=== lcfork[m] is now known as forkliang[m] === forkliang[m] is now known as lcfork[m] === Polo is now known as Guest1976 === themill_ is now known as themill === _filename is now known as filename === k4rma is now known as K4rMa [00:36] I had to boot my dual boot laptop into windows to recover my sons iPad the other day. When I tried to reboot back to ubuntu grub didn't appear instead it went to a grub minimal recovery screen or something. I've done a bit of searching around google and have tried some things but I'm way out of my comfort zone would this be a good place to ask? [00:36] I have managed to boot it with a live USB drive, but I haven't been able to use any of the things i've found on Google to help out. [00:36] not sure about that one. When you reboot into Linux, have you done a `sudo update-grub`? [00:36] GNU BRUB version 2.06 [00:36] Minimal BASH-like line editing is supported.... [00:37] that's what i see on boot up. Just booting it into ubuntu now to retry the update-grub command. I have tried it can't recall what it said [00:39] cool, let us know what it says: sudo update-grub | nc termbin.com 9999 [00:40] I bet Windows overrode your UEFI boot selections or something [00:42] \/usr\/sbin\/grub-probe: error: failed to get canconical path of '\/cow'. [00:43] hmmm had to add the \s in there to make it show [00:43] /cow? That's a new path. [00:43] I don't recall that error searching for that one now. [00:43] (yeah, IRC commands start with a slash) [00:43] yup that's what it says [00:44] /cow sounds a bit like the live boot environment [00:44] Oh, you're booted from a different Linux system, that's why. [00:44] (cow = copy-on-write, aka, your rootfs is read only) [00:45] That sounds right. I can only booth the live environment [00:45] Can you mount your /boot partition? [00:46] Also, is this UEFI or BIOS/CSM? [00:46] UEFI (I think) [00:46] Try this: https://www.ubuntupit.com/how-to-repair-the-grub-bootloader-using-a-ubuntu-live-usb-drive/ [00:46] Use the flash drive's rescue mode [00:49] Hi. I just ran `sudo apt update; sudo apt upgrade`. There were a bunch of xorg and nvidia driver updates. It seemed to be fine, but a few minutes later my desktop froze. I can still move the mouse, but the desktop is unresponsive. The Xorg app is taking 100% of my CPU. Is there any way I can restart X without rebooting? [00:49] I'm ssh'd in from my laptop. [00:50] waltman: can you tell if apt is still running in the background? Killing X or Wayland would likely take out apt as well. [00:51] after tose updates you would need to reboot, IIRC [00:51] Xorg is still running [00:51] apt finished maybe 10-15 minutes before this happened. [00:52] oh, okay, then try: sudo systemctl restart gdm [00:52] that's the nice-ish way to do it [00:52] No response. [00:52] waltman: maybe systemctl restart display-manager would do it, that's probably very similar to signofzeta's command; but imho restarting X11 isn't all that different from the reboot :( so you might as well go all the way down and back.. [00:53] .. as you supposed to.. [00:53] I don't get an expert mode or rescue broken system just - Try or install ubuntu - ubuntu (safe graphics) - OEM install - Boot from next volume - UEFI Firmware settings [00:54] That didn't help either. [00:54] windows might have switched on Fastboot in uefi.. [00:54] rjft197: try Method 2 on that web site. I admit I've never had to do this. [00:54] This has happened to me before after an nvidia update. It should really force a reboot after that. [00:55] OK, I'm just going to reboot it. I'm be back in a few minutes. [00:58] signofzeta Oops i didn't scroll down to see method 2. I have tried boot-repair it doesn't have any suggestions. I will get the output from it though. [01:06] https://paste.ubuntu.com/p/RV7KrXz7hY/ [01:09] everything seems normal. let me look. [01:11] can you mount your Linux partition? /dev/nvme0n1p7 [01:11] Also I discovered that a bit of the fdisk -l output looks like this [01:11] Device Start End Sectors Size Type [01:11] Where as parted -l reports [01:11] Number Start End Size File system Name Flags [01:11] 1 1049kB 263MB 262MB fat32 EFI system partition boot, esp [01:11] 2 263MB 397MB 134MB Microsoft reserved partition msftres [01:12] I'm back. Rebooting seems to have fixed things. [01:12] I'm guessing your /boot is on either partition 7 with your other Linux files, or on the ESP. [01:12] waltman: excellent! [01:13] rjft197-lv: You might try mounting your system's disk to /mnt using "sudo mount /dev/ /mnt", then bind-mount critical paths ("sudo mount --bind /dev /mnt/dev". repeat for at least /proc and /sys, maybe more than those). Then "chroot /mnt", and "update-grub". [01:13] basically, GRUB is starting up but it can't find your grub.cfg file. [01:14] rjft197-lv: Oh, you might also need to mount the EFI system partition with "sudo mount /dev/ /mnt/boot/efi". [01:15] > sudo mount /dev/nvme0n1p7 /mnt [01:15] NTFS signature is missing. [01:15] Failed to mount '/dev/nvme0n1p7': Invalid argument [01:15] The device '/dev/nvme0n1p7' doesn't seem to have a valid NTFS. [01:15] Maybe the wrong device is used? Or the whole disk instead of a [01:15] partition (e.g. /dev/sda, not /dev/sda1)? Or the other way around? [01:15] NTFS signature is missing. [01:15] Have I tried to mount the wrong thing there? [01:15] rjft197-lv: That's the wrong partition - that's where Windows is. [01:16] no, looking at his partition table, p7 is the only one marked as Linux Data [01:17] Oh, I can't see the pastebin, so... Weird. Try "sudo mount -t ext4 /dev/nvme0n1p7 /mnt". [01:17] Replace "ext4" with whatever filesystem you're using (if you're using BTRFS, use "btrfs" instead of "ext4", for instance). [01:18] signofzeta: Any clue why on planet earth "mount" would be trying to mount NTFS when it's a Linux fs? That's so weird. [01:18] I think that's the issue. [01:18] arraybolt3: I don't know. It's guessing wrong? [01:18] (And I'm realizing "NTFS signature is missing" doesn't mean it's a Windows partition necessarily - sorry.) [01:19] fd... seems to think it's linux [01:19] but par... seems to think it's unknown [01:19] Are you able to get into your files with the file manager? [01:19] I can't mount the drive [01:19] I've just tried those options. I'd assume it was ext4 but I've tried ext2 too. [01:19] ext3 I mean [01:20] Ubuntu defaults to ext4 these days. Do you remember what file system it was? [01:21] Device Start End Sectors Size Type [01:21] ... [01:21] \/dev\/nvme0n1p7 389177344 1961932799 1572755456 749.9G Linux filesystem [01:21] that's from fdi [01:21] sk [01:22] rjft197-lv: I wonder if maybe Windows pulled a fast one on the partition table, so that the partition table has the partition marked NTFS when it's actually ext4? TestDisk might be able to do the trick here, but that could be somewhat risky, and I've never used it, so... [01:22] your partition table looks normal: EFI System Partition, bunch of Windows partitions, then one Linux one. [01:22] I'd say to try running fsck, but now I'm getting uncomfortable [01:23] Most of my data is backed up. I was just hoping that windows had managed to do something iffy without losing all the data to negate the need to reinstall the OS and all the software [01:23] what the heck happened while you were in Windows? Feature update? [01:24] And trying to mount it with the "-t ext4" option in there didn't do the trick? [01:24] Happy to try anything one last time. Probably I hadn't booted into windows for a long time. I don't use it anymore. It's only because my sons iPad was set to unavailable so needed a restore [01:24] I tried -t ext4 [01:24] ext3 [01:24] ext2 [01:24] and the other one you suggested [01:24] Bah. Try popping open GParted and see what partition type it thinks it is. That should tell you the filesystem. [01:25] How can you write that. [01:25] GParted [01:25] ha, parted -l [01:25] It wasn't letting me write that earlier. I was trying to provide the output from parted -l [01:25] It's not a terminal command, it's a GUI app. If you're in a live Ubuntu desktop, you should have it in the Application Menu. [01:25] Try this: fsck.ext4 -n /dev/nvme0n1p7 [01:25] The -n switch will make no changes [01:26] gparted and fdisk have differing opinions [01:26] And yes, you were muted for spamming. Use a paste bin site like paste.ubuntu.com, termbin.com, etc. for multi-line stuff [01:26] Whatever FS gparted thinks, put that after the -t and see what happens. [01:28] fdisk -l https://termbin.com/b2x2 [01:29] parted -l https://termbin.com/tc7bd [01:29] gparted has some red exclamation marks next to p1, p2, p3 and p7 [01:30] p7 has an unknown file system in gparted [01:30] yeah, that filesystem is hosed. You can try fsck -n on it to see what can be done, but damn -- what did Windows do? [01:32] fsck.ext4 ..... command https://termbin.com/y58s [01:33] So weird. Best I can figure at this point would be to run the commands suggested by the fsck, but that could end in disaster. Guess now I know to NEVER dual-boot Windows and Linux... [01:34] yeah, I don't even know, man. Something happened while you were in Windows, and that's definitely not normal. Dual-booting should be fairly painless. [01:35] If you're not comfortable causing worse damage, I'd take a full disk image of the NVME device with dd before trying anything fancy with fsck. But if you're fine with losing everything on the drive (including Windows possibly), then I'd try the suggested fsck commands and maybe TestDisk and see what happens. Might work, might blow up. [01:36] fsck can't even find an ext2/3/4 superblock on that volume, so it's more than hosed. Unless it's something else like ZFS or btrfs. [01:37] I would have thought I'd gone with the default system when installing. [01:37] yeah, nothing wrong with ext4. I picked ZFS for auto-snapshots and stuff -- not that that would have helped you here. [01:37] and the laptop is only a year old or so. [01:38] Wipe it, reinstall Linux giving it the whole entire disk, then plop Windows in a virtual machine. [01:38] I don't know what happened to the contents of that partition, but it's gone. You can try booting into Windows to see what it thinks, but it'll probably just say "RAW" [01:40] At this point I'll try anything I did try the commands that fsck suggested they just gave the same output [01:40] I can't boot into windows. I don't have a boot screen to do so. [01:41] ideally I'd keep the dual boot. I don't have instalation media for windows to create a virtual machine. And I do have one project that I'm working on that would need me to use windows. [01:41] Since it's UEFI, you might be able to use your firmware's boot menu to select Windows Boot Manager instead of ubuntu. Or is that also broken? [01:43] your device ran into a problem and needs a restart. we'll restart for you [01:43] and it doesn't [01:43] ah, damn [01:44] Well, I'm not sure about this one. Startup Repair (if you can get that far) might be able to get Windows working, but your Linux partition seems to be toast. [01:45] I've heard of Win10 breaking GRUB when it updates, but that's usually a quick fix. I've never seen it hose a partition like that. [01:45] There's a dell supportassist os recovery. I wonder if that will allow it to reinstall. [01:46] can't speak for that, but do you have a backup of your Windows data (if you care about that)? [01:46] thanks anyway. Yeah I don't have the best luck with computers. I had a nas for a bit of a better chance of not losing data. Two hard drives died within 10 minutes of each other :( [01:47] that happened to me, too! RAID 5 with faulty Seagates. [01:47] I have a backup of anything that is important. Just no ability to install windows i don't think. [01:47] can you borrow another computer? Grab a flash drive and download Microsoft's Windows 10 (or 11) Media Creation Tool. [01:47] Then get that working and make an Ubuntu install disk. [01:50] I haven't tried this since the Win7 days, but supposedly, the way to dual-boot natively is to do Windows first, then Linux. [01:51] I have another computer that's not the issue. And I have an ubuntu live usb key that I can install ubuntu from. [01:51] Yes I've only ever done windows first then linux. [01:51] at least you have a backup. I commend you for that! [01:52] (my backup wasn't working at the time my laptop broke, had to swap motherboards but i managed to get that working. [01:52] just need to get my laptop back. I just don't know how to get windows onto the laptop now. [01:53] the easiest way would be to make a Windows installer, clean the disk, then start from scratch. [01:53] I'd use the Windows Media Creation Tool. I'm not familiar with SupportAssist. [01:53] If it lets you partition the disk, just save some space for Ubuntu [01:56] I'm heading to bed. Good luck, rjft197! [01:57] Thanks, Just don't know how I'm going to install it. It came preinstalled so i don't have a certificate. I don't know how that works these days [01:58] oh, it's automatically activated. Your Windows key is burned into the ACPI tables. [01:59] just install the right edition (Home, Pro, or other) and it takes care of itself [01:59] Oh cool. there is a download and install the factory operating system. So time to start the reinstall. Thanks very much [01:59] anytime! === fmirkes_ is now known as fmirkes [02:40] hi guys, i just turned on my laptop and now have no sound devices in pavucontrol [02:40] any ideas on how to fix? [02:41] jwash: Did you just finish rebooting after an update? [02:42] yes [02:42] this is my soundcard: Audio device: Intel Corporation Sunrise Point-LP HD Audio (rev 21) [02:43] owner@Latitude:~$ alsamixer [02:43] cannot open mixer: No such file or directory [02:43] jwash: Almost certainly the result of a broken kernel update. Try these instructions: https://askubuntu.com/questions/945403/how-to-downgrade-kernel-after-bad-update-16-04 [02:47] installing a new kernel, will see if that fixes [02:48] jwash: Pretty much any time you reboot and then something doesn't work, it's a kernel problem and a kernel downgrade (or maybe a kernel upgrade) will do the trick. Just keep trying new kernels as they come out to see if they work yet, and if not, fall back to the old one. [02:51] whole computer is lagging now [02:52] startup took 2 min before login screen appeared, usually 5 secs [02:52] had to use onboard to login, then it took 2 more min before it connected to wifi and keyboard started working.... [02:53] jwash: That's... weird. That shouldn't happen. How did you install the kernel? [03:02] hi [03:03] how to install ubuntu lts without gui on my VM ? [03:03] !server [03:03] Ubuntu Server Edition is a release of Ubuntu designed especially for server environments, including a server specific !kernel and no !GUI. The install CD contains many server applications. Current !LTS version is !Focal (Focal Fossa 20.04) - More info: http://www.ubuntu.com/products/whatisubuntu/serveredition - Guide: https://ubuntu.com/server/docs - Support in [03:03] i have iso file of ubuntu 20.4 [03:03] what vm exactly? [03:04] oracle VM [03:04] Virtual Box [03:04] If you use the iso for ubuntu server, it won't install a GUI by default. [03:05] but installation looks gui [03:05] But if you're trying to use the regular iso to install w/o dekstop, dunno if thats possible... [03:05] you can see virtualbox, right? [03:05] enigma9o7[m], yes [03:06] oerheks, yes [03:06] enigma9o7[m], which the correct link for it? [03:06] Yes? I dont recall asking a question.... [03:06] indeed server needs no gui or monitor [03:06] means [03:06] downloads on ubuntu.com? [03:07] i downloaded from ubuntu.com [03:07] yes [03:07] https://ubuntu.com/download/server [03:08] https://mirror.ette.biz/ubuntu-releases/22.04/ubuntu-22.04-live-server-amd64.iso for example [03:09] From untu.com/download/server select option #2 for a direct download of the iso. [03:09] (ubuntu) [03:10] enigma9o7[m], i am downloading your suggested version [03:24] what does [m] mean on peoples name [03:25] descent1: Pretty sure it means they're on Matrix chat bridged to IRC. Don't know that for sure, but I know Matrix is bridged, and I've seen users who have both a normal nick and a [m] nick, so that's my best guess. Anyone want to confirm? [03:26] ah ok [03:26] Usually, yeah. Some people see that and think it means mobile, so use it for that. But typically it's Matrix. [03:49] yes confirmed it means I'm using a matrix client, not real irc client. === JoeBk_ is now known as JoeBk [05:07] I need help with apt policy if someone has a minute [05:08] when I am installing chromium there is another policy [05:08] 101.0.4951.64-1~deb11u1 500 [05:08] 500 http://deb.debian.org/debian-security bullseye-security/main amd64 Packages [05:08] *** 99.0.4844.74-1~deb11u1 700 [05:08] 300 http://ftp.debian.org/debian stable/main amd64 Packages [05:08] after the version.... the 700. Not sure where that comes from, but it blocks the upgrade to a newer version [05:10] tgoyen, #debian [05:10] /etc/apt/preferences.d [05:13] @enigma9o7[m], Thanks! [05:24] yeah there was a special preferences file in there someone had left me.... [05:36] Greetings people, My Dell Inspiron 7000 series laptop does not wakeup with touchpad interaction... [05:36] It used to wakeup in Ubuntu 21.10 [05:36] But no longer works in Ubuntu 22.04 Linux Kernel 5.15 [05:36] Just to clarify, [05:36] suspend and wakeup IS working [05:36] touchpad after wakeup IS working [05:39] anyone ? === Punna is now known as Pwnna === compuguy5 is now known as compuguy [06:45] is anyone there ? [06:45] yeah [06:45] Hi [06:45] hi [06:46] I have some problem with Ubuntu 22.04 [06:46] I have "a" problem [06:47] I'm here too! [06:47] What's gone wrong? [06:48] my laptop does not wakeup on touchpad interaction [06:48] power button for wake up works [06:48] but not touchpad trigger ... [06:48] also, does anyone know how to contact Kai-Heng Feng ? [06:48] I found this https://patchwork.kernel.org/project/linux-input/patch/20200618145515.5055-1-kai.heng.feng@canonical.com/ [06:49] which is the closest I think i have come to identify problem. [06:49] The touchpad used to wakeup laptop from suspend up till Ubuntu 21.10 [06:49] in 22.04 its not working [06:50] nightBulb, I know how I'd contact (Canonical/Ubuntu) people, but I can't see other passing on that detail (if it's not publically available) [06:50] Hmm. Not sure how to fix the touchpad, if it's not working anymore, but it used to work, then it's quite possible it's a kernel problem. You can fall back to 20.04 for the time being. [06:50] As far as contacting a Canonical employee, that's not how this sort of stuff is usually done. You file a bug report. [06:51] It appears the issue is with Linux Kernel 5.15 .. [06:51] I filed a bug report [06:51] s/other/others ^ , but I agree completely with arraybolt3 [06:51] guiverc, can you ask ? [06:51] https://patchwork.kernel.org/project/linux-input/patch/20200618145515.5055-1-kai.heng.feng@canonical.com/ [06:51] what happened to that patch ? [06:51] in that case I'd file a bug report too [06:51] Then just wait, and check your email daily. Contacting people directly is not a good way to get a problem fixed, it's a good way to make people annoyed. [06:52] guiverc, arraybolt3 https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1971933 [06:52] Launchpad bug 1971933 in linux (Ubuntu) "Unable to wake laptop using trackpad input" [Undecided, Confirmed] [06:52] I was hoping Kai was here on IRC ... [06:52] Looks like a good report to me. Just give it time. [06:52] no-one is responding in #ubunuu-kernel [06:52] to get bugs noticed; test to see if it occurs in the current development release... that tends to get a developers attention more than a released product that requires SRU; they'll backport if possible (ie. kinetic fixes will reach jammy in time via HWE at minimum) [06:53] I've been doing bug reporting stuff, it takes some time, but I've witnessed a bug get fixed (actually, two bugs), so the reports don't go unnoticed. These people just have gobs of work to do (me included, though I'm not a Canonical employee). [06:53] Yup [06:54] I saw the buglist was quite big [06:54] that's why I was trying to get it fixed as soon as possible, by identifying which version caused trouble .. [06:54] 5.13 -> 5.15 [06:55] Ubuntu 21.10 -> Ubuntu 22.04 [06:55] nightBulb: Like guiverc said, try using the latest development release (Kinetic) and see if it happens there, too. The way Ubuntu works is, new changes get merged into the development release most easily. That's the one everyone works on. [06:55] Here's Ubuntu Kinetic downloads: http://iso.qa.ubuntu.com/qatracker/milestones/433/builds/248206/downloads [06:55] arraybolt3, thanks :) [06:56] Note, don't use development releases as your daily drivers - they may have bugs that the devs are still sorting out. Use them for testing (like seeing if the trackpad issue happens). [06:56] arraybolt3, it boots off of USB ? [06:56] of off [06:56] of of ? [06:57] Depends on how you want to do it. You can burn it to a DVD, flash it to a USB with Startup Disk Creator, dd, or balenaEtcher. [06:57] But yeah, should work off a USB no problem if that's what your asking. [06:58] Yup, I test it (of of) USB first... [06:58] then report on bugreport ? [06:59] Yep. If you can reproduce the problem on the bug, stick a comment on the bug report indicating that it happens in the latest Kinetic. [06:59] okey-doke [06:59] (You can also add a "Kinetic" tag once you've verified it still happens, but I'm happy to do that for you.) [06:59] sure .. [07:00] off I go ... [07:00] thanks arraybolt3 guiverc [07:00] :) [07:01] You bet. I'll also do some testing on my end, and pitch in some help. I've already subscribed to bug notifications for the touchpad bug. [07:01] :) :) :) === dtg01101 is now known as dtg01100 [07:14] wow will pulseaudio ever be fixed? it is extremely buggy since years, unable to output analog audio and now it seems even fails over hdmi. killing it sometimes cures it, but thats not really a solution [07:14] de-facto: hdmi is a bit problematic, especially with displayport; which pulseaudio version do you have? [07:14] with analog audio it makes the volume slider wildly wiggle on itself sometimes, other times it works [07:15] the one from 22.04 [07:15] de-facto: please share 'fuser -v /dev/snd/*' [07:16] its listing /dev/snd/controlC0 and /dev/snd/controlC1 [07:17] pastebin it please [07:17] nothing really interesting in there but ok [07:18] https://paste.debian.net/1241248/ [07:18] i replaced username [07:19] de-facto: np, looks good; pulseaudio does not 'wiggle' volumes unless you have flat-volumes enabled in /etc/pulse/daemon.conf, but that setting is disabled by default for ages [07:20] hi all, is there any way to specify for a package to be installed from the main channel and not the -updates channel? [07:20] I'm having issues installing libssl3-dev in a jammy docker container [07:20] de-facto: so if that is still 'wiggling' it is either firefox operating output volumes (it is a known issue) or you have some other app changing volumes on alsa devices [07:20] Keep getting 'libssl-dev : Depends: libssl3 (= 3.0.2-0ubuntu1.1) but 3.0.2-0ubuntu1.2 is to be installed' [07:21] i-garrison, how can i disable any app to control volume? [07:21] (sorry i mean libssl-dev not libssl3-dev) [07:23] i dont want any volume changes by any app, its too buggy [07:23] can i disable that interface completely? [07:23] de-facto: there is no option to "disable" that for all apps, but here is a described workaround specifically for firefox https://bugzilla.mozilla.org/show_bug.cgi?id=1422637#c19 [07:23] Mozilla bug 1422637 in Core "Volume does not stay where it is manually set using pulseaudio" [S3, New] [07:24] de-facto: also if you have something changing volumes on alsa level there is nothing pulseaudio can do with that [07:24] i dont know what is causing this, but i dont need any program to change volume at all, hence my desire to get rid of this annoyance [07:25] de-facto: and please check if you are no affected by this 22.04 bug https://bugs.launchpad.net/ubuntu/+source/pipewire-media-session/+bug/1953052 [07:25] Launchpad bug 1953052 in pipewire-media-session (Ubuntu Jammy) "In Jammy sound level reset after reboot" [Medium, Fix Released] [07:26] de-facto: well disabling volume controls on alsa level is separate topic; on pulseaudio level you can use that firefox workaround but set the virtual sink as "default" which will isolate pulseaudio apps from device volume controls [07:26] this bug affecting me already exists for many years [07:26] its not new in the current ubuntu or such [07:27] probably more like a decade old or such [07:27] de-facto: maybe yours is different, but 22.04 bug I linked is new and it looks like it could apply for you [07:27] de-facto: still check if you do not have flat-volumes enabled in /etc/pulse/daemon.conf [07:29] Unode: Try setting the version you want manually, like so: "sudo apt install libssl3=3.0.2-0ubuntu1.1" [07:29] idk if that has anything to do with the sound slider wildly wiggling around, e.g. many times a second [07:29] ; flat-volumes = no [07:29] all defaults [07:30] arraybolt3: thanks, that works. Switching the order of packages installed on the system also seems to work. If libssl-dev is the first thing I install then apt is happy. Didn't yet check what is the final result but at least no build failures. [07:30] Unode: Awesome! [07:30] arraybolt3: thanks for the suggestion. I was trying to avoid pinning the version as that will limit future uses. [07:31] also killing pulseaudio fixes it sometimes, hence i blame pulseaudio itself [07:32] Unode: Glad you got a working solution. [07:33] Unode: Does using "=" pin a package version? I didn't know that, and the info I'm finding doesn't tell me that, so that's important to know. [07:33] de-facto: ok if you can reproduce the issue at will and have some time to debug this in detail, hop onto #pulseaudio on OFTC and we will check what exactly is causing volume jumps on pulseaudio side [07:34] thats the problem i did not find out what triggers this yet [07:34] sometimes it just ends in that state, other times it works without any sign of problem [07:34] arraybolt3: sorry by pinning I meant writing a fixed version in the recipe. I misused the word pinning there as it has a different meaning under apt. [07:34] de-facto: if that happens e.g. daily you can set up debug logging and we will catch it [07:35] Unode: Oh, ok, that makes sense. [07:39] ubuntu server 22.04 has not netinstall/minimal iso. installation on HP servers via ILO is really hard, even with USB stick take long time to boot for first installation any idea? [07:39] mini iso on older version is about 50-100 MB, instead full ubuntu server is about 1.5 GB. [07:40] sweb: You could try using a preinstalled server image. [07:40] http://cdimage.ubuntu.com/ubuntu-server/jammy/daily-preinstalled/current/ [07:41] how can i confugure, network, apt-proxy and etc using this method ? [07:42] The image is still beefy, but it's not as large as what you're using currently. You can boot your server from a lightweight OS like Alpine, and then use xzcat and dd to flash the image to your server. [07:42] sweb: You'd probably need to configure the server manually after installation. Once you've figured out exactly what all config files to change, you can make a Bash script that does it automatically, then be able to deploy the server image to multiple servers. [07:42] sweb: Once the script is built, just boot, flash, reboot, run script. [07:43] it's really bad decision to deprecate mini iso, it's very usefull and fast [07:44] sweb: (If you need mass deployment, you could also try using DRBL and Clonezilla. Once you've got one master image built (which you can probably do in a VM), you can then use Clonezilla to mass-deploy the image over the network. [07:51] ok so the latest ubuntu 22.04 desktop update screwed up my nvidia drivers again [07:52] the last time it was in the proposed-repo, I couldn't find a way out so I reverted with timeshift [07:53] now it's in the main repo and it completely screws up the works as it did when it was in proposed, so "great" [07:53] I'm exactly in the spot where I was at when the proposed-repo killed my nvidia drivers [07:53] got a 30-series nvidia [07:53] all my work's stuck [07:54] I spent a day trying to troubleshoot this when the proposed-repo screwed things up, the only thing I concluded that would fix it in the end was to revert to my daily timeshift backups [07:56] baviano, whilst in proposed did you file a bug report? that's part of why it sits in proposed; so people can test & report issues... if none are reported & it proves successfuly in testing it'll go to main repositories. [07:57] nvidia-smi says no driver installed, back when this happened when this was still in the proposed-branch; which was somewhere last week, I suppose, the problem was that there was no corresponding nvidia kernel whatevers to the version [07:57] can't fucking remember, pardon the french [07:58] guiverc: lol so the previous guy gave me a ton of crap for using anything from the proposed-repo altogether and said I shouldn't have used that (as in: "you shouldn't have worn that dress") [07:59] proposed can help... but if you didn't have problems before; and the problems only occurred when you enabled proposed - that's not a good thing & needs reporting generally.. It maybe I'm mising something in your history & prior advice sorry [07:59] so ... the last time (= last week) when this happened with the proposed repo, I did manage to get some "life" back into my nvidia card, i.e. I could get it to show up on nvidia-smi, but the graphics would still be forced on to my internal gpu's output [07:59] atm it's the same square one as it was with the proposed-repo's update [08:00] sorry for being a bit salty but that's a production machine that I run neural network training on and I'm ... yeah [08:00] I try to do the right thing, not clutter up things too much, ... could it be that I have Nvidia's developer repo enabled? [08:01] I mean am I barking at the wrong tree here ... let's see [08:02] /var/log/dpkg.log:2022-05-18 06:43:13 install nvidia-dkms-510:amd64 510.60.02-0ubuntu1 510.73.05-0ubuntu0.22.04.1 [08:02] ok so that was the latest thing that's been installed according to '/var/log/dpkg.log' [08:03] which is odd, since I _swear_ I saw a bunch of packages on ubuntu's software updater nagging me about how they need to be installed ... but /var/log/dpkg.log says that's the only thing that's been installed in the past few days [08:04] let's see, running [08:04] 'sudo apt install nvidia-dkms-510:amd64 --reinstall' just to see what the hell, [08:04] oh yep it couldn't probably figure out a way to handle the UEFI secure boot [08:06] okay ... alright. [08:06] so the auto-update screwed it up because the dkms package needs you to set up the UEFI secure boot, and it got compeltely omitted, so ... [08:06] *blood pressure maybe returning back to normal* [08:09] WORKS! [08:09] guiverc: works, when I reinstalled the dkms package that had been installed last night and re-entered a new password and validated it on boot ... [08:09] the auto-update process on dkms drivers, woo-hee.... [08:09] thanks for the help and sorry about the gun [08:09] ;::D [08:11] hiii [08:15] hii [08:15] soy turganti [08:22] guiverc: ok, so to recap; the dkms driver package got auto-updated last night via ubuntu's software updater (the gui one); obviously since I have secure boot enabled, it needed me to set a password for the new driver and whatnot, which was completely omitted by the software updater (that only wanted me to reboot) [08:22] hello ! [08:23] guiverc: so yeah, reinstalling the package via apt like I said over there solved the issue, but that's overall a bit of a... let's say, if you're not much into package handling, well, sure glad I don't do end-customer endpoints. [08:24] you might want to see if it could be possible to i.e. open up the terminal prompt for dkms driver installation via software updater whenever it's required, or somehow pipe the output through the software updater. [08:24] because otherwise you'll just end up with a black screen with no video output on reboot. I'm glad this workstation has a built-in GPU as well [08:24] others might not be that lucky. [08:24] k thnx bye [08:25] arraybolt3, I just checked on Kinetic [08:25] bug exists [08:25] on Kinetic "current" build [08:28] How to mark ? [08:47] Hello, When I start the system the desktop looks like this: https://i.ibb.co/fQdLdvg/j.jpg How can I make the wallpaper "full screen" by default? [08:48] The system starts like when the 'super key' is pushed. [08:54] exit [09:24] nightBulb: You should see an area marked "tags" underneath your bug report (before the comments, I believe). You should be able to edit that - add the word "kinetic" to the line of tags, separate from the other tags. [09:24] I'll do it in a bit (if I remember, which I hopefully will) [09:26] nightBulb: Got it tagged, will comment once I get it tested. You may want to comment as well. [09:29] arraybolt3, thanks, (The long bug report kind-of / somehow threw me off, and couldn't see the tags option at the bottom) Just saw it ... [09:29] Thanks for tagging [09:31] 👍 [09:34] hello. question: I want to know, using a raid-1 on Ubuntu on a server. how can I get knowledge if there`s a failure on some hdd drives? maybe get an email if so. [09:35] or see if there is a prefail condition? [09:37] dreamon, depends how you're implementing raid-1 -- in software via mdadm? [09:40] (my hardware raid knowledge is likely massively out of date at this point so I can't help with that, but I can provide some hints on monitoring with mdadm and smartmontools) [09:40] waveform, want to use a brand new mainboard. so think its hardware designed? never used a raid befrore [09:41] okay, if it's hardware raid, and if hardware raid still operates like it did about 10+ years ago, then basically monitoring is up to your raid hardware and whatever management tools it provides because the OS won't "see" the underlying drives anyway -- just the device the raid hardware provides. But as I say, that's how it *used* to be and I've no idea if that's still relevant (I'd heard rumours it's all soft-raid these days anyway, but what that [09:41] means in practice I don't know) === polymorp- is now known as polymorphic [09:44] waveform, how does software raid work? is this a OS thing to do? so using mdadm its a linux management software? [09:44] maybe lower performance? [09:47] yes, mdadm is the OS doing software raid; with the spare capacity on modern processors (and the vast disparity between the speed of processors and the speed of mass storage) I'd be extremely surprised if there was any performance difference there, even with lower-end CPUs. The main worries are (or used to be?) things like corruption in the case of power loss (hardware raid used to have battery/capacitor backup systems for that, software raid can rely [09:47] on journalling-esque systems which does indeed lower performance but is optional -- if you have a decent UPS setup you *might* consider it unnecessary for example) [09:49] obviously in the case of the OS managing the raid setup, it *can* see the individual drives, so in this case I strongly recommend installing smartmontools and setting that up to do periodic self-tests on each drive in the array (staggered so you don't go hitting 'em all at once), with e-mail sent in the case of errors. Further, mdadm should be configured to performance periodic "scrubs" of the array (I *think* that's actually the default in ubuntu/ [09:49] debian's packaging of mdadm, but worth confirming) [09:51]