[00:06] systemd-analyze -> Startup finished in 2.190s (kernel) + 7.471s (userspace) = 9.662s — does this mean that ssh should be available within 10 seconds of bootup? I'm confused - because that does not seem true? [00:14] pycurious: try systemd-analyze critical-chain ssh.service === genii is now known as genii-core [00:33] sarnold: am trying to find a place to pastebin the output - ubuntu paste seems to now need login! [00:35] sarnold: systemd-analyze critical-chain ssh.service gives 43ms and shows the other services that are dependent on ssh - but it does not tell me why it takes 30 seconds on this machine to let me enter it using ssh? [00:35] pycurious: hrmn, strange :/ you can try | nc termbin.com 9999 [00:36] pycurious: on my system there's 14 other units after the ssh.service [00:41] sarnold: any chance you can cut and paste that list? [00:41] pycurious: https://paste.ubuntu.com/p/Dmthd87Fhk/ [00:41] pycurious: sigh I meant to, but got distracted. figures. [00:42] sarnold: you dont have docker service enabled, correct? [00:44] sarnold: mine is 43ms- how do you find out - how long it took to get the ssh server on your machine up and running to accept ssh connection? [00:45] sarnold: in your picture - is that the total time that ssh server takes to become active? do i have to add all those numbers? [00:46] pycurious: no docker here [00:48] pycurious: if you want to know the actual time since boot when ssh.service was started, perhaps the systemd-bootchart package is a better starting point [00:58] sarnold: I just got the svg - not sure how that tells me when I can ssh into the machine. It's x-axis stops at 20seconds. I know my machine can't be sshed into before 30 seconds. [00:58] pycurious: weird. :( [01:16] sarnold: my critical-chain output https://askubuntu.com/questions/1255501/ubuntu-20-04-slow-boot-issue looks similar to this [01:16] any ideas what this graphical-chain does? [01:17] pycurious: I don't see graphical-chain in that [01:20] sarnold: critical-chain output "graphical.target @2min 25.876s" [01:20] sarnold: https://askubuntu.com/questions/1255501/ubuntu-20-04-slow-boot-issue [01:21] pycurious: ah, graphical.target is whatever display manager you're using [01:23] sarnold: @2min 25.8sec - seems to be the start time for that service. When did that get done? [01:24] pycurious: it should be about two and half minutes after boot started [01:52] sarnold: So that 2.5 minute is what graphical.target starts - when does it end / or become active? [01:52] sarnold: when you pasted your ssh.service - I had the same question on that. [01:53] In your machine - https://paste.ubuntu.com/p/Dmthd87Fhk/ - the ssh service starts at 29ms, but clearly its not active and started at that point. How do I find out when is it ready for connections? [01:54] pycurious: that 29ms means that it took 29ms to start it; it sadly doesn't have the @ time to say when exactly it started [01:55] sarnold: ah i see - and the staircase - are those services that depend on it? [01:55] pycurious: the other way around -- ssh.service depends upon network.target, which depends upon NetworkManager.service, etc etc [02:06] My critical-chain has docker.service @4.168s +1.796s - any ideas if this can be optimized? [02:11] probably, though I've never gone that far down the systemd rabbit hole [02:11] I think it'd be along the lines of adding a *new* default target, one that Wants your graphical.target *and* Wants your docker.service, so that it'd get started somewhere along the way; and also remove a Wants that is currently starting it [02:12] why do they keep closing ##freenode? [02:13] oops wrong chan [02:13] hi! I'm using xfce and thunar on Ubuntu focal. the places bar shows my newly created, ext3 formatted partition "derp", but when I click on it, I'm prompted for sudo credentials to mount it. is my user account lacking a group or something? which? [02:17] solved it via fstab for now === j2bv17 is now known as j2bv16 === sig_9 is now known as bebop [04:21] hello [05:19] hi [05:19] hi 1 === uni is now known as Guest3948 [06:30] does anybody know if its possible to add the time next to the output and input displayed on the terminal? === LnxTen_ is now known as LnxTen [06:45] shailangsa: do you mean for every program that writes something to the screen, or reads from the keyboard, you'd also see the time? AFAIK no, you'd need a special terminal program for this, for example journalctl also tracks the time when programs write to it [06:46] You can modify the shell prompt to display the time also. [06:47] i mean every line on the terminal shows a timestamp next to it === TechMonk_ is now known as TechMonk [06:48] See also the `script` command which runs programs while keeping the time information of their terminal input/output [06:50] shailangsa_: while if you lower your expectations and say e.g. "I have this script, can I run it and see the date of every line in its output?" then that is possible [06:51] how? [06:52] Example: ls -1 / | while read -r line; do echo "$(date): $line"; done [06:56] oh thought the ubuntu terminal would have the option to display the time at least or thought some existing packages could do this [06:57] shailangsa: you could set it using PS1 variable i think [06:58] The prompt only takes effect when the program finishes, not for every line in its output [06:58] yeah i mean about the input part [06:59] If a program reads from stdin, the prompt won't show. It will only show when bash reads the next command from stdin [07:00] true, not sure what he wants really [07:04] shailangsa: fwiw, https://unix.stackexchange.com/questions/26728/prepending-a-timestamp-to-each-line-of-output-from-a-command === WaV2 is now known as WaV [07:34] /msg NickServ VERIFY REGISTER rhcm123 eSYp2Lkqymcf37Ev === lordrishav1 is now known as LordRishav === Etua1 is now known as Etua === Etua1 is now known as Etua [09:03] any ideas why grabc doesn't work when clicking a pixel inside a firefox window (but works with other apps) ? [09:04] * stevenm__ assumes most now ask... what the heck is 'grabc' [09:04] grabc (i.e. apt-get install grabc) when run just lets you pick out any pixel on the screen and it'll tell you the color of it [09:05] but in a firefox window it'll seemingly give you a random color [09:31] printer works fine via usb, but via network it says lpstat sas printer can't be localized. (in german though, so i'm not sure it's the correct translation). [09:34] lavaball: I have several networked Brother printers and had to use their software to get network printing working [09:36] I did reinstall a computer a few days ago and was able to get it working on 20.04 desktop so a few things are working better [09:36] so distro upgrade. [09:36] i tried that, said i should install all upgrades first. but apt-get update upgrade ... said nothing to install execpt the held back packages. [09:37] you see, you solve one issue for me, i give you 12 new ones. [09:37] anyway, thanks for helping me out. i appreciate it. [09:38] lavaball: printer model may help to find more info [09:38] oh, nah, i bet the 20.04 thing works. [09:38] this should be 18.04 i think. [09:39] so i'll do the distro upgrade. [09:39] good luck or look for OEM software [09:39] when i do do-release-upgrade it still says install all updates. [09:41] usually a good idea to upgrade before release upgrade [09:43] i did, but i also had to apt-get dist-upgrade. [09:43] didn't know. well, now it works. [09:43] thanks again. [09:49] Is anyone familiar with this document: https://wiki.ubuntu.com/KernelTeam/BuildKernelWithChroot [09:52] Given it's referencing Ubuntu versions over ten years old (😱), I was hoping I could get a y/n on whether it's worth pursuing this avenue or not. [09:52] Trying to install the iwlwifi backport for 18.04 on Linux 4.9 onto a debootstrap'ed rootfs using chroot [09:52] oldgalileo: it would be best to state your goal first... [09:52] Compile a kernel for what, another ubuntu version? [09:54] Which Ubuntu version are you targeting? 18.04 has kernel 4.15... [10:00] anyone know how to get apt to correctly pass options to dpkg as per "man apt.conf" DPkg:: options? This is failing in 2 ways (1) argument not passed and (2) with Debug:: the dpkg command isn't executed :s " sudo apt-get -o=DPkg::Options=--force-all -o=Debug::pkgDPkgPM=true -f install" [10:02] alkisg: lol it's a custom kernel and a bit of a janky setup but yeah. 4.9 with ubuntu 18.04.05 [10:04] alkisg: I'm not asking with support on any of the weirdness going on here. I guess a different (though not necessarily better) question would be "has anyone had any luck getting dkms to work with chroot" [10:05] oldgalileo: can you do it via making it a DKMS packaged source? Makes building much easier (although if I recall correctly iwlwifi is too interwined to extract it) [10:05] TJ-: I'd have that problem in some debian packages from outside the repos, the problem in that situation was something to do with the package build options wheere a metadata/parameter hadn't been set in the package to 'take input' so although apt supported it, that specific package didn't [10:08] ikonia: how do you mean? These options are for the dpkg tool but it looks like DPkg::Options isn't the correct option at all; in the apt source I see DPkg::Options::Tools::... for example [10:08] TJ-: 😢 https://github.com/dell/dkms/issues/145 [10:08] Issue 145 in dell/dkms "dkms does not appear to be chroot safe" [Open] [10:08] lmao [10:09] oldgalileo: that bug report is expected really; if procfs and devtmpfs aren't also mounted into the chroot [10:09] TJ-: perhaps I'm missunderstood your problem, I've tried to pass in package arguments but the package that was 'recieving' the argument wasn't build correctly to take them [10:10] oldgalileo: "for n in proc sys dev; do mount --rbind /$b /chroot/$n; done" then it would work [10:10] ikonia: no, this isn't passing arguments to a package, it's (supposed) to add an option to the /usr/bin/dpkg command line (in this case --force-all) [10:11] ikonia: but looks like the documentation is hopelessly wrong; can't even figure out the correct incantation from the apt source [10:11] ikonia: for example mentions of DPkg::Options in /usr/lib/python3/dist-packages/aptdaemon/worker/aptworker.py [10:13] TJ-: my apologies, I missunderstood [10:14] ikonia: no need ... it is a weird one [10:14] ikonia: I deployed the typical workaround ... deleted conflict files! [10:15] ikonia: hit this trying to install a debug/patched version of systemd. Built it locally last night (amd64) and dpkg --install -ed the .debs only to find there are a bunch of :i386 packages (thatI hadn't built) also required and therefore apt wanted to remove about 30 :i386 libraries! [10:17] ikonia: finally worked around that this morning by pushing the source up to my bug-fixes PPA and installing the packages from that (amd64 and i386 built in PPA) and that causes a local /usr/share/doc/$PACKAGE/changelog* conflict. Was trying to use --force-all to override that [10:18] oldgalileo: just noticed a bad typo in the mount command! [10:18] oldgalileo: sohuld be $n not $b - the original would recursive-bind / and every mounted file system under it to the chroot!! "for n in proc sys dev; do mount --rbind /$n /chroot/$n; done" then it would work [10:19] TJ-: Haha yeah. All good, had to change it a bit anyway. Also TIL about rbind vs bind. Usually try to understand what things are doing before running it. [10:21] oldgalileo: I used rbind so devpts is also mounted rather than having to list it [10:26] Yeah, I hadn't ever considered a bind mount wouldn't recursively bring in other mounts [10:32] I'd also want to set it up as an rslave, yes? [10:32] TJ-: So that when recursively unmounting it will only effect the slave [10:33] TJ-: (correct me if I'm wrong, not the most experienced with this stuff) [10:34] I think it should end up looking like `mount --rbind --make-rslave /sys /chroot/sys` and then `umount -R /chroot/sys` [10:34] oldgalileo: "for n in dev sys proc; do umount --recursive /chroot/$n; done " [10:35] Agreed. But that has to be done with the top mount being a slave right? [10:36] Because otherwise that would also unmount the sub-mounts on the host [10:43] oldgalileo: no it doesn't; not for me anyhow [10:43] oldgalileo: hmmm, has behaviour changed? I've used those constructs for years === keypushe- is now known as keypusher [10:45] TJ-: This was purely me approaching this in thought, I haven't run this yet. [10:46] TJ-: I guess I'm a bit confused WHY it isn't true that unmounting on the non-slave rbind wouldn't affect the host [10:46] 😭 rip TJ [10:53] does rdesktop still work on ubuntu 20.04? i'm getting handshake error. [11:02] Need a new TJ [11:02] Anyone want to help me better understand bind mounts? 😃 [11:10] upgraded to 20.04. still says printer can't be found in lpstat -t [11:12] oldgalileo: what's not clear ? [11:16] ikonia: How could you reasonable unmount recursively without making the bind mount a slave? [11:16] *reasonably [11:17] what do you mean making it a slave ? [11:17] mounts are just a tree [11:17] you unmount from the top downwards [11:17] Hello all... I have a fun one that I've spent the last 5 hours on. We have 2 identical 20.04 hosts acting as firewalls (active/standby using ucarp). One of these hosts has stopped renaming the network interfaces after a reboot this afternoon - all interfaces are now just the old ethX style, instead of ens9fX for the 10g nics, and ens5fX for the 100g nics. The misbahving host accidentally got upgraded to kernel 5.4.0-80-generic. Running ` [11:17] udevmdm` in debug mode gives me this error: "eth7: Failed to query name_assign_type: Invalid argument eth7: Policies didn't yield a name and Name= is not given, not renaming." and indeed I am unable to read that file in /sys (same "Invalid argument" error). Rebooting into kernel 5.4.0-65-generic makes everything work again. Anything I can do to fix, or should I report as a bug? [11:19] ikonia: By default, a bind mount propagates mount events both ways. So if I have /var/orig/ and I mount it into /tmp/new/, anything that gets mounted in /tmp/new/ will propagate right back to /var/orig/ and vice-versa [11:20] ikonia: So why wouldn't an unmount do the same thing? [11:20] It should, _right_? [11:20] There isn't a really big point I'm trying to make here. I'm just surprised that TJ- would've been able to use this without it mucking stuff up [11:22] oldgalileo: ahhh I see what you mean [11:22] it's designed to work that way [11:23] I'm not seeing the problem, a bind mount is the equivilent of a mounted symlink (over simplifying) [11:23] that's the desired behaviour, so what's the issue ? [11:30] solved it. hostaddress name thingy was weird. changed the lpd: ... to ip and boom. [11:41] ikonia: I guess what I'm saying is, "I don't know much about thiss. From what I've learned, I'm surprised that TJ- mounting without setting it to be a slave would work. I don't think it should work, and if I am right then that's validation of what I've learned" [11:46] I'm still not sure what you mean by slave [11:46] oldgalileo: a bind mount is like an advanced symlink [11:48] if you use it in that context, it's amazingly powerful [11:57] ikonia: Setting up a bind mount to be a slave will limit the propagation of mount events within the bind to be not bi-directional [11:57] what do you mean by slave ? [11:57] it doesn't make sense why you keep saying 'slave' [11:57] there is no slave [11:57] ikonia: mount --make-slave [11:57] it's just a mount point of an existing resource [11:57] You can make a slave [11:58] So you can do mount --bind --make-slave and this will enforce the behavior I just described above [11:58] but that's because you 'want' that behaviour [11:58] I'm struggling to understand the issue/question here [11:59] Lol exactly. I'm sorry, it's a clarification basically. I want to make sure my understanding is correct, and that TJ- missed something when he described his thoughts on the solution he proposed earlier [12:00] so what are you trying to actually do ? [12:06] At this point, I have it solved. Basically, dkms uses anonymous pipes which try to write to /dev/fd which I didn't have mounted in the chroot. I was running into an issue trying to install the iwlwifi backport because dkms tries to configure it after install and was erroring out. [12:20] Hello [12:25] webchat71, can we help you ? [12:38] Ubuntu 20.04 - After updating libreoffice 4.6.6 -> 4.6.7, a Calc document (40kB) with 4k rows and roughly every other row BOLDED becomes slow to load (about minute to open) and edit. Removing all BOLDING from the document, makes it fast again. Any workarounds? Have tried everything with the "hardware acceleration" and anti-aliasing in [12:38] Options->View. [12:43] Can anyone suggest ways to debug why installing `backport-iwlwifi-dkms` (https://packages.ubuntu.com/bionic-updates/backport-iwlwifi-dkms) wouldn't be enough to get iwlwifi to load? [12:44] doesn't correctly detect the hardware that triggers the dependency to load it ? [12:45] ikonia: That was my understanding as well :/ [12:46] if you manually load it does the wifi wor ? [12:46] work [13:01] greetings, how can I tell the drive type (hdd,ssd,nvme,etc) of a drive that is in a bios raid? THe usual tools (lsblk,smartclt,hdparm,etc) give false readings. [13:07] hi :) I have a server with Ubuntu focal and want to share files via sshfs. In the past, for this I had to create a user account and set its shell to scponly. is this the best way to go today, too? [13:08] Sven_vB, #ubuntu-server? [13:08] ok [13:18] After my laptop (Thinkpad P14s) wakes up from standby the touchpad is functioning in a handicapped state - sensitivity when moving the cursor and when tapping-to-click is down. Any idea how to troubleshoot this? [13:36] Hi folks [13:38] hello BluesKaj [13:39] hi jabberwock === genii-core is now known as genii === stevenm__ is now known as stevenm [14:42] hello [14:57] I am using the following IP tables rules but for some reason people outisde Canada can still reach my website: https://dpaste.org/Y4ZP Does anyone see an issue with my rules? [15:02] does anybody know if having directories and files within /home/ as opposed to home/admin1/ has limitations and differences in anyway? [15:04] shailangsa: a plain user can't create dirs directly under /home. If you want to create a /home/shared for all users, and share group permissions etc, or a /home/photos writeable by one user and readable for others, sure, you can [15:05] So to create these dirs you'd use sudo mkdir, chmod etc [15:07] hoppity: I don't think all traffic/ip's are geotagged [15:07] leftyfb: what do you mean? [15:07] when using a command which is added to the path like python or androids' repo command, does it work within home/ the same way it would in home/admin1? [15:40] Yes [15:52] but of course i would need to add it path by first placing the binary in somewhere like usr/bin right? [15:57] Happy 30th birthday, Linux! This is the day when Linus posted on Usenet advertising his new OS. [15:58] !discuss | signofzeta [15:58] signofzeta: Want to talk about Ubuntu, but don't have a support question? /join #ubuntu-discuss for non-support Ubuntu discussion, or try #ubuntu-offtopic for general chat. Thanks! [16:31] Hello. My ubuntu suddenly started freezing and dropping to command line on restart. Running fsck was required to get it running both times. But when I run smart check on my drive it says it is fine? Does anyone know what I should do [16:32] also `apt update` now fails [16:34] hob60: fails how? [16:36] hoppity https://pastebin.com/3SgSbCLs [16:37] interesting, never seen that before. Sorry [16:38] OK thanks anyway. I think part of my SSD must have gotten corrupted but idk [16:38] but if part got corrupted the smart check would show it I would think [16:38] hob60: consider booting into safe mode via Grub and apt update from there [16:39] its tricky to get wifi working in safe mode though so I would connect to your router with a cable [16:47] hob60: verify the hashes of all package-installed files to being with. "sudo dpkg --verify" should not report any errors (although it is permissible for admin-altered files under /etc/). That doesn't check data files generated after install of course. [16:48] s/being with/begin with/ [16:48] hoppity: your pastebin points to at least two files corrupted under /var/lib/ [16:50] oops, sorry hoppity that was for hob60 [16:50] no worries [16:51] hoppity: looks to be safe to delete the two corrupted files and let them be re-downloaded/built [16:51] oh!!! [16:51] hob60: looks to be safe to delete the two corrupted files and let them be re-downloaded/built [16:51] hob60: "sudo rm /var/lib/apt/lists/partial/security.ubuntu.com_ubuntu_dists_focal-security_restricted_binary-i386_Packages /var/lib/command-not-found/commands.db.metadata" [16:54] hob60: you can rebuilt the command-not-found database with "sudo /usr/lib/cnf-update-db" [17:00] `dpkg: error: parsing file '/var/lib/dpkg/updates/0003' near line 0: [17:00] newline in field name '/.'` [17:00] That seems ominous [17:00] (after `sudo dpkg --verify`) [17:04] and I can't remove that package because it says no such file. But it has a hash that is mismatching?? [17:04] also I am getting prompts to report system problems throughout this [17:04] TJ- [17:05] Maybe I just need to say goodbye to this install of ubuntu :( [17:05] hi, anyone from canonical here? [17:05] whats your issue webchat81 [17:06] oh no technical issue, just wanted to ask about wallpapers in future releases [17:07] one of my photos were included in 17.10 so just wanted to offer more with a ui person [17:08] hob60: that doesn't look good - has the PC lost power before doing a controlled shutdown, or have you forced powered off ? [17:13] Is it possible to use rm to remove all files with a certain extension recursively from current folder? [17:14] on stack overflow, I see find . -name "*.txt" -type f -delete [17:14] I wonder why rm is not used? [17:14] TJ- I don't remember doing that but it is possible I used to power button at some point. Since the smart check thinks the disk is OK (actually now it says "OK - 2 bad sectors") do you think it is safe to reinstall to this same disk? [17:15] hob60, check dmesg ; paste it, if you want [17:15] hob60: well, '2 bad sectors' might be a sign that something is going wrong. First thing I'd do is back-up any valuable data and configurations [17:16] hob60: look at the SMART data carefully; especially pending sector/reallocated sector counts [17:16] TJ- yeah code for my project is all pushed. I only installed a couple weeks ago so not much to back up [17:17] ioria I don't know how to read dmesg but there is a lot of red and no timestamps. Will try to make a paste [17:17] hob60: things you might forget like GPG and SSH configs and keys in $HOME, or package-specific configs under /etc/ [17:17] webchat71, dmesg | nc termbin.com 9999 [17:18] hob60, dmesg | nc termbin.com 9999 [17:19] ioria https://termbin.com/hyjj [17:20] hob60: that drive is failing; its dead [17:21] hob60: that, or the cables or motherboard link is disturbed [17:21] not good hob60 [17:21] hob60, maybe this time it's only a corrupted cache (maybe we can fix it), but no way [17:22] hob60: looks to be ata6.00: ATA-9: Samsung SSD 840 PRO Series, DXM04B0Q, max UDMA/133 [17:22] TJ- yeah 840 Pro. Disappointing, it really hasn't seen much action. But I will check the cables before giving up [17:23] hob60, change port but ... [17:23] hob60: sometimes unplugging and reseating everything firmly helps amazingly [17:23] 1710 [17:23] hob60: you've also got a USB issue "usb usb5-port3: Cannot enable. Maybe the USB cable is bad?" [17:24] I remember seeing that message somewhere. Maybe I did a shoddy job connecting stuff to my new motherboard [17:24] hob60: definitely got data corruption "systemd-journald[394]: File /var/log/journal/8a713c0755714552ada3819022018d01/system.journal corrupted or uncleanly shut down, renaming and replacing." [17:25] hob60: it looks like a connection issue since most of the hardware errors are ATA bus [17:25] TJ- I get the feeling it is barely running right now.... lots of slowdowns [17:25] TJ- OK that is hopeful then [17:26] hob60, run 'sudo journalctl --verify ' but usually it fixes itself [17:26] TJ- but since the corruption has already happened is the install probably doomed anyway? [17:27] hob60: well, "dpkg --verify" didn't look bad, so it may only be limited to files under /var/ like logs and so on so recoverable. Fix the cabling if that is it, then worry about fixing the data :) [17:28] TJ- roger that. I'll sign off and do some tinkering in a bit here. Thank you TJ- and ioria [17:28] ok === keypushe- is now known as keypusher [17:59] Ubuntu 20.04 - After updating libreoffice 4.6.6 -> 4.6.7, a Calc document (40kB) with 4k rows and roughly every other row BOLDED becomes slow to load (about minute to open) and edit. Removing all BOLDING from the document, makes it fast again. Any workarounds? Have tried everything with the "hardware acceleration" and anti-aliasing in [17:59] Options->View. [18:19] guest_4534: do you mean 6.4.7? [18:20] guest_4534: also, the snap version is at 7.1.5.2 [18:22] yes, ment 6.4.7, sorry about the mixup === diskin is now known as Guest6369 === diskin_ is now known as diskin === EriC^^_ is now known as EriC^^ [18:48] hi, there [19:38] so, I've installed xubuntu in a cheap laptop, but for some weird reason it doesn't detect neither my speakers nor my microphone. any ideas? [19:38] (it doesn't even detect the wireless card but I fixed that with a usb wifi adapter for the time being) [19:42] Is there a way to make this iptables rule block a specific port only? -A INPUT -m geoip ! --source-country US -j DROP [20:02] hoppity: how about --destination-port [20:03] tomreyn: thanks, tha's exactly right [20:04] Exterminador: run alsa-info to gather information on your system and post it online (do upgrade it first if it asks about it). [20:04] Exterminador: about the wireless chipset, post lspci -knn | grep -EA3 '(Network|Ethernet) controller' | nc termbin.com 9999 [20:06] hi??? [20:07] tomreyn: the alsa-info output: https://termbin.com/il6k ; about the chipset, I have no output, but for the moment the audio is the most important [20:10] I know the pseudo wireless chipset (it seems to be some kind of internal usb thing) is a Realtek 8723DU Wireless Lan 802.11n USB NIC [20:12] the most important thing to me at the moment is to have audio+mic working, honestly [20:18] Does cron notify the admin in case of failure? [20:19] Exterminador: i'm afraid i can't review the output right now - no time. one thing you can try is to switch from the GA kernel you are currently running to the HWE kernel (5.11) https://gist.github.com/tomreyn/8d7675840d7bc7389b32e4d8887ca449#how-do-i-switch-from-the-ga-to-the-hwe-stack [20:20] hoppity: if you make it to [20:20] matsaman: how can I make it email a specific local user? [20:20] let's say "pancho" [20:21] hoppity: make what email to a specific local user? (I just joined, maybe I can help [20:21] ) [20:22] hoppity: look for MAILTO in crontab(5) [20:22] SimPilotAdamT: I have a cron job that runs at midnight. I would like a user named "pancho" to be emailed (local email) upon cron failure [20:22] sarnold: thanks [20:22] aahhh oki [20:23] tomreyn: I'll give it a try for sure. thanks for your time, anyway [20:38] hello guys [20:39] i need help to install ubuntu without usb or cd drive can this work? [20:39] i already downloaded ubuntu-20.04.2.0-desktop-amd64.iso [20:40] changing to the HWE kernel seems to not fix any of the issues :/ [20:41] guys anyone can help? [20:42] sam21: could you explain little more [20:43] i dont have usb or dvd drive and i want to install ubuntu [20:43] Exterminador: can you show us " pastebinit <( journalctl -k ) " [20:43] sam21: what /do/ you have on the PC you want to install to? [20:43] i have windows 11 now [20:44] i want to delete windows and install ubuntu [20:44] wubi? [20:44] sam21: does it have a virtual machine hypervisor, and can the hypervisor pass the raw storage device to a guest virtual machine? [20:44] is wubi still alive? [20:45] pasiz: wubi died long ago [20:45] oh [20:45] TJ-: seems to weirdly require login to view? https://paste.ubuntu.com/p/DwszxsqcMc/ [20:45] TJ- sorry i dont know whats that mean [20:47] sam21: without an installer device, it is possible to do, if (a) use Windows Disk Management to free up space by shrinking Windows NTFS file-systems and partitions to make way for Ubuntu and (b) the raw storage device/disk can be mapped into a virtual machine guest, said guest booting from the Ubuntu installer ISO file [20:47] sam21 did you even try googling it? [20:47] Exterminador: there are a lot of "kernel: [Hardware Error]: event severity: fata" [20:48] yea i'm watching some youtube videos i find a website using app called ( kde neon) but i dont know if it safe or not yukiup [20:49] https://linuxhint.com/install-ubuntu-without-usb/ [20:49] I guess that the laptop is having some issues (altough it is kinda new) [20:49] Exterminador: there's a kernel stack-trace later too, looks to be related to UVC (USB Video Class) so likely an internal USB-attached camera [20:49] yukiup thanks bro :D <3 [20:51] TJ-: yeah. I need to take the laptop to a repair shop to see. I suspect that I might be facing some hardware issues [20:51] sam21: be VERY careful with that guide; it'll overwrite the Windows install whilst it is still running [20:52] Exterminador: show us " pastebinit <( lspci -nnk ) " [20:52] I've switched to Xubuntu because Windows was being a pain (no wifi after update) [20:53] TJ- yea thats look if i made a mistake i will stuck without pc lol [20:53] I'm getting a Unable to read from: /dev/fd/63 [20:54] Exterminador: hmmm, is 'lspci' installed and working? [20:55] sam21: maybe this? https://netboot.xyz/ [20:55] Exterminador: I think due to that motherboard being very new, you'll need to be using the very latest mainline kernels for a while. Ubuntu does build mainline kernels [20:56] !mainline | Exterminador [20:56] Exterminador: The kernel team supply continuous mainline kernel builds which can be useful for tracking down issues or testing recent changes in the Linux kernel. More information is available at https://wiki.ubuntu.com/Kernel/MainlineBuilds [20:56] TJ- i'm switching to ubuntu because i'm starting to learn coding and i wanna dive deep on it and i want to stay on the mood :P [20:56] sam21: you can run a dhcp server and tftp server locally if you want to rely entirely on your own infrastructure, but this can probably save a few hours [20:56] sam21: that's great! Don't break it on the way though. Without a USB 'thumb' drive to install from it is going to be tricky [20:57] TJ-: I'll take a look at that. regarding the lspcin, it is installed and works [20:57] Exterminador: maybe I typed the command wrong ... checking [20:57] TJ- ur so friendly thanks for being nice [20:58] TJ-: https://paste.ubuntu.com/p/dx4ChdcBSr/ [20:58] Exterminador: command works here, weird [20:59] Exterminador: errrmmmm it doesn't have a PCI network device [20:59] Exterminador: kindof ironic seeing as how many "Communications Controller"s it reports [20:59] sorry, "Signal processing controller" [21:00] the laptop is a Thomson Neo N14C4WH64 (the wireless card is some sort of internal usb adapter?) [21:01] Exterminador: USB!? oh really! that explains it. "pastebinit <( lsusb )" [21:02] Exterminador: I mistakenly thought it was an ultra-new device due to the timestamp on the firmware (april 2021) [21:02] TJ-: https://paste.ubuntu.com/p/pcFJ4jmMjs/ [21:03] Exterminador: am I correct in thinking this was unique to the French market? [21:03] probably, since I'm in France [21:03] Exterminador: the internal device with the problem is "RTL8188EUS" ? [21:04] TJ-: nope. should be the Realtek Semiconductor Corp. 802.11n WLAN Adapter. the other one is the external wifi adapter I'm using iirc [21:05] Exterminador: ahhh, OK [21:06] I had to be creative to connect this laptop to the internet. fortunately I have an wifi usb adapter that works perfectly [21:07] Exterminador: cannot see any kernel module that claims an alias for that device ID. " grep -rni '0bda.*d723' /lib/modules/*/modules* " [21:09] hey [21:10] @tj [21:10] TJ-: no output from that command [21:10] TJ- yea it didnt work :P [21:12] Exterminador: there's some out-of-tree source-code, some with DKMS build wrappers, which may be worth trying but I suspect the issue you're going to have (based on my brief searches) is that device is usually on the PCI bus rather than USB so the drivers probably won't be designed to work with USB. Having said that, give it a try anyhow. See https://github.com/smlinux/rtl8723de === sharperguy__ is now known as sharperguy [21:13] if I just switched from an Nvidia system to an AMD system [21:13] what do i have to do so my GPU is recognized [21:14] goddard: try sudo ubuntu-drivers install [21:16] damn No drivers found for installation. [21:16] i have a 6800 xt [21:16] :( I was hoping that'd Just Work :( [21:17] thanks for trying [21:18] isn't that GPU supported by amdgpu in-kernel driver? [21:18] i thought so [21:18] maybe i have noveau installed still or something? [21:19] some random reddit page suggests that it might be too new for amdgpu from 20.04 [21:19] goddard: have you checked with 'lspci -nnk ' that it has a driver in use? [21:19] !hwe | goddard might need this [21:19] goddard might need this: The Ubuntu LTS enablement stacks provide newer kernel and X support for existing LTS releases, see https://wiki.ubuntu.com/Kernel/LTSEnablementStack [21:19] no let me try [21:20] 04:00.0 PCI bridge [0604]: Advanced Micro Devices, Inc. [AMD/ATI] Navi 10 XL Upstream Port of PCI Express Switch [1002:1478] (rev c1) [21:20] Kernel driver in use: pcieport [21:20] i see it [21:20] goddard: this will limit output to GPUs " lspci -nnk -d ::0300 " [21:20] my guess is that's unrelated to the video part .. [21:20] goddard: that isn't it, that's just a port [21:21] goddard: e.g. on mine I get "05:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Picasso [1002:15d8] (rev c1)" [21:22] https://pastebin.ubuntu.com/p/wyHR8rb7vW/ [21:22] goddard: "Kernel driver in use: amdgpu" [21:22] hmm i wonder why steam hangs my session then [21:23] past its boiling point? [21:23] maybe [21:23] check kernel logs for signs of problems; check Xorg log (if not using a Wayland compositor) [21:25] k [21:25] TJ- hey i got some error msgs do i need to change anything on the bios? [21:27] TJ- the msg was " windows faild to start a recent hardware or software chage might be the cause [21:28] TJ- File :\ubnldr.mbr status: 0xc000007b [21:28] TJ- [21:28] TJ- info : the application or operating system couldnt be loaded because a requierd file is missing or contans errors. [21:30] sam60: no idea - not used MS Windows since 2004 [21:31] TJ- lucky u lol [21:31] TJ- thanks anyway i will keep trying tell it work [21:31] TJ-: unfortunately the out-of-tree source-code for the realtek didn't worked. but that's not a big issue for me. currently my biggest issue is audio/mic not working [21:34] Exterminador: in the kernel log I couldn't see any sign of the snd_hda_intel codec being detected [21:36] I'm wondering what that means now, lol [21:36] Was about to order a wireless keyboard mouse, but wondered if they're secure. Wouldn't a wireless keyboard using 2.4Ghz just be broadcasting for the world to listen to? [21:37] Exterminador: there's a generic interface to sound devices using snd_hda_intel. To actually use the devivce behind that interface there needs to be a codec driver and the pinctl (connections) need configuring. Your kernel log doesn't show those last two things, which makes me think that device has a unique codec [21:39] Exterminador: kernel log only show a couple of lines, starting with "snd_hda_intel 0000:00:0e.0: DSP detected with PCI class/subclass/prog-if info 0x040100" [21:41] Exterminador: output of your lspci suggests you may need to use an alternate kernel module "Kernel modules: snd_hda_intel, snd_soc_skl, snd_sof_pci" by blacklisting snd_hda_intel to let one of the others have a shot [21:42] how do I do that blacklisting? [21:48] Exterminador: before blacklisting test it manually. First check the reference count is 0 on snd_hda_intel with "lsmod | grep snd_hda_intel". If so, "sudo modprobe --remove snd_hda_intel" then try one of the other modules "sudo modprobe snd_sof_pci" and check for pleasing kernel messages "journalctl -k -n 20" [21:50] TJ-: doesn't seems to be 0: https://paste.ubuntu.com/p/SXzccWWMTP/ === cacti is now known as ice [21:55] Exterminador: that output looks promising! I wonder if your kernel log lost something due to the stacktrace [21:56] might be possible. it's just that it annoys me that I have no sound, not even with headphones :/ [21:57] Exterminador: just re-looked at your ALSA debug stuff. The device is HDMI audio only! [21:57] huh? but it does have speakers [21:57] Exterminador: so it looks like the internal sound chip is indeed not found [21:59] Exterminador: not sure if this is relevent but sawit earlier "Invalid pltconfig, ensure IPC1 device is enabled in BIOS" [22:00] dang. i need to check the bios. any ideas of what that "IPC1 device" is? [22:03] Exterminador: Just looked at the kernel source - don't think it is related. I think IPC == Intel Performance Counters and its part of the telemetry debugfs [22:03]