[00:06] Can anyone explain why installing nvidia-driver-410 would cause linux boot to hang with this message https://i.imgur.com/ytRLcYL.jpg [00:22] trying to make kitty show up as my default terminal. how do i get it to appear as a choice for sudo update alternatives? 2) after creating a .desktop file for kitty how come it doesn't appear on launch menu? i already ran sed -i "s/Icon\=kitty/Icon\=\/home\/$USER\/.local\/kitty.app\/share\/icons\/hicolor\/256x256\/apps\/kitty.png/g" ~/.local/share/applications/kitty.desktop === wodencafenvy is now known as wodencafe [00:28] also; when i run ctrl + alt + t i no longer see gnome terminal [00:35] did you issue update-desktop-database ~/.local/share/applications ? [00:43] Hello guys [00:44] Do anyone knows how to recover a broken install when recovery mode fails? [00:44] I tried to update nvidia drivers on my laptop [00:45] Like, how could I repair packages from a live usb? [00:47] Gr1zzly: Tried booting into recovery mode and purging the Nvidia driver ? [00:47] yeah, won't boot [00:47] pc hangs in without error [00:48] Gr1zzly: Try then booting a liveUSB to the boot menu and choose " boot from first hard drive" . [00:49] k === [1]MrMobius is now known as MrMobius [01:36] I am a bit confused with package naming, I need "libgcc++:i386", on one VPS it's available (linode.com) on another, it says cannot find any package by this regex (scaleway). Both are bionic(18.04). What name should I use ? [01:37] kinda can't figure out why same distro on the different hosting services would have different package name [01:44] X-Kent: GCC support library ? [01:45] !info libgcc1 [01:45] libgcc1 (source: gcc-8 (8.2.0-1ubuntu2~18.04)): GCC support library. In component main, is required. Version 1:8.2.0-1ubuntu2~18.04 (bionic), package size 46 kB, installed size 131 kB [01:45] !info libgcc2 [01:45] Package libgcc2 does not exist in bionic [01:45] Bashing-om, yea it seems so, not sure why on some systems the anem is different [01:45] s/anem/name [01:46] that's the gcc lib for i386 when the host is amd64 I think [01:49] X-Kent: There are a *LOT* of those libraries. see: ' apt search libgcc1 ' . [01:51] Bashing-om, the naming is a complete mess, no idea why but on that problematic vps, the package "libstdc++6:i386" worked and provided what I required [01:52] X-Kent: :) .. You the better man than I . [01:53] X-Kent: only reason is there are i386 programs needing it [01:53] hggdh, yep, i use "aapt" and it requires it [01:53] !info aapt [01:53] aapt (source: android-platform-frameworks-base): Android Asset Packaging Tool. In component universe, is extra. Version 1:7.0.0+r33-1 (bionic), package size 413 kB, installed size 1440 kB [01:54] then aapt is a i386 binary, I guess [01:54] yep it's a 32bit one [01:55] interesting. On one instance I have (Bionic) apt-cache policy returns it as a amd64 [01:56] hggdh, I use manually built aapt that is 32bit, "legacy" reasons. [01:56] ah [01:56] hope that libstdc++6:i386 will be available on both VPSes [01:57] then yes, it might require the i386 gcc libraries === amos1 is now known as amosbird [02:06] figured it out, the problematic vm should have "sudo dpkg --add-architecture i386" then it finds the i386 packages [02:16] I'm seeing an error similar to this one : https://hastebin.com/iperabonih.sql at boot time before it flashes away into the system. any insight would be lovely as there's nothing similar in windows [02:16] syslog also tells me that my bios is broken ... === khronosschoty is now known as not_dbrown_I_swe === not_dbrown_I_swe is now known as khronosschoty [03:26] Can anyone guess why installing nvidia-driver-410 would cause linux boot to hang with this message https://i.imgur.com/ytRLcYL.jpg === stormbard is now known as stormbard_away === stormbard_away is now known as stormbard [03:33] JustAPerson: hmm, we had a user with that driver version into boot problems aswell [03:33] JustAPerson: where did you get the driver? [03:43] I just installed kubuntu alongside Windows but there wasn't an efi entry created in bios, what do i need to do to be able to select os to boot to? === capellaDies is now known as capella|away [04:21] Heya guys, I have a problem with connecting my Magic Keyboard 2 A1644, the GNOME Settings way doesn't work, bluez-tools way is working but have to manually reconnect it every time it looses connection and I have to use on screen keyboard on gdm to log in :/ The issue is better described here: https://community.ubuntu.com/t/apple-magic-keyboard-2-a1644/8591 [04:38] Ok, I believe that my Kubuntu live installer installed in MBR mode, is there an easy way to convert it to efi so grub can take over the booting duties? [04:51] Ok, my Kubuntu installer ran in MBR, is there any way to convert it to EFI safely?' [05:02] lotuspsychje: installed from the ppa:graphics-drivers/ppa [05:02] JustAPerson: what does ubuntu-drivers list reccomend for your card? [05:05] lotuspsychje: `ubuntu-driver devices` lists nvidia-driver-390, 396, and 410 (recommended) for my newer card (GTX 1070ti) and the older binary nvidia-340 for my GTX 770Ti. I believe the older card is still supported by the latest drivers [05:05] JustAPerson: perhaps try 396 latest from the graphics ppa [05:07] JustAPerson: the user yesterday got stuck on 410 also [05:08] lotuspsychje: ah. I think I have tried 396 but I forget what goes wrong. 390 definitely causes random kernel hangs. Give me a few minutes to try 396 again [05:11] JustAPerson: your system up to date to .38 too right? [05:17] lotuspsychje: currently running 4.15.0-36-generic so let me try updating that too before restarting [05:18] JustAPerson: yeah its important, we test on right kernel [05:29] lotuspsychje: now cannot even get to the recovery prompt on 4.15.0-38 w/ nvidia-drivers-396. See this picture of https://i.imgur.com/dv4pRlQ.jpg [05:30] Trying some of the older kernels I have installed to see if they can reach recovery prompt [05:30] JustAPerson: recoverymode is before systemd loading, enter grub [05:32] JustAPerson: hold shift at power on to enter grub, or ctrl alt F1 to enter tty at systemd boot [05:33] lotuspsychje: in grub, do I need to use `e` to change one of the boot commands? [05:34] JustAPerson: no, boot commands is only for debugging, but if you cannot enter desktop you could try the nomodeset option [05:36] JustAPerson: another way to get out of black screens nvidia is: sudo apt purge nvidia* to fallback to nouveau, and enter your desktop again [05:37] JustAPerson: after that, you can install another nvidia driver version again [05:38] lotuspsychje: I'm familiar with `apt purge nvidia*` which I've had to run from the recovery root shell many times. Hence my remarking that I couldn't even reach the recovery shell [05:39] With `nomodeset` I can actually boot into my desktop [05:39] Does that disable the nvidia driver? what are the semantics of that? [05:39] JustAPerson: nomodeset, sets no mode and is just for testing purposes [05:40] JustAPerson: an nvidia card, should be running with an nvidia driver [05:40] JustAPerson: or opensource or the proper driver [05:41] lotuspsychje: `lspci -v` indicates my cards are using the nvidia drivers instead of nouveau. Am I all set? Do I just permanently add `nomodeset` to my grub config? [05:43] JustAPerson: no, as i said nomodeset isnt the best way to use your card [05:43] JustAPerson: if an nvidia driver doesnt work properly its not right [05:44] bbl breakfast JustAPerson [05:44] lotuspsychje: thank you for the help [05:44] try switching nvidia drivers till it works ok [05:44] on kernel .38 [06:26] any one here can help me with a Vinux question [06:26] ? [06:30] OERIAS: I know it's slim, but your best chance is likely #vinux - really only support Ubuntu and official flavours here. [06:39] i want to install 18.04 on my mums packard bell easynote LE, but running the live system from usb the keyboard does not work at all. under the installed windows10, it does. [06:39] any ideas? [06:50] pragomer: did you check usb is enabled in bios? === Kon- is now known as Kon_ [07:00] it works when booting with uefi..mm [07:20] HI, trying to add a nomodeset argument to grum menu, Where can I permanently add this on ubuntu 18.04 ? [07:26] tomvolek: edit /etc/default/grub and run update-grub to rebuild grub.cfg [07:27] hateball: on ubuntu 18.08 i edited /etc/grub.d/10_linux and it doesnt take affect [07:27] tomvolek: that is, sudo nano /etc/default/grub, edit the line with GRUB_CMDLINE_LINUX_DEFAULT, ctrl+x to save and exit, sudo update-grub [07:28] tx [07:29] on my grub i have few flavors of ubunut, die to a graphic card issue i need to add nomodeset to the boot linr , that will fixes my issue === russellsara08 is now known as russellsara0875 === russellsara0875 is now known as russellsara08 === russellsara08 is now known as russellsara0875 === neunon_ is now known as neunon [08:07] What is with the incessant hash sum mismatches during apt-get updates? [08:08] It's only been happening for the last couple of months. [08:26] hello, I have a question: do you know in which configuration file the chosen window manger choice is stored, for each user? [08:37] Hi, i'm doing conversion of PST file and its havy on the disk, so my system is really slow...its all stuttery [08:37] i've checked and sda and sdb (2 disk that i use) are using cfq [08:38] is there a better scheduler [08:38] sda is ssd disk; sdb is spining disk [08:38] is there a config file to change scheduler per disk [08:38] memphisto: and you're reading where and writing where? changing the scheduler is probably not the best approach. [08:39] reading and writing to same sdd disk (sda) [08:39] same file system also? [08:40] which utility do you use to convert? [08:40] readpst [08:41] i assume PST is the source format then. so what's the target format? [08:41] mbox file [08:42] i've changed shceduler for sda disk online to deadline and system seems responsive [08:42] and top shows that the bottleneck is i/o? [08:43] well top says readpst has few threads all 100% CPU usage [08:43] memphisto: okay so its actually cpu bound, not disk i/o [08:43] since its reading/writing to disk it has to be disk [08:44] well i don't know how to check that, i can see 100%cpu and readpst is causing it [08:44] htop shows cpu load per cpu core [08:45] and since its reading/writing on a disk , and i've previously noticed when copying large files i have sluggines in linux [08:45] i changed scheduler and it works [08:45] what'S the hardware, which ubuntu release is this? [08:45] how do i change scheduler per disk , persistantly [08:46] xenail, 4.15.0-36-generic [08:47] sdd disk is KINGSTON_SKC400S37256G [08:47] spining disk is HGST_HTS725050A7E630 [08:47] laptop is hpe elitebook 8570w [08:49] you can use the 'elevator' kernel parameter to set the scheduler, but i'm not sure about single storages. [08:49] that i know, but i'm interested per disk [08:50] as i have different disk types [08:51] 18.04 provides tuned / tuned-utils which can set it. [08:53] tomreyn: thanks, i'll check wheather 16.04 has it [08:53] it doesn't. i guess you just need to create a systemd service, @reboot crontab or similar [08:54] ufff [08:55] see the 2nd answer for an example https://serverfault.com/questions/612949/changing-the-i-o-scheduler-for-a-specific-disk [08:59] tomreyn: will look at it rightaway [09:00] a udev rule would also work i guess [09:00] something akin to SUBSYSTEM=="block", ACTION=="add|change", KERNEL=="sd[a-z]", ATTR{queue/scheduler}="deadline" [09:02] ideally oyou'd do it based on storage device type (ssd vs hdd), not just based on the device node sd[a-z] [09:05] tomreyn: yes, i've done it now via systemd service, but i've just found some article to do it using udev [09:05] ACTION=="add|change", KERNEL=="sd[a-z]", ATTR{queue/rotational}=="0", ATTR{queue/scheduler}="deadline" [09:06] but too late, i've already done it systemd way [09:07] alright, nice find though [09:08] memphisto: if you use udev KERNEL=="sd[a-z]" isn't so good now we have nvme devices [09:09] hmmm, could be interesting ... does an MD or LVM device show a rotational node? [09:10] hmmm, never a spinning disk around when you need one :) [09:10] TJ-: i don't have one to test :) [09:11] loop devices don't offer scheuduler, but none [09:12] memphisto: interesting conundrum that is [09:13] TJ-: my md0 does [09:13] it inherits properly, i compared ssd to hdd [09:14] just what will happen if you span an md across ssd + hdd? [09:14] yes; just checked on an LVM device-mapper node too... interesting if the underlying is hybrid [09:14] hello seniors [09:15] tomreyn: it'll show a superposition of 0 and 1 - a Quantum drive! [09:16] quantum drivography is beyond me [09:16] please help, how to stripped down boot process, for removing qemu-kvm, virtualbox, boxes from boot [09:19] adikwok: you identify the matching packages and uninstall / purge them [09:20] ok tomreyn i try it now [09:20] example: dpkg -l virtualbox [09:20] example: dpkg -l '*virtualbox*' [09:20] TJ-: tomreyn: what happens when its hybrid ? [09:20] tomreyn: dpkg -l '*virtualbox*' [09:21] memphisto: i don't know. [09:22] pff [09:22] adikwok: yes this was the proper example [09:22] it gave me list of virtualbox tomreyn [09:22] I do have one more problem not related to previous one. When comming back from suspend i get just cursor [09:23] and can't kill x; nor cant go to tty? [09:23] and its all on 4.15 kernel [09:23] 4.13 was working correctly [09:23] nvidia card but nouveau driver [09:24] adikwok: disable the hyervisor services [09:25] TJ-: im still in waiting aptitude purge virtualbox [09:25] memphisto: what happens if you suspend from a TTY ("systemctl suspend") rather than from the GUI or lid-close event? [09:26] Strange; 18.04 (server) has an /etc/sysctl.d/xxxx that contains "net.ipv4.ip_forward=1" but it isn't being set ("systemctl status systemd-sysctl") shows the service ran. How to debug!? [09:34] TJ-: can't go to tty [09:35] nothing works, just mouse coursor is moving.... [09:35] memphisto: I meant when you have normal control, switch to a TTY to do the suspend [09:35] I'm using KDE;just to mentinon [09:36] TJ-: ctrl+alt+F? doesn't do anything [09:36] memphisto: if you have another PC you could (try to) SSH in to investigate [09:36] nope [09:36] memphisto: Not from a normal clean boot GUI session? [09:37] TJ-: on normal it works, but when i come back from susspend nothing works; can't go to tty screen [09:38] is it so,a forward proxy would sit in front of clients,and a reverse one would sit in front of server? Sorry for basic question but it got me thinking [09:40] TJ-: does it have a .conf ending to not be ignored? [09:40] memphisto: after a reboot check towards the end of the last boot log ("journalctl -b -1") [09:40] TJ-: I assume it is not literally xxxx right? [09:40] TJ-: i can only powerit off forcefully [09:40] cpaelzer: it's applied manually or restarting the service [09:41] cpaelzer: but either was applied and then something else cancelled it out - but there isn't much configured and the only suspect there might be netplan/networkd [09:41] memphisto: that sounds like an ACPI firmware bug may be the root cause [09:42] cpaelzer: or, for some reason it ignored it. Let me check some more...ooohhh! it's a symlink to /usr/local/ which is a separate FS mounted later. Duh! [09:42] * TJ- kicks himself around the room [09:52] How do I install Mono SDK 5.12.0? [09:53] good morning every one. I now am an experienced user of ubuntu desktop. Now i want to learn ubuntu server. I followed instructions and i have now running a software raid ubuntu server. But im not so experienced with commands and im getting frustrated. May i install a desktop (gui) enviroment over my ubuntu server fresh install? thank you very much [09:54] it is generally not recommended to use a GUI on Ubuntu server however, you can install anything via apt [09:54] wendico: yes; you can install any packages you choose [09:54] security is not a big concern [09:54] can i just install on top of it "ubuntu-desktop" [09:55] wendico: Yes, that's the recommended way [09:55] wendico: may I recommend a book The Linux Command Line [09:55] it's free online [09:55] i also advise against installing a gui on a server [09:55] I've upgraded from 16.04 to 18.10 on my ThinkPad X1 Carbon - the way Ubuntu handles the high DPI screen is aweful. It worked fine in 16.04 but now certain apps are unusably small, and my mouse pointer changes size between apps. Any ideas of how to make it usable? [09:55] tj: thanks. leonardus: please do so [09:57] wendico: http://www.linuxzasve.com/preuzimanje/TLCL-09.12.pdf [09:57] hey peeps. a link in https://help.ubuntu.com/community/SerialConsoleHowto is a link to some scamware site [09:58] leonarduds: thank you, bookmarked [09:58] this one : hxxp://www.rajeevnet.com/linux/grub_serial_console.html [09:58] change the xx to tt if you want to get hacked [09:59] I went to all the trouble of registering with ubuntu-one only to find the page is immutable. sadface. kthnxbye [10:01] mlh: you can easily join the wiki editors team [10:04] Hello is there a way I can remove a static route using Network manager? To be more specific I have two routes like this https://pastebin.com/9HyNm2H0 , and want to remove the route from device eth0 [10:06] afx_, was this route created via networkmanager? [10:06] I tried to create a new static route for this device with higher metric (like this https://pastebin.com/ZGe2kcCt) but , the gui does not let me apply the changes [10:06] Deknos, no these routes were created by default [10:08] afx_: those look like the default routes for each interface [10:09] TJ-, correct . I would like to remove the default from eth0 and let only the specified in the network manager [10:10] afx_: in the NM Connection Editor, on the IPv4 tab, in the Method selector choose "Automatic (addresses only)" - that will only set the IP address via DHCP. It will not set a default route nor set any DNS servers provided by DHCP [10:11] Hello. I've connected external display to Ubuntu 16.04, it gets detected by mu lap top doesn't output anything after that. How may I fix that ? [10:11] hey, I am trying to compile something with CUDA (through clang). I've never done this before. I keep getting a lot of 'cannot find libdevice for sm_30. Provide path to different CUDA installation via --cuda-path, or pass -nocudalib to build without linking with libdevice.'. What should be the cuda path for ubuntu? [10:12] afx_, after doing it with the GUI, look into /etc/NetworkManager for any changes, with these keywords you can set the attributes with nmcli as well. [10:12] fugitive: you should have some FN,key combination to make the port aktive [10:12] oh, never mind, it lists cuda < 7 and cuda > 9 as unsupported, and I got CUDA 9.1. [10:12] SimonNL: let me see [10:13] TJ-, Unfortunately I have to use static ip not DHCP [10:13] afx_: so you've got eth0 in NM set to "Manual" ? [10:13] fugitive: one of them om top row should show something that looks like a monitor [10:13] TJ-, yes [10:14] actually, there is only underscore ``_'' on laptop screen. Just noticed it [10:14] and black screen [10:14] OPERATORS: mlh LINK is a spam, malicious web page. please check and ban [10:14] fugitive: laptop brand and type ? [10:14] fugitive: or a link to user manual [10:15] http://termbin.com/eygm , <- xorg.conf ; Lenovo ThinkPad W530 [10:16] And I am using KDE, if that makes some difference.. gpu - GK107GLM [Quadro K1000M], but i am using only nouveau atm [10:16] afx_: in which case just don't set a default gateway. In the IPv4 tab, press the "Routes" button and check the option for "Use this connection only for resources on its network" [10:16] Deknos, NetworkManager/system-connections there are compiled files , can't view the files [10:17] afx_: I agree entirely that option os so badly described when 'everyone' knows the term to look for is "default route" [10:19] I can't get Ubuntu to run on my laptop on a non-unity version. It was always happy on Unity (16.04) but since updating to 18.10 I simply can't get high DPI scaling to work. Is there anything I can do to get Java applications scaling with everything else? [10:21] TJ-, "Use this connection only for resources on its network" , will only let me if set to DHCP . pfff [10:22] tomreyn: sorry for late reply. i tried to apt purge . and reboot could not finished. i checked my X11 is gone [10:22] fugitive: If I'm not mistaking it's FN+F7 [10:23] tomreyn: apt purge virtualbox. i meant [10:23] afx_: really? in which case I don't think NM is setting the default route for eth0; Either that value has been sitting around for a while or something else is affecting it. If you disconect the eth0 connection does "ip route show" report that default route via eth0 or not? [10:23] fugitive: might be smart to read your manual yourself. [10:24] hello, my ubunu 18.04 stopped asking for LUKS password, any ideas how can i diagnose/enable? [10:24] tried adding cryptedevice to kernel parameters with no luck [10:24] SimonNL: nop, didn't worked. The monitor seems turned on, as I see underscore in top left corner. I believe it is related to the Xorg.. [10:25] TJ-, no the route is not there with eth0 disabled. [10:25] tomreyn: now virutualbox has gone from boot process. [10:26] fugitive: Oh ok so the monitor does have a signal I assume. [10:27] fugitive: can't help for the xorg, no knowledge [10:27] TJ-, With DHCP , static routes and 'Use this connection only for resources on its network' on eth0 , i get the desired static routes, besides that I have to use a static IP too. [10:27] bb asap [10:27] TJ-, I think I ll have to use a script ... [10:29] TJ-: do i have to disable the hypervisor.service too? [10:31] TJ-, Ah wait the default gateway on the manual setting was the problem . By deleting the default gateway (adding only static ip /netmask) , I can save the configuration . Let me check if I have the desired connection.. [10:32] TJ-: how to setting xorg.conf to have refresh rate? .. my resolution is 1280x800 N/A . i ever saw it had refresh rate when i tried install liquorix kernel. when i back to mainline kernel 4.19 . refresh rate is gone. and youtube laggy. especially when running hd. [10:33] afx_: What you're wanting to do is something I do frequently; so I'm not sure why you're having problems [10:33] TJ-, hmm [10:35] TJ-, me settings are: Manual ip (ip address, netmask) , dns, routes automatic (disabled) , 6 static routes configured, Use this connection only for resources .. (enabled) [10:36] afx_: screenshot from another PC http://i.imgur.com/bGd9lMn.png [10:36] afx_: if there is still a default route being set, then its by something else [10:36] TJ-, can I pm you ? [10:38] afx_: I disable PMs; support is in-channel [10:39] afx_: this link shows it set: http://i.imgur.com/yEdMm8s.png [10:39] TJ-, no problem [10:40] Hi folks [10:40] so you have selected "use this connection..." , you have NO routes , you have NO gateway set [10:40] TJ-, also I don't have "Ignore automatically obtained routes" . I am on 18.04 [10:41] afx_: Yes, this is 18.04 [10:42] TJ-, in gnome desktop I don't have this setting [10:43] afx_: does you user have sudo/adm group membership? [10:43] afx_: I'm wondering if options are restricted for unprivileged users [10:45] TJ-,https://imgur.com/a/908k5DU . Once I remove the default gw , the "Add" button gets green [10:47] TJ-, my account is sudoer [10:50] afx_: looks like gnome dumbed-down the standard network manager connection editor! I don't use gnome so didn't realise they changed it so much [10:51] TJ-, like I said , with GW removed from manual ip configuration I get the desired static routes [10:52] and I think defining static routes and gw might be redundant [10:52] after all I get the desired result :) [10:52] let me restart and check connectivity === tinyhippo is now known as spookyhippo [10:53] I will log in #ubuntu from my phone for this [10:53] brb TJ- === spookyhippo is now known as spockyhippo [11:05] hello, dificult question. I have a label printer conected to a windows machine. The support webpage got only drivers for windows and mac, i dont find the printer on the list if i connect the printer to an ubuntu machine (only usb connection available on the printer) i can keep my windows machine to use the label printer, but i would like to print from ubuntu. What would be the easiest way to make my usb windows connected printer available for my ubuntu to [11:05] print? thank you for advice. My printer http://www.dymo.com/en-US/labelwriter-450-twin-turbo-label-printer#tabContainer [11:07] I played long time with the printer connected to ubuntu and i could make it print, but format goes out of label margins i could not make it make a correct print [11:08] printer is shared and i can print on the network from other windows machines [11:08] but when i add printer, even givin the machine IP address ubuntu cant find my printer [11:09] my ubuntu machine can find other printers on the network wich work perfectly (wifi and network different printers working) im just missing my label printer [11:09] TJ-, it works :) [11:09] Thank you very much for any clue or tip you can give me [11:10] TJ-, really appreciate your time [11:10] will be afk for 30-45 min [11:14] Found this: http://www.dymo.com/en-US/dymo-label-sdk-and-cups-drivers-for-linux-dymo-label-sdk-cups-linux-p--1 but it is an empty webpage lol [11:16] wendico: did you download and install the drivers from dymo.com ? [11:24] !info printer-driver-dymo | wendico have you installed these? [11:24] wendico have you installed these?: printer-driver-dymo (source: dymo-cups-drivers): printer driver for DYMO label printers. In component universe, is optional. Version 1.4.0-8 (bionic), package size 53 kB, installed size 170 kB === spockyhippo is now known as spookyhippo [11:29] QUIT === kite_ is now known as james === james is now known as Guest56265 === Guest56265 is now known as fff === fff is now known as kite_abcd === Butterfly^ is now known as Guest18148 [12:05] is there a good option for keeping a python script running basically all the time, restarting if machine reboots.. IF i dont have access to sudo, which means no systemd afaik [12:05] Nelluk: systemd unit (for the user) [12:06] i feel like i read somewhere that 16.04 doesnt use systemd at the user level but i cant find that again [12:07] Nelluk: hmmm, easy to check "ps -efly | grep 'systemd.*--user' " [12:08] yeah looks like its running [12:08] Nelluk: in which case see "man systemd.unit" :) [12:08] im looking at this guide which indicates i'd need a sudoer to enable-linger for my user https://www.brendanlong.com/systemd-user-services-are-amazing.html [12:09] Hey how can I get GRUB to not show up at all (unless I hold down a key like ctrl or shift) [12:09] Nelluk: UNIT FILE LOAD PATH Table 2 [12:12] lantizia: set the timout to zero [12:12] Nelluk: if you cannot enable login linger, how about using tmux or screen? [12:13] gpunk, doesn't work [12:13] gpunk, apparently there is some weird conflict if your dual booting [12:13] did you lunch update-grub ? [12:13] and it sets it to 10 [12:13] yeah [12:13] tbh i'm using 'GRUB Customizer' which does it all for you [12:13] after modifing which file ? [12:14] read/edit /etc/default/grub [12:14] yeah you don't know - fair enough [12:14] you are very polite [12:14] your welcome [12:14] well i need to talk to someone who understands the actual issue! [12:15] lantizia: usually you would set "GRUB_TIMEOUT_STYLE=hidden" in /etc/default/grub [12:15] yeah it its - but it doesn't do it [12:16] another guy has the same issue here ...https://askubuntu.com/questions/1036091/how-to-set-grub-timeout-to-0-on-ubuntu-18-04 [12:16] lantizia: that's due to /etc/grub.d/30_os-prober [12:16] hidden doesn't work if you are using os-prober [12:16] lantizia: when you've other OSs it changes [12:16] i'm about to make a manual entry for windows instead and tuen os probe off [12:17] * lantizia reboots [12:17] lantizia: the key part you can probably affect in config is this test: " if [ "$quick_boot" = 1 ] ..." [12:17] lantizia: so, find out how/what sets quick_boot and set it to 0 [12:42] does 18.04 server have significantly higher requirements than 16.04? (ie if im spinning up a cheap VPS with 512mb of ram, should 18.04 run fine?) [12:43] https://help.ubuntu.com/lts/serverguide/preparing-to-install.html.en [12:43] depends on what you are doing with the vps [12:44] not much, some python scripts, maybe host a simple django site [12:44] then i would recommend upping it to 1gig if you can just to be on the safe side === emerson is now known as GDPR [12:47] Hey guys I'm switching to Ubuntu 18.04 from and old version of Ubuntu (15.10). In the installer if I select erase 'Erase Ubuntu and install' will that take care of using the same home and swap partitions or do I have to manually do that with the ' else' option. I really don't want it to wipe my /home. [12:48] no that will erase everything [12:48] ddoobb: it will delete everything on the selected target [12:49] Alright glad I asked here before going ahead [12:49] ddoobb, no need to to wipe ./home , just set the mountpoint for it and use manual partitioninmg [12:50] When I select the / partition that Ubuntu will be upgraded in, does it matter if I don't select 'format this position'? [12:50] Partition [12:50] it will format it anyway [12:51] best to do so [12:51] Alright thanks guys [12:51] ddoobb: first of all, you really need to have a backup before doing this installation (and in general). [12:51] there is an option to not to ? [12:51] is /home a partition ? [12:51] gpunk yeah there's a checkbox when manual partitioning and you can leave it turned off [12:52] gpunk yeah my/home is a separate partition [12:52] usw manual in oreder to set your /home partition mountpoint [12:52] ddoobb: yes please backup your important data files prior to installing [12:53] I live life on the edge [12:53] hi all i was hoping someone could tell me what the latest and greatest kernel is and if its a good idea to upgrade mine i have 4.15.0-36-generic-Thank you [12:53] ok you'r good, but better do backups ... [12:54] just make sure you don't choose to format /home, ddoobb [12:56] conjo: Are you using 18.04? you can install the 4.18 kernel from cosmic via bionic-proposed: linux-image-generic-hwe-18.04-edge [12:57] i am thanks for the reply [12:58] Vähintään tarvitaan se yhdistyksen yleishallinto. [12:58] Sorry - wrong channel. [13:17] If I e-mail an ICS meeting appointment to someone in a different timezone, will it convert to their time? [13:22] Iarla: I'd hope the time is stored in UTC === bs_ is now known as bs [13:23] can anyone tell me why i shouldnt update kernel to 4.19 ? am running ubuntu 18.04 | 4.15.0-36 generic kernel at present [13:23] if the meeting is important send a test appointment and verify Iarla [13:24] conjo: because you might break something is one reason [13:25] to unstale ? [13:25] stable [13:25] conjo: if it isn't an Ubuntu kernel it will be missing some patches Ubuntu carries (e.g. for apparmor, and ZFS) [13:26] how to distinguish what is for ubuntu and what isnt am using ukuu [13:26] conjo: even our mainline builds don't include those patches, so things can break (e.g. lxc/lxd containers) [13:26] TJ-: I'll look into that, thanks. cryptodan it's important to me, but not necessarily to them :) But if I can't be certain, I will ask them to confirm - good idea. [13:27] where can i find a page with information on what is stable and what isnt noob friendly [13:27] and what is for what flavor etc === bs is now known as zombs [13:28] conjo: as I said earlier, kernels shipped in the Ubuntu distro and archive are the only 'stable' packages we support [13:29] conjo: is there a particular reason you feel you need to run a newer kernel? === lotuspsychje__ is now known as lotus|NUC === predmijat_ is now known as predmijat [13:33] no reason other than security possible performance improvement and bug fixes (what i have read about) [13:35] tj can you break that down a little more for me namely "and archive" whats that where do i find this archive you speak of [13:36] Linux T440 4.15.0-38-generic #41-Ubuntu SMP Wed Oct 10 10:59:38 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux [13:36] https://bugzilla.redhat.com/show_bug.cgi?id=712019 I'm getting this bug exactly but it says that it was closed upstream already like 4 years ago. How do I fix this? [13:36] bugzilla.redhat.com bug 712019 in kernel "kswapd0 using 100% CPU" [High,Closed: upstream] [13:36] im serious i am learning linux not well versed [13:38] conjo: http://a.co/d/2vqRrik that book helped me get started [13:39] conjo: as for your kernel, the kernel available in the LTS releases of ubuntu (18.04) are kept up to date with security patches for 5 years. You're good. [13:39] cheers mate will check it out [13:41] so just to clarify if i run sudo apt-get update && sudo apt-get dist-upgrade -y my computer will be updated as much as need be and all good for 5years cus its a lts version [13:41] correct [13:41] thank you [13:44] and looking up support like https://forums.opensuse.org/showthread.php/487620-Kswapd0-high-CPU-usage/page2?s=15cc53dc51d7e71af549c2ff138c15c3 shows no answer [13:47] How can I toggle the terminal using a keybind? For example I want to toggle the terminal using ctrl+space but I also want to minimize the terminal using ctrl+space [13:50] The issue only exists for me for KDE Neon and Ubuntu. In Manjaro, arch and fedora it works fine. It's super annoying because I can't play games [13:52] Does anyone have experience with Ubuntu on a Dell Precision 7710? I have issues with hardware hang (but no kernel panics) and it seems to be video related. Upon sleep, detecting external monitor, suspend, shutdown, the system hangs and has to be manually shutdown by holding power button. [13:56] bfdi533: sounds like firmware ACPI bugs; see/try http://iam.tj/prototype/enhancements/Windows-acpi_osi.html [14:00] I had read something about ACPI and someone else said to put "acpi=force" into my grub config which I did but it is not any better. [14:02] bfdi533: does "sudo shutdown -h now" power off the system or does it halt it? [14:03] I just tricked a "microsoft-helpdesk-representative" that called me into talking with me for seven minutes explaining to him that i was system administrator, and that I do not run windows. in the end I just started to let him ask his questions. [14:04] When He asked me what I saw on the background of my computer I said that I see the emblem for Interpol on the background of my dekstop.. [14:04] lapion: you got a scammer [14:04] the very second actually [14:04] not a real microsoft tech [14:04] lapion: The topic belongs in #ubuntu-offtopic please... try to keep things for Ubuntu support questions and solutions here [14:05] bfdi533: acpi=foce wouldn't help since ACPI is already enabled :) [14:05] cryptodan, I know... [14:07] I have tried the acpi_osi.sh script and will see how it goes ... [14:07] Certainly the article describes my behaviour. === w0dencafe is now known as wodencafe [14:25] I'm not sure what to google! in osx when I use super/alt/ctrl +left/right I can easily navigate sentences and paragraphs. I'm having trouble w/ the equivalent in gnome firefox google docs === Class7_ is now known as Class7 [14:29] falieson: there might be something about gnome shortcuts. [14:30] https://help.gnome.org/users/gnome-help/stable/keyboard-shortcuts-set.html.en ? [14:33] Hi all, I [14:33] I too [14:36] I'm planning to reinstall my laptop with Ubuntu 18.04, and I'm looking for a suggestion to achieve some disk redundancy. I've a 250 GB SSD, and a 1 TB HDD. The SSD is not a top quality one, so I want some kind of (async) mirroring of its data to the bigger HDD, and the rest 750 GB is for not that important data. Is this mdadm the only solution to achieve this: https://askubuntu.com/a/567765/6341 - what do you think [14:44] antivirtel: How about using the SSD as one large bcache device in writebac mode in front of the HDD? see https://wiki.ubuntu.com/ServerTeam/Bcache [14:49] Hm, this sounds great TJ- - do you think writeback mode is safe enough? Is it supports home encryption? [14:50] I have a logitech m705 mouse on ubuntu 18 and the wheel scroll speed is very fast. Are there any tools to make adjustments to this? Not available in standard settings -> mouse [14:53] geodb27, thanks! thats not it but I do have some ideas [14:59] meh :-/ using the new gnome for a day was nice and fancy but I'm going to have to go back to fluxbox. 15-years and nothing comes close to being as simple and powerufl [15:05] Here's my fluxbox keyshortcut file incase anyone wants to have an idea of how easy it is to setup: https://gist.github.com/Falieson/090af79f734371e64daeb75f989756a3 === ianychoi_ is now known as ianychoi [15:09] Anybody an idea how to minimize the memory consumption of zfs? arcstats shows over 33 GB of RAM usage which is not acceptable. === stormbard is now known as stormbard_away [15:12] has anyone here used macOS? do you prefer it or ubuntu? === stormbard_away is now known as stormbard [15:13] blackandblue: This is a support channel. Go to #mac or #ubuntu-offtopic for such opinions [15:21] NTQ: That seems to be a bit of an advanced question... Do you know of a ZFS specific forum or channel that might help? [15:21] pragmaticenigma: I think I can find a channel for that. [15:25] NTQ: i think #openzfs might work [15:26] pragmaticenigma: I am in #zfsonlinux for now === capella|away is now known as capellaDies [15:37] I'm trying to rescue a hobbled system. I can run startx, and the WM comes up, but nothing responds to input. The only error in the log is about systemd-logind, which is expected, since I ran with init=/bin/zsh [15:37] I can switch to another VT and kill the WM, which then exits X [15:38] d-rock, CTRL ALT F2 [15:38] This is the only error: [ 3286.010] (EE) systemd-logind: failed to get session: PID 24048 does not belong to any known session [15:39] thechitowncubs: that doesn't work. I have to SSH in from another host and kill -9 twm [15:40] d-rock, ps aux | grep twm [15:40] kill the pid of twm [15:41] Yup [15:41] I can do that all day long. I'm trying to figure out what the problem is [15:41] maybe check X log [15:41] The only error I see is above [15:42] Trying an strace on twm in xinitrc now [15:42] Huh. strace produces nothing [15:42] Literally an empty file :( [15:44] d-rock: if you're trying to rescue it, what from? it sounds like logind is relied on by what you're using, which would make sense if systemd isn't running [15:44] d-rock: probably because the underlying tooling relies on dbus to do IPC [15:44] systemd and udevd are borked, but otherwise the system is running fine [15:44] All of my services are up and running [15:45] I don't have the time right now to reinstall or figure out what the deeper problem is. I just want X up and running [15:45] d-rock: I'd tend to focus on fixing udevd - if that's broken there'll be a lot of knock-on effects [15:46] TJ-: You and I discussed this last week. My udevcookie isn't getting decremented, which then prevents everything else from starting [15:47] d-rock: oh, that one! crickey. did you determine the problem is definitely happening at the initramfs stage? [15:47] I booted with init=/bin/zsh, and /proc/mdstat says all of my arrays are active. vgchange -ay brings up all volumes properly, but then hanges waiting on udev to complete the cookie. udevadm shows add and change events for all of the block devices [15:48] d-rock: this was after moving NICs etc wasn't it? [15:48] I removed a NIC and enabled SMART on two disks in my BIOS [15:49] I undid those changes (put the NIC back in, disabled SMART), but it still hangs on boot [15:49] Not that either of those should cause problems [15:52] TJ-: when you say "definitely happening at the initramfs stage", I'm trying to find in the chat log what we were discussing [15:52] init=/bin/zsh comes up fine, along with all block devices [15:52] d-rock: agreed. Have you investigated /lib/udev/rules.d/95-dm-notify.rules (possibly that is copied into the initrd.img too) [15:53] Yeah, I looked there. Honestly, not completely grokking rules [15:53] d-rock: I was telling you that udev operates in 2 stages; once initially in the initialramfs, then again once the pivot to the real root file-system and systemd has started as PID 1 [15:54] d-rock: because of where you report the problem it seems to be stuck in the initialramfs [15:54] I was able to do an apt upgrade, which pulled in a new kernel and rebuilint initrds [15:56] Honestly, at this point I really just want to rebuild the system, but I'm slammed with work for the next 2-3 weeks and was just trying to get X up and running to get something done in the interim [15:57] It does come up. I see my desktop and everything. I just can't move the mouse or type anything :( [15:59] I've been using Ubuntu since Breezy and this is by far the most frustrating experience I've had [15:59] Sorry, shouldn't vent [16:01] d-rock: You can, even I am having issues with tmux on Ubuntu 18.04 === stormbard is now known as stormbard_away === stormbard_away is now known as stormbard === stormbard is now known as stormbard_away === stormbard_away is now known as stormbard === stormbard is now known as stormbard_away === stormbard_away is now known as stormbard [16:04] d-rock: from re-reading my log of our conversations it feels to me like this is an LVM issue. no input devices is probably due to udevd not running, because I think the event sub-system relies on it to discover input devices [16:08] d-rock: you've got a single VG with multiple PVs on 3 MDs I think, (1 MD = SSD, 2 MDs = 2 x 2HDD). MDs seem to come up fine, but VG doesn't. That tells us there's a problem with one or more of the PVs [16:15] How would that manifest? vgscan shows everything OK, pvscan shows three volumes. vgchange brings all volumes online, but then hangs waiting on udev to complete the cookie [16:15] fsck on all volumes is clean [16:18] I can not boot live flash drive on my windows 10 machine, I get the following messages [16:19] Failed to open \EFI\BOOT\mmx64.efi - Not Found Failed to load image \EFI\Boot\mmx64.efi: Not Found Failed to start MokManager: Not Found [16:19] Any ideas? [16:21] is there any official documentation and best practice for using network namespaces on ubuntu 18.04? [16:23] does netplan support network namespaces? [16:25] I have a bizarre problem on my laptop, which happened to me this morning. I locked my machine, unplugged it from the dock, and took it to a meeting. In the meeting room, I opened it up, and got the login prompt twice. And my session from earlier was gone [16:25] This happens to me fairly frequently [16:26] What should I do? === SimonNL is now known as SimonNL_Afk [16:29] It's very annoying. I was all set to demo something, but had to scramble to get all the pieces up again === capellaDies is now known as capella|away [16:39] Don' [16:40] Some_Person: Recommendation, don't change power states when you are not logged into an active session [16:41] pragmaticenigma: So I should not lock my machine before going to a meeting? Just pull the plug and go? [16:41] Some_Person: Pull the power, then lock [16:42] Pulling the dock means I lose my keyboard and monitors, so that doesn't really work [16:42] I leave my laptop closed at my desk [16:42] Some_Person: Just found your problem [16:42] ? [16:43] I don't see how I'm doing anything wrong [16:47] Some_Person: Ask your colleauges how they do it [16:49] pragmaticenigma: I did ask one. He says he does basically the same as me, but doesn't have this problem [16:50] Some_Person: Then you should bring that up to your IT department [16:53] so. i upgraded a machine to 18.04 LTS and it went well. rebooted just fine when it was done. now, do i dare doing it with a machine that is 300 km away that i can't access physically for some time? [16:53] no [16:53] thank you for your concise and straight-forward answer, pragmaticenigma [16:53] :p [16:54] quem: Sorry, it's reflexive... I personally never remote update/upgrade any machine without someone standing by at the remote location to monitor it [16:55] yeah, my intuition tells me it's a bad idea. but there's an itch that tells me the opposite, too. [16:55] pragmaticenigma: I have done so. We looked through syslog and a bunch of other things, but were unable to pin down a root cause. Best we were able to tell, it seems like the X server is going away for some reason, but it's not clear why. [16:55] Unable to use tap and drag gesture of touchpad on Kubuntu 18.04. I am able to select the icon but it doesn't get dragged. I am able to use all other functions of touchpad. Please help. [16:56] i could ask my mother to stand by and monitor it, but i doubt that would be of much use. [16:56] quem: As long as the machine is getting it's security fixes, it will be fine until you can get there [16:57] yeah. i suppose so. i should really just remove the upgrade reminder from the login motd. [16:58] pragmaticenigma: hence why I'm here [16:59] that's our company's IT at least... I'm not going to bother going to the enterprise's IT since they're incompetent [17:04] Some_Person: There is no solution to your problem that could be addressed via this channel. Other than the suggestion I made earlier, to open the laptop screen while docked, then remove the laptop from the dock. Or don't lock the machine, remove it from the dock and open the lid [17:05] TJ-: when you say there's a problem with one or more of the PVs, what should I be looking for? There's nothing in dmesg that looks suspicious, pvscan shows all three PVs just fine [17:06] d-rock: did you check syslog or kern.log for any issues. dmesg is just current boot errors [17:06] Yeah, the only thing that pops up is a message about lvmetad not running [17:11] any ideas on how to adjust the mouse wheel on ubuntu 18? Not the speed of the cursor movement, the wheel itself [17:18] Some_Person: you open a bug report :) [17:18] should* [17:20] . Now even rescue mode doesn't work [17:20] I'm just going to have to reinstall [17:30] Some_Person: minimally, we'd want to see some logs here. It does sound like X is crashing so xsession_errors would be a good start [17:31] tgm4883: ~/.xsession-errors ? or do you mean a different file? [17:32] Some_Person: that's the file [17:32] Only have two lines in there [17:32] openConnection: connect: No such file or directory [17:32] cannot connect to brltty at :0 [17:32] Some_Person: that seems odd to me. Is there another file named similiar (eg. .xsession-errors.1) [17:33] tgm4883: there's .xsession-errors.old [17:33] Some_Person: anything good in there? [17:33] https://pastebin.com/TvrzkJaD [17:36] tgm4883: The file does date from the right time [17:38] Hey, I've got some trouble with systemd: I've got a couple of services that are static, so I can't enable them. These are daemons, though. [17:39] So I can turn them on, but every time I reboot, I need to activate them again. [17:39] Some_Person: sorry, something came up at work. You'll want to state a few more details. Ubuntu release and version (ubuntu/kubuntu/etc), graphics card/drivers [17:39] !systemd | joeyjo0 start here [17:39] joeyjo0 start here: systemd is the default init system for Ubuntu 15.04 onwards. For information on transitioning from upstart to systemd, see https://wiki.ubuntu.com/SystemdForUpstartUsers For a guide to basic service management with systemd, see https://www.digitalocean.com/community/tutorials/how-to-use-systemctl-to-manage-systemd-services-and-units [17:41] tgm4883: ubuntu 16.04, the machine is a laptop with both Intel graphics and Nvidia graphics, and I'm actually not sure which one is in use [17:43] @ubottu: I've checked the articles. But I've still got some trouble. [17:44] I'm running postfix and some milters. However, these milters do not start on their own. [17:45] First of all, spamassassin crashes because it's unable to create a directory in /var/run(other problem). spamass-milter does not start up, either. [17:46] joeyjo0: they're likely static since they have no [Install] section; they're designed to be dependencies of some other unit, via a .wants symlink [17:46] hi [17:46] is there a version of usb-creator that .) supports GPT .) still has the option to store data in the flash drive ? [17:47] joeyjo0: you can create those symlinks yourself, e.g. under /etc/systemd/system/some.{service,target}.wants/ [17:50] TJ-: So what if I have a static service with no wants? Would it just not start? === hein is now known as heineagle === heineagle is now known as hein [18:20] hey I have an old ubuntu 12 system, migrating away from it but I need to raise shmmax. so I found this calculation on a blog and was wondering if you guys approve of how it calculates a new shmmax value? https://bpaste.net/show/18bacaaecb1e the system has 12G of RAM and shmmax is still set to its default of 32M. [18:22] need to raise it to dump postgres so it can move to the new ubuntu system. [18:24] stemid: if you just need it to get to dump, why not make it some very large value? rather than trying to find a perfect value === Fuchs is now known as zombiefox [18:26] nacc: a couple reasons. partly because the migration is of a few systems and will likely last util xmas. to dump postgres is only one of the early steps. also it's worth doing right to learn. [18:26] until* === katamo is now known as zajac === zajac is now known as realzajac === realzajac is now known as iamzajac === iamzajac is now known as thezajac === thezajac is now known as zajacd === zajacd is now known as katamo [18:32] stemid: afaict, all it's doing is setting shmall to 1/2 of system memory and then shmmax to the same? [18:34] ok, I wasn't sure what PAGE_SIZE and _PHYS_PAGES were or how they related but 1/2 of system memory is a general rule of thumb according to the posts I've read. [18:35] stemid: `man sysconf` === SimonNL_Afk is now known as SimonNL [19:01] whats a good package for sharing one monitor with multiple servers? Is synergy still the only solution ? [19:01] how plausible/easy is it to migrate an existing, working install to a new hdd ? [19:02] latk: clone with dd or rsync [19:02] lotus|NUC: There must be more to it than that? [19:03] latk: explain us a bit what you wanna do exactly [19:03] new to virtual box, running it on Ubuntu 18.04, Win10 client, how do I share the USB drive with the virtual pc? [19:04] so I have a machine which currently has two drives, one running windows one ubuntu. The ubuntu install is on a small disk. I have a new disk now, and would like to copy my current install to this new disk, and replace the old one with it. [19:04] bieb_: check the #vbox channel mate [19:05] latk: yeah just clone it mate [19:05] lotus|NUC, my bad.. posted that in wrong channel [19:05] lotus|NUC: Hah, awesome. [19:05] lotus|NUC: Will do. [19:05] latk: or clonezilla [19:06] How can you make cp never create a symlink [19:07] I'm fed up of copying things so that I can remove them from the source directory, only to see that it creates a symlink [19:10] tomvolek: ubuntu-server? [19:10] lotus|NUC: 1 Ubunut desktop, 2 ubuntu-server [19:11] tomvolek: perhaps ask in #ubuntu-server they might have some ideas [19:11] ok tx === patr0clus is now known as s3nd1v0g1us [19:30] cp just duplicates what you tell it to, if you tell it to copy a symlink, it will copy the symlink [19:30] -L flag might interest you though [19:31] Anthaas: ^ [19:32] unless you have aliased cp to cp -s or similar [19:33] but, anyway, for full options on 'cp', see 'info cp' === stormbard is now known as stormbard_away [19:39] Good day all. [19:40] How can I adjust disk caching (Read/Write caching for different devices), in Ubuntu? [19:42] I'd like to increase a read cache on an external drive. I'm trying to play media from it but the media keeps stalling or skipping frames. It's not that big of a problem when I use a media player that caches the media for playback like mpv_mplayer but it chokes with VLC, for example. [19:43] But ultimately I need to know how to adjust the way Ubuntu caches read and writes to devices. [19:43] the app reading the file could hint that it will read the file sequentially [19:45] ...and not what various programs do for cache on their own. [19:45] Euph0ria: does it connect via usb 2 or 3? [19:46] lotus|NUC: USB 3, but that's not the issue. Even if the storage is internal or ESATA, for example, I just need help finding out how to adjust storage device caching in Ubuntu [19:47] read/write caching, that is. [19:48] Euph0ria: this what you want? https://linuxconfig.org/improve-hard-drive-write-speed-with-write-back-caching [19:48] Euph0ria: have a look at https://www.kernel.org/doc/ols/2007/ols2007v2-pages-273-284.pdf [19:49] hey [19:50] lotus|NUC: Read caching, as well as write caching. But I'll thankfully look at the resources you shared. [19:50] so I added a extra path in ld.so.conf.d/01_libs.conf in order to override some system libs [19:50] but ldconfig -p still lists the old libs [19:51] shadeslayer: for best help, explain your whole story mate, ubuntu version, what are you doing === stormbard_away is now known as stormbard [19:52] Ubuntu 18.04, and like I said, I just created a dir in /usr/lib/aarch64-linux-gnu/extralibs, and added my libs and symlinks there [19:53] and then added that path to /etc/ld.so.conf.d/01_libs.conf [19:53] shadeslayer: did you update the cache? "ldconfig -p" only prints its state [19:53] tomreyn: yep done already [19:54] and the timestamp on /etc/ld.so.cache is current? [19:54] tomreyn: yeah [19:54] "sudo ldconfig -v" will show what it does while updating the cache [19:55] Hey guys. Using 18.04 LTS 64-bit desktop. Resolution set at 1920X1080, but the screen is double that. CTRL-ALT-F2 to get to console mode makes the text microscopic. I tried some stuff from google (vi /etc/default/grub, update-grub2, etc), but it didn't make any change. Ideas? [19:57] shadeslayer: is aarch64 actually a supported architecture? [19:57] tomreyn: of course [19:58] Imma reboot, brb [20:00] I'm back. Last thing I tried also didn't work. [20:01] shadeslayer: not according to https://help.ubuntu.com/community/SupportedArchitectures , to http://archive.ubuntu.com/ubuntu/dists/bionic/ or to http://ports.ubuntu.com/dists/bionic/ [20:02] wad: whats your native screen resolution and your graphics card chipset please? [20:02] hm [20:02] oh well [20:02] shadeslayer: can you show your sources.list? [20:02] https://paste.ubuntu.com/p/9MkHvybYrP/ [20:03] shadeslayer: what do "dpkg --print-architecture" and "arch" report? [20:04] Native res: 3840x2160 [20:05] lscpi: 00:02.0 VGA compatible controller: Intel Corporation HD Graphics 530 (rev 06) [20:05] Imma reboot again, trying something else. brb. [20:07] I'm back, my latest attempt didn't make any difference either. [20:08] tomreyn: moment [20:08] tomreyn: root@localhost:~# dpkg --print-architecture [20:08] arm64 [20:09] wad: change your resolution in system settings? [20:10] I did that, it's at 1920x1080. [20:11] I'm just wanting to get console mode to be the same resolution. It seems to be running at 4k instead. [20:13] Changes to /etc/default/grub, followed by update-grub, have no effect. Weird. [20:13] wad: what happens if you leave it on native res, and play with scaling? [20:13] I tried that, but when I'm plugged into my external monitor, it gets all confused. [20:13] shadeslayer: so it's a different architecture than aarch64, i guess. [20:13] Display config is tricky. :) [20:13] tomreyn: nope ;) [20:14] the arch triplet is aarch64-linux-gnu [20:14] wad: your 2 monitors have different res? [20:14] Yes [20:14] wad: thats tricky indeed [20:14] I know. Ubuntu does an amazingly good job, given the complexity! [20:14] shadeslayer: okay, maybe it's just over my top [20:14] ok [20:14] I think I'll ask in ubuntu-devel [20:14] someone might know [20:15] I think I'm just going to leave my console-mode resolution alone. Looks like it's more complicated. [20:17] tomreyn: thanks anyway :) [20:18] This is a laptop, with a 4k display. I plug it into an ultra-wide curved external monitor when I'm at my desk. Ubuntu remembers the right config in both cases. [20:20] wad: if that helps some, I decided to lower the resolution on my laptop (also 4k) to 2560x1440, keeping scale at 100% (until we get fractional scales) [20:23] hey all - quick question - I've got an ubuntu 16.04.5 box that Im doing some router virtualization on - on boot up - when I try and fire the vm - it tells me it cant enable hugpages - however - with no other changes - after reboot and login if I manually restart libvirt-bin and then attempt to start the virtual - it all works 100% fine [20:23] but - I have to do that manual restart of libvirt-bin every time hte box restarts - and even telling rc.local to restart libvirt-bin after a 30 second pause on bootup - doesnt seem to fix it [20:23] any ideas as to how I resolve this? [20:24] Symmetria: might your start of the guest race with the mount of hugetlbfs? [20:25] cpaelzer possibly - but Im really not sure how to fix this - its really frustrating - because it means that I cannot set these things to restart automatically on reboot of the system [20:25] Symmetria: how have you set to start the guest - via virsh autostart ? [20:25] cpaelzer - no - its a script supplied by juniper which starts them - I call that script from rc.local [20:25] that script does a wholeeeee bunch of other things in the mix - system checks - interface checks etc [20:26] wad checkout setupcon [20:26] (also sets up all the interface bridges and cpu pinning of those etc) [20:27] heh the wierd thing is - even if I login to the freshly rebooted box and do a sudo /etc/rc.local - which I have at the moment set to restart libvirt-bin - it comes up straight away - it just wont do it on boot [20:28] Symmetria: /etc/systemd/system/rc-local.service only has After=network.target [20:28] so it might really just be too early [20:28] mmm any ideas how I can make it do this *last*? like - as the last thing it does when the system boots? [20:30] Symmetria: create a drop in config for it at /etc/systemd/system/rc-local.service.d/override.conf [20:30] Symmetria: and add at least: [20:30] [Unit] [20:30] After=network.target libvirt-bin.service [20:31] not sure which target it would be for the hugepage mount [20:31] but that you'll find in search engines [20:32] Symmetria: to check the drop in then do [20:32] systemctl daemon-reload [20:32] systemctl cat rc-local [20:33] Symmetria: dev-hugepages.mount is the other one you want to add there [20:33] overall that should ensure better ordering for your case [20:33] * cpaelzer falls into bed [20:35] hello - i am looking for some guidance to get networking up and running on a virtual server (remote). right now i can only ssh into a recovery system and chroot into my server. i upgraded yesterday from 16.04. to 18.04. [20:36] hey, come back again with the same issue, would you happen to know why installing the nvidia drivers and even trying to force them in the xorg.conf won't use them? (nouveau is blacklisted) [20:37] thanks cpaelzer lemme go play a bit [20:38] mefistofeles: check the system logs; first that the kernel module nvidia.ko is loaded ("lsmod | grep nv") then the Xorg log ("/var/log/Xorg.0.log" usually) to check the Xorg nvidia driver is loading [20:38] TJ-: it's not loaded [20:38] lemme check the xorg log [20:39] mefistofeles, you may have some sort of version mismatch [20:39] hi, i'm trying to set the root home directory the same as my main users to have the same dotfiles. everything works well but everytime i run a command as root i get errors from ecryptfs, specifically "you do not own that mountpoint" and "signature not found in user keyring". how would i resolve these? i tried running ecryptfs-insert-wrapped-passphrase-into-keyring and that got rid of the keyring error but did [20:39] not persist after reboot. [20:39] [ 41.469] (EE) Failed to load module "nvidia" (module does not exist, 0) [20:39] WoC-: it's possible, I'm using nvidia's repos because I need latest CUDA support [20:40] it used to work before latest upgrade [20:40] i c [20:40] What kind of cuda app are you trying to use ? [20:40] mefistofeles: well if the kernel driver isn't loaded the org driver cannot [20:40] I restarted my pc and plymouth screen shows nothing when entering passphrase (typing it + ENTER also does nothing) recovery mode + resume gets me to a low resolution login screen, keeps on asking for password (login loop) i tried apt install -f, dpkg-reconfigure xserver-xorg not getting anywhere [20:41] mefistofeles: so, figure out why the nvidia module has failed. Do "uname -r; dkms status" - is the nvidia module for the current kernel version "installed" [20:41] WoC-: gromacs [20:42] $ dkms status [20:42] nvidia, 410.66, 4.15.0-38-generic, x86_64: installed [20:42] that's weird [20:42] and yes, that's the current kernel [20:42] "lshw -c video" configuration line only shows "latency=0" is this normal? [20:43] nvidia-smi output looks good? [20:43] WoC-: no [20:43] it says no nvidia driver/gpu found something like that, I can paste the exact msg if you like [20:43] no need [20:44] are you sure you have the right version of kernel mod for your card ? [20:45] WoC-: I don't understand the question, sorry [20:45] You may have a version of the kernel driver (mod) that is incompatible with your specific card [20:46] WoC-: oh, I don't think so, it's a GTX 1060 which should be supported fine with 410 mod version [20:46] would you mind doing a lspci | grep ^VGA and paste ? [20:46] WoC-: actually I have another machine with archlinux using that exact GPU with that driver 410.66 [20:46] WoC-: sure, one sec [20:46] k [20:47] WoC-: now I see a difference with the arch machine (which is the same machine setup) [20:48] Good :) [20:48] WoC-: lspci | grep -i vga in the arch box detects just the NVIDIA card, whereas ubuntu detects both the integrated (intel) and the NVIDIA card, is it trying to use optimus technology or something like that maybe? [20:49] where does transmission store a copy of a downloading .torrent file so to share with more machines [20:50] kjartan: hey. my old dentist's name was kjartan [20:50] :D [20:50] damn shame i had to move to a new city [20:50] he was excellent. [20:50] hello #ubuntu [20:50] I recall it being in some hidden folder [20:51] WoC-: any ideas? haha [20:52] leosemilie: find / -type f -name '*.torrent' -print [20:52] though this isn't needed because transmission is meant for using with more machines [20:52] mefistofeles, dang i have a similar setup, but i never remember [20:53] but my nvidia is ancient [20:54] WoC-: I wonder if I can disable the integrated one from BIOS/UEFI, gotta check that later [20:54] this is a desktop machine so no real need of the power saving benefits from using the integrated card [20:54] worth a try [20:54] seems like kernel 4.4.0-138 is the problem, booted with -137 and all works [20:56] gotta go afk for now, thanks for the help [20:58] Anyone know if libinput 1.12 will get backported to 18.04 HWE ? [20:59] Latest kernel 16.04.3 kernel update borks my system. It either hangs boot, or boots in a really funky resolution and I can't login via the console. (Though I can login via ssh). Booting old kernel works. So my question is: How do I make sure that the old kernel is never removed? Is it possible to mark a kernel as "don't delete even if someone runs 'apt autoremove'"? [21:00] DaffyDuck_: having a similar issue, hat is the broken kernel version? [21:01] 4.4.0-137 works, but 4.4.0-138 doesn't. I'm using amdgpu-pro [21:01] DaffyDuck_: exactly same issue here, must be a regresssion in 138 [21:02] sruli: Have you reported it upstream? [21:03] I only use Ubuntu as a workstation so I'm not quite sure were one reports such issues. :/ [21:03] DaffyDuck_: when i joined the room someone was having issues with their graphics, not sure exactly what the issue was but it was nvidia, [21:04] DaffyDuck_: just wasted 2 hours trying to get back up and running, dont have time to report bug now... lets see if more people complain [21:04] DaffyDuck_: you could manually install the package which you don't want autoremoved [21:04] While it sucks, at least I know my hardware isn't breaking -- which was my first fear. :) If others are seeing it I assume _someone_ will report it. [21:04] tgm4883: Ah, thanks! [21:05] DaffyDuck_: IIRC, there's 3 === stormbard is now known as stormbard_away === stormbard_away is now known as stormbard [21:11] i am trying to specify my question a bit more - my network is not working with netplan.io setup. i can not find any errors in the logs. after reboot running networkctl status --all gives me no output [21:21] anyone here use lenovo x1 carbon? [21:22] hi [21:22] I have added a xenial apt-repository but I keep getting messages about no release version found for xenus-pgpd and I cannot find this being mentioned in any file on the file system === Butterfly^ is now known as world [21:36] gambl0re: this is the support channel, you should ask your support question [21:41] how much gb does ubuntu os take? [21:43] gambl0re: that depends on how much you install, but I generally do a 20GB partition for / [21:43] for desktop installs [21:45] tgm4883: ok. i want to dual boot windows and ubuntu. i have 256gb hd. how much gb should i allocate for ubuntu? [21:45] gambl0re: depends on what your use case is [21:46] im gonna use ubuntu for software development only [21:46] rest i will probably do on windows [21:47] gambl0re: 40GB? [21:48] if i run out of space later, i can always resize the partition? [21:49] gambl0re: you can always resize, the complexity of it depends if you use just ext4 or luks or lvm [21:49] gambl0re: yes, I'd install it in LVM though which would make resizing a little easier [21:50] so when i install ubuntu i should create new partition as lvm? [21:50] gambl0re: IIRC there's an option during install to install it to LVM [21:50] will ubuntu boot my default? [21:50] by [21:51] gambl0re: You can set grub to boot linux or windows by default. My guess is that it would boot Linux by default after install (I don't dual boot so IDK) [21:52] tgm4883: great. how difficult would it be for me to remove ubuntu os and just boot windows? [21:53] usually the grub menu will give you a certain amount of time to choose which os you want to boot. [21:53] gambl0re: pretty easy. You could delete the parititon in windows and resize your C drive. You might have to reinstall the boot loader [21:54] tgm4883: ok how difficult is it to install bootloader? [21:54] gambl0re: you'd do that from windows [21:55] gambl0re: which is outside the scope of this channel, but IIRC it was fairly painless [21:55] I don't want to file a bug report, can someone just tell me if I'm going crazy or not or if they've seen something similar before? Used nautilus to make an SMB mount (gvfs). If I delete a folder on the smb share through nautilus with shift+delete it disappears until I refresh the view. If I use a terminal it works right away, no errors. [21:56] ....and by terminal I mean using gnome terminal, cd'ing to the mountpoint on the same client system and running rm -rf $dir [21:58] tgm4883: thanks === WoC- is now known as WoC === world is now known as Butterfly^ [22:22] hey guys what is the best player to handle a 1080p x265 video file [22:23] to be able to play it without lag and all that === dingir___ is now known as dingir [22:30] rhoks: it depends on the hardware, and if you have hardware acceleration available, [22:30] ...but start with vlc or mplayer === patr0clus is now known as s3nd1v0g1us === ShaRose_ is now known as ShaRose [22:53] this command did not work for me but it has worked for other i wonder if a slight adjustment will make it work for me, is supposed to fix brightness not working : [22:53] xfconf-query -c xfce4-power-manager -p /xfce4-power-manager/handle-brightness -keys --create -t bool -s true [22:53] Option parsing failed: Unknown option -keys. [22:56] UBuxuBU: remove the space, it is "handle-brightness-keys" not "hand-brightness -keys" [22:57] ok ill try that thanks [23:00] here is whathappeend : [23:00] xubu@xubu-NV59:~$ xconf-query -c xfce4-power-manager -p /xfce4-power-manager/handle-brightness-keys --create -t bool -s true [23:00] Command 'xconf-query' not found, did you mean: [23:00] command 'xfconf-query' from deb xfconf [23:00] Try: sudo apt install [23:00] UBuxuBU: it is "xfconf-query" (note the 'f' [23:00] UBuxuBU: and yes, if the package isn't already installed, you obviously need to install it [23:00] what should i do [23:01] UBuxuBU: it tells you: "sudo apt install xfconf" [23:01] the slider and power stuff is here it just doesnt work [23:01] even my fn key pulls up the brightness thing [23:01] it just cant talk to it [23:02] It may be the system has more than one backlight controller, and the GUI has attached to the wrong one. That is a pretty common issue due to firmware bugs in some PCs [23:03] my keyboard has no lights only the monitor [23:04] it attach to wrong one why dont i see anything getting dim [23:07] how do i query it to see the settings [23:07] the guy who said this command works said they simply have it off in the OS [23:08] he said his was set to false ans he made it true and it works [23:10] UBuxuBU: there can be multiple backlight controllers under /sys/class/backlight/ - and if that is the case, as I said, it may be the GUI is using the wrong one. Generally only 1 will work correctly, so when there is more than one, it requires disabling the non-functioning version from loading [23:11] which is the easiest way to upgrade from 18.04 -> 18.10 ? [23:13] WoC: first; do you /need/ 18.10? If you do release upgrade, then in 9 months you'll have to do it again to 20.04 [23:13] i watched the other small lights and none of them dim or brighten [23:16] SysGhost, do you know a goo guide on how to enable hardware acceleration === capella|away is now known as capellaDies [23:16] I have mpv and it lags/stutters a lot when the bitrate goes up === nshireTimeout is now known as nshire [23:20] rhoks: which GPU do you use? [23:22] SysGhost, cat /proc/cpuinfo shows me for the 'model name' - AMD E1-2500 APU with Radeon(TM) HD Graphics [23:24] AMD it is [23:26] which graphivs drivers are you using? === daniel is now known as Guest52204 [23:28] it's a rather old chip, so I'd guess amdgpu drivers aren't gonna work for it. [23:29] so it's the older radeon drivers... [23:30] Can ubuntu write to an HFS+ drive [23:31] !info hfsprogs [23:32] hfsprogs (source: hfsprogs): mkfs and fsck for HFS and HFS+ file systems. In component universe, is optional. Version 332.25-11build1 (bionic), package size 132 kB, installed size 336 kB [23:33] bray90820: Looks like one can make it so ^ . Homework required :) [23:33] SysGhost, I see. [23:33] rhoks: that chip uses Radeon HD 8000 series GPU, so either use catalyst xvba or libva mesa drivers to get hardware acceleration layers in. once that's done you have to tell your media player to use the libva output to get harware acceleration going [23:35] why it lags is because the CPU itself is too weak to do the software decoding in real time. [23:36] rhoks: remove all doubts of what card the system recognizes: lspci -vnn | grep -i VGA . [23:36] This might give a bit more insight: https://wiki.archlinux.org/index.php/Hardware_video_acceleration (I wish Ubuntu had a good wiki page on this matter. I just can't find any if there is) [23:36] 00:01.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Kabini [Radeon HD 8240 / R3 Series] [1002:9838] (prog-if 00 [VGA controller]) [23:37] I think that confirms that its using Radeon 8240? [23:37] while that wiki is for arch, the general way of doing it is the same, packages might have different naming though [23:39] indeed. Radeon HD 8240 (A rather weak GPU, but sufficent to do video playback accelration) [23:40] if you go with libva-mesa or catalyst xvba, depends on which graphics drivvers you have installed in the first place. [23:42] I'm gonna try and find a ubuntuforum or askubuntu question on the specific gpu and see how others have solved it really, cuz I wouldn't know where to start [23:43] but to be honest I might just watch my video with the stuttering, since I don't really watch that much video on this laptop. [23:44] it should be a matter of installing the right libs, and tell the media player to use it. [23:44] which and how depends on what you have installed. [23:44] rhoks: https://help.ubuntu.com/community/RadeonDriver says you use the radeon driver . What shows ' sudo lshw -C display ' on the configuration line ? [23:45] driver=radeon latency=0 [23:45] Bashing-om, ^ [23:46] rhoks: Then the right driver is installed . What one can do for optimizing is beyond my experience - I run nvidia . [23:46] rhoks: https://help.ubuntu.com/community/BinaryDriverHowto/AMD [23:47] See section 6. [23:47] TJ- sound broke in my 18.04, i did a separate install of 18.10 just to see if the sound would be working and it is, in 18.10, therefore it would be nice to upgrade so i get sound again. Not sure which method would mess up my settings/configs the least though... [23:48] There used to be a really handy tool to reconfigure the sound, but i dont think there is any such thing anymore [23:51] SysGhost, that article/guide is outdated because all 3 of those "xvba-va-driver libva-glx1 libva-egl1" are not found by apt-get [23:52] I'm googling ubuntu 18.04 libva glx [23:52] Only thing you need for your radeon is a gallon of napalm and a zippo ;) [23:55] apparantely there is only libva-glx2 available through apt-get