: no [18:15] arraybolt3[m], nobody [18:15] brkcore: Where did you download Ubuntu from? [18:16] ubuntu.com [18:16] brkcore: if you don't trust your install, reinstall from scratch and restore from backup [18:16] I don't have anything compromised yet, it seems like its been a bit over a month now [18:16] brkcore: /join #security if you'd like help with forensics [18:17] leftyfb, ok, will try with security channel [18:17] brkcore: The thing is, it came from somewhere. Someone messed with your system or you forgot installing it. [18:17] brkcore: if you have a remote access application installed on your computer without your knowledge, you have been compromised [18:17] So keep that in mind, and realize how utterly creepy it is to have software like that installed without your knowledge. Then backup and reinstall. :) [18:18] Or have some forensics guys help you figure out what happened if you'd like to dig further. [18:18] I don't remember installing it, and im not sure how to distinguish activity from just the regular modules and services on boot === money is now known as polo === polo is now known as money === money is now known as polo === polo is now known as money === money is now known as polo === polo is now known as money === Bencraft14192 is now known as Bencraft1419 === dionysus70 is now known as dionysus69 === Guest3811 is now known as DD3my [19:58] trying a do-release-upgrade and stuck on a "An unresolvable problem occurred while calculating the upgrade." [...] "Please use the tool 'ppa-purge'" [19:58] is there a way to identify which ppa/packages are causing this [19:59] et09: so which ppa do you have active? [19:59] the tool writes a log to /var/log/release-upgrades/ (or similar) [19:59] yeah there's two, apt.log and main.log... i do see some "broken" package references [19:59] et09: you would need to see if there is a repo for the newer version each one of them [19:59] in the log, there will be one or more lines starting "Foreign:" which will probably indicate the root cause [20:00] there's like 50 packages under foreign [20:00] if this was a contest, i'd congratulate you [20:00] what, most fucked up system, or worst support inquiry [20:00] that's a high number of foreign packages [20:01] a bunch of postgres stuff mostly [20:01] Is there a way to check ones number of foreign packages without upgrading? [20:01] apt list --installed | grep ',local\]$' is not identical, i think, but similar [20:02] there's a .distUpgrade for everything in sources.list.d [20:02] pretty sure ppas are not local installs oO [20:02] et09: generally, you're supposed to remove all 3rd party applications and repositories before upgrading releases. There's no way those can be supported as part of the upgrade [20:02] though this won't list packages installed from active third party repositories [20:03] i'll try it [20:03] wouldn't be the end of the worst if this instance got destroyed [20:03] world [20:03] do-release-upgrade disables third party repositories itself. but this won't remove the packages [20:04] it's odd that ppas can't support an upgrade path [20:04] if you want the upgrade to succeed, remove both [20:04] not odd, unsupported ppa's are unsupported [20:04] i mean, hypothetically, maybe the ppa spec should include some info that would enable a minimal upgrade support [20:05] i understand that'd be almost impossible for cross-ppa issues, but relative to the main repos... [20:06] et09: how would that work, if the ppa is for example non maintained anymore, and the package doesn't compile on the newer ubuntu? [20:06] that obviously wouldn't. but if there's a ppa for each ubuntu release, it should be able to act like a repo extension [20:08] jsut thinking out loud there [20:09] I can't follow. as long as the package is not in ubuntu, ubuntu can't really do anything about the package (and it can break the upgrade). but I do see that if the package is within the ubuntu ecosystem (I package systemd for example) then it should upgrade automatically without issues (as long as the versioning and naming is aligning) [20:56] I have a flash drive (Samsung Bar 128GB) that can reach 330MB R/65MB W under Mac and Windows, but only 25MB R/W under Linux. Is there anything I can do to remedy this? [20:56] To add to the case, each test was performed under the OS's native FS (APFS on Mac, NTFS on Windows, ext4 on Linux) [20:58] how did you test? [20:58] Windows - CrystalDiskMark, mac - AmorphousDiskMark (clone of the former), Linux - dd and hdparm [20:58] write test: dd if=/dev/zero of=test bs=1MiB count=1024 oflag=dsync [20:58] read test: hdparm -t --direct /dev/sda [20:59] also tried read with dd, same result as hdparm [21:00] as I understand it, crystal disk mark will saturate the device queues with long pipelines, multiple threads of execution, etc; dd will just issue one IO at a time [21:01] if you leave off that oflag=dsync then the OS will return success sooner and let dd issue another overlapping write [21:01] writing zeros is a bad test, a lot of controllers will recognize zeros and compress those away [21:02] if I leave the dsync flag, won't caching mess with the result though? [21:02] five61, "Disks" has a "Benchmark Parition" feature. never really tried it but is shows fancy graphs [21:02] *partition [21:03] buffering, but yes; fsync will issue an fsync on the file that'll wait before exiting, I think that'll help [21:04] if I time "sync; dd ...; sync" and divide that by amount written, would it be more accurate? [21:05] or rather the other way around, divide amount written by time taken [21:07] https://p.haxxors.com/96nbmovx.png [21:08] ravage: does this tool have a CLI? This is a server and I haven't set up X server on it.... [21:08] nope. fio may be a good alternative [21:09] https://arstechnica.com/gadgets/2020/02/how-fast-are-your-disks-find-out-the-open-source-way-with-fio/ [21:09] there is also a mac version. just for comparison :) [21:11] do be careful of eg fio on mac, https://twitter.com/marcan42/status/1494213855387734019 [21:15] ravage: Ah, an upscale version of hdparm. Looks useful [21:21] so slightly better results with fio, 40MB R/70MB W [21:21] still far from the 65/330 results on mac and windows though :/ [21:22] I did notice the following in dmesg, could it be related? [21:22] [ 3529.627217] usb-storage 2-1.3:1.0: USB Mass Storage device detected [21:22] [ 3529.627880] usb-storage 2-1.3:1.0: Quirks match for vid 090c pid 1000: 400 [21:23] anyone know about "Composes Key" ? [21:24] https://askubuntu.com/questions/1274176/how-to-access-keyboard-preferences-activate-the-compose-key-insert-german-um [21:24] Relates to this question which I share. [21:24] I've already installed other language packs but it's not about changing entire keyboard. [21:25] When typing in my en keyboard occasionally I want to write foreign words in their correct spelling which means having certain accents on certain letters. Is that possible in Ubuntu? [21:26] In OSX, it is. For e.g. if I simply hold down the 'u'key, many options of various accents around the 'u' pop up to select from. Same with the 'a' and other letters. Is that possible in Ubuntu? [21:27] amaroq, https://i.imgur.com/007uju3.png [21:27] https://help.gnome.org/users/gnome-help/stable/tips-specialchars.html.en#compose [21:27] There seemed to be some documentation around that but it was out of date when the guy in Askubuntu asked about it on a previous Ubuntu version [21:28] ravage, thx. So, that's in Settings? In the thread it was suggested that its in Gnome-Tweaks but I can't see it there [21:34] but, where is the "Compose" key on the keyboard when I am typing and need it? [21:37] amaroq: I think you need to choose which key you want to use as the Compose key (like left Alt). [21:38] yes, but I'm not finding where to define that. It says to go to Tweaks and click disable next to Compose Key, but i don't see it [21:39] 4 Click Disabled next to the Compose Key setting. [21:39] what desktop environment are you using? [21:39] ...but, I don't see [21:39] Ubuntu jammy [21:42] amaroq: alright; there's probably a "settings" app somewhere. maybe it's called "control panel"? maybe it's on the dock? maybe it requires clicking through an applications menu to find it? do we still let you hit f2 or alt+f2 and then start typing to search? [21:43] amaroq: if we still have an f2 or alt+f2 to type, try that and tyep 'settings' and see if you get something that looks like a settings page [21:43] https://imgur.com/a/yZoTKGb [21:44] sarnold, I have both Settings and Gnome-Tweaks...I just can't find the Compose Key the link is pointing me to [21:44] amaroq: what link? [21:45] ravage's screenshot shows a settings app that looks pretty different from your tweaks app https://i.imgur.com/007uju3.png [21:46] amaroq, https://p.haxxors.com/84xujt9i.webm [21:53] ravage, Danke schón [21:53] gerne [21:54] the default 'left alt' wasn't doing anything. I changed it to left Super key and now something is happening. But, still, how to find all the exact accents? [21:55] OSX seems a lot easier in this regard. Out of the box, we just hold down the letter and all the different possible accents associated with that letter pop up and we can select it [21:59] Schön...so, ist besser [22:03] ravage: awesome :D === bernard is now known as niber [22:20] I'm having an issue with vfio and the updated kernel...  running 22.04, kernel 5.15.0-38 worked fine, every kernel after that doesn't allow vfio for me, virtmanager keeps saying host doesn't support passthrough [22:26] sorry, 5.15.0-39 worked fine, since then it doesn't [22:39] !info vfio [22:39] Package vfio does not exist in jammy [22:44] jhutchins: vfio is basically how you pass through hardware into a VM. [22:57] arraybolt3[m]: Yeah, I figured it was internal. [23:00] any idea?  I'm running x99 xeon e5 2630v4 [23:04] webchat14, and what GPU? [23:05] quadro p400 [23:06] it's not a gpu specific problem, I don't think, I have vfio-pci.ids in the bootloader commandline to disable the GPU, as of > 5.15.0-39 it doesn't even do that [23:07] file a bugreport? i finnd no simular issues, yet [23:07] -n [23:08] Hello from lubuntu, ubunters around the globe. [23:09] up to 5.15.0-39 the display goes blank after the bootloader, when vfio takes over the device, but after 5.15.0-39 linux still has hold of the gpu and virtmanager says the host doesn't support pci passthrough when I try to start the VM [23:09] when I boot in to 5.15.0-39 manually at boot time everything works as expected [23:10] webchat14: Just to confirm, the VM environment is all stock Ubuntu packages, not direct/3rd party, right? [23:10] correct [23:11] oh wait, I do have docker and kubernetes [23:12] webchat14: Right, but that's not involved in the VM is it? [23:12] nope [23:12] webchat14: Can you still boot to the last functional kernel to confirm that's the change that caused it? [23:12] we're talking about the host right?  not the vm itself [23:13] yes I can boot to 5.15.0-39 and everything works [23:13] !es [23:13] En la mayoría de los canales de Ubuntu, se habla sólo en inglés. Si busca ayuda en español entre al canal #ubuntu-es; escriba " /join #ubuntu-es " (sin comillas) y presione intro. [23:13] webchat14: Well, I suppose we could build a "clean" vm that didn't use them just to see. [23:13] webchat14: I think you've nailed it down well enough for a good bugreport. Reproducable, cured with previous release. [23:14] we could look at journalctrl.. [23:15] I would suspect the vhost layer, but reverting the kernel fixes it.