[01:12] ^ [01:28] How do I find my IP address on my ethernet connected 20.04 machine? I have followed all kinds of instructions from Google fails or is unbelievably complicated so I won't bother trying it cuz I know it's a simple thing. [01:28] ip a [01:29] Thanks sarnold. [01:29] there's also hostname -I -- I don't know how it works, but it is much more readable and concise :) [01:30] I get a lot of output but I don't know what it means in which numbers to pick out of the many numbers that I get [01:30] I'm still on the same thing of trying to find out what happens just before my computer freezes [01:31] You were helping me with that in the past. (I used Firefox and it held okay for a week. But then when I went to look after it froze I had rebooted my phone and lost the terminal response. Terminal on my phone) [01:32] There's no tab keys or up arrow keys on my phone so I don't know how to ask terminal what my previous commands were [01:32] 'history' may do it [01:32] history | more or history | less are also good [01:34] Ah ha sarnold i am going to install "hackers keyboard" [01:34] Ok will try [01:35] when my phone was rebooted The history didn't persist [01:35] bummer [01:35] you can configure bash to write history more frequently, but perhaps it comes configured that way to avoid writing to the flash too much [01:36] Could this work 127.0.0 1/8 [01:37] thinkofher7e rest are all inet6 [01:37] Gotta catch a bus.. away for a bit. [01:40] tripelb: 127.0.0.1/8 is the localhost 'loopback' address range; every linux system will have those, and they all mean the local system [01:40] tripelb2: if you're trying to cross a network, those won't work :) === zbenjamin_ is now known as zbenjamin [01:48] Hi [01:48] How do you convert .fsb audio file to something else? What in ubuntu reads fsb audio file? [01:52] Hash: http://manpages.ubuntu.com/manpages/focal/man1/gh_fsb_decrypt.1.html [01:52] https://github.com/gdawg/fsbext ok I compiled this. I can get a wav out now. [02:02] How can I fix: clang : Depends: clang-10 (>= 10~) but it is not going to be installed. I'm on ubuntu focal and I have run apt-get update [02:03] greg: apt-cache policy clang | nc termbin.com 9999 [02:04] leftyfb, im running this on travis-ci so will take sec [02:06] leftyfb: is that apt-cache policy clang or apt-cache policy clang-10 ? [02:07] leftyfb, https://termbin.com/8k8j [02:07] oh right, that would make more sense [02:07] greg: apt-cache policy clang-10 | nc termbin.com 9999 [02:09] ok gimme minute again [02:10] leftyfb, https://termbin.com/1m3q [02:17] greg: this looks like it should have worked. maybe these will explain why it didn't: [02:17] greg: grep -hEv '^([ ]*#.*)?$' /etc/apt/sources.list{,.d/*.list} | nc termbin.com 9999; apt-cache policy | nc termbin.com 9999 ; cat /etc/apt/preferences{,.d/*} | nc termbin.com 9999 [02:21] tomreyn, https://termbin.com/bnyz https://termbin.com/j4n7 cat: /etc/apt/preferences: No such file or directory [02:21] cat: '/etc/apt/preferences.d/*': No such file or directory [02:22] it's a *guess* but that bionic mongo repo stands out to me [02:23] mixing and matching releases tends to lead to funny things like this [02:23] i don't add any repos this is travis-ci's image [02:24] if theres some command to default the apt file i can run that [02:25] oh weird, I'm also just now noticing that POWER repo is ... amd64. hah. [02:26] greg: I really only know how to solve these things interactively -- apt install clang-10 and see what errors apt throws back; usually apt install clang-10 ... iterate that process until you get to the *cause* of the problem [02:27] sarnold, well this is in a container that runs a script fresh. I can't really interact with it on the fly [02:28] I can put whatever in the script though [02:28] greg: yeah.. that's why I was really hoping leftyfb or tomreyn would have magic insights :) hehe [02:30] not really, no. there are some Debug:: options you can pass to apt, which can maybe help [02:30] see apt.conf(5) [02:30] apt-config dump may also help [02:30] in the past, travis-ci's containers had been modified a lot, i assume this may still be so. [02:31] but things seems to have changed quite a bit since i last read the past equivalent to https://docs.travis-ci.com/user/reference/overview/ === sauvin_ is now known as Sauvin [02:34] Sarnold hostname -I .. will try this when I get home again. [02:34] tripelb: I think you'll prefer it :) [02:35] tomreyn, is the defult apt.conf somewhere i can just overwrite it with? [02:35] greg: i wouldn't recommend this. rather get support from travis for this system [02:41] greg: maybe give this a try: apt-get -y -o Debug::pkgProblemResolver=yes install clang &> /tmp/aptlog; cat /tmp/aptlog | nc termbin.com 9999 [02:42] (probably clang-10) [02:44] the original message was "clang : Depends: clang-10 (>= 10~) but it is not going to be installed.", which was probably (we weren't told) the result of running "apt install clang". [02:44] hrmmn i didnt get a link from that cmd [02:45] this means the resolver did consider installing clang-10, but chose not to. the above command should tell us why it chose not to install it. [02:45] greg: you'll need sudo for apt-get [02:46] oh ok [02:48] Hello everyone! Anyone used network in initramfs? then please share any doc [02:50] tomreyn, im still not seeing anything from that cmd [02:52] greg: hmm, maybe your /tmp is not writable. the problem there is that you will likely need to write the output to a file first, because apt-get will likely run longer than the one or two seconds the termbin.com server is happy to wait for input. [02:52] you can try pastebinit instead [02:52] I know I can write to /tmp [02:53] sudo apt-get -y -o Debug::pkgProblemResolver=yes install clang 2>&1 | pastebinit [02:53] but pastebinit may not be installed, yet [02:54] another option: sudo apt-get -y -o Debug::pkgProblemResolver=yes install clang 2>&1 | curl -F 'f:1=<-' ix.io [02:58] tomreyn, http://ix.io/3lsk [03:01] does apt-mark showhold output anything? wehat about apt list --installed | grep ',local\]$' [03:01] i think the above output describes whatthe problem is, but i'm too tired to interpret it properly [03:02] i dont need sudo? [03:02] not for these [03:05] the output looks to me like you have broken packages already before starting to install. sudo apt-get -f -y -o Debug::pkgProblemResolver=yes install may fix those, or discuss what the problem is. [03:05] tomreyn, https://travis-ci.org/github/enigma-dev/enigma-dev/jobs/769100424 [03:12] greg: what do i see there exactly? [03:12] tomreyn, well i run your cmd after update [03:12] soo w/e u looking for idk [03:13] apparently lines 271 to 291 are output of apt list --installed | grep ',local\]$' [03:13] ~line 270 [03:14] apparently, this libc6 package version doesn't exist on mirror at this time. [03:14] 2.31-0ubuntu9.3 is in -proposed [03:14] it existed last week [03:16] that's the same problem jochensp reported here and later in #ubuntu-devel yesterday [03:16] is there a bug report? [03:17] bug 1912652 [03:17] bug 1912652 in ubuntu-release-upgrader (Ubuntu Hirsute) "Upgrading libc6-lse breaks on systems it is in use" [Undecided,Confirmed] https://launchpad.net/bugs/1912652 [03:18] I am trying to configure network in initramfs. I got some error. logs ----> https://paste.ubuntu.com/p/z4J4hSdnyV/ [03:19] tomreyn, guess i can just wait then [03:20] greg: you could also try enabling -proposed with low priority [03:22] not sure what that means [03:22] but sounds like itll cause ma another issue in 2weeks :P [03:27] greg: untested, but i guess it can work: echo -e "Package: *\nPin: release a=*-proposed\nPin-Priority: 400" | sudo tee /etc/apt/preferences.d/proposed >/dev/null; sudo add-apt-repository -y -u 'deb http://archive.ubuntu.com/ubuntu focal-proposed main universe multiverse'; [03:28] more info: https://wiki.ubuntu.com/Testing/EnableProposed [03:30] i'm assuming you'll start each build with a "fresh" build VM, so activating focal-proposed now should not be something you'll actually need to revert on the system later (i.e. you'd just need to remove the above oneliner off your travis-ci.yml) [03:34] tomreyn, well that line didnt seem to help [03:36] maybe this would in combo with the above line sudo apt install -t focal-proposed clang-10 [03:36] but this bug should never have crept into the latest LTS release in the first place. :-( [03:36] bugs happen [03:37] lot of moving parts [03:37] tomreyn, sudo apt-get -y install clang lldb lld libc++abi-dev libc++-dev do i replace thar? [03:37] or just put yours above [03:38] so that's the command you were running before you asked for help here? [03:38] yes [03:38] ah, good that you're actually telling us [03:38] maybe try making is less complex for starts, this can help with debugging [03:39] remove everything but clang [03:40] then install the others one by one, and see which one actually causes the problems [03:40] if clang still won't install, add -t focal-proposed [03:41] i should catch some sleep, good luck! [03:41] ya it still failed [03:41] ill try -t [03:46] tomreyn, that seemed to do it it. thanks for all the help [03:46] gnight === UniCode is now known as NoCode === popey6 is now known as popey === guillaumedsde018 is now known as guillaumedsde01 === antoine8 is now known as antoine === jose- is now known as jose === Perk7 is now known as Perk [06:48] Good morning [06:48] mornin' [07:09] mornin cheetos [07:54] <[VMGuy23]> Mornong [07:55] <[VMGuy23]> *morning [07:57] Hi, can someone guide me in the right direction to resolve this error? [07:57] Enter password: [07:57] ERROR 2013 (HY000): Lost connection to MySQL server at 'reading initial communication packet', system error: 104 [07:58] When trying to " sudo mysql -u mysql -p " [08:14] <[VMGuy23]> It looks like the server is not running or is unreachable [08:14] <[VMGuy23]> Or you haven't selected a server [10:11] i am sitting at an ubuntu 20 server login screen tty1 being prompted for login, i type my username, hit enter, it immediately re-prompts for login. I am able to SSH in no problem, but not at the terminal. This is running in vmware workstation. I did notice a .list file was corrupted earlier during package updates. Suspect corruption. fsck comes back OK. Any thoughts? Even tried different [10:11] keyboard. [10:13] baconology: check "journalctl -n 100 -u systemd-logind" (from ssh of course) for clues. Also look at the end of "/var/log/auth.log" [10:16] starting login service, new seat seat0, watching system buttons, new session 1 of user baconology [10:16] i assume that's the SSH [10:16] i dont even get prompted for password at the terminal, immediate reprompt for login, doesn't appear as if there are entries generated in journal [10:16] thanks for the tip though [10:17] sudoers and sshd. should I use the .d folder to preserve setting during updates? [10:17] ubuntu 20.04 server [10:20] seeing entries in multipathd (sda: add missing path, failed to get udev uid, failed to get sysfs uid, failed to get sgio uid (invalid argument) [10:21] (in journal) [10:21] baconology: hmmm, I wonder if you've made changes to the PAM config (/etc/pam.d/) [10:21] oh my, login: segfault at 8 ip ________ error 4 in ld-2.31.so [10:21] baconology: those may be something else but will also need resolving or ignoring, but I don't think they're related to a login attempt [10:22] baconology: AHA! [10:22] the poor kernel [10:22] im leaning towards corruption of my vm filesystem given the earlier .list file corruption today [10:23] funny that i can login via ssh [10:24] <[VMGuy23]> I have a Wayland issue: On 21.04, the Ubuntu Dock takes a while to update whats on there or not. It's on time when something opens or closes but takes about 30 seconds to show if an opened app is not pinned, and the icon stays for 30 seconds after closing. The dot indicators are on time. === xamua is now known as xa-mua [10:48] baconology: because sshd does the login, not logind [10:54] got it [10:54] gonna wipe and reinstall [10:55] i blame vmware or windows updates random reboots [11:02] i am running gnome and accessing it over VNC. how come I can't create desktop shortcuts? [11:02] i have an 'Activities' button on the upper left that allows me to open applictions; why can't i put those on my desktop [11:07] i am using tightvnc with gnome, how can I create desktop shortcuts? [11:13] how can i set my system to automatically disconnect from the internet within the context of innactivity [11:13] ? [11:14] shimbles: I haven't used Gnome in over a decade, but as I understand it, there was a deliberate decision to remove the ability to have launchers on the 'desktop' (which used to be provided by the file-manager but the code was brittle apparently) [11:14] tmm88: how does it connect to the Internet? via a local WiFi service, or directly via a cellular modem? [11:19] TJ- i was able to fix it with gnome-tweaks [11:20] im not sure why gnome went full retard [11:23] as i understood it, the point of gnome was to offer a finer (less advanced) selection of options; not to remove options that you definitely need [11:26] TJ-: via wifi [11:28] shimbles: removing 'stuff' has been the general theme of Gnome for a long time, according to its disgruntled users at least [11:29] tmm88: so do you want to retain the Wifi connection to other local devices but prevent routing out to the Internet, or simply disconnect wifi? [11:40] Ubuntu Server 16.04 updated to Ubuntu Server 18.04 without issues for first time :) [11:42] !yay | mrkubax10 [11:42] mrkubax10: Glad you made it! :-) [11:43] mrkubax10: checkout the !18.04 releasenotes anyway for known transitions on -server [12:13] So, I'm installing Lubuntu on this laptop which doesn't have an ether interface, and the wifi drivers doesn't get automatically recognized during install. I need to offline install them, I tried a lot of different articles on the internet. [12:13] In the end most of the articles seem to be very outdated, and I run into dependency problems. The closest I got was having fwcutter trying to connect to an online service, probably to download some blobs, but since I do not have another connection, this is a no go. [12:15] usb ethernet card? [12:16] baconology: I don't have any, and buying one is not a reasonable option in the short term. [12:19] CyL: do you have a "smart hpone"? [12:19] *phone [12:20] tomreyn: Yes I do, I mean, I guess an Android phone count as that [12:20] right, connect it to the computer using a usb cable [12:20] then enable tethering on the android device. [12:21] Hmmm, are all the default drivers for tethering installed on a fresh install? [12:21] yes [12:21] I'm reinstalling Lubuntu right now, I messed up so much with it yesterday that I'm probably in a very inconsistent state with apt [12:23] I appreciate a lot you advice, but still it's hard to grasp Ubuntu developer's wouldn't have provided a more reasonable solution to this kind of chicken and egg problem. [12:28] CyL: there's apt-offline also. but i think the solution i provided is the most simple one, other than acquiring hardware which is well supported out of the box. [12:29] tomreyn: I did try apt-offline, but it has depency problems, and sorting that out on a fresh install is hard. Also, I'm not sure why, but many, many packages failed to download yesterday [12:29] CyL: running lspci -nn | grep Network should return some [0123:4567] like output to the end of the lines. this identifies your network device chipsets, and should help identifying the needed driver. [12:29] TBH it's because of apt-offline I'm a little scared of my current apt state [12:30] personally, i'd always prefer the smartphone / tab / other computer / usb-ethernet/wireless dongle approach [12:33] Hi folks [12:33] oh no not BluesKaj [12:33] tomreyn: yeah, it is simpler and more effective where available [12:33] ;) [12:34] greetings summonner :-) [12:36] CyL, have you looked into the "proprietary drivers" window? is the hardware listed there? [12:36] summonner: No I have not, I was used to Ubuntu prompting me to install proprietary drivers [12:37] cheers [12:39] summonner: can I have Lubuntu live session load proprietary drivers before proceeding with installations? This would be a life saver [12:40] can anybody recommend a good laptop to run ubuntu on? [12:41] CyL, that would be the same as doing it on a standard installation, the diff b/n live and install is you have it on a disk and not an unpacked ram drive and "mounted iso" (not 100% correct) [12:42] hello [12:43] ubuntu starts but fails to start, [ok] reached target Timers. it is frozen. what can it be? it is ubuntu groovy [12:43] shailangsa: one with a cpu, ram and a storage device. for more concrete recomendations, please /join #ubuntu-discuss and provide more details. [12:43] jason234: did it work previously? [12:44] never [12:44] jason234: so this is a fresh 20.10 installation? [12:45] CyL, install the driver, then follow the instructions here to load it https://help.ubuntu.com/community/Loadable_Modules [12:46] CyL, that way you can test it, ... but I thought you already had it working? (from reading the scrollback?) is the card supported? [12:54] Is there a way to prevent systemd from unmounting a filesystem before stopping a service? [12:59] Forza, if you look into the systemd configs, you can see the answers. you're looking for things similar to "start before" "stop after" [14:05] Why doesn't the hexchat and notification plugin go into the notifications area? Instead it is outside of the theme's shaded area where it would look better. [14:07] NoCode: did you try hexchat-indicator [14:09] Pretty sure that's what I use. [14:11] NoCode: i didnt try myself yet, but you could try installing the indicator, and disable gnomes hexchat notify, then set highlights in hexchat on your nick, see what it does [14:33] hello, which Network manager version has ubuntu 16.04? [14:34] !info network-manager focal [14:34] network-manager (source: network-manager): network management framework (daemon and userspace tools). In component main, is optional. Version 1.22.10-1ubuntu2.2 (focal), package size 1811 kB, installed size 7472 kB [14:34] DJAnonimo: ^ [14:34] !info network-manager xenial [14:34] network-manager (source: network-manager): network management framework (daemon and userspace tools). In component main, is optional. Version 1.2.6-0ubuntu0.16.04.3 (xenial), package size 1911 kB, installed size 9592 kB (Only available for linux-any) [14:35] DJAnonimo: my bad it's actually the one right above ^ , thanks lotuspsychje [14:35] wow. nice bot. thank you [15:13] Hi [15:13] Hi, I giot issues on upgrading bionic to focal. [15:13] udevd doesn't upgrade and no longer works. [15:14] How can I disable colors in terminal? I'm using Mate. export TERM=xterm-mono doesn't work. [15:14] I hate colors in my shells [15:14] systemd-udevd.service: Cannot add dependency job, ignoring: Unit systemd-udevd.service is not loaded properly: Exec format error [15:14] /lib/systemd/system/systemd-udevd.service:26: Executable path is not absolute: udevadm control --reload --timeout 0 [15:17] When I change ExecStart=udevadm to ExecStart=/bin/udevadm then it changes back instantly to ExecStart=udevadm in [15:17] /lib/systemd/system/systemd-udevd.service:26 [15:17] Where should I fix this? Unfortunately I'm not very familiar with systemd. [15:37] floogy, the exec line for that unit file is : ExecStart=/lib/systemd/systemd-udevd [15:50] floogy: firstly, how are you making the change? using "systemctl edit systemd-udevd.service" ? [15:52] Hi, has anybody heard of any tool that can change the pwm frequency of the lcd's backlight in linux? [15:52] edits to systemd unit files using systemctl create override files at /etc/systemd/system/.d/override.conf [16:00] naftilos76, intel ? [16:01] ioria: ryzen 5 2400 [16:01] ah [16:22] ubuntu rocks! <3 === somethingspicey is now known as somethingspicy === LabMonkey is now known as Mechanismus [16:27] AAAAAAAABBBBBBBCCCCCDDDDDDDD [16:30] Hello! I have this laptop: https://www.amazon.co.uk/Lenovo-IdeaPad-Creator-15-6-Laptop/dp/B08BJ8TFPQ (No advertising intended, just showing the full specifications list) and I cannot make the brightness control to work. I am using Ubuntu 20.04. [16:30] Can anyone help me please? [16:41] @catalin I guess you already tried the Fn key in combination with F8/ F9 to change the brightness? [16:43] you did not succeed at responding with the 3 minutes catalin persisted [16:43] *within [16:45] @tomreyn we are all volunteers and users here, not a professional helpdesk as far as I know. If he pays me I will respond sooner [16:46] pointless to respond when they left VvdBergh [16:46] VvdBergh: yes, sorry, i didn't mean to blame you, just to point out that catalin's gone. [16:49] I know. I am actually surprised people actually think they can get an answer here within 1 minute after asking a question [16:50] I leave irc open in the background so I can look back and see if there was an answer or I look back in the logs as this channel is logged [17:07] Noob mistake: I have auto-login when I boot up my computer, but I forgot the actual password, so I can't change it or do anything with sudo. What are my options for changing/retrieving the password? How does the autologin do it? [17:14] @jfcaron try this https://itsfoss.com/how-to-hack-ubuntu-password/ [17:19] <[VMGuy23]> Hello, why does the dock take a long time to update which icons are visible? The open indicator is fine, it's just that closed apps' icons stay for 30 seconds and icons take 30 seconds to appear. [17:19] <[VMGuy23]> Only on Wayland™ [17:20] you gotta love wayland [17:21] Loving Wayland is Mandatory [17:21] <[VMGuy23]> 21.04 [17:22] THanks VvdBergh, I ended up being able to guess the password after all. [17:22] Just took about 10 tries [17:23] [VMGuy23]: what kind of specs do you encounter those slow animations? [17:23] First thing to do now is disable autologin @jfcaron I have it off for just this reason (besides the fact I work in an office where my co workers tend to do pranks with each others computers) === trohrt is now known as oginome === oginome is now known as trohrt [17:24] <[VMGuy23]> its not slow animations, its the dock icons taking time to appear/disappear [17:26] [VMGuy23]: ok, and your specs? [17:28] <[VMGuy23]> AMD Ryzen 3 3200U with Radeon Vega Mobile GFX [17:28] <[VMGuy23]> AMD® Radeon(tm) vega 3 graphics [17:29] <[VMGuy23]> Copied from About menu in settings [17:29] [VMGuy23]: lets have a look inside your dmesg please? maybe volunteers find something useful [17:29] ioria, TJ, thank you a lot! I think this will be helpfull, because I noticed those ... .d folders under /lib/systemd/system/ e.g. like user@.service.d/timeout.conf and already wondering, if that would also override those wrong entries! [17:30] <[VMGuy23]> lotuspsychje: All of it? I can send easily [17:30] ~$ for service in {systemd-udevd.service,systemd-hwdb-update.service,systemd-udev-trigger.service,systemd-udev-settle.service}; do cat -n /lib/systemd/system/"$service" | egrep '26|25|22|27' | egrep -i 'exec' ; done [17:30] 25 ExecStart=/lib/systemd/systemd-udevd [17:30] 26 ExecReload=udevadm control --reload --timeout 0 [17:30] 25 ExecStart=systemd-hwdb update [17:30] 22 ExecStart=udevadm trigger --type=subsystems --action=add [17:30] 27 ExecStart=udevadm settle [17:31] <[VMGuy23]> lotuspsychje: https://termbin.com/i29q [17:32] [VMGuy23]: your secureboot is enabled, did you try comparing yet with secureboot off? [17:32] So I haveto edit those 4 entries by `systemctl edit ` instead of using vi on the physical file, to create those overrides under the .d folders. [17:33] <[VMGuy23]> I don't think it's secure boot related, and I remember secure boot being off... [17:33] <[VMGuy23]> But I'll go check now [17:33] <[VMGuy23]> Just restarting [17:33] !paste | floogy [17:33] floogy: For posting multi-line texts into the channel, please use https://paste.ubuntu.com | To post !screenshots use https://imgur.com/ !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. [17:34] Ok, I thought limiting on six lines might be sufficiant, sorry [17:35] <[VMGuy23]> wait how did secure boot disabling fix the issue [17:35] <[VMGuy23]> it seemed unrelated [17:36] <[VMGuy23]> nevermind it didnt fix the issue [17:36] [VMGuy23]: can we see a sudo lshw -C video plz [17:37] <[VMGuy23]> Or at least, it seems to have slightly reduced the issue [17:38] TJ-, ioria, after ExecStart appears the line ExecReload=udevadm control --reload --timeout 0 [17:38] <[VMGuy23]> https://termbin.com/2c5ft [17:38] hey I was on kubuntu and was asking to upgrade. Mid upgrade before installation was happening, someone told me to stop because of a major bug. So I did, and now my sources are now all stuck on hirsute. [17:39] How do I make my sources go back to groovy? [17:39] <[VMGuy23]> Acts10-13: Check your EFI version. If it's 2.0 or higher then you should be fine installing Hirsute. [17:40] Acts10-13, check your /etc/apt/ directory; you should have a sources.list.save or something in there [17:40] [VMGuy23]: okay [17:40] [VMGuy23]: that looks good, driver loaded [17:40] <[VMGuy23]> The bug is for EFI 1.20 or something [17:40] floogy: well --reload makes sense [17:40] <[VMGuy23]> lotuspsychje: yeah I reinstalled 21.04 just to get those [17:41] [VMGuy23]: did you test other ubuntu releases on wayland to compare that? [17:41] <[VMGuy23]> Did not happen with 20.10 on Wayland iirc [17:43] [VMGuy23]: so with secureboot off, how many secs do they appear/dissapear now? [17:44] <[VMGuy23]> After a few minutes, instantly [17:45] [VMGuy23]: so you are saying, wayland is kinda stuttering until desktop has loaded and then starts working better? [17:46] <[VMGuy23]> No, perfectly smooth, they just don't appear for a bit. They then appear/disappear normally after that time has passed [17:48] [VMGuy23]: check my bug #1922437 [17:48] bug 1922437 in gnome-shell (Ubuntu) "login into desktop is stuttering on wayland" [Low,Incomplete] https://launchpad.net/bugs/1922437 [17:49] TJ, I created by edit --force a override.conf file, but it throws still the same error [17:49] im getting also a waiting time before able to work properly [17:49] <[VMGuy23]> lotuspsychje [17:49] <[VMGuy23]> mine different [17:51] [VMGuy23]: you got a spinner or ssd? [17:51] TJ-, ~$ cat /etc/systemd/system/systemd-udevd.service.d/override.conf [17:51] ExecReload=/bin/udevadm control --reload --timeout 0 [17:51] <[VMGuy23]> SSD [17:51] <[VMGuy23]> 512GB NVMe SSD with 16GB dual-bank ram [17:52] <[VMGuy23]> i think [17:52] <[VMGuy23]> i mean, 2 8gb sticks [17:52] floogy: what's the error reported? [17:53] TJ-, the same as before. /lib/systemd/system/systemd-udevd.service:26: Executable path is not absolute: udevadm control --reload --timeout 0 [17:53] So the override is not in action [17:53] floogy: which is correct! you've not given a path to udevadm. should be /usr/bin/udevadm [17:54] ~$ which udevadm [17:54] /bin/udevadm [17:54] [VMGuy23]: ok well, maybe you should file a new bug for this [17:54] floogy: units do not use a shell so there is no PATH search, you need to specify the full path to an executable always [17:54] floogy: /bin/ is a symbolic link to /usr/bin/, see "ls -ld /bin" [17:55] TJ-, here ist a directory [17:55] ot's [17:55] [VMGuy23]: or check the existing ones first; https://launchpad.net/ubuntu/+source/gnome-shell-extension-ubuntu-dock/+bugs?orderby=importance&start=0 [17:56] ~$ ls -ld /bin [17:56] drwxr-xr-x 2 root root 12288 Mai 1 19:46 /bin [17:56] floogy: ahh, you said earlier this is a 16.04 > 18.04 upgrade [17:56] floogy: in new installs /bin/ became a symlink to /usr/bin/ [17:56] No, 18.04 to 20.04, butt began with hoary [17:57] ;) [17:57] floogy: ha! I'm on 20.04 but had forgotten its codename [17:57] focal [17:57] TJ-, ~$ cat /etc/systemd/system/systemd-udevd.service.d/override.conf [17:57] ExecReload=/bin/udevadm control --reload --timeout 0 [17:57] floogy: yes [17:58] But it's not active. Ithrows still the former error [17:58] floogy: did you do "systemctl daemon-reload" ? [17:58] No, thank you. As I said I don't have any clue on systemd [17:58] floogy: that ought not to be needed when you use "systemctl edit ..." but maybe [18:00] <[VMGuy23]> lotuspsychje: found this: https://bugs.launchpad.net/ubuntu/+source/gnome-shell-extension-ubuntu-dock/+bug/1890891 [18:00] Launchpad bug 1890891 in gnome-shell-extension-ubuntu-dock (Ubuntu) "Dock icons don't appear or appear delayed if you have none pinned" [Low,Triaged] [18:01] TJ-, unfortunately that doesn't work. [18:01] floogy: mmm, can you show us "pastebinit <( systemctl cat systemd-udevd.service )" [18:02] still the same error. /etc/systemd/system/systemd-udevd.service.d/override.conf still not noticed by systemd [18:03] TJ-, it's empty [18:03] No files found for systemd-udevd.service [18:04] TJ-, -rw-r--r-- 1 root root 55 Mai 1 19:50 /etc/systemd/system/systemd-udevd.service.d/override.conf: ExecReload=/bin/udevadm control --reload --timeout 0 [18:05] floogy: if there were no files found that suggests the main .service file is missing or broken completely [18:05] Or does it need udev to read it in? [18:05] Okay, yes it's disappeared [18:06] floogy: what does this report? "pastebinit <( find /lib/systemd/system/ /etc/systemd/system /run/systemd/system -name 'systemd-udev*' -ls ") [18:06] grrr typo [18:06] floogy: what does this report? "pastebinit <( find /lib/systemd/system/ /etc/systemd/system /run/systemd/system -name 'systemd-udev*' -ls ) " [18:07] https://paste.ubuntu.com/p/dSTNJmwBQt/ [18:07] Your command: http://paste.ubuntu.com/p/Wrx3MtMqwR/ [18:09] Ouch. it was /etc instead of /lib [18:10] floogy: thanks. one more: " pastebinit <( dpkg --print-architecture; which -a udevadm; file $(which udevadm) ) " [18:10] floogy: overrides are ALWAYS in /etc/systemd/.... [18:10] okay [18:10] http://paste.ubuntu.com/p/n3PKy7DNvZ/ [18:10] floogy: files in /lib/systemd/system/ are installed by the packages [18:11] floogy: let's see what is being reported: " pastebinit <( journalctl -n 50 -u systemd-udevd.service ) " [18:12] http://paste.ubuntu.com/p/ndnSGmbQ8r/ [18:14] ~$ ls -l /etc/systemd/system/systemd-udevd.servic* GIVES -rw-r--r-- 1 root root 55 Mai 1 19:50 override.conf [18:15] floogy: I think "Cannot add dependency job, ignoring: Unit systemd-udevd.service is not loaded properly: Exec format error" may relate to a stray control code in the .service file, not something to do with the ExecXXXX= lines [18:15] Maybe, uit's the ol18.04 running kernel? [18:16] https://paste.ubuntu.com/p/dSTNJmwBQt/ [18:16] floogy: I have to go to dinner now, but I wonder what program you used (before systemctl edit) to alter those files? I think it's inserted an unexpected, possibly invisible, control character [18:16] vi [18:17] But it has reverted all changes instantly [18:17] floogy: hmmm [18:17] okay have a nice dinner. Thank you [18:17] let me look at the systemd source-code, see why it reports that error [18:18] The last Paste is the content [18:21] TJ-, Maybe this is important: https://paste.ubuntu.com/p/M3GfqwdrsW/ [18:23] floogy: no, that is expected and normal [18:23] floogy: can you manually start the service ? [18:27] floogy: collect some debug logging from systemd itself so we can see what it is doing when reloading the unit. "sudo kill -SIGRTMIN+22 1" (that enables log_level=debug). Then reload: "sudo systemctl daemon-reload", then relax log_level "sudo kill -SIGRTMIN+23 1" and finally, check the log: "journalctl -b 0 -n 1000" to get last 1000 reports [18:28] floogy: if you're lucky, somewhere in that log, will be a clue to some syntax error in some unit file, hopefully the systemd-udevd.service but could be a related file [18:44] TJ-, https://paste.ubuntu.com/p/Zd3Kscx4XN/ [18:52] floogy: back. so, all the jobs you mentioned earlier related to udevd are reporting Exec format errors. That comes from the kernel, so I'm suspecting you've got some 32-bit executables mixed up in the 64-bit install if it isn't a control code issue in the .service files themselves [18:53] floogy: what does this report? "systemd --version" [18:54] That might be. How can I figzure out which packages are in the wrong format? [18:54] floogy: as I'm on 20.04 amd64 too, and not seeing the errors you're seeing, it seems pretty clear something on your system didn't get upgraded correctly, or some change you made manually is causing this [18:55] As I said before(?) I upgraded this installation from hoary, and also upgraded it from 32bit to amd64 with a helper script. Okay. That's not the standard way. I didn't thought that kind of exec format could be meant. [18:55] apt-show-versions might help out, then [18:56] Also this happened at ~ 10% upgrade procedure to 20.04 [18:57] systemd 237 [18:57] +PAM +AUDIT +SELINUX +IMA +APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD -IDN2 +IDN -PCRE2 default-hierarchy=hybrid [18:59] floogy: AHA! that isn't 20.04. 20.04 has "systemd 245 (245.4-4ubuntu3.6)" [18:59] !info systemd 18.04 [18:59] '18.04' is not a valid distribution: bionic, bionic-backports, bionic-proposed, cosmic, cosmic-backports, cosmic-proposed, disco, disco-backports, disco-proposed, eoan, eoan-backports, eoan-proposed, focal, focal-backports, focal-proposed, groovy, groovy-backports, groovy-proposed, hirsute, hirsute-backports, hirsute-proposed, kubuntu-backports, kubuntu-experimental, kubuntu-updates, partner, precise, precise-backports, precise-proposed, stable, testing, [19:00] !info systemd bionic [19:00] systemd (source: systemd): system and service manager. In component main, is important. Version 237-3ubuntu10.46 (bionic), package size 2845 kB, installed size 12524 kB (Only available for linux-any) [19:01] floogy: what does this show? " pastebinit <( uname -a; lsb_release -a ) " [19:03] Some apt-show-versions output https://paste.ubuntu.com/p/XsZ6SB3gCR/ [19:04] floogy: "libsystemd0:amd64/focal-updates 237-3ubuntu10.46 upgradeable to 245.4-4ubuntu3.6" [19:04] No LSB modules are available. [19:04] http://paste.ubuntu.com/p/MTmBgdwNdg/ [19:04] floogy: so the release-upgrade from 18.04 to 20.04 has not completed [19:05] Yes. It stoped with the systemd/udev error [19:05] at ~10-12% [19:05] floogy: right, so that is complicating the issue of fixing it. I had missed that part earlier (been in and out of the channel) [19:06] I think I had the same issue. found a website that said to remove two LSB packages [19:06] floogy: try the obvious fix for packages. "sudo apt -f install && sudo apt upgrade" [19:06] compdoc: can you recall which? [19:07] Yes. I see. I also got a german thread in ubuntuusers with some additional info (sorry for that sort of crossposting) https://forum.ubuntuusers.de/topic/upgrade-bionic2focal-scheitert-an-systemd-udev/#post-9249505 [19:08] compdoc, unfortunately I didn't found such site with hints how to solve the issue. [19:08] Im looking [19:09] Exec format... I wonder if it is related to the binfmt-support (shouldn't be but who knows!?) [19:10] I found some stuff with the same error, but on bash scripts for systemd without a shebang [19:10] Therefor I thought, it doesn't have to do with the arch i386/amd64 [19:11] I have to go turn on the server that I upgraded a few days ago, to look at the lastlog [19:12] floogy: I have the glimmering of an explanation now :) systemd v237 may not have supported ExecXXX= without absolute paths (as we see in the errors about absolute paths). But later systemd (v245) does. your system is currently running the old v237 systemd but has the latest udevd files. I need to check if the 18.04 udevd.server used absolute paths [19:12] Okay, thank you very much. [19:13] floogy: ok, looking in a local 18.04 container, the systemd-udevd.service doesn't even have a ExecReload= command and no non-absolute paths [19:13] Yes, that seems to be a good explanation, if it's accurate. [19:14] Yes, so that seems to be the case. Isn't systemd the default in 18.04? Ah, maybe leftovers from way older releases? [19:14] floogy: so, one solution here may be to forcefully downgrade systemd-udevd temporarily (if there's still a 18.04 package in the local /var/cache/apt/archive/ (unlikely ) or fetch the package manually and install it using dpkg --install [19:14] floogy: I think what happened is udevd upgraded before systemd which broke things [19:15] since its .service file now contains 'stuff' that systemd v237 can't handle [19:15] sudo dpkg --force-all -i /var/cache/apt/archives/udev_245.4-4ubuntu3.6_amd64.deb postrm tries to restart systemd-udevd.service [19:16] and fails [19:16] floogy, sorry, it was a poblem with lxd and lxd-client. [19:16] compdoc, thank you. So I guess it's not related? [19:16]