[00:34] TJ-: thanks! I will look into it [01:21] so apparently an xorg extension vnc server is more efficient than a scraping vnc server. Does anyone know why? In my mind having another xfce session start is just more RAM load. === sveta is now known as Sveta === zbenjamin is now known as Guest34344 === zbenjamin_ is now known as zbenjamin [01:45] hi, having trouble with xrdp, I am getting the following error. "KDSETMODE failed: Bad file descriptor". not finding a solution, any ideas? [02:00] Can someone please help me with a monitor issue I'm having? I'm using Ubuntu 18.04 with i3-wm, and nvidia-driver-430. After Ubuntu puts itself to sleep after inactivity, and I log back in, only 2 out of my 3 monitors turn back on. My primary monitor (display port), and my second monitor (hdmi), but my third display port monitor does not turn on. If I look in nvidia-settings it's listed as "disabled". If I enable it in there, it does [02:00] nothing. If I unplug the monitor and plug it back in it then gets recognized and functions as normal. Any idea where I can go from here? Thanks. [02:10] i've tried execute skypeforlinux command via cron everyday 16:00 but no luck ... the command executed but the window not appear [02:24] Ubuntu keeps crashing with a message, "kernel panic - not syncing: Fatal exception in interrupt" [02:25] Can anyone help me figure out why it is doing this and how to stop it? [02:31] what's your environment golden_ticket, how to produce that error? [02:32] i'm just going to do a fresh install [02:32] thanks anyway chieta [02:32] when did it start? [02:32] today [02:32] which ubuntu release to be installed? [02:32] I really haven't done anything with the OS either [02:33] It's version 18.04 [02:33] how about 19.04? [02:33] I don't want to be so bold [02:33] noted, what's your box? [02:34] i don't understand that question [02:34] how abut ubuntu livecd for check your hardware supported or not? [02:34] it should be. I've done this 100 times before on the same thing [02:43] Best video/,media player for Ubuntu 18? [02:54] if you want tried the latest 18.04 [02:55] to get the latest kernel... hope it help golden_ticket [02:57] how much longer is 19.04 supported? [03:01] truexfan81: 19.04: 19 == 2019. 04 == April - is an interum release ( LTS is every 2 years ) .. so 19.04 ha 9 months of support == EOL Jan 2020. [03:02] hmm maybe i'll upgrade it to the next lts when that comes out and then stay on that for awhile [03:02] truexfan81: Next LTS is 20.04 ,, Be a while yet :P [03:03] true, i guess there will be a 19.09 i'll have to upgrade to first? [03:04] truexfan81: 19.10 . yep .. that is the path, in your case :) [03:04] that works thank you [03:09] chieta, are you saying if I change my kernel, that might fix this error? [03:16] golden_ticket: Have you tried booting a different kernel from grub's boot menu ? [03:20] Bashing-om, no [03:23] golden_ticket: The kernel panic "could" be something in that current booting kerenl. Will only take a minute or so to try an alternate installed kernel. That is why the system does not remove old kernels :) [03:24] should `sudo usermod -a -G "www-data" myusername` add myusername to the www-data group? [03:24] because its not and i am not sure why [03:25] Bashing-om, i'll give it a try === golden_ticket is now known as amazoniantoad [04:02] how would i go about setting my graphics driver target to 435? [04:03] I've added the graphics ppa and its target is 430 [04:09] magic_ninja_work: What relese is intalled and what card ? could be the release and/or the card does not support the recent 435 version driver. [04:09] installed* [04:09] 19.04 and 960M [04:10] I'm sure 435 supports the card. It is intel/960 hybrid graphics. Bashing-om [04:10] I see the package nvidia-graphics-drivers-435 [04:11] Bashing-om, seems you are correct, though, and the system can't find the 435 drivers for 19.04. They just show them for 18.04. [04:14] magic_ninja_work: Strange: https://launchpad.net/~graphics-drivers/+archive/ubuntu/ppa // yepper no 435 version for 19.04 .. I find that odd :( [04:14] me as well. They have 18.04 but don't have the intermediate release one, which is the one that is supposed to get it. [04:19] magic_ninja_work: Just a surmise on my part - the developers pulled 435 for more tweaking ? [04:20] Well, I've check it several times and have not seen it on there for 19.04 at all, but it absolutely could be. [04:26] On 18.04, I have two monitors, with the left one being detatchable. I want to make the right one 'primary' such that all windows say there/ open there and I only put things on the left one manually. By default, it seems like as soon as I plugin the second, left one, all previous windows get shifted over there. How can I swap it around? [04:27] there is a setup in the system settings app to handle your displays. [04:28] In 'displays' I have good settings for moving and enabling / disabling them, but nothing about primary [04:31] https://www.google.com/search?client=ubuntu&channel=fs&q=select+primary+monitor+on+gnome&ie=utf-8&oe=utf-8 [04:31] Try some of those. [04:38] What is recommended way to get ZFS newer then 0.7 for Ubuntu 19.04 (like 0.8.1 ) ? [04:39] I used to compile my own packages locally, and it worked, but used to complain on them on every package operation, so that's now nice solution I think. [04:40] I see here is packge for 19.10 (btw I dislike release code names) , but not for 19.04: https://launchpad.net/ubuntu/eoan/+source/zfs-linux [04:50] nikolam: hi! does it have to be zfs-linux package? I don't see that at https://packages.ubuntu.com/search?keywords=zfs&searchon=names&suite=all§ion=all [04:51] nikolam: what package name do you want to upgrade to the newer version? [04:52] nikolam, for example, 'zfs-dkms' 0.7 is available in disco [04:52] nikolam, https://packages.ubuntu.com/disco/zfs-dkms [05:08] I would like to use newer features of OpenZFS/zfsonlinux, that are in 0.8.1 and not in 0.7 , like transparent encryption. [05:09] I also don't care for Ubuntu code names, I think 19.04 , 19.10 means more. [05:14] I have build my own packages again, with dpkg-source -x and dpkg-buildpackage -us -uc , but I envision again problems while updating if I install them (e.g. constant labeling locally-built packages as somehow wrong or broken during update) [05:15] Maybe I should just use another Linux distro, if I want newer packages is supported release. [05:49] also, doe Ubuntu fail to boot if some of the mounts in /etc/fstab fails to mount? [05:58] nikolam, adding the nobootwait option on non-critical mounts will make it not worry about mount failures too much [05:59] there's also nofail [05:59] but default behavior is it fails to boot if mount point is not there? [06:01] How do I identify exactly what packages are security updates that /usr/lib/update-notifier/apt-check reports? Every method I've found on the web hasn't matched its count. [06:01] I think auto and default options make it fail to boot if it fails to mount [06:02] not something I run into often, not sure [06:02] Would some body please help in connecting wifi whose usr name and pswd I know? [06:06] wifi doesn't have a "usr name" sohail-ahmed ? [06:06] gbellinoz: correct, wifi authentication I meant [06:07] Are you using a GUI or the command line? [06:07] doug16k, it is really sad default behavior .. I think if I have single drive failing in /etc/fstab that shouldn't be the reason to halt whole machine and give unexpecting user a ROOT prompt. [06:07] gbellinoz: right now gui, but I can use command line [06:07] Gnome desktop (standard Ubuntu)? [06:08] gbellinoz: yes I think its standard [06:08] So do you have the little WiFi networking symbol in the upper right corner? [06:08] gbellinoz: yes [06:09] And when you click on that does it list your WiFi network's name? [06:09] gbellinoz: yes [06:09] Then what happens? [06:11] I clik on the select network, and then the network I want to connect to... a gui appears with many boxes, authetication, usr name, pass, etc ... I tried all options but it trys connecting but then says password is incorrect [06:11] Screenshot? [06:11] gbellinoz: ok. [06:11] I don't think I've seen it ask for a username. [06:11] Maybe you have some setup I don't know about. [06:12] gbellinoz: no. It does not ask for a usr name... I put it in the blanks [06:12] gbellinoz: in the gui that appears after clicking over the avaiable networks [06:13] Oh, you said usr name again up there. A screenshot would help here. I don't have a lot of experience, but that might make your issue clearer and some else will chime in. [06:13] doug16k, thanks for advice on mount options [06:13] sohail-ahmed, if you press print-scrn on your keyboard, a screenshot should appear in your home Pictures directory [06:13] nikolam, np [06:16] gbellinoz: where should I upload the ss? I dont think pastebin allows it [06:16] imgur? [06:17] I like this one, no BS -> http://pasteall.org/pic/ [06:19] They wait until it gets popular, then suddenly, login req'd, paid plans available, hey, pop-up, do you want 30 days free? [06:19] been free for years [06:20] it's primarily for sharing blender files, images are secondary there [06:20] That's amazing. And no GA or FB scripts.... [06:21] Like the web used to be :) [06:21] exactly [06:23] gbellinoz: sorry Imugr was taking time, so choose something else: https://prnt.sc/p2i6vv [06:23] if only someone had mentioned that. [06:25] gbellinoz: sorry? [06:25] read the chat above. [06:26] gbellinoz: ok, here it is:http://pasteall.org/pic/59c5566be6850b01d2d5fc1c89570578 [06:30] sohail-ahmed, were you given instructions telling you to select Tunneled TLS by someone running that network? [06:31] same question about the inner authentication protocol [06:31] doug16k: no. The system automatically selected it and though I tried different combinations with both [06:35] sohail-ahmed, try Protected EAP [06:35] doug16k: peap version? [06:36] don't know, try higher and go down from there if it doesn't work [06:39] doug16k: yes Its connected. Thank you very much to you and @gbellion for all the help. Have a very blessed day!!!! [06:39] nice [06:48] is there anything wrong with this fstab entry? because when I put it in and boot into lubuntu, I end up in recovery mode. [06:49] UUID=aee19d0f-9fb7-4339-bc81-a998d7449055 /home/caleb/D ntfs auto rwx,user,auto 0 0UUID=aee19d0f-9fb7-4339-bc81-a998d7449055 [06:49] oh no [06:49] excuse the part at the end there, that's not in the original version and is due to something funky with emacs I think [06:49] UUID=aee19d0f-9fb7-4339-bc81-a998d7449055 /home/caleb/D ntfs auto rwx,user,auto 0 0 [06:50] this was the entry [06:50] actually maybe I just double tapped C-y by accident [06:50] and didn't delete it all [06:51] (had to grab the line from /etc/fstab~ and copy it into fstab) [07:03] anyone know why grub would refuse to put windows 10 into the menu? this is a new build. im running 2 ssd's. linux on one, windows on the other. and not matter what i do, it will not add windows to grub. advice? [07:04] glitchd: Not even update-grub adds Windows 10? [07:05] EoflaOE, nope, even ran os-prober, but that didnt even find windows. [07:05] EoflaOE, but if i select that drive that windwos is on at boot, it loads it just fine [07:07] glitchd: According to "fdisk -l", which partition did you install Windows 10 on? And is your system UEFI? [07:09] EoflaOE, heres lsblk, https://pastebin.com/TsGH0dUs [07:09] and here is fdisk -l [07:09] sudo fdisk -l [07:09] Disk /dev/loop0: 14.8 MiB, 15462400 bytes, 30200 sectors [07:09] Units: sectors of 1 * 512 = 512 bytes [07:09] Sector size (logical/physical): 512 bytes / 512 bytes [07:09] I/O size (minimum/optimal): 512 bytes / 512 bytes [07:09] [07:10] [07:10] [07:10] Disk /dev/loop6: 54.4 MiB, 57069568 bytes, 111464 sectors [07:10] glitchd: fdisk -l on pastebin, not here [07:10] Units: sectors of 1 * 512 = 512 bytes [07:10] Sector size (logical/physical): 512 bytes / 512 bytes [07:10] I/O size (minimum/optimal): 512 bytes / 512 bytes [07:11] glitchd use a psastebin [07:11] *pastebin [07:11] /dev/sdb2 1085440 1288191 202752 99M Microsoft basic data [07:11] /dev/sdb3 1288192 1320959 32768 16M Microsoft reserved [07:11] /dev/sdb4 1320960 414515199 413194240 197G Microsoft basic data [07:11] [07:11] [07:12] Disk /dev/loop8: 42.8 MiB, 44879872 bytes, 87656 sectors [07:15] nikolov@ubuntu-ivan:~$ inxi -F [07:15] System: Host: ubuntu-ivan Kernel: 5.3.0-050300rc7-generic x86_64 bits: 64 Desktop: Gnome 3.33.91 [07:15] Distro: Ubuntu 19.10 (Eoan Ermine) [07:16] cool [07:17] Nice Blade, but any support for Eoan should go to #ubuntu+1 [08:03] Hi. WTF? E: Failed to fetch https://mirrors.edge.kernel.org/ubuntu/dists/disco-security/universe/dep11/by-hash/SHA256/6f356e9b4a093e30c24721cc37a85da760b342041d4dc6c8f0586dfc89c3a517 404 Not Found [IP: 147.75.101.1 443] [08:06] anyone alive here? [08:26] hello [08:27] I've just installed 18.04.3 on my laptop for the first time, the system hangs with fans spinning on logout and sleep [08:28] Kolas, is that an apt update? no idea what you are doing when that failed [08:28] giaco: You might have to set some "acpi" or similar option as kernel boot paramater. [08:28] or upgrade? or what [08:28] it hangs also on shutdown too [08:29] giaco: Google your laptop model, ubuntu and acpi. [08:29] vlt: thanks [08:29] giaco: Then try a few options. [08:29] oops not Kolas should have been KOLANAICH, who left. sorry for ping kolas [08:30] giaco: If you don't want to edit /etc/default/grub everytime you can add the option right at the start in the GRUB menu for testing. [08:30] we didn't meet KOLANICH's hard realtime response time of < 3 minutes [08:31] vlt: I know, thanks :) do you think that it can also be related to gpu drivers? [08:34] :D [08:34] giaco: I have no idea, sorry. === abbiya_ is now known as abbiya [08:45] computer freeze on "apt-get install nvidia-driver-430". Damn, this is not cool [08:48] giaco: Is that an Ubuntu package? [08:49] giaco, easy way to install that is to run Software & Updates, go to the Additional Drivers tab and select it there [08:49] vlt: thats a non-free driver afaik [08:53] yes it is an ubuntu package [08:53] see `apt show nvidia-driver-430` [08:59] giaco: hanging on logout is a weird one; that does suggest some kind of issue with the display drivers but it shouldn't hang the entire PC. When that happens can you switch to a TTY with Ctrl+Alt+F2 / F3 / F4 ? [09:01] giaco: if the PC is hanging with general commands I'd suspect a hardware<>driver problem; there may be clues in the kernel log from boot-time ( "journalctl -b 0" ) [09:01] TJ-: it doesn't switch TTY when it hangs. It doesn't switch when it is not hanged too, because changing to TTY hangs [09:01] giaco: Whilst it is freshly booted can you show us "pastebinit <( journalctl -b )" we may be able to find some clues [09:02] suspecting video driver, I'm installing the nvidia one. I went into the secure-boot wizard to add the key in uefi, but it failed that too "can't load /var/lib/shim-signed/mok/.rnd into RNG" [09:03] it created the private key, it asked me a password, but it didn't output any key for me to put into uefi [09:03] damn, never been so hard to configure linux on a new laptop [09:04] giaco: Doesn't surprise me; as they become more integrated and the ACPI code more tied to Windows Linux suffers [09:06] giaco: is it a model with dual GPUs that switches between them. Those cause terrible problems and it hasn't got better [09:08] ok, but it fails in trying to setup a proprietary driver [09:14] ok it seems to have worked correctly anyway. MOK key enrolled, nvidia driver loaded, TTY change now works and also logout and reboot [09:15] it was all about gpu driver [09:19] ok, quite satisfied. Next problem, installing torbrowser. I do know that the gpg keys have been poisoned months ago, but the installation on ubuntu 18.04 is still not fixed [09:21] Does anybody know a good way, so I can write different alphabets in different windows? Like in one write japanese, and other one english? [09:23] dbugger: We can't support Windows here. ##windows [09:24] EoflaOE, i didnt mean "windows" as in "the OS" [09:24] I meant, while switching between apps [09:24] in one, I want to write in English [09:24] and other in Japanese [09:31] dbugger do you know the locale for the japanese? analogous to en_US? [09:32] if you run it from a terminal like this: `LANG=en_US some-program-name-here` then it will run just that program in english USA locale [09:32] do equivalent thing for japanese (or just run it normally if your system is japanese by default) [09:32] tor problem solved, so nevermind my last question [09:35] doug16k: Does LANG= change the keyboard layout used? [09:36] vlt, not sure [09:37] I never observed that behaviour. [09:38] sometimes people show pastes in some non-english language, and get very little help, I've often suggested they force it to english that way and repost. as far as input method, there is usually some icon in the system tray to switch, and there are keybinds to go through the input methods, but I am just an english speaker and haven't had a lot of experience with input method switching [09:42] dbugger: A (common) keyboard shortcut to switch between layouts is Alt + (left) Shift. If you can't find a way to link one layout to a window/program maybe setting such a shortcut helps a bit. [10:21] hi [10:22] somehow my google search today goes to google.com instead of google.de and google.com blocks me weirdly [10:22] how do you switch firefox to google.de? [10:23] julius: that would be something the target web server is enforcing with a redirect I'd expect [10:23] couldnt find any setting to tell firefox what "google" search means [10:23] i can only switch between different search engines [10:24] clear your browsers data and cookies, and restart the browser [10:25] in ~/.cache and ~/.config and ~/.mozilla [10:25] that sounds like a windows reboot [10:26] but ok, lets try that [10:29] i can only see ~/.mozilla... and ~/.cache.... dont know where to find firefox in .config [10:31] no, in config you would find google chrome , if you have more browsers [10:31] oke [10:38] ../nick only-test [10:38] hi [10:44] im looking to fix firefox, dont use chroome [10:46] julius, like how do you propose to?: [10:52] julius: did you try from the command-line to see if it is Google redirecting your user agent? [10:53] julius: as in "wget -S -O /dev/null http://google.de/" [11:00] I've just booted up Ubuntu 18.04.3 LTS from USB on my Dell Latitude, and the key mappings are all wack. i hold down the ctrl key and the cursor goes all over the place/highlights dehighlights entire words.. arrow/shift keys do all types of weird things & tab across fields etc. is this something known, or something fixable? [11:02] EoN: sounds like you're unique! :) [11:03] EoN: I wonder if you've accidentally enabled accessibility options like sticky-keys and other options [11:05] hmm... it works fine when using ubuntu through VirtualBox on Windows [11:07] EoN: if holding down the Ctrl key causes the cursor to move that sounds like it is causing some kind of pseudo-mouse input. There is a hotkey sequence that can enable a special NumLock mode that could be the cause. [11:07] EoN: I presume the keyboard doesn't have a separate numeric keypad so the main keys double-up as numeric keys when NumLock is enabled? [11:08] how long before 19.10 is out, in terms of days :) [11:08] and for eg just now i pressed 'ctrl' key within text editor, and the 'right click context dropdown' menu opens up. something is very wacky! lol. maybe the touchpad is glitchy or osmething [11:09] TJ- is that special numlock/accessibility mode accessible through the ui? [11:10] ahhh it's the touchpad! as soon as i 'disabled' the touchpad, keys are working perfectly. how weird! [11:11] EoN: oooh.. could it be the touchpad is detecting your hand ... it could be too sensitive. There should be an option to disable touchpad whilst typing [11:12] even when i kept my hands way away from it, was occurring.. maybe i was breathing too heavily lol [11:31] TJ-, i did see a option to disable the touchpad while typing [11:32] EoN: vibration from pressing the keys or resting on the frame possibly... or maybe just a ghost in the machine! [11:33] immu: releases are usually toward end of the month so .10 would be expected end of October [11:33] gnome tweak > keyboard and mouse [11:33] okie [11:33] yeah. it looks like there are a number of google reports of oversensitive touchpads on dell machines (and others) [11:33] seems like its not an uncommon issue [11:33] immu: see https://wiki.ubuntu.com/EoanErmine/ReleaseSchedule [11:38] I am trying to install AMD Ryzen 3200g with graphics drivers on ubuntu LTS. can somebody help? [11:40] TJ-, tnx [11:41] ubuntu-drivers autoinstall ? [11:41] or perhaps with the free driver … [11:45] how do I find out which driver is running atm? [11:46] no drivers found for installation [11:52] BennyBron: there's not really a proprietary driver for AMD anymore. [11:52] BennyBron: so there's nothing for it to install. [11:53] BennyBron: which LTS? [12:04] I have a ssh connection to a server X opened from /dev/pts/0 ; how can I access that connection and send commands to the server X, but from /dev/pts/1 ? [12:07] redecas: why would you want to? ptys are allocated dynamically as needed for shell logins and other things that require them. only the owner of it can really do anything with it. [12:09] ryuo: there is a ssh connection open to the server; it is opened from another device, but from the same user. I cannot create more ssh conenctions to the server because the /etc/passwd file is misconfigured; so I want to fix it from the ssh connection [12:10] so you have 2 connections open? [12:10] Howdy all [12:11] " because the /etc/passwd file is misconfigured" explain please ? [12:13] I have a ssh connection from a device [12:13] then how do you have access to another pty? [12:14] those are only created for sessions, local or otherwise [12:14] my mistake [12:14] so what keeps you from using the existing connection? [12:15] pc1 --- vm1 ----ssh---- server [12:15] pc2 ---- vm1 [12:15] I am pc2 [12:15] and have a connection to vm1 [12:15] yeah, this sound more you have a school assignment, or too lazy to walk over to the other device [12:15] vm is the virtual machine [12:16] so... it's a pty in the vm? [12:16] yes [12:16] pc1 and pc2 use the same user from vm1 [12:17] (the root user) [12:17] this would be possible if screen or tmux or byobu was in use, but i don't see how you could do it otherwise. [12:18] no, they are not used :| [12:18] oer: you're right. [12:18] oer: i can't see how someone would get into this situation unless they were put into it on purpose. [12:19] the shell was changed from ksh to bash [12:19] and it does not work any more [12:19] to connect to the server [12:19] redecas: here's the thing.. if you try to write to the pty from another one, you'll just be writing to the screen. [12:20] yes, I know [12:20] that won't do any good... you have to be able to trick it into accepting input from another source... but clearly ssh is in control her. [12:20] this isn't really a support question. to #ubuntu-offtopic? [12:20] it is linux related [12:21] maybe you are right [12:21] i'll continue it there. [12:21] or ##networking ? [12:35] Hey Guys, I'm using ubuntu core with an old official kernel-snap (4.4.0-135-generic). I track program network traffics by adding classid in net_cls cgroup and then mark it with iptables. However I found a bug in many devices: The OUTPUT Bytes which have the classid mark stop increasing after several hours (one stops at 77G, one stops at 29G ). However the program still work and sending packets. === ImAdPEr|wORkInG is now known as iMadper [12:37] It looks seem the classic field of skb is cleaned somehow. [12:37] The issue looks like: https://lists.linuxfoundation.org/pipermail/containers/2014-January/033848.html [12:37] Does anyone face the same issue? [12:39] iMadper: that sounds incredibly focused so unlikely, but I'd ask the question in #ubuntu-kernel and possibly #ubuntu-server - would also be worth your examining the commit history of that kernel version to see if the patches you've linked to are included [12:41] TJ-: Thanks a lot. I'm check the kernel patches.. And still tons of patches to check... I just try if I'm luck enough that someone met the same issue here. :-b [12:42] iMadper: you should be able to quickly zero in if they're there with "git log --pretty=oneline -- drivers/net/veth.c" (assuming you've the 4.4.0-135 tag checked out) [12:43] iMadper: Am I correct that 4.4 would be from ubuntu-trusty kernel series? I can refresh my local clone of that and check too [12:45] TJ-: IIRC, I stolen the kernel snap from ubuntu core 18.04... I'd check it as well. [12:48] iMadper: if it is 4.4 it'll be in the ubuntu-trusty kernel repository === jelly-home is now known as jelly [12:50] TJ-: 4.4.0-135-generic ... Okay it must be trusty then. Hmmm, It seems I have to upgrade kernel for 14k devices.... :-/ [12:50] iff I'm sure that new kernel fixes the issue... [12:51] iMadper: did you find the patch in a later than -135 release ? [12:51] iMadper: aha, my mind is like a sieve! trusty was 3.3 ! [12:52] TJ-: I'm not even sure that the patch fixes the issue... [12:52] TJ-: currently I have totally no idea about how to reproduce it. [12:52] iMadper: looks like ubuntu-xenial is the 4.4 tree... I'll get there eventually :) [12:53] TJ-: Okay, It makes sence. As I still remember I stolen it from ubuntu-core 18.04. [12:53] iMadper: the counteres you're referring to that stop incrementing are those reported by iptables? [12:53] TJ-: Yes. [12:54] TJ-: Something like: 546 27418 MARK all -- * * 0.0.0.0/0 0.0.0.0/0 cgroup 1111120 MARK set 0x10f450 [12:54] iMadper: so the issue could be with iptables itself... so would be good to know where it reads those values from and check if the kernel's copy have frozen or not first [12:54] the 1111120 is the cls.classid. [12:55] TJ-: Must read from sock's skb. [12:55] TJ-: If I move the pid to another net_cls and move it back, the iptables works again. [12:57] Hi can anyone please help? I tried to install wrong version of amdgpu and now I'm stuck in login loop. Ie I enter my PW and it loops back to login screeno [12:57] iMadper: what doesn't make sense is why it'd stop - if that code were the cause you'd expect the counter never to increment [12:57] When I drop to terminal I can log in [12:58] hello how i fix it i use with ubuntu 16 error while loading shared libraries: libgnutls.so.26: cannot open shared object file: No such file or directory [12:59] iffraff: try "find $HOME -not -user $USER -ls" which may show you've got files owned by root which your user therefore cannot change which are needed for the GUI, e.g. $HOME/.Xauthority [12:59] TJ-: Yep. When the counter stops I'm pretty sure that: 1. the userspace program is still sending packets. 2. The pid of the program is still located in the netcls.tasks. 3. the classid is still set. 4. the iptables rule is exist. [13:00] ?? [13:01] Tj seems I own everything there. Xauth is in my user name [13:02] someone? [13:02] Apt says unmet dependencies try --fix broken. But that doesnt work [13:02] mircx1, what did you do when you got that error? [13:03] nobody can answer your question with such sparse info [13:03] install ssl [13:03] !details | mircx1 in a pastebin please [13:03] mircx1 in a pastebin please: Please elaborate; your question or issue may not seem clear or detailed enough for people to help you. Please give more detailed information; for example, we might need errors, steps, relevant configuration files, Ubuntu version, and hardware information. Use a !pastebin to avoid flooding the channel. [13:03] So I can't uninstall the broken packages and I can't updat [13:04] Upgrade the apt. Not sure if that's related to my login problem or not [13:04] !pastebin [13:04] 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. [13:04] iffraff: dependency issues you get mostly when adding external ppa's and they conflict with apt [13:05] iffraff: right, so you've cleared the most obvious and usual issue (bad ownership) so it is likely the GPU drivers, so now find the Xorg.0.log and see if it helps, also look at $HOME/.xsession-errors [13:05] K one sec [13:05] openssl is already installed, what ssl exactly? can you pastebin the install attempt with command and all errors? [13:05] https://paste.ubuntu.com/p/FhCFfc6ZT2/ [13:06] lol [13:06] mircx1: the full apt output please [13:06] no, all.. command and everything else [13:06] i try install that ssl_gnutls [13:08] https://paste.ubuntu.com/p/FhCFfc6ZT2/ this a error [13:08] again without the command used... [13:09] Tj xsesserr all looks ok. Don't understand it all but nothing seems obviously wrong [13:09] Don't know where the xorg.0.log is [13:09] Sorry for short hand on phone [13:11] iffraff: it used to be /var/log/Xorg.0.log but Gnome moved it to under $HOME and I can never recall where since I do not use Gnome [13:12] Tj there is one dir owned by root in $home. It's .21 I think. Writing is super small and blue [13:12] xorg: /usr/lib/xorg /usr/include/xorg [13:12] iffraff: ".21" isn't a core directory so ignore that, but not sure what process would create that [13:13] iffraff: try "find $HOME/.local -name 'Xorg.*log' " [13:15] how i fix it? [13:15] iMadper: I've checked the most recent mainline kernel and the ubuntu-xenial 4.4 series; the patches in that mail-list are not applied [13:16] mircx1: share your full apt & command output first please [13:16] mircx1: volunteers dont know what you are trying to do [13:16] sudo apt-get install libgnutls-dev [13:17] mircx1: ok and pastebin all the text coming after your command too [13:17] hello all, does anyone know what might be happening on the server installation filesystem setup where almost every move I make to select a disk, my selection is reset back up to the top? [13:18] TJ-: I see. And there is no similar fixes... [13:18] i send it [13:18] iMadper: I'm currently checking for that [13:18] https://paste.ubuntu.com/p/FhCFfc6ZT2/ this a error [13:19] this is infuriating, the file system setup screen is absolutely unusable. I checked the whole disk for errors and it reported good. [13:19] Tj right so I have 0 and 1. Looked at 1. Tone of stuff but at end a lot of systemdlogind releasing I'd for some number [13:20] iMadper: in 5.3-rc7 (latest mainline) "gitlog --grep CONFIG_NET_CLS_CGROUP" only reports 3 commits matching and they're all 2013 or earlier so nothing stands out [13:20] TJ-: It's a p2p network programming. New connections create and destory fastly. So the new alloced skb doesn't contain the classid as well. Or, all skbs that contains the classid is not captured by netfilter. [13:21] TJ-: Yes. There is no much classid related patches at all... [13:21] iMadper: could it be a race condition in that case? [13:21] TJ-: No idea... I'd try to check iptables related patches... [13:21] mircx1: lets see pastebin of: lsb_release -a && uname -a && sudo apt update && sudo apt full-upgrade [13:23] Tj actually in .0 file, at end there is a sent fault error at 0x0 not sure if that helps [13:23] * iMadper is creating compile_commands.json for ccls now... [13:23] iffraff: if you have the path to the file you can share it with us by doing "pastebinit path/to/file" and telling us the URL [13:24] TJ-: Thanks for your kindly help. It's too late in my timezone. I'd check it tomorrow. I really appreciate your kindly help. :-D [13:25] * iMadper mostly because it will take too much times to compile the kernel with my poor computer... [13:28] iMadper: in 5.3-rc7 drivers/net/veth.c::veth_xmit() there's a stanza that increments the counters, which includes "stats->packets++;" that only triggers with if (likely(veth_forward_skb(rcv, skb, rq, rcv_xdp) == NET_RX_SUCCESS)) { ... so in current mainline at least, counters would stop if veth_forward_skb() failed [13:28] wow I thought the problem was only with the 18.0.4.3 installer but the same exact issue plagues the 19.04 installer. the filesystem setup screen is utterly broken and unusable how is this possible? [13:28] iMadper: have to see what the 4.4 code does in that same place [13:29] vbgunz: bad GPU config [13:30] TJ-: I tried the default and safe graphics and only on the filesystem setup screen is this a problem with constantly resetting my selection [13:30] vbgunz: I presume you're (trying to) use the GUI server installer not the traditional debian-installer TUI based one? [13:31] vbgunz: All I've heard about theat subiquity (server) installer is bugs and problems [13:31] I can make selections on all previous screens just fine. I downloaded ubuntu-18.04.3-live-server-amd64.iso direct from ubuntu and tried both default and safe graphics [13:32] iMadper: in 4.4.0-135 the same basic code is there, triggered by if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) { [13:32] it's just unusable, impossible to get pass this screen [13:32] iMadper: so without further details we'd have to presume the packets are no longer being forwarded successfully [13:34] TJ-: Wow! I need to read more context to understand how it works... [13:39] TJ-: I'm afraid the bug will ruin my weekend... Orz... [13:39] Tj got dropped last I read was pastebin doc.. [13:39] My response was. [13:39] Tj, I'm sorry but I'm irc on phone and terminal on actual laptop. So I don't think I can do that. If you give me an idea of what to look for I can dig.iffraff 08:35:00The context of issue is that I have a laptop with onboard Intel you, and Nvidia gpu. I added an egpu and raedon and tried to install drivers but the drivers were only for 18.04 and I' [13:39] m on 19.04. And now cat login and can't apt upgrade due to missing dependencies === iMadper is now known as ImAdPER|SLeePINg [13:40] ImAdPER|SLeePINg: Here's a list of the commits in 4.4.0-135 mentioning "classid" https://kernel.ubuntu.com/git/ubuntu/ubuntu-xenial.git/log/?h=Ubuntu-4.4.0-135.161&qt=grep&q=classid [13:41]