[01:26] join #systemd === Sven_vB_ is now known as Sven_vB === zbenjamin is now known as Guest9330 === zbenjamin_ is now known as zbenjamin [01:46] I have a focal machine on which, to get network working, I have to either unplug and replug my USB ethernet adapter after openbox is ready, or issue command "sudo modprobe ax88179_178a". I created a file /etc/modprobe.d/usbnet.conf with "ax88179_178a" + linefeed in it, rebooted, but "lsmod | grep ax" still is empty. dmesg has no messag about it, syslog says "systemd-modules-load[1193]: Failed to look up module alias 'ax88179_178a': [01:46] Function not implemented". how can I fix it? === skr is now known as Guest30609 === HappyHotDog is now known as CAPPYHOTDOG === Guest30609 is now known as skrull === syb0rg_ is now known as syb0rg === CAPPYHOTDOG is now known as SarcasticHotDog === SarcasticHotDog is now known as HappyHotDog === OEIRAS is now known as OERIAS === HappyHotDog is now known as scruffles5 === scruffles5 is now known as scruffles7 === scruffles7 is now known as HappyHotDog [02:55] okay, what's going on [02:55] I put the new beta on my pi & now I can't change back to xorg in gdm [02:55] why can't I change the default session in gdm [02:55] I checked /usr/share/xsessions & gnome-xorg.desktop is literally right there [02:56] someone here needs to tell me how to beepboop gdm the right way to re-enable the ability to change sessions, because I am *not* using wayland if you outright refuse to fix it working in framebuffer_depth=24 mode [02:56] I'm not going to use ubuntu as a production OS with amiga graphics [02:56] NO ONE WILL [02:57] I can't believe you even left framebuffer_depth to default to 16, seriously, amiga [02:57] the graphics I'm looking at right now as I'm typing this should look better than the best of what the 1980s has to offer [02:57] omg, & the worst part [02:58] the worst part about all of this [02:58] is wayland was working just fine in 20.10 in 24-bit color mode === skr is now known as skrull === skrull is now known as Guest57087 [03:01] see, notice how there's no one here [03:01] they're all on their private canonical company.....thing [03:01] all the cool kid volunteers left 10 years ago [03:01] you deserved to die, ubuntu [03:02] I bet the forums were taken offline & all archives destroyed for legalise liability reasons [03:07] atari2600a, yes the default flavors are optimized for more usual use cases. this helps ensure Ubuntu makes enough money to keep going. which in turn allows for people like me to pick just the good parts. [03:07] atari2600a, what install method did you use anyway? [03:07] 'makes enough money' [03:07] preinstalled image [03:07] the only one [03:08] I'm just saying, arch is run by 6 guys in an irc server [03:08] do you mean the cloud image? [03:08] no the desktop image [03:09] contrary to popular belief, the raspberry pi is a computer [03:09] yeah. still, desktop is geared towards "desktop users". [03:09] btw there is absolutely no need to call it 'preinstalled' when it runs the OEM installer on first boot [03:10] yeah, & a deskop user won't want to look at a 16-bit image [03:10] you might be better off starting with the cloud image (https://cloud-images.ubuntu.com/focal/) and then use ansible to install what you want and not install any clutter. [03:10] this isn' AOL, this is Ubuntu. the thing you install to do image editing & watch youtube [03:10] omg [03:11] you guys getting so EEE'd by microsoft with all this azure cloud shit, I'm not doing it. I install a desktop os & run update on the package manager, just like I have for the past 20 years [03:11] no containers, no virtualization, just a barebones x server [03:11] I'd be contempt with weston IF IT WORKED IN 24-BIT MODE [03:11] afaik the "cloud" image is just a very barebones Ubuntu. [03:12] yeah, so I'd install the ubuntu-desktop package myself, I get it [03:12] of course you can also multistrap to get it even more barebones [03:12] but that still wouldn't change the fact that your defaults don't allow us to change sessions at gdm [03:12] well, the ubuntu-desktop package has some dependencies I don't like, so I forgo that. [03:13] yeah but if I'm doing all that, why not arch or deb you know? [03:13] upstream [03:13] dunno, I never encountered those defaults in focal, due to the way I install. [03:13] I use Ubuntu because the support is nice usually. :) [03:13] I'm just saying, things are getting a bit KDE4/GNOME3 Classic in here [03:14] *was [03:14] I mean, here we are [03:14] over 1000 users & no one to talk to [03:14] micsoroft took the slow & steady route to kill ubuntu [03:14] yeah it's easter holiday [03:14] oh, shit, right [03:14] fuck [03:15] sorry apparently the server lags me when I swear [03:15] no problem [03:16] but, it's not like an op's gonna ban me. & even if u r op, it's like, whatevs I'm not getting much help anyways :) [03:17] indeed. however, that might be a matter of approach. :) [03:17] anyway, I gotta go now. happy ranting :) [03:19] hi people! === skr is now known as skrull [04:46] hello, so i would like to monitor my disks running in lvm raid1 , what tool is best to make sure the disks are ok at all time? smarttools or do you recommend something else? [04:52] hey, im currently trying to use an isp programmer but im running into some issues with ubuntu handling it, im currently running ubuntu 18.04.5 and when i go to plug in the usb isp programmer it never makes it to a tty, in the journalctl -xe im seeing https://pastebin.com/3FQcgyyP [05:01] Hello! & on behalf of everyone at Canonical, Ltd I'd like to WELCOME you bothto the null void! [05:02] pagios, apt-cache search smart [05:02] jamie_1, you need to something something add serial to the right user/group [05:03] there should be a whole wikipedia page about getting serial interfaces up & running [05:03] sorry not wikipedia, what I meant to say is there should be a whole outdated ubuntu page on he matter [05:04] when in doubt, pretend it's an arduino, because it's going to interface the same way probably. https://google.com/search?q=arduino%20not%20working%20ubunu [05:04] you know what the sad thing about ubuntu is? [05:05] 24hrs later, no commits to the repositories [05:05] everyone's getting down with jesus on sundays [05:09] atari2600a: i'll take a look, bit new to screwing with devices [05:18] atari2600a: why not use zabbix or nagios [05:28] honestly.... nevermind... im just gonna build a quick programmer out of an arduino and save myself the nightmare im having right now [05:29] the further i looked into the more i realized on linux i would actually have to reprogram the programmer as the firmware on it is garbage and im too lazy to do that right now, ill reflash it later and make it an easy to use programmer. But i want my printer back up so im just gonna build one [05:35] not really an ubuntu topic, but here you go: https://wiki.fryktoria.com/doku.php?id=arduino:how-to-fix-a-usb-isp-programmer-and-make-it-work-with-arduino-ide-on-linux === LabMonkey is now known as Mechanismus [06:31] I have created a boot script. There is just a simple line `echo "SET=0" > "$rootmnt"/boot/boottime.rc` . But update-initramfs updates successfully. The file is not begin created in boot dir. [06:31] Any suggestions please? [06:32] look at me [06:32] LOOK AT ME [06:32] systemd is the init system now [06:34] atari2600a: I want to run the script before mounting rootfs and wants to include the script in initrd. === jrg_ is now known as jrg [07:37] hi people! [07:37] I am running 2 vm but this is just a beginning [07:52] atal, you aren't getting it bruh [07:52] the kernel runs initramfs [07:52] initramfs drops directly into systemd [07:53] we're talking late-80's monolith kernel-&-monolith structure [07:53] it controls init, services, networking & scheduling [08:01] If I want to run a script in initrd then what should I do? [08:05] atal: https://unix.stackexchange.com/questions/460219/how-to-run-a-script-before-switch-root-in-initramfs 2nd answer [08:08] Hi there. Having a bit of an issue after dd-copying a system SSD for my Raspberry Pi 4B. The copying went without any problems, but the new SSD boots to BusyBox and I can't mount the SSD's Linux partition due to an unknown error? [08:09] since it's been copied with dd "as-is", I'm wondering if I should i.e. force the filetable or something like that to the new SSD that won't boot. [08:10] toffo: can you also share the dd command you used to clone, perhaps volunteers know something [08:10] lotuspsychje: sure. [08:10] # sudo dd if=/dev/sde of=/dev/sdf bs=1M status=progress [08:11] toffo: the disks where same size? [08:11] 'bs=1M' just because it's faster that way, 'status=progress' does nothing to the copying but shows the status [08:11] lotuspsychje: yep, but different manufacturers. the other is a WD MyPassport and the other's Samsung T5 USB3 SSD. [08:11] both are USB3 SSD's tho. [08:11] 500GB [08:12] toffo: did you reboot for a second time too? [08:12] lotuspsychje: yes. [08:12] It just goes to BusyBox [08:12] hmm [08:12] I do see the partitions oknp [08:13] I read from somewhere that at times you have to copy the MBR separately, or use fdisk to force the filetable/partition structure on the dd-cloned disk [08:13] EriC^^: ^ one for you [08:14] now I'm wondering if the results might've been better with i.e. Filezilla [08:15] I guess I should A/B compare the geometry of the disks. One thing that I've heard is that if there's a mismatch in i.e. sector sizes or so, that can be a problem [08:16] oh, okay, so gparted says "Can't have a partition outside the disk!" === ace_me^^1 is now known as ace_me^^ [08:17] (actually, it said it twice for that disk, so I'm kinda wondering if dd wrote "out of boundaries") [08:17] toffo: can you pastebin the output of 'sudo parted -ls' ? also did you run 'sync' after dd, that's pretty crucial as lots of stuff would still be in the cache if you just unplug the ssd after dd [08:18] however, according to 'fdisk -l', the source drive should be somewhat smaller than the destination [08:18] EriC^^: I have created a script in the same directory but the script is not doing it's job. https://paste.ubuntu.com/p/CF5zMhxT7J [08:18] there's plenty of empty space on the source drive, so another option would be to recreate the file tables and just rsync the files? [08:18] atal: perhaps try to let it send its log somewhere as it runs [08:19] maybe add /bin/bash -x and let the output go to /somedir/myscriptlog [08:20] toffo: that would work, you'd need to edit the uuid's for the fs, and then chroot and reinstall grub [08:21] I find it a little bit odd that the process failed like this, I've made plenty of migrations on Raspberry Pi SSD's before [08:22] toffo: as i said, pastebin parted's output and did you run sync? [08:22] It must have something to do with the fact that they're both SSD's of the same size but from different manufacturers, and that the filesystem is expanded to the boundaries of the source disk. But then again, dd gave no warnings or errors. (i.e.: writing out of bounds, out of space, etc.) [08:22] doing it asap [08:24] EriC^^: Please elaborate I don't understand. [08:25] atal: in the script's shebang, use #!/bin/bash -x (it lets it be very verbose/debug) and when you run the script in that initrd, put the script somewhere else and let the line be /path/to/script > path/to/log 2>&1 [08:26] so that when it runs it outputs everything it does, then you can debug why it's not working [08:31] EriC^^: https://pastebin.com/PKYD2YNs so there's the destination drive's 'parted -ls' and 'fdisk -l' output [08:32] EriC^^: interesting how 'parted -ls' refuses to co-operate altogether, but I do get at least the structural listing out of 'fdisk -l' [08:33] toffo: the disk has 976773168 sectors, sde2 ends at 976781278 [08:34] toffo: ultimately the source disk was slightly larger than it [08:34] EriC^^: ah. hmm. funny, because even dd usually warns when it's writing out of bounds [08:35] EriC^^: the source drive does have like hundreds of gigabytes of empty space [08:35] toffo: see the source drive's sectors [08:35] EriC^^: but of course the main fs is expanded to the drive's size [08:37] EriC^^: dammit, the source indeed was .x1 GiB larger :D [08:38] palm => face [08:39] it happens [08:40] After a couple of days of uptime, my, fairly newly-installed, Ubuntu 20.04 system becomes quite unresponsive, with a few seconds delay when changing windows or doing most things. Here is the output from "top", sorted by CPU and Mem usage, and also "free". Can anyone spot what's causing it, or what else I need to do to troubleshoot it, please? [08:41] toffo: if you have a backup of the source one and still want to dd, you could shrink the fs just a tad, and dd again, and on the destination one you could then make a fresh table pointing to the fs locations, but one that doesnt exceed the disk space, remember to use sync after dd though [08:41] https://www.dropbox.com/s/yuuy05iwq45cvfi/Selection_253.png?dl=0 [08:41] https://www.dropbox.com/s/th3cf9w7gt5i509/Selection_254.png?dl=0 [08:41] https://www.dropbox.com/s/pb49cldb4cmiy62/Selection_255.png?dl=0 [08:41] okay, so, given that the source drive is around half-empty ... What would be my options? https://pastebin.com/PKYD2YNs [08:42] EriC^^: I've updated that url ==^^ [08:43] EriC^^: yep, the source drive is working oknp 100%, just wanted to make a "hot-swap(ish)" backup of it now that I got all my junk together [08:45] is it risky to resize a filesystem on an ssd? I've read on dd and its recommended use from StackExchange and God knows how many different *nix forums, every proposed solution is always countered by ten(s of) nay-sayers [08:46] i wouldn't resize without a backup if it's important, i mean it's supposed to be safe but ive had some bad encounters especially with ntfs-resize on linux [08:46] kinda lols when the opinions on 'dd' to begin with are split between 'the only thing you'll ever need' and 'grandpa, that was good for the time when you did backups on large magnetic reels' [08:47] toffo: if you want to save time copying the data since its mostly empty, you could delete the linux partition on the dest, recreate it to use proper space, format a fresh fs (and pass mkfs.ext4 -U here to save time editing fstab) then rsync the files [08:48] EriC^^: yes, that's my other woe, not to bash on extfs, but what I've read, to this day it still has some flaws in it that can become -- [08:48] EriC^^: Yes. [08:48] if you're using uefi, the PARTUUID of the efi should be the same as your source, so if you ever plug it in, it should look for it and boot the efi file there and all should be good since you're using the same uuid for the root fs [08:49] EriC^^: someone once complained something that even rsync isn't a "failsafe" method when you're cloning a primary Linux partition ? [08:49] if you're using msdos/legacy you'd need to chroot and reinstall grub to the mbr, easy to see on the source what you're really using, ls /sys/firmware/efi , if it shows files its uefi [08:50] toffo: i dunno how that's possible, rsync copies and runs md5 checks [08:50] so beats me, juse make sure to use the proper options with it, it's found on the archwiki i think on backups [08:50] EriC^^: in other words, what I've read of rsync is that some fundamental files can easily be copied in an order where they are not supposed to have been created or smth [08:54] this OS/2 program is TUI-based. [08:54] EriC^^: I remember reading a spin-off thread on that from StackExchange or some of these other forums when I was looking into finding the right method to mirror a filesystem [08:56] EriC^^: or, could it have been just for rsyncing live systems... [08:56] (which of course should never be done) [08:57] (at least on that level that is) [08:57] toffo: yeah for live systems it makes sense [08:58] i dont think for a non running system it would/can mess up the file locations [09:02] gparted does complain about "partition being outside the disk" twice for the target SSD btw ... I'm wondering if I should just "nuke the entire thing from orbit" and i.e. copy the partitions with gparted [09:02] I mean at least the msdos boot one [09:03] the sector size is different between the two as well. other one has 512/512 bytes, other one 512/4096 bytes [09:04] the sector difference i think is just that physically its 4096B/sector but it acts as if 512, so it's all good there [09:05] EriC^^: I'm just kinda worried when gparted does warn about a partition being outside the disk _twice_ ... [09:06] I disabled the kernel auto update in my Ubuntu 18.04.5 PC by configuring linux-image-generic in held state.. Is there a way to verify kernel auto update is disabled or not? [09:07] yeah, it's not usable as is, or dependable toffo , use rsync as i mentioned above [09:07] it's off by about 4M right now missing data from the end of the disk, which might be related to the ext4 fs or who knows, you want the backup to be 100% [09:09] EriC^^: Yes, that's what I'm already in the process of getting ready to do :) I was just wondering if I should i.e. wipe both the fat32+ext4 partitions from the dest. SSD (= make it blank), then recreate the partition tables in gparted for both the system-boot and writable, just to be on the safe side .. [09:09] I might be a bit paranoiac here but when it comes to *nix and filesystems, no amount of caution can be exaggerated too much :D [09:12] toffo: ah this is for a rpi or something? [09:12] EriC^^: yes.. I think I mentioned it? RPi4B, Ubuntu 20.10 64bit. [09:12] EriC^^: it has its own quirks [09:12] ah, did not see that, i think those use labels to mount stuff 'writable' and 'system-boot' and stuff [09:13] yeah exactly :D [09:13] EriC^^: well they would use but I've edited the fstab to use UUID's instead, it's imho a bit of a bad practice in a lot of circumstances to point to drive labels alone [09:15] toffo: i'd say leave the system-boot one as is, since it's not wrong, has the same UUID and label, the ext4 one you need i guess to have the 'writable' label on it and when you mkfs the fs use the old UUID there so it stays the same [09:15] im not sure if rpi also use the partuuid somehow, maybe you can also set that to the same one when you create the partition for ext4 [09:15] might as well [09:16] EriC^^: the problem is that since at least the ext4 partition is out of bounds, gparted doesn't want to co-operate unless I wipe the whole drive. [09:16] toffo: use gdisk [09:16] i mean, fdisk [09:18] EriC^^: ty. one question; if I wipe the drive's partition records and re-run dd for the partition#1 (boot), it will leave the rest of the drive unallocated, right? [09:19] let's say; dd if=/dev/sda1 of=/dev/sdb1 , that should just copy the first partition and leave the rest unallocated? [09:20] yeah, but it wont copy the partuuid, only the fs [09:21] toffo: the disk identifier is the same for both right now, right? [09:21] EriC^^: yes. [09:21] yeah, you want those to be the same so the partuuid is the same [09:21] EriC^^: but I can always just edit the cmdline.txt and /etc/fstab to match the new ones? [09:21] if they get changed, that is [09:22] also, is clonezilla any good for operations like these? I've never used it much [09:22] usually for uefi in the motherboard it looks for a partuuid to get the efi, its just that if you want it plug and play without having to add it in case you ever need to use the backup [09:23] not sure how rpi boots really, just speaking out of standard uefi [09:23] EriC^^: I've usually just edited the /boot/ or /boot/firmware/cmdline.txt and /etc/fstab to match the UUID's of the drive. [09:25] toffo: aha, if you want my advice, leave the disk as is, only delete the 2nd ext4 one, and recreate it in fdisk so it uses the whole space, and doesnt go beyond it [09:25] then mkfs a fresh ext4 using the original UUID of the fs from the source, and rsync that sucker over [09:25] everything should be pretty identical, partuuid,uuid, files etc [09:26] EriC^^: should I wipe what's in there first with fdisk or such ? [09:27] (and then dd the boot partition, and then create the primary writable with i.e. [g]parted ) [09:27] no [09:27] it wont help, it'll just create extra steps to do to make them identical [09:28] toffo: the only fault right now in the partition table is that its saying the 2nd partition ends at sector.... which is outside the disk, remove that partition and create it in a right way, all good [09:29] EriC^^: fdisk already thinks it's all unpartitioned space ... [09:30] toffo: type 'sudo fdisk /dev/sde' [09:31] EriC^^: yup, apparently gparted already worked some magic onnit :D [09:32] EriC^^: but it's okay, I mean that boot partition's size is trivial, I just dd that one alone === ace_me^^ is now known as ace_me [09:40] toffo: alright just make sure the disk identifiers stay the same [09:40] and uuid [09:41] EriC^^: apparently if I just try to dd the first partition, it expands to fill the entire drive and disk identifier stays at 0x00000.. etc. [09:42] guys, my Ubuntu 20.04.2 LTS decided last night that the bluetooth dongle doesn't exist. [09:42] any suggestions? I did google before I came here and asked. [09:43] EriC^^: partclone would support ext4 filesystems ... I really dread fdisk tbh. [09:43] eliyahutbr: can you see your dongle with 'lsusb' ? [09:44] nope. io see two linux foundation root hubs plus other hardware I recognize [09:44] io = I do [09:45] eliyahutbr: tried switching ports? tried the dongle on another computer? [09:45] there is no dongle. its part of the laptop. sorry i wasn't clear. I typed what the error message told me. [09:46] eliyahutbr: ah, well that changes things, a dongle is a usb device [09:46] happy to dump lshw if that would help [09:47] or grep it for something [09:47] eliyahutbr: or at least that's the common name for a usb stick.. so, yeah, it's a built-in bt [09:47] (with some help.) [09:47] eliyahutbr: did you run any apt updates before it stopped working ? (i.e. kernel updates , etc.) [09:48] very possibly. I also click update when asked but the GUI [09:48] how can I check that in my logs? [09:48] eliyahutbr: pastebin or termbin usually is ok for data dumps [09:48] eliyahutbr: The apt history is in /var/log/apt/history [09:49] eliyahutbr: run 'dmesg' (usually required as 'sudo') also to check for system error messages that might be related [09:49] https://paste.ubuntu.com/p/KGKMYtBQgJ/ [09:50] eliyahutbr: sorry, /var/log/apt <= that directory, and there should be the most recent history log as 'history.log' (older ones might be present as gzipped numerical "history.log.x.gz" files [09:51] https://paste.ubuntu.com/p/pXH2nMCVVw/ 'dmesg' [09:52] Upgrade: libopenexr24:amd64 (2.3.0-6ubuntu0.3, 2.3.0-6ubuntu0.5) [09:53] Hi, how to reduce the size of a .vdi on ubuntu, I have a .vdi using 70GB but the contents inside is ~30GB, how would I do this ? [09:54] i suspect Upgrade: libopenexr24:amd64 (2.3.0-6ubuntu0.3, 2.3.0-6ubuntu0.5) is the culprit. Can I 'undo' it? === skr- is now known as skr [09:55] eliyahutbr: it should just be the openexr runtime files for the EXR format [09:56] eliyahutbr: kernel updates would make more sense, or if you've recently installed anything wi-fi related for example [09:57] eliyahutbr: if it's a built in bluetooth enumerator in a laptop, it goes to one of those "eternal woes" category, even on windoze side [09:58] eliyahutbr: 'inxi | grep Kernel' says what ? [10:00] eliyahutbr: I doubt tho that the problem would be in the libopenexr, since that's not related to any wireless stuff [10:01] CPU: Quad Core Intel Core i5-1035G1 (-MT MCP-) speed/min/max: 896/400/3600 MHz Kernel: 5.8.0-48-generic x86_64 Up: 27m === GWM-Deus is now known as GWM [10:03] eliyahutbr: Actually, when I was running to similiar strange issues, i.e. some devices not getting recognized at times etc with one of my x86/64 desktops on 20.04.x LTS, I just finally backed everything up and upgraded the distribution to 20.10. In my case, that helped. [10:04] eliyahutbr: obviously, there's no telling if that might help and if you'd rather stick to a LTS Ubuntu, then obviously that's not an option. [10:04] i was actually thinking of doing a wipe and reinstall of LTS [10:05] but maybe an upgrade would fix it quicker [10:05] eliyahutbr: personally, things have worked out way better in all of my Ubuntu installs on the 20.10 in comparison to 20.04.x [10:05] so whats the command line process? [10:05] I keep everything backed up all the time [10:05] (well dailY) [10:08] what is the smallest ubuntu install I can do [10:08] ? [10:08] eliyahutbr: first, on your desktop, go to the 'Software Updater', then, once it's done checking for updates, "Settings" change the "Notify me of a new Ubuntu version:" to "For any version", close [10:08] eliyahutbr: refresh, and it should notify you of 20.10 being out [10:08] i just want the bare minimum, in terms of MB / GB? [10:12] Newb101: no window manager etc? ubuntu server probably [10:14] with a window manager, seems like the standard ubuntu comes in at 8GB? [10:14] anything below the 4 GB mark ? [10:14] the lower the better [10:16] Newb101: lubuntu or xubuntu are the lightest as for having a window manager [10:17] Newb101: https://cdimage.ubuntu.com/lubuntu/releases/20.10/release/ the ISO file is ~1.8GB [10:18] thanks [10:18] well toffo, i'm upgrading. i'll come back on the flip side and let you know what happened. [10:18] thanks [10:18] Newb101: lubuntu and xubuntu are very light weight since those are probably the two ones, in addition to Ubuntu MATE, that you'd want to run on i.e. a SBC (Raspberry Pi, etc..) [10:19] thanks [10:19] Newb101: Ubuntu MATE does expand to minimum 8GB as for the space requirements, though. [10:20] eliyahutbr: np and good luck. hope it gets sorted out! [10:20] on a seperate question, how would I migrate an anbox installation from ubnutu 20.04 stock to Lubuntu, which files would I need to copy across? [10:21] it's a snap [10:22] hi I am thinking about reducing 5% reserved blocks for 14TB drive which I use to store a data. AFAIK default is 5% I was thinking to lowering it to something small like 0.1 which is still 14 GB, although seems that could be too small for a large drive and I worry if fragmentation or fsck won't have enough space [10:32] how do I migrate a snap from one installation to another with all of it's configuation [10:32] ? [10:33] e.g. ubuntu stock 20.04 to Lubuntu 20.04 [10:45] I have this socket which I need to forward locally. However, the user for the remove need to run sudo or root to acces the socket. How do I do that? [10:46] e.g. `ssh -L /home/john/bird.sock:/var/run/bird.ctl john@10.129.2.2`, how do I access root on a ubuntu machine or at least run `sudo sus` before socket forwarding? === jelly-home is now known as jelly === skr- is now known as skr [11:19] I've written a script to output "top" to a file. See the script at https://termbin.com/153f When I run it, the processes, tracker-store and tracker-miner-f, show very high CPU usage. See the "top" output at https://termbin.com/usvs However, when I run the same "top" command, 'top -o "%CPU" -b -n 1 | grep tracker > tracker.txt' from the terminal, the processes' usage is low or zero. What could be the difference between these two executions that [11:19] could cause processes to behave differently? === OERIAS is now known as CristobalColon === skr- is now known as skr [12:03] 'Morning all [12:08] Morning!!! (Sorry, Jacked up on Caffeine, and the sun is out; happy vibes). === CristobalColon is now known as OEIRAS [12:40] I have a directory which I want 'shared' multiple times, but in some kind of layered way. I have a master copy, and I want something like a new directory where if the file exists in the master copy you have a ln -s, files that get overwritten need to no longer be a ln -s, but a new file [12:41] The closest thing I can imagine it being close to is something like a wordpress uploads folder where you want the production uploads folder on the dev/uat env. If someone uploads a file in the dev env, it shouldnt be added to the production upload. [12:41] I suppose a cp -r is the easiest with the tradeoffs being space and 'up to date-ness' [12:42] you wouldnt want to rsync it since it will overwrite modified files in the dev env [12:46] hard links rather than soft links sound like they'd provide what you discussed in the first part. [12:47] sorry, that made no sense. i mean: what you discussed in the first part sounds like it can be solved by hard links rather than soft links. [12:48] tomreyn, I think I'm just being difficult. I should just cp -r everything and if it needs updating, I rm -Rf and cp -r again [12:49] tomreyn, Iw as thinking of something like a docker system where you have a docker image. When you do a add/modify/delete a file to the container, it stays in the container. the next instance of the same image will still have the master copy. [12:50] I'm running Ubuntu 20.04. When I press the windows key and search for an app, activities appear on my main monitor, and the search box and results to my search appear on my second monitor. How do I make both stay on my first monitor? [12:51] ignorand: maybe an overlay filesystem would help, too. i think i'm not awake enough to understand the full picture. but if you have a simple solution, as you just discussed, you might as well work with that for now. [12:51] If I press the "show applications" button in the task bar, it works as expected, with everything on my main monitor [12:52] tomreyn, sorry, got disconnected [12:52] tomreyn, I was thinking of something like a docker system where you have a docker image. When you do a add/modify/delete a file to the container, it stays in the container. the next instance of the same image will still have the master copy. [12:53] ignorand: maybe an overlay filesystem would help, too. i think i'm not awake enough to understand the full picture. but if you have a simple solution, as you just discussed, you might as well work with that for now. [12:55] tomreyn, an overlay filesystem is what I would essentially want yes. === skr- is now known as skr [13:02] ignorand: https://www.kernel.org/doc/html/latest/filesystems/overlayfs.html https://wiki.archlinux.org/index.php/Overlay_filesystem === skr- is now known as skr [13:06] tomreyn, thanks [13:07] gonna try it [13:07] Hi! When I boot with the laptop lid down and a large HDMI screen plugged in, the login screen (lightdm I think) displays a very small resolution (1280x1024), which is less than the full "display" resolution of 1920x1080, so I only see a fraction of the login screen. Is there a way to set the default resolution for lightdm per screen? Or how is this resolution chosen? [13:08] I would like to set the default monitor resolution to 1920x1080 so that it matches the "display" resolution and I can see the whole thing on the screen, and it looks less stretched [13:09] (after I log in, the display settings kick in and I get my 3440x1440 resolution working on the display; but that doesn't happen on the login screen) [13:10] (lightdm I think) ---> depends on what version and flavour of Ubuntu you use cousteau_ [13:16] Hello! Is there a way to export default Ubuntu Calendar (3.38.1) events? Running Ubuntu 20.10. Thanks! [13:17] Syncing with online accounts is not an option. [13:19] probably using evolution, you could https://help.gnome.org/users/evolution/stable/exporting-data-calendar.html.en === skr- is now known as skr [13:22] pizzaburger: does not seem to be implemented for (gnome) calendar itself: https://gitlab.gnome.org/GNOME/gnome-calendar/-/issues/31 [13:23] tomreyn: alright, thanks! Gonna go look for a calendar which does have this function. Or back to ye old pen & paper one I guess [13:24] pizzaburger: also https://askubuntu.com/questions/1123603/how-to-export-a-calendar-from-gnome-calendar-to-a-file [13:26] tomreyn: awesome, thanks for the help! [13:28] you're welcome, pizzaburger [13:28] on ubuntu 20.04, how do I ensure that the dash only appears on my primary monitor? Currently, I'm seeing half of it on my primary monitor, and half on my secondary monitor. [13:29] I've already configured the settings, 'org.gnome.shell.extensions.dash-to-dock multi-monitor false' and 'org.gnome.shell.extensions.dash-to-dock preferred-monitor -1' [13:31] clarkk: Settings - Apperance - Dock. There's a "Show on" drop-down list to select which monitor you want it to appear on [13:34] Maik: well, it's Xubuntu 18.04 and I'm pretty sure it's lightdm [13:34] (or that's what Google suggests. lightdm is installed, that's for sure) [13:35] pizzaburger, that doesn't seem to do anything. Do I need to reboot for it to take affect? [13:37] cousteau_: yep, lightdm, and why xubuntu18.04? It goes EOL this month. [13:38] you mean, 16.04? [13:39] Ark74: no [13:40] all other Ubuntu LTS Flavours are only supported for 3 years [13:41] pizzaburger, forget that. thanks for your help [13:41] Maik: yikes, didn't it have 5 year support? [13:41] anyway, because I installed it in 2018 and I'm lazy with updates, that's why [13:42] 18.04 should be supported up to 2023: https://ubuntu.com/about/release-cycle [13:42] not xubuntu 18.04, though [13:43] But it will only receive Maintenance updates from this year forwards. [13:43] cousteau_: nope: "Supported LTS release: 18.04, Bionic Beaver The 18.04 release, codenamed Bionic Beaver, is a Long Term Support release and has support for 3 years, until April 2021." https://xubuntu.org/download/ [13:43] anyway, lightdm not being part of xubuntu specifically, I'd expect it to continue being supported. I'm not quite sure what it means for xubuntu support to end while the rest of ubuntu continues being supported, honestly [13:44] eliocamp: main ubuntu yes, other flavours not [13:45] Don't they receive basically the same maintenance updates that mainline ubuntu, though? Security patchs and all that stuff? [13:45] I was trying to understand the support periods, too. I believe it simply means, "base" packages are supported for 5 years, and "extended" packages are often supported for only 3 years. [13:45] anyway, how would I fix this? (I understand that if the easiest solution is "just upgrade to a newer version; that bug was fixed ages ago" then interest in fixing this will be low, but I don't think that's the case) [13:45] https://help.ubuntu.com/community/EOL#Ubuntu_18.04_Bionic_Beaver https://xubuntu.org/release/18-04/ "End of Life: April 29, 2021" [13:47] eliocamp, remline: the specific packages needed to make a flavor differ, are usually community supported, in the 'universe', not 'main' pocket, so different policies apply to them in terms of security updates. === skr- is now known as skr [13:48] Thanks, I didn't know that. I guess that'd be another reason to prefer main ubuntu over flavours. [13:49] apt policy SOMEPACKAGE lets you check the pocket a package belings to [13:49] tomreyn: As a consequence, a package like xfce4-terminal would be supported for only 3 years whether you installed Ubuntu, Xubuntu, or Kubuntu, correct? [13:49] eliocamp: in the past main Ubuntu LTS was also only 3 years supported, they changed it to 5 years. [13:50] and the interim releases went from 18 months to 9 months of support [13:50] The documentation at https://wiki.ubuntu.com/LightDM doesn't seem to apply to my version. It's funny how it says "oh btw, from versions 15.10 onwards this changes", and then goes on showing how it's done in 15.04 and earlier >:( [13:51] cousteau_: Unfortunately, the ubuntu wiki can be pretty out of date. You might need to reference other sources. [13:52] * cousteau_ wished this documentation was kept as much up to date as he is expected to keep his machine up to date [13:52] remline: it would receive community support during the full lifespan. though community support basically means self-support, unless the security team or the flavor operators get around to provide updates. which, more of then than not, they do. [13:52] remline: that's the problem; all sources I've found seem to apply to an older lightdm [13:53] remline: but when flavors state their supprt lifetime, they also state they'll provide support for packages during this lifetime. [13:53] they all refer to a /etc/lightdm/lightdm.conf I don't have, specifically to a line I can't find in any /etc/lightdm/* [13:56] cousteau_: I'd try to set the desired resolution through the xorg config rather than through the lightdm one though. [13:56] cousteau_: Which line are you referring to? [14:00] Good news is I found /usr/share/doc/lightdm/lightdm.conf.gz which clarifies the thing about [Seat:*] (specifically, (1) it still exists, and (2) yes, you're supposed to type a literal asterisk) [14:03] remline: the one that would say something about display-setup-script=... (that's what the tutorials I found told me to change. Not the Ubuntu wiki though; couldn't find that listed there) [14:04] kryten: any reason why? I'm having trouble figuring out what happened to my xorg.conf too (I'm using the Nvidia drivers, because there's no reason for things to be easy) [14:05] in any case, `xrandr --output HDMI-1-1 --preferred` seems to default to 3440x1440, so I have no idea why lightdm would choose 1280x1024 [14:08] hm, https://itectec.com/ubuntu/ubuntu-how-to-adjust-display-settings-for-lightdm-greeter-on-18-04/ seems to address my specific version (I googled for 1280x1024 to see if that was the default for some obscure reason, and somehow landed on that page) [14:09] (btw in case you're wondering, cousteau is on the X session where I can copy-paste stuff and cousteau_ is on irssi on a tty which is very limited but which I won't have to kill whenever I logout and back in) [14:15] is there a little lite UBUNTU for Raspberry pi Rpi 3, model B ? [14:16] cousteau_: Cause I feel like making the display manager run some script on its startup is kind of a hack.. [14:17] unixbsd: I know it typically uses something called Raspbian, which is Debian-based and not Ubuntu-based but it's pretty much the same. But I bet you can install Lubuntu, or Xubuntu, or maybe even Ubuntu on it [14:17] ah I look ubuntu + ufs2 type on kernel for the PI... [14:17] (ufs.ko). [14:18] unixbsd, https://discourse.ubuntu.com/t/how-to-install-ubuntu-server-on-your-raspberry-pi/14660 [14:18] kryten: so is modifying the global system-wide X config only because lightdm does whatever it wants while other environments work fine. Ideally the solution would be to make lightdm set the resolution directly, but there doesn't seem to be such an option [14:20] unixbsd: ...well, Lubuntu/Xubuntu/Ubuntu for ARM (or AArch64, not sure) [14:20] I am downloading. I am curious abiout this kernel inside [14:20] ok this trick didn't seem to work. Gonna reboot to see if I needed to reboot for it to work [14:21] unixbsd, should be something 5.4ish [14:21] I get this: https://cdimage.ubuntu.com/releases/20.04.2/release/ubuntu-20.04.2-preinstalled-server-arm64+raspi.img.xz [14:22] oh, by the way, I made a ubuntu-live with uefi/ and legacy, including virtualbox. It allows to backup / clone / install Windows 10. If you want, let me know. [14:24] unixbsd, if kernel age is your concern i'd go with the 20.10 one ... [14:24] (beyond that 20.04.2 should indeed be fine too) [14:24] ok, xzcat ... we will see it. [14:24] you can get the same kernel version in 20.04 LTS [14:26] yay it worked! Turns out I needed to reboot AND add `display-setup-script=` before I wrote the command (I forgot to add that part to the line, whoops) [14:27] but yeah, creating a /etc/lightdm/lightdm.conf.d/99-hdmi-resolution.conf file containing [Seat:*] display-setup-script=xrandr --output HDMI-1-1 --mode 1920x1080 did the trick. [14:27] (hacky, I know, but hey, at least it worked) [14:27] Just restarting the lightdm service would have been enough though. [14:28] And yeah, I think it's sufficiently proper for what you got. === skr- is now known as skr [14:33] good it is a kernel 5.4.x [14:37] wow [14:37] ufs.ko is in there!! [14:37] that means that I can use it to copy my freebsd a5 harddisk to use it !!! [14:40] I made a copy of the kernel, I can install it with debootstrap methode. I prefer to install by hand ubuntu: https://gitlab.com/openbsd98324/ubuntu-raspberry/-/tree/master/releases/v1 [14:40] does anyone have an idea why ubuntu keeps messing external hd drives formatted as NTFS? [14:41] it is so g-d annoying [14:41] even floppy disks are more reliable [14:41] it is likely fuse and autofs, including ntfs-3g of fuse ntfs [14:41] you can delete purge autofs processeses... [14:41] OEIRAS: do you also have Windows 10 on this machine? [14:42] should I remove ntfs-3g and reinstall it? [14:42] yes I have Windows 10 [14:42] but that is not relevant [14:42] OEIRAS: https://help.uaudio.com/hc/en-us/articles/213195423-How-To-Disable-Fast-Startup-in-Windows-10#:~:text=To%20disable%20Fast%20Startup%2C%20press,the%20column%20on%20the%20left. [14:42] I have not booted windows on this machine in over a week [14:42] OEIRAS: define "messing up" [14:42] I just recently formatted this HD [14:43] leftyfb, corrupting file system to the point that is it not recognisable [14:43] OEIRAS: did you disable fast boot on Windows? [14:43]