=== i73770K is now known as Dete === Dete is now known as DetectiveTaco [00:25] * jimcooncat hasn't been here in a long long while [00:26] defected to Mint, but I do run an Ubuntu server at Linode. [00:26] But I remember when this room had a brilliant 16 year old leader named John Dong. [00:27] !support | welcome back, jimcooncat [00:27] welcome back, jimcooncat: The official ubuntu support channel is #ubuntu. Also see http://ubuntu.com/support and http://ubuntuforums.org and http://askubuntu.com [00:28] there are other channels, maybe #ubuntu-offtopic for social chat === oer is now known as OerHeks === zbenjamin is now known as Guest51175 === zbenjamin_ is now known as zbenjamin [01:20] I installed sendmail in my local vm and unable to send emails [01:20] sudo nmap -sT -O localhost | grep 25 [01:20] 25/tcp open smtp [01:25] rexwin_: Depending on where that VM resides, you may need a local MTA like postfix to relay your email to another server for delivery. === akem__ is now known as akemhp === brendantcc is now known as totallynotbrendo === totallynotbrendo is now known as brendantcc [01:59] sometimes when I click a link, it causes my browser to crash [01:59] How do I fix this? [01:59] it's driving me crazy [01:59] when I click links elsewhere, outside the browser. [02:00] Firefox? What Ubuntu version? [02:01] HI I'm Trying to install nvidia drivers, I've followed several guides none with any luck. I've tried disabling wayland and also setting nomodeset in grub with no avail. Either way I can't seem to boot into a graphical session with nvidia drivers? any suggestions? [02:03] mathgrad: Wayland and Nvidia drivers are a work in progress. Are you wedded to Wayland ? [02:04] no [02:05] mathgrad: then at the login screen change the DE back to Xorg, and pastebin - lspci -k|grep -iEA5 'vga|3d' - . See what we have to work with. [02:07] https://pastebin.com/SdRqmHBP [02:10] Bashing-om and thank you for trying to help [02:13] I've never had a problem with the nvidia drivers before the thing I'm doing different this time is a zfs native root install on 19.04 === brendantcc is now known as nahhj === nahhj is now known as nahhhh [02:14] mathgrad: Nvidia recommends the 430 driver for this card: https://www.nvidia.com/Download/driverResults.aspx/151568/en-us. What have we now for drivers installed ? pastebin ' dpkg -l | grep -i nvidia ' . [02:15] there are no nvidia drivers installed yet [02:16] mathgrad: So, what have you done to install a driver ? OEM install is a no no . [02:18] I've followed the suggestions on https://askubuntu.com/questions/1135869/ubuntu-19-04-wont-boot-with-nvidia-proprietary-drivers-418 [02:19] I can try again and just revert the snapshot but I've tried at least twice today === nahhhh is now known as brendantcc [02:21] akemhp, firefox, chrome [02:21] akemhp, Ubuntu 19.04, occurred with LTS as wel [02:21] mathgrad: Is secure boot a factor here ? That 1st reply "should" have been effective - and what release is this as the 430 driver is in the software repository in 18.04. [02:22] No I haven't enabled secure boot. After install of the drivers I do get a pkcs#7 error spamming my shutdown log however [02:22] 19.04 [02:23] mathgrad: Back to : dpkg -l | grep -i nvidia :) [02:23] Should I try to install again? [02:23] Nothing is returned from that command [02:24] gonna try it one step at a time [02:25] mathgrad: What results ' sudo apt update ; sudo apt upgrade '? Then we see about proceeding in what direction, [02:25] I think the ppa isn't enabled yet since I used debootstrap to build this system [02:26] 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. [02:26] mathgrad: What release is this As the PPA may not be needed. [02:26] disco [02:27] One thing is if I enable the ppa and run ubuntu-drivers devices the drivers are listed. but ubuntu-drivers devices returns nothing without the ppa enabled [02:28] mathgrad: PPA not needed here. [02:28] !info nvidia-driver-430 disco | mathgrad [02:28] mathgrad: Package nvidia-driver-430 does not exist in disco [02:29] !info nvidia-driver-430 bionic | mathgrad [02:29] mathgrad: Package nvidia-driver-430 does not exist in bionic [02:30] i need help in setting up a atftpd server. thought that i understood how to use it but the client can't retrieve the file. i used atftpd --daemon - zenmap shows that it isn't running though [02:30] mathgrad: ^^ Bot is tellingan untruth >> "apt list nvidia-driver-430 >> nvidia-driver-430/bionic-updates 430.26-0ubuntu0.18.04.2 amd64" . do ' aptlist nvidia-driver-430 ' and see that the driver is there. [02:31] mathgrad: ** apt list nvidia-driver-430 [02:32] nothing [02:32] mathgrad: checking ! [02:33] nvidia-driver-430/disco 430.50-0ubuntu0.19.04.1~gpu0 amd64 [02:33] after adding the ppa [02:36] Bashing-om ^ [02:38] mathgrad: Odd that the 430 driver is not in disco. However, is the PPA source now availabale ' tail -v -n +1 /etc/apt/sources.list.d/* ' . [02:39] no, should I comment it? [02:40] I mean uncomment it [02:40] mathgrad: Appears we do need that PPA source. If it is not there we will have to add the PPA. [02:40] it's there, just commented out. I uncommented it [02:41] Should I try installing it now? [02:41] mathgrad: Ho-kay ... with secure Boot disabled - run ' sudo ubuntu-drivers autoinstall '. [02:42] how do I make sure secure boot is disabled? [02:43] mathgrad: Newer systems are EFI ... and all implement the firmware differently _ I can not advise on what your firmware is. [02:44] https://pastebin.com/HZch21ez [02:46] that was with apt install nvidia-driver-430, same message for autoinstall except it references 435 packages [02:50] mathgrad: checking the PPA. [02:52] mathgrad: 435 version should also be fine - did the driver install ? [02:52] no [02:54] mathgrad: Then pastebin the output of the attempt - As it "should" have installed- we need to look at what the system reported. [02:57] one sec, brb [03:09] ok reverted, back [03:10] mathgrad: "nomodeset" still in effect ' cat /proc/cmdline ' ? [03:11] https://pastebin.com/JsViY2Ch [03:12] No, nomodeset is not in effect [03:13] ok 435 is iinstalling [03:13] will post log in a sec [03:14] wow something wrong, it's going 26kb/s [03:17] must be the server, speed test show normal speeds [03:17] this gonna take a while [03:17] from an ubuntu mirror? ya I see that too from canadian one [03:17] yeah launchpad.net [03:18] main one is full speed for me [03:18] well, ppa.launchpad.net [03:22] oof, 55min 24s [03:24] I'm surprised it's even working though [03:26] hmm it was just that package almost back to normal sped [03:37] Bashing-om no luck, got a black screen and had to remove the nvidia driver to get into desktop [03:39] the installation did indicate though I am not in secure boot [03:40] https://pastebin.com/7kps9REh [03:41] mathgrad: No way yet to tell - what shows ' /var/log/gpu-manager.log ' ? [03:41] https://pastebin.com/63rvjJ7v [03:42] wait could it be booting to my integrated card should i disable the intel graphics? [03:43] not sure if it's enabled but I should probably check [03:47] manhnt9: Optimus ! my initial request for info (lspci -k|grep -iEA5 'vga|3d') must not have been complete :( [03:48] Bashing-om: sorry? [03:49] manhnt9: sorry -bad highlight - one I meant to address left :( [03:50] Bashing-om: have we talked before? I don't understand. [03:51] manhnt9: No we have not conversed - I had meant toa ddress "mathgrad" and I failed to observe my tab completion :( [03:52] Bashing-om: ah OK, no problems [03:55] Bashing-om: ok so disabling integrated graphics allows nvidia to display the loading screen but doesn't actually boot the graphical logic [03:55] which is better than the black screen previously [03:55] gonna try 430 instead of 435 [03:56] did you see anything in the logs that might help? [03:57] mathgrad: with nouveau - all looked good. [03:58] going to try 430 will be back shortly [04:03] Kernel driver in use: nvidia success [04:03] So 435 is borked on a 1070 [04:04] that's bashing-om [04:04] thanks* [04:04] mathgrad: thats hard to believe [04:04] 435 is pretty latest driver [04:04] I couldn't get to a graphical login with it. I couldn't even get to a login with ctrl-alt-f2 [04:04] should work nice on GTX cards [04:05] but i'm running a deboostrap install with a weird zfs setup, not sure why that'd make a difference though [04:05] the normal installation usually is no problem but this has been an adventure [04:14] mathgrad: but its working on 430 now? [04:15] yeah [04:15] mathgrad: :D All's well that ends well. [04:16] mathgrad: didnt follow the origin thread, but what kind of errors did 435 give you? [04:17] Firstly it wouldn't show the boot/login on the nvidia card when integrated graphics were enabled. So I disabled that when I rebooted it shows the boot screen on the nvidia card but went black except the loading dots and wouldn't progress, purge 435 then installed 430 first boot with 430 it switched to showing the boot messages, ctrl-alt-f2 then ctrl-alt-f1 it shows graphical login [04:18] mathgrad: on GTX cards its normal to have a black screen without the nvidia driver installed, workarounded by nomodeset [04:19] nvidia 430 driver works perfectly for me. what did I miss? [04:19] mathgrad: but it surprises me 435 doesnt work on 1070 [04:19] 1060 here [04:20] doug16k: mathgrad says 435 is borked on his card [04:21] ah, it's an optimus laptop? [04:22] no, desktop with gtx 1070 [04:23] I very much doubt there is any measurable difference between 430 and 435 on 1070 [04:23] on 20xx, maybe [04:23] it is not in their business interest to make older cards go any faster [04:24] fixing major game malfunctions, sure [04:41] Hello all, I have an iphone that i have quite a bit of video on and I would like to put it on my ubuntu 18.04 pc. I am using shotwell because that is what came up by default. I am using shotwell to import it but it has been about 7 minutes and it is still at 0%. Do you think it just requires more time or is this not a workable action for file so big. The video is 22 minutes long It's taking up several gigs on my ipad and that is why I want [04:41] to remove it. Shotwell has no problem importing photos from my iphone. [04:42] jkoofer: first import your pics to your harddisk, then let shotwell import quicker [04:44] best hypervisor for a single VM running windows? [04:45] @lotuspsychje ironically it just completed it's import. It went from 0% to 100% instantaneously [04:45] Apparently it was importing but the load bar was not illustrating that. [04:48] mathgrad, qemu can be configured to give excellent performance. I remote desktop into it and use RemoteFX and get excellent performance [04:49] for visual studio that is. with qxl drivers installed on the guest you get autoresizing [04:58] is there a "tile all the windows" thing ? [04:59] !rootirc | Artemis37 [04:59] Artemis37: It's not technically our business, but we'd like to tell you that IRC'ing as root is a Very Bad Idea (tm). After all, doing anything as root when root is not needed is bad, and especially bad with software that connects to the Internet. [05:04] What's the shift key that if I hold it down for a few sec will bring up a window of all the hot keys? [05:05] CarlFK: all hotkeys are under gnome systemsettings/devices/keyboard [05:10] lotuspsychje: thanks [05:43] hi guys, given a PID what's the best way to get the PPID of it? [05:57] doug16k qemu seems nice [06:19] whats the latest ubuntu that using network/interfaces and not netplan [06:25] windows boots faster in qemu with a raidz2, ssd cache and log drives than bare metal [07:00] i see only bars not numbers in htop details . https://i.imgur.com/RN86ZY7.png [07:02] Maybe the color scheme of your terminal? [07:02] Try in xterm or rxvt. [07:02] ok i see that [07:03] It looks like they are slightly visible on your screenshot. [07:03] yeah [07:05] akemhp: thanks :) [07:05] Np. === mathgrad_ is now known as mathgrad === mathgrad_ is now known as mathgrad [08:32] Does anyone know why journaling of file systems is bad for SSD's? [08:34] That was advice from the very early days of SSDs when their lifespans were much shorter. [08:34] aiena it's old crappy advise, when SSDs were new and expensive, the idea was to reduce the amount of read/writes to extend the life of the SSD [08:35] even back then lifespans were in decades for commercial SSDs, I've always used journaling FSs [08:35] there is absolutely no doubt that it's completely fine to use a journaling fs like ext4 on any SSD released in the last 8 years or so [08:36] and the older ones are so small and slow in comparison that i'd be surprised you wanted to use one at all [08:37] here's an article from 10 years ago debunking the "wisdom" of the time that you should never put swap, journal etc on an SSD https://robert.penz.name/137/no-swap-partition-journaling-filesystem-on-a-ssd/ [08:40] hi [09:09] what's the right click atlternative for ubuntu 18.04 desktop? [09:12] juanonymous: the window button on the keyboard [09:13] or shift+ f10 [09:14] so i just need to point the mouse on the area where i have to perform then either use shift + f10 or window button [09:14] let me see [09:18] yep [09:18] thanks [09:24] Im struggling getting my script to run without root that edits a file owned by root. Adding both the script and the file to sudoers, the script is still not allowed to write to the file. Any ideas? === kostkon_ is now known as kostkon [09:31] flog: I’d say the permissions are derived from the user *running* the script. [09:31] vlt: it semes like it doesnt though [09:32] flog: Show us. [09:32] flog: can you change the permissions of the file, maybe add an acl so your user can write to it? it would be better than adding a script to sudoers [09:32] EriC^: that does work. but it reverts back to beeing owned by root after a reboot. [09:33] flog: which file is it? [09:33] /sys/class/backlight/amdgpu_bl0/brightness [09:33] flog: oh [09:33] flog: have you tried using "xbacklight" to change the brightness? it doesnt require root permissions [09:34] xbacklight returns "No outputs have backlight property" [09:34] But writing to the earlier mentioned file does change my backlight. [09:35] flog: ok paste the sudoers line you added [09:35] you're running the script via "sudo /path/to/script" right? [09:36] I can run that script without issues with sudo [09:36] so what's the problem [09:37] you want to run it without a passwd so it can be automated? [09:37] Give me a sec [09:37] https://paste.ubuntu.com/p/4KcDrtngST/ [09:37] thats the script [09:37] ok [09:38] log ALL=(root) NOPASSWD:/usr/local/bin/t495-bl.sh [09:38] flog: or you could add a udev rule to set the GROUP/MODE on that node in sysfs [09:38] flog not log [09:38] TJ-: have not tried that approach [09:39] flog: so as the user log, if you type "sudo /usr/local...../bl.sh" it doesnt work? [09:41] EriC^: sudo t495-bl.sh works just fine. [09:41] flog: im confused, you said earlier adding it to sudoers doesnt allow to write [09:42] The script runs fine. [09:42] Well just a sec [09:43] https://paste.ubuntu.com/p/jJj6ft9RKV/ [09:43] Im allowed to execute the script but the script is not allowed to write to the file. [09:44] flog: something like SUBSYSTEM=="amdgpu0_bl0",ACTION=="add|change",GROUP="sudo",MODE="0664" [09:45] after upgrading from disco to eoan the restart and shutdown buttons are grayed out. what could be the reason? [09:45] TJ-: in a new .rules file i assume. [09:45] the power manager says there is no authorization to shutdown: http://dpaste.com/197484T [11:44] [09:45] ERC> [09:45] flog: actually, no, SUBSYSTEM=="backlight",KERNEL=="amdgpu_bl0" ... [09:46] Never touched any udev stuff. [09:46] flog: yes, /etc/udev/rules.d/99-backlight-amd.rules maybe, with SUBSYSTEM=="backlight",KERNEL=="amdgpu_bl0",ACTION=="add|change",GROUP="sudo",MODE="0664" [09:47] single equalsign before "sudo"? [09:48] flog: you could even set the brightness by adding ATTR{brightness}="" [09:48] flog: yes, it's assignment whereas the preceeding are all tests of equality [09:48] same for mode? [09:48] flog: yes [09:49] Do i need to reload the rules to make it take affec [09:49] ~ [09:49] ~ t? [09:50] flog: it reads as If subsystem == backlight AND kernel (device) == amdgpu_bl0 AND action is 'add' OR 'change' THEN DO GROUP=sudo; MODE=664 [09:50] flog: you'd need to re-trigger the device discovery [09:51] I am probably doing something wrong since the script still fails. [09:51] flog: "udevadm trigger -t subsystem -s backlight" [09:51] *fails to write to the brightness fi [09:51] ~ [09:51] ~ le [09:51] ~ [09:51] ~ [09:51] Sorry for the random pastes... [09:51] oops, typo [09:52] flog: "udevadm trigger -t subsystems -s backlight" [09:54] TJ-: no change unfortunately... [09:54] flog: can you show me the .rules file ? [09:55] SUBSYSTEM=="backlight",KERNEL=="amdgpu_bl0",ACTION=="add|change",GROUP=="sudo", MODE=="0644" [09:59] flog: OK, tested it here, and trigger isn't triggering this rule (I added RUN+="/usr/bin/touch /tmp/backlight" to create a tell-tale file) [09:59] flog: so maybe it would work on a boot but trigger needs additional/different command-line options [10:00] Ill try a reboot [10:00] flog: indeed that is the case; adding the verbose option to list devices that'd be triggered doesn't list any: "sudo udevadm trigger -v -t subsystems -s backlight " [10:01] So in my backlight rules file ill add the RUN+= at the top? [10:01] flog: hang on! go it [10:01] oh? [10:01] use this instead: "sudo udevadm trigger -v -s backlight" [10:02] flog: the rule (i am testing with) here isn't firing though so it needs tweaking in some way [10:03] Well i guess the trigger worked since it spat out the path to amdgpu[Flogpad:flog:1] [/etc/udev/rules.d] [10:03] i also need to do something about my trackpad randomly pasting crap. [10:08] flog: I'm trying to figure it out here with an intel_backlight node... still not got the rule to fire (no tell-tale file created) [10:08] Maybe i should take a different approach. [10:09] the whole reason for me doing this is that the defuault scripts and tutorials online for controlling backlight in i3 didnt work. [10:09] So i managed to hack together the erlier mentioned script. [10:10] However if i login on the default desktop environment the hotkeys works just fine. But i have no idea what script/program is triggered by the hotkeys. [10:10] If i could find out what the default DE uses maybe i could use that instead. [10:10] flog: yeah... the 'best' way is via udev but it can take some messing to get it to work. As a FYI, I have in a separate shell "udevadm monitor -upk -s backlight" so I can see the events being generated when triggering [10:12] flog: AHA! there's a bug in .rules. I removed the GROUP/MODE and RUN+= fires the touch [10:14] Sorry but i am completely lost now. What do i need to change? === dijitol1 is now known as dijitol [10:18] flog: I've found someone else dealing with the same issue, and a solution, hang on and I'll give you a new .rules file to use [10:18] Cool, thanks! === dijitol1 is now known as dijitol [10:23] flog: I'm looking at the last reply in this question, but with intel_backlight, I'm not being able to set the group/mode in that way but the tell-tale file is touched, so it is possible my system doesn't behave the same way as AMD may do. https://unix.stackexchange.com/questions/459874/udev-doesnt-want-to-run-chgrp-and-chmod [10:23] hi, i am new to linux, I have installed 19.4 on to my Lenovo MIIX 320 and I can# [10:25] hi, i am new to linux, I have installed 19.4 on to my Lenovo MIIX 320 and the screen is allways 90 degrees to the left rotated [10:25] flog: a question. Are you sure the device name has a lower-case L in it, or could it be a ONE? "amdgpu_bl0" in your info above but in the link I gave you it shows ..._b10 (so Bl0 or B10) ? [10:25] can anyone help me with it? [10:26] could someone help me solve my mounting problems? i got an .img with 3 partitions, trying to mount the 2nd like 'sudo mount -o offset=1001470 /home/noudle/disk-drive-virtio0.raw.img /tmp/mount' but i get 'failed to setup loop device for /home/noudle/disk-drive-virtio0.raw.img.' [10:26] TJ-: it is a lowercase L. [10:26] flog: OK... so not that then! [10:28] Still wont allow the script to write to the brightnessfile. [10:28] Im about to give up. [10:28] Feel a bit bad about it since you've put so much time into trying to help me :/ [10:28] flog: it is weird [10:30] flog: the commands work manually so I suspect those files within the device node aren't (re)created at the point our commands run [10:31] But isnt it strange that both sudoers and udev doesnt work? [10:33] flog: using sudoers how is it not working, when you did sudo it didnt give a permissions denied [10:35] flog: almost have it working, just tidying it up for you now [10:36] EriC^: the script runs, but no permission to edit the target file. with sudo(manually in terminal) the script keeps permission to edit target file. [10:37] flog: and it works... pushing to a pastebin now [10:37] TJ-: you are ace! [10:41] flog: I'm setting the group to "video" since that is more appropriate (use "adduser $USER video") to allow users to adjust brightness. Here it is http://paste.ubuntu.com/p/zsn9Yh9ccP/ [10:42] here it causes: -rw-rw-r-- 1 root video 4096 Sep 29 11:19 /sys/class/backlight/intel_backlight/brightness [10:44] flog: obviously you need "sudo chmod +x /usr/local/bin/backlight.sh" too [10:45] the script is already executable [10:45] Hello I would like to give chromium which was installed via snap only permissions to ~/Downloads. Is this possible. I see that snap connections says home chromium:home :home manual but how can I change the settings. [10:47] TJ-: so how this is supposed to work is to modify the permissions for the backlight file when something want to edit it? or is it a permanent permission change? [10:48] is there any possibility to put command xrandr -o right into a boot file? my Lenovo Miix 320 starts everytime with the screen rotated to left [10:48] flog: permanent. It'll operate on boot and set the perms so anyone in 'video' group can write to 'brightness' -- if you want change that to 'sudo' in the script's chgrp of course [10:49] but running udevadm trigger will read all rules and apply them without having to reboot? [10:49] !info iio-sensor-proxy | zuzka_ sorry, meant to give you this hint earlier. It handles things like accelerometers and can handle rotations [10:49] zuzka_ sorry, meant to give you this hint earlier. It handles things like accelerometers and can handle rotations: iio-sensor-proxy (source: iio-sensor-proxy): IIO sensors to D-Bus proxy. In component main, is optional. Version 2.4-2 (bionic), package size 45 kB, installed size 191 kB (Only available for linux-any) [10:50] zuzka_: but I believe your issue is something to be fixed in systemd. I patched systemd for this on another device some time ago, I'll find the link to it [10:50] TJ-: doesnt look like im in the video group [10:50] thank you [10:51] zuzka_: this was what I had to do: https://github.com/systemd/systemd/pull/10554#pullrequestreview-169283317 [10:52] flog: ahhh... usually the 1st user gets added to that and others... or maybe that's just on my network! [10:52] flog: well, use 'sudo' then :) [10:52] thank you very much TJ [10:53] zuzka_: the challenge is figuring out the correct matrix, as in the ACCEL_MOUNT_MATRIX= [10:53] TJ-: the issue is that I'm running the script from i3 on hotkeys. [10:53] Easiest wolud just be for me to chmod the file every time im rebooting manually [10:53] flog: so make the user a member of video group then the script doesn't need sudo perms [10:55] Maybe i dont even have a video grop [10:55] flog: "getent group video" [10:56] flog: it should be a system group, probably 44 [10:56] Hm im in that one. [10:57] flog: OK but if you only just added yourself to it, it requires a log-out/log-in to take affect, compare "groups" (what you're in right now) with "groups $USER" (what the system has you listed in) [10:57] brb relogging [11:01] tj, i think this is a bit too difficult for me as i have only been using linux for about 7 days. Is there no option to write a script with xrandr command and execute that script at startup? [11:02] TJ-: the udev rule does not seem to be triggered :/ [11:03] TJ-: i give up for now. but thank you for your effort in trying to solve this for me. [11:04] flog: that seems weird ... the only reason would be if some other udev rule is over-riding it [11:05] zuzka_: is that on gnome your screen starts rotated? [11:05] flog: originally I included in the .rules file the tell-tale to be sure the rule triggered, as in ,RUN+="/bin/touch /tmp/backlight" AND in the .sh I removed the comment in front of the 'ls' that writes the current details to /tmp/bl.log so I could be sure both udev and script were working [11:06] zuzka_: which way do you need to rotate? I may be able to help you determin the correct hwdb entry [11:07] lotuspsychje yes [11:07] TJ to the right [11:07] zuzka_: another idea, could be comparing with unity desktop, see if that boots normally [11:09] lotuspsychje: it's a udev/hwdb issue since the display is installed incorrectly (orientation wise) [11:10] TJ-: think unity would do the same in his case then? [11:11] lotuspsychje: nothing to do with the GUI [11:11] kk [11:11] we can compensate using a stop-gap xrandr command but the fix is in the udev/hwdb [11:11] zuzka_: can you show us "pastebinit <( journalctl -k )" [11:13] zuzka_: Here i have a local over-ride which we might be able to create for you [11:16] So I'm having a stupid problem. I use my ubuntut machine roughly once a week. And everytime I use it FSTRIM starts and the machine is incredibly sluggish and slow. What could be a solution to this problem? Thanks. [11:16] !trim | kjartan_ [11:16] kjartan_: Many Solid State Drives support TRIM, which allows the drive to do garbage collection and improves performance. Ubuntu 14.04 activates it by default. For older versions, see http://askubuntu.com/a/19480 for information on activating it. [11:17] kjartan_: there's a systemd unit "fstrim.service" which presumably is being started [11:18] TJ-, lotuspsychje, yes thanks. I think I don't want to disable it, just make it run slower ...? It's hogging 99.9% of the disk while running. [11:19] kjartan_: I'd presume fstrim.service is triggered by its timer, see "systemctl cat fstrim.timer" [11:21] TJ-, Yes, it's running weekly. Just as I suspected. [11:23] kjartan_: you might be able to enhance that with an OnBootSec= or RandomizedDelaySec= [11:24] kjartan_: I *think* adding an OnBootSec=3600 would effectively be ANDed with the OnCalendar= so you'd have an hour :) [11:26] TJ-, That's excellent! It's just really annoying the first 5 minutes of boot and initializing the system, makes every time I sign in very slow. OnBootSec=600 it is. [11:26] TJ-, thank you. [11:29] kjartan_: in which case something like : sudo mkdir /etc/systemd/system/fstrim.timer.d; echo -e "[Timer]\nOnBootSec=3600" | sudo tee /etc/systemd/system/fstrim.timer.d/local.conf [11:30] TJ-, oh. I just edited the file directly. Your suggestion would be better when/if it for some reason is being patched by apt? [11:31] kjartan_: this way you don't alter package-installed files and you only write the changes you want to make which makes it easier to reason with later when you've forgotten all about it [11:31] TJ-, smart. [11:32] kjartan_: see "man 5 systemd.unit" and "Example 2. Overriding vendor settings" [11:33] kjartan_: specifically the last option "Alternatively, the administrator could create a drop-in file ..." [11:37] TJ-, yes I see. Very practical. [11:45] Hey folks === lotus|i5 is now known as lotuspsychje [12:08] I face this "https://paste.ubuntu.com/p/nZyr6PgjQx/" when I want to update , my repo list is https://paste.ubuntu.com/p/wSpj4Sgy5W/ [12:09] Hey Guys, I'm running a ubuntu server on in a virtualbox environment. Somehow the disk (10GB) is out of the sudden full... do you have an idea how I can find where this diskspace is used and how to clean up? [12:09] pabed - debian is not ubuntu, uses toy story characters as releases, and there is no 'Bionic' toy-story character, it's a Ubuntu release, not a debian release [12:10] guiverc: what should I do ? [12:11] pabed: deepakdeshp? [12:11] pabed, I would remove the debian line; but I don't know the reason why you added it [12:11] also a line with ubuntu xenial [12:12] yes the xenial should be changed to bionic - IF you are using bionic [12:13] I just tired to remove an old kernel (hdd full), but unfortunately I need free space on the hdd to remove that... does anyone has an idea how to get that space? [12:13] I don't know where all that space is used... [12:17] there was hetzner repo and I didn't install docker and I removed it and replace https://paste.ubuntu.com/p/nZyr6PgjQx/ [12:24] i don't know how removing a kernel would require disk space Autchirion, if you're using a file manager, maybe your wastebasket is full [12:24] try to remove things with apt, for example apt autoremove removes out of date deb packages [12:33] humpled I tried apt-get autoremove already [12:33] and I did sudo apt-get remove linux-image-[version]-generic [12:34] Autchirion: try removing the files from /var/cache/apt/archive [12:35] it is empty [12:36] Autchirion: any thing in you Downloads folder that isn't needed? [12:36] no, I don't think so, I don't even have a download folder [12:37] Autchirion, sudo du -haxd1 / | sort -h [12:38] swap.img 2gb [12:38] I don't think that's something out of the ordinary [12:39] var has 4.5GB [12:40] Autchirion, sudo du -haxd1 /var/log/journal | sort -h [12:40] yes, I did that [12:40] ah, my bad [12:40] missread [12:40] 417MB [12:41] for /var/log/journal and one /var/log/journal/[random letters and numbers] [12:41] Autchirion, trim the journal maybe ? ; sudo journalctl --vacuum-time=2d [12:42] 200MB disk space free, let's see if I can start working with that [12:42] Autchirion, and remove swap if needed [12:43] still can't remove the kernels [12:43] Autchirion, can you paste dpkg -l | grep linux-image [12:43] sure, rebooting right now, give me a sec [12:44] Autchirion: what apps do you use in the vm? [12:44] why rebooting .... [12:44] You can use an app calle filelight to see usage [12:44] maybe the tmp folder is full from say browser cache [12:44] I only use bitwarden in this VM [12:44] called [12:45] does bitwarden create temp files? [12:45] hmm that looks like a password manager its highly unlikely [12:45] use filelight its pretty weird that your out of space [12:45] yes, it is a password manager [12:46] the server for it... [12:46] @ioria output: https://ibb.co/8DGN5Qy [12:46] sorry, copy paste doesn't work in the VM [12:46] hmm you could use filelight to investigate the cause [12:47] just 4 kernels, not sure it's the issue [12:47] try clearing browser cahce etc and see if makes a difference [12:47] filelight requires a setup? if so, I don't know how to install it... [12:47] it's a windows server terminal only... [12:47] apt-get install filelight [12:47] there is no browser [12:47] its an ununtu package afaik what DE are you using for the ubuntu u have [12:47] yes, but without space on the HDD I can't install a programm... [12:47] there maybe a gnome alternative swhich uses gtk [12:48] Autchirion: is your host a linux machine [12:48] no, windows machine [12:48] I mean the host on which virtualbox is running [12:49] yes, host is win10, guest is ubuntu server [12:49] you could setup the VM as a samba server and connect to the guest with the Ip from windows explorer and investigate [12:49] use the root of your home directery as a samba share [12:49] hi [12:49] then I would need to install smb protocoll on the guest, which requires space on the HDD [12:49] usually samba is installed by default so no extra software needed [12:50] Autchirion, try to manually remove the -60 kern : https://help.ubuntu.com/community/RemoveOldKernels#Safely_Removing_Old_Kernels [12:50] Autchirion: usually smaba is installed already on the guest [12:50] just run apt-get install samba and see if it says it's already installed [12:50] i am not sure if my question is related with ubuntu but on my thinkpad TLP power manager doesnt work [12:50] ir just look for a /etc/samba folder [12:50] https://linrunner.de/en/tlp/docs/tlp-linux-advanced-power-management.html [12:51] if it exists most likely samba is already installed since it is there already you dont need to install it makes sense [12:51] i set battery threshol 75%-%85 but it keeps charging to 100% [12:51] so no additional disk space [12:51] threshold* [12:51] ioria I just deactivated the swap which gave me 2GB, so now I can uninstall the kernel via apt [12:52] good [12:53] but kernel still shows up [12:53] after sudo apt-get remove linux-image-[version]-generic [12:55] I need to figure out how to extend the Virtual HDD for a linux guest [12:55] Autchirion: you need to reboot the vm [12:56] sry, dc [12:57] Autchirion: you need to reboot the vm [12:57] thank you [12:57] for the kernels to dissapear [12:57] will do it [12:57] what does (b & 1) mean? https://linx.li/binpow.cpp [12:58] aaand they are gone, thank you [12:59] m92lk: ask in #c++ [12:59] that looks lie a c plus plus question [12:59] *like [12:59] and something to do with binary operators [12:59] Autchirion: your welcome [13:00] *you're [13:00] now check if samba is already installed in the vm [13:00] oops sorry was meant to post over at #programming [13:02] no, was not installed [13:02] but I would rather figure out how to increase disk size or create a new setup, 10GB is insufficient :D [13:04] but first, food, my Girlfriend is hungry and that's a life threatening situation... :D [13:04] ha [13:04] tank you again for the help [13:04] well I got a powe failure so well missed most of the convo [13:05] but eat now :) [13:05] I want to figure out how to increase hdd size for the VM [13:05] that's my next step, 10GB is insufficient [14:34] I asked this the other day but unfortunately poofed and likely missed a response, but in having issues of airplane mode being hard-locked on in ubuntu and debian, could the fact I have to launch w/ acpi=off have some relation to this? [14:35] nate: do you have some details to share please, this will help volunteers to help you better [14:35] !details [14:35] 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. [14:39] lotuspsychje: I sorta just did define the details, airplane mode is locked on lol. I was going over it the other day with a couple of people, done all the standard stuff I can think of. It's just an HP laptop that has been very anti-linux lol [14:39] took me like a week to figure out that to even install it I had to run the installer even with acpi=off [14:40] nate: i mean, ubuntu version, kernel version, wifi chipset, etc [14:43] lotuspsychje: I've tried the 18 LTS on latest 5 kernel as well as Debian 10 (which I believe is still using latest 4.x kernel, I'd have to reinstall to check). It's an Intel 7260 Combo (Wifi+BT), using the latest 7260-17 driver straight from the kernel git repo [14:45] nate: so, you enable airplane mode, and cant go back anymore till reboot, is that what you experience? [14:45] I've tried all the rfkill switches I could think of, tried doing key-remapping for it to 'read' the airplane control key on the keyboard (monitoring key presses doesn't even show that key being pressed though all the other function control keys do), no options in the BIOS for the control, etc [14:46] lotuspsychje: Airplane mode is being enabled period, from the start [14:46] oh [14:46] that sounds like the kernel could block your wifi chipset to me [14:46] It shows up as hard blocked in rfkill [14:46] nate: can you pastebin your dmesg please? [14:47] Yeah but I don't really understand why, especially across multiple distros. And one sec I'll have to get the laptop out and booted up [14:47] There curiously does seem to be a LOT of complaints of this in google in relation to HP laptops for all debian based distros, but yeah brb [14:48] nate: same drivers are used across several Oses [14:48] -distros [14:51] rfkill hard block info is read from the UEFI/BIOS [14:51] nate: ^ [14:51] nate: so you know about those hp PCI device whitelists? i forgot the details, and i think i read they no longer do it, but at least with older models they used to only allow those parts they sell [14:52] My HP is about a year old, no BIOS whitelist like my Lenovo [14:54] jeremy31: Yeah but there aren't any options for this in the BIOS/UEFI, it would be weird if they had it rigged somehow to explicitly hard lock on non-windows, particularly with the secure boot/uefi stuff all off I would think [14:54] tomreyn: This is all the standard hardware of the laptop [14:55] Nothing is non-OEM [14:55] nate: can we see your dmesg please, volunteers will have a look for you [14:56] working on it, had to fix grub again, it apparently did an update last time I had it on and removed my acpi=off switch [14:59] nate: are you editing /boot/grub/grub.cfg directly, instead of editing /etc/default/grup and running update-grub? [15:00] Yeah I did, but I also did it before doing the initial apt upgrade which was probably a derp idea. Also I should have looked at dmesg way ago, it looks like it's the driver throwing the RF_KILL bit. Gonna try and roll back the ucode to an earlier one [15:01] * jeremy31 nate, rather than acpi=off, try acpi_osi=" [15:02] hey MonkeyDust [15:02] lotuspsychje [15:04] jeremy31: Blank? I had tried to 'detect' what acpi_osi to use at one point but it seems that with acpi=off you can't actually read the ACPI strings. I'll try that though [15:04] nate yes, leave it empty === dionysus70 is now known as dionysus69 [15:10] Ho man that was a lot of ACPI errors in the bootup, but it does appear to be booting lol [15:10] ... and airplane mode is no longer locked on. It was the god damned acpi=off lol [15:11] Thank you much jeremy31 :P [15:11] nate, good that it works [15:11] !cookie | jeremy31 [15:11] jeremy31: Wow! You're such a great helper, you deserve a cookie! [15:13] hey hey that appears to also have made the bluetooth driver load up as well, sweet [15:14] now if only I could remember who originally mentioned to me it could have been the acpi=off switch so I could thank them too... [15:14] nate, I think the 7260 wifi/bluetooth was supported in 3.13 kernels [15:18] in windows, win+arrows allows you to move windows from left to right, and across monitor thresholds, I noticed in ubuntu 19.04, that it works within a single monitor, but won't jump across. am I missing a keystroke? [15:20] jeremy31: Yeah but when I was in the acpi=off mode it wasn't even registering bluetooth hardware at all despite it being part of the intel network chipset [15:20] nate, it likely shows in lsusb results with a 8087 ID, might be 07dc [15:21] kenperkins: I believe it may be shift+super+arrow. [15:22] ah, i see, so the've separated it, move between displays, and manage within display [15:23] it appears workspaces are disabled by default in 1904? [15:24] no, they're there? dynamic, vertical ones, by default [15:24] I'm not finding the hotkey for it, at least in the keyboard settings [15:25] except if you have an external monitor, that one doesn't use workspaces by default [15:25] press and you'll see the workspace switcher on the right [15:25] ctrl+alt+up/down to switch using the keyboard [15:25] (also hold down shift to take the current window with you) [15:28] hmm, I can't figure out how to have both monitors use workspaces [15:28] nvm [15:28] workspaces are amazing [15:39] IIRC there's a setting in gnome-tweaks for that [15:40] (personally, I like having one monitor always display the same stuff while I can switch the other between workspaces) [15:46] Hi Guys,I installed ubuntu but when i start my laptop , my ubuntu doesn't up and i can't use it, and same my windows,but when i plug in my flash memory(ubuntu live) i can use it. Is it about failing boot loader? [15:47] ** I use Debian , that's a problem for my friend [15:56] @mgedmin I'm doing a bit of both, I'm having slack stick to all workspaces, but the rest of both monitors change [16:02] [16:02] I'm trying to build bsdmainutils on bionic. I installed the sources via apt and ran apt-get build-dep bsdmainutils - though when running make I get error sys/capsicum.h: No such file... How do I find the correct package to install? === BrianG61UK_ is now known as BrianG61UK [16:06] rkta: there's no capsicum on linux afaiu, it's a bsd thing [16:06] rkta: that's why it's removed by debian/patches/col_cap.diff in the bsdmainutils source package [16:07] how are you building it? use debuild instead of running make [16:07] (or dpkg-buildpackage, which is a bit lower level) [16:08] mgedmin: thanks, will look into these. [17:10] How do I find statistics program R in Ubuntu 18.04 Applications store? [17:13] Hi I have a vm that is contacting my host's localhost via 10.0.1.1. I'd like to I guess see a list of availavble networks, and see what ips they are using so I can try and set the correct internal network for the vm [17:13] is there a cli command that will show this? or does it sound like I am on the right track? [17:17] iffraff: try `ip addr` or `ip route`, but there's no guarantee that'll show all the networks used in your intranet, some could be reachable via the default gateway [17:18] I answer to myself: you can't. One letter is not enough to be a string, the installation package is called r-base. [17:18] mgedmin: ok, so I did try ip addr but, it didn't shed much light. does what I'm trying to figure out make sense? I could give you more context. [17:19] maybe? [17:23] mgedmin: so i have an emulator. it actually shows up as a vm in virtualbox. it must connect to the local react native server to get the code for the app. but the app must call out to an api running locally. It connects to the react native server fine, but when the app tries to call out to the api it's using 10.0.1.1 with the correct port. and that doesn't work. there are two network adapters in the vm settings. I'm thinking the [17:23] issue is there somewhere. [17:23] but I don't know much/anything about network adapters so idk [17:26] hi, which font should I install paired with ohmyzsh and urxvt? I am having trouble getting icons to show? [17:26] mgedmin: I get it if you don't know, it's a pretty niche problem. [17:29] iffraff: so the app is running inside the VM, which has ip 10.0.1.X, and the api is running on the host which should have ip 10.0.1.1? [17:30] iffraff: is the api's listening socket bound to 0.0.0.0, or is it bound to 127.0.0.1, or is it bound to something else? [17:31] (it's probably not the problem; I suspect you may need to tweak the virtualbox networking settings for that VM to allow it to access the host, but I'm not sure how) [17:32] mgedmin: so, sort of, the app, inside the vm, needs to connect to two servers on the host. one is running the code for the app, and this one is being reached. It is running the code and then trying to make an api call to the other serice on the host and failing. I can see that it's trying to call out to 10.0.1.1 but the other server is running on localhost. [17:33] It could be 127.0.0.1 or 0.0.0.0 I'll try all three if/when I figure out where to put it [17:33] iffraff: so the 'other service' needs to bind to a non-local interface, else it won't work [17:34] iffraff: by definition a service bound only to 'lo' can only be reached from other services on the same 'lo' [17:34] iffraff: it's called the 'loopback' interface for a reason, and it isn't routable off-host [17:35] meaning not localhost or the other two. that makes sense. Ok here's the kicker however the second service (we'll call api ) however the api is setup, should work. because there is a different emulator that does not run on VBox that is able to hit both servers. [17:36] however now that I type that, perhaps the fact that it doesn't use vbox means it is on local network thus lo whoud and does work [17:36] hmm. so would 0.0.0.0 work? [17:41] well, crap, the api service is running in docker. I'm sure this is sufficient, but the service is a rails app and is running on 0.0.0.0. maybe I need to do something to docker as well though [17:50] yeah, doesn't docker use network namespaces or something? [17:50] and you have to explicitly export ports? [17:57] mgedmin: yes, but I can hit the api via localhost:3000 from the host. for some reason the app in the vm is trying to hit 10.0.1.1:3000 I"m not sure if that is sort of generic ip like 127.0.0.1 or something, but it' snot reaching the container [17:58] iffraff: might be worth checking if the api is reachable at 10.0.1.1:3000 from the host [17:58] but yeah, docker and a vm, sounds like fun [18:00] well it doen't just come back unreachable. it's hanging, which is interesting but it 's not coming back so ... no [18:00] hi people. I want to install ubuntu on my laptop. it's an hp laptop. I cannot get it to boot from the cd/dvd rom. I don't have an extra flash drive. Can someone please guide me, can I install from windows itself? [18:00] i tried to find secure boot in the bios, but it's not there [18:00] i try to boot from the cd/dvd drive [18:00] it still boots into windows [18:02] vigilant: try pressing f9 when the pc starts [18:02] i did [18:02] it still boots into windows [18:02] let me try again brb [18:02] vigilant: Does the DVD drive work? [18:02] cd must not be made right [18:03] In my pc is f12 or f10 but it can vary from pc to another [18:03] i burnt the cd on windows [18:03] or dvd* [18:03] the latest ubuntu install [18:03] i verified it too [18:04] vigilant> may be you didn't burn the iso but copied it directly to the dvd [18:04] vigilant: can you change the boot order from BIOS? [18:05] Oh, kadiro has a point here: can you read the contents of the DVD? Do you see a directory structure? [18:06] wait what the hell [18:06] ther is no data on the dvd drive [18:06] i just chcked in windows on the lapotp [18:06] laptop [18:06] brb [18:06] :) [18:06] That might be a minor issue... :) [18:07] hehe yeah [18:07] sir [18:07] this is a ubuntu-18.04.3-desktop-amd64 [18:08] that's what i burnt [18:08] so on my other computer [18:08] i can see contents in the dvd drive [18:08] is is one file or many directories [18:08] many directories [18:08] Ok. Good. [18:08] .disk , boot casper [18:08] and so on [18:09] ok [18:09] there are only two other files [18:09] in the main directory of the dvd [18:09] :/. [18:09] i think i am doing something wrong [18:10] with witch application did you burn it and it complete the burning or it have some errors? [18:10] burn image [18:10] right clicked the iso [18:10] in windows [18:10] and burnt it [18:11] from the right click menu [18:11] :) [18:11] something is wrong with your software/iso [18:12] hhhhhhhhhhm [18:12] okay [18:12] can you get into the bios menu, vigilant? [18:12] yes [18:12] let me [18:12] make the iso image [18:12] from another program [18:12] if not you have to access the bios and make the pc boot first from the dvd drive [18:12] uh, you shouldn't have to make ISO images [18:13] hmmmmmmmmmmmmmmz, which program should I use in windows? btw my bad about the flood :/ [18:14] i need to make a bootable drive right? :o [18:14] no just burn the right iso [18:15] okay [18:15] ty [18:15] vigilant: https://tutorials.ubuntu.com/tutorial/tutorial-burn-a-dvd-on-windows#0 [18:15] How do you run Firefox with the wayland backend by default? I can run it easily enough from the command line (GDK_BACKEND=wayland firefox), but beyond making a desktop file is there a way? [18:17] GentlemenDemon, try Exec=env GDK_BACKEND=wayland firefox (not sure btw) [18:18] I believe there is a command that starts with "neon" that shows system stats/specs... do you know what I am talking about? [18:18] ioria> yes but in desktop file he have to use : Exec=sh -c ... [18:18] have to say, super impressed with steam/dota performance on linux [18:19] kadiro, why ? [18:19] ioria> because of the variable he must use a shell i think [18:20] maybe, but i don't think so [18:20] kadiro: or /usr/bin/env, which is what ioria suggested [18:20] I think you are both right [18:23] "neofetch" was what I was looking for :) [18:23] So, the mechanis would be to create a desktop file with the command and place it .local/share/applications? [18:24] yeah [18:25] how does ubuntu do on mbp hardware [18:28] how come neofetch says my CPU is 4.600GHz, when /proc/cpuinfo says 1.80GHz? [18:33] mgedmin: What is neofetch? [18:36] Ok. Found it. [18:36] elias_a: it prints a brief system info [18:38] inxi is a gooog app for showing some info i think [18:38] good* [18:40] mgedmin, probably this : https://forums.gentoo.org/viewtopic-p-8327546.html?sid=ef9c07e237a19440378618014d7a74e5 [18:42] Hello everyone [18:42] My microphone doesn't seem to be working properly? 16.04, discord is giving me choppy squeaks back, and audacity records veeeery slowly, like everything is being compressed, and I sound like a chipmunk [18:44] Looking for a little help to fix my mbr for windows. I think I've some how created a wonky mbr on a different drive that what windows was installed on when I installed grub. Not really sure how I managed that but regardless when I try to boot into windows from grub I see the blue screen error. [18:44] my my bootloader info. http://paste.ubuntu.com/p/ZMXD5kK4Xp/ [18:44] heres* [18:46] JoshChernoff: try booting in Legacy mode and not UEFI [18:46] sda1 looks to be where my windows install is but the mbr is on sdc1 [18:46] I think I am in legacy mode [18:47] I only say that because when I tried to use boot-repair is said it could not repair due to being in legacy mode [18:47] not formula with the diff between UEFI vs legacy [18:48] JoshChernoff> why the recovery windows is in sda1 and the windows system is in sdc1 ? i can't understand that [18:48] In the end I can change out my boot load order in my bios to boot back into windows but grub seems to try and use a different drive with a diff mbr when trying from gurb [18:48] IDK [18:49] I think windows is really in sda1 not sdc1 [18:49] and the bad mbr is on sdc1 [18:49] which ubuntu version are you running there? [18:49] 18 [18:49] !YY.MM [18:49] Ubuntu version numbers are: YY.MM (YY=release year,MM=release month). Each year sees two releases, so just specifying YY is imprecise. See also https://www.ubuntu.com/about/release-cycle [18:50] Try repair the windows first to fix the mbr/efi thing and try again instaling grub but following the instruction about how to install grub [18:50] yeah I figured I would have to use window to fix that, thought I could do it from the grub side [18:51] I' [18:51] thanks [18:52] also 18.04.06 [18:55] Quick question. I just added a cronjob to run on reboot (@reboot user cmd) that runs a python script. I can verify it's working and running in the background, but I don't see the process listed anywhere. How do I keep track of it? [18:56] also when in grub windows points to /dev/nvme0n1p1 [18:57] this [18:57] ``` [18:57] ```sdc1: __________________________________________________________________________ File system: Boot sector type: Unknown Boot sector info: Mounting failed: mount: /mnt/BootInfo/sda3: unknown filesystem type ''.mount: /mnt/BootInfo/sdc1: unknown filesystem type ''.sdc2: [18:57] __________________________________________________________________________ File system: ntfs Boot sector type: Windows 8/2012: NTFS Boot sector info: According to the info in the boot sector, sdc2 has 3518803967 sectors, but according to the info from fdisk, it has 7813771263 sectors. [18:57] Operating System: Boot files: ``` [18:57] blaaa sorry [18:57] is there any reason not to upgrade from 18 -> 19 ? [18:58] !YY.MM | naribia [18:58] naribia: Ubuntu version numbers are: YY.MM (YY=release year,MM=release month). Each year sees two releases, so just specifying YY is imprecise. See also https://www.ubuntu.com/about/release-cycle [18:58] oops [18:58] ** 18.04 -> 19.04 [18:59] because 18.04 is LTS [19:00] a-magpie, have you tried 'ps aux | grep -i python' ? [19:01] a-magpie, or ps auxf | cat | grep python [19:02] naribia: unless you like to upgrade the version every 6 months, no, no reason at all. 18.04 is a Long Term Support (LTS), while 18.10, 19.04, and 19.10 are supported for just 9 months [19:04] ioria, and there it is! Thanks for that :) [19:04] a-magpie, ok [19:09] Does preseed response file (answers.yml) from GRUB work with Ubuntu Live (subiquity installer)? === kostkon_ is now known as kostkon [19:12] cmihai, https://askubuntu.com/questions/1063393/error-creating-custom-install-of-ubuntu-18-04-live-server [19:14] ioria: I'm familiar with di/preseed, but since newer versions of Ubuntu ship subiquity, I'm trying to figure out how to use answers.yaml with that instead. However, this is not documented anywhere, that link just points at the git repo https://github.com/CanonicalLtd/subiquity [19:14] when i hit record in audacity, the sample rate is set to 1024000hz, where is it getting that number? [19:14] https://ubuntuforums.org/showthread.php?t=2390710 suggests "(1) you can unsquash the file casper/filesystem.squashfs in the ubuntu-18.04-live-server-amd64.iso , and get the folder " squashfs-root", then mkdir subiquity_config in the folder of squashfs-root [19:14] Which is nonsense, it's checked with gpg key, and it's not a useful unattended install if I have to change the ISO.. [19:18] cmihai: my understanding is that 18.04 subiquity does not support preseeded installations. i think this is currently in the works for 20.04 LTS, though. note that ubuntu server support is in #ubuntu-server [19:18] tomreyn: thank you! That makes sense :-) [19:19] cmihai, not use the live-version , i guess [19:20] Yeah, http://releases.ubuntu.com/18.04/ only points at live, but http://cdimage.ubuntu.com/releases/18.04/release/ still has the debian installer version for 18.04.03. [19:20] the alternative 18.04 LTS server installer (d-i) should still work for preseeding. [19:20] https://discourse.ubuntu.com/t/please-review-design-for-automated-server-installs/11923 [19:21] cmihai, http://cdimage.ubuntu.com/releases/18.04/release/ [19:21] tomreyn: thanks, that's handy. Old installer still works, tried it. Wanted to investigate this new installer thing and see if I should panic for the future of server or not ;-) [19:22] sy, http://old-releases.ubuntu.com/releases/bionic/ [19:22] but, again, server support is in #ubuntu-server ;) [19:23] Noted for future questions. Problem solved - that page you linked gives me everything I need for the design, and it looks solid. Cheers! [19:24] be sure to add to it if you have further suggestions / needs [19:24] arecord only records squeaks too [19:25] Ben64, tested a previous kernel ? [19:25] how previous? [19:26] Ben64, what's your actual ? [19:26] 4.15.0-58-generic at the moment [19:27] maybe a later one then ;) [19:27] Ben64, 4.15 is now 4.15.0-64-generic [19:27] Ben64, are you fully upgraded ? [19:27] no, but it won't fix the problem (i bet) [19:28] i'll give it a go though [19:28] Ben64, you know; apt update && apt full-upgrade [19:28] yep i know :D [19:28] time arecord --duration=1 --format=dat blah.wav [19:28] Recording WAVE 'blah.wav' : Signed 16 bit Little Endian, Rate 48000 Hz, Stereo [19:28] real 0m11.958s [19:29] 1 second = 12 seconds [19:29] Hey, can someone tells me if with lightdm it is possible to select the desktop environnement at login like with gnome-dm, and if it detects when a second screen is plugged? [19:29] akemhp_, i guess so, why ? [19:30] Hello, how to enable wifi at boot time on ubuntu 18.04, without the requirement of logging in [19:30] ioria, I'll try it then, cause i would not if it didn't. It's for an older computer, i want light stuff on it. [19:30] relik, Make available to other users [19:31] akemhp_: switching window manager should be https://i.stack.imgur.com/2G9ti.jpg [19:32] tomreyn, This is your lightdm? [19:32] ioria that only works if someone logs in, but i want to access the device through internet [19:32] akemhp_: no [19:33] Thanks, i'm installing it right now, it looks i can still easily switch back to gdm in case of problem. [19:34] sudo dpkg-reconfigure gdm3 (or "lightdm") should allow for switching [19:34] relik, Automatically connect to this network [19:36] akemhp_: my source of this image was https://askubuntu.com/questions/747806/change-icon-for-window-manager-at-login ("user contributions licensed under cc by-sa 4.0 with attribution required") [19:36] tomreyn, Yeah i got it, i think if you don't know you can click this icon, you might as well miss it. [19:37] tomreyn, ioria it works just fine, i'll switch my other Ubuntus to lightdm, thanks. [19:37] ok [19:43] hi, do u know a news aggregator like news.google.com where I can make searches ? [19:44] Is anyone using bacula? I'm trying to backup to a USB HD using Ubuntu 18.04 but get the error: [19:44] init_dev.c:123 [SE0001] Unable to stat device FileChgr1-Dev1 at /media/mguy/BackupDrive: ERR=Permission denied [19:45] mguy: maybe you have mounted it with wrong permissions [19:45] mmystic: It automatically mounts when I plug it in. I did a chmod 777 and chown bacula:bacula [19:45] /dev/sdd1 on /media/robert/BackupDrive type ext4 (rw,nosuid,nodev,relatime,uhelper=udisks2) [19:46] make a ls -l /media/robert [19:46] drwxrwxrwx 5 bacula bacula 4096 Sep 28 16:45 BackupDrive [19:48] /media/mguy/BackupDrive /media/robert/BackupDrive [19:49] I changed the path in the first paste [19:49] Different username on that machine [19:50] I am looking to test 19.10 as mu daily driver when it is released (19.04 today). How can I best/easiest backup my complete system to an external drive or network drive? [19:51] ioria: still chipmunky [19:51] Ben64, sy about that, no other ideas [19:51] :( [19:51] it's so weird [19:51] mguy: maybe u have enabled apparmor or selinux ? [19:52] !backup | Aavar [19:52] Aavar: There are many ways to back your system up. Here's a few: https://help.ubuntu.com/community/BackupYourSystem , https://help.ubuntu.com/community/DuplicityBackupHowto , https://wiki.ubuntu.com/HomeUserBackup , https://help.ubuntu.com/community/MondoMindi - See also !sbackup, !borg, and !cloning [19:53] tomreyn: whank you :) [19:53] mmystic: selinux is not being used but I didn't know about apparmor and it appears to be on [19:54] !borg | Aavar: personally i prefer restic (less so) and borg (more so) [19:54] Aavar: personally i prefer restic (less so) and borg (more so): borg is a fast backup tool with incremental backups: https://www.borgbackup.org/ (Ubuntu packages: borgbackup, borgbackup-doc, and borgmatic) [19:54] mguy: try with apparmor_status as root [19:54] it shows u if your app is armored or not [19:54] It's on [19:54] Let me disable it and restart [19:58] It's still on [20:00] mguy: https://www.cyberciti.biz/faq/ubuntu-linux-howto-disable-apparmor-commands/ [20:00] but are u sure is enabled for bacula ? [20:00] mmystic: I have no idea [20:01] I might just quit messing with it [20:01] I am going to try to get it working at work this week on centos [20:01] Just wanted to useit at home real quick [20:01] tomreyn: borg looks great, but can I use it to backup my entire install? [20:01] with apparmor_status it show u two gruoups, one compliant and one enforce. If bacula profile is under "enforce" then it's enabled [20:03] Since right now we just have a cronjob that runs a tar command for our backup solution [20:04] mguy: what kind of backup u need ? [20:05] it's nice backuppc, bacula it's pretty difficult to use [20:05] Something that will do full and incremental backups nightly to USB drives [20:06] Running ubuntu 19.04 on a Dell Insprion 7591, doesn't boot unless I set pci=noacpi. However, when setting pci=noacpi the trackpad stops working (strangely, it works fine with acpi=off, but then of course a whole bunch of stuff doesn't work properly. I think it's related to "can't find IRQ for PCI INT A; please try using pci=biosirq" messages that I'm getting in kernel log? Any general advice? [20:06] mguy: I suggest u to try backuppc, it's very nice and pretty simple to use. http://backuppc.sourceforge.net/info.html#features [20:06] (by the way, I get the pci=biosirq message even when that option is set) [20:07] mmystic: okay, will give it a shot [20:08] mguy: it's for professional/company purpose. But if u need just for personal use there is something much more easy [20:09] mmystic: we have a number of servers we need to backup and should be backing up to network if I have any say in it [20:11] mguy: ok, then backuppc is for you. It backup in many different way, via samba/windows share or via ssh. Pay attention at the permissions. [20:11] colimit: i always try bios updates when i have compatibility issues. [20:12] colimit: failing that, any solutions will tend to be model specific. [20:12] colimit, you did try pci=biosirq ? [20:13] doug16k: Yes, but I just get the same recommendation again (that is, to set pci=biosirq) [20:14] colimit: but seriously, check for bios updates. it'll make sure you aren't wasting your time on a problem that may already be solved. [20:17] Well, a 18.04 live works with the mic [20:17] but i'm not ready to upgrade yet [20:19] So I noticed there's "PCI: Unknown option `biosirq'" in the kernel log? [20:21] colimit: did you put biosirq in /etc/default/grub [20:22] no, I edited (pressing) the record in the boot menu [20:23] jeremy31: Could that make a difference? [20:24] colimit: don't know for sure but it is a possibility [20:32] colimit: looks like it should be pci=biosirq [20:32] colimit: and that might not be valid any more [20:34] I indeed set pci=biosirq. What do you mean this option might not be valid anymore? (It was removed? if so, why does the kernel ask me to set it) [20:38] colimit: Are there any BIOS updates available? [20:39] I thought not, but let me check again [20:44] There's an update from august 2019 [20:45] colimit: Dell does try to help with Ubuntu as they sell some computers with Ubuntu installed [20:46] dmidecode -s bios-version report my bios version as 1.0.1, the update is labeled 1.1.0 [20:46] so I probably should do the update [20:46] I think I'll try that, thanks [20:50] test [20:55] i have a headless cloud vm, and i want a persistent x session and vnc hosted access to it... is there any good methods? i used to just use x11vnc and vnc4server but they don't seem to be working [20:55] it's a headless system, no video outputs, and my usual method isn't working and having it always on/open makes more sense than ad-hoc sessions [21:10] I updated my bios, but it didn't really help [21:10] colimit: Might want to ask on a dell forum [21:16] jeremy31: Yes, probably. Ignoring the IRQ errors for now, it looks like xinput -list shows my trackpad but as a generic mouse (PS/2 Generic Mouse Device). With acpi=off it's also shown but the difference is that it actually works to some extend [21:19] colimit: have you tried booting a distribution that uses a newer kernel? [21:19] i've seen situations where that fixes issues. [21:20] colimit: i would suggest trying manjaro or arch, for testing purposes. [21:20] [21:20] colimit: did you post a full kernel boot log here, yet? [21:21] ryuo: Yes, might be a good idea, there's an arch wiki page about this model (where it is claimed to work) [21:21] colimit: if so, it may simply be a matter of waiting for 19.10. [21:21] the kernel for it may work where this one does not. [21:22] tomreyn: No, but I can post that if you're interested [21:22] if you are wanting to still use ubuntu. it's not unusual for there to be problems with ubuntu and bleeding edge hardware releases. [21:22] so, it seems to only happen after I open audacity, if I skip audacity and record using aplay, it works fine [21:22] that's where a distribution like Manjaro can be useful. [21:23] Aavar: you can use it to backup all the files you need backed up, i.e. everything you'r eneed to rebuild this system from a fresh installation. which is normally not the 'entire install' (i.e. luckily not a storage image / bare metal recovery approach as on windows, for example) [21:23] colimit: i could have a look at it if you posted it if you were interested [21:23] alright, hang on [21:24] colimit: you could just journalctl -b | nc termbin.com 9999 if oyu like [21:26] tomreyn: https://termbin.com/aosq [21:27] Issues with iwlwifi and nouveau are not too important at the moment (although I guess they are also related to IRQ issues) [21:28] pci=biosirq is a X86-32 option only according to https://www.kernel.org/doc/html/v5.0/admin-guide/kernel-parameters.html [21:30] I should note that, despite the fact that the kernel reports that biosirq is not an option, it does have an effect: when the option is set, I can change the screen brightness and otherwise I can't [21:33] how far does it boot when you don't have either of the pci= options? [21:34] freezes, usually before the loading screen is rendered [21:35] without "quiet" the last I got was something about wmi_bus [21:35] so that would be right after "Run /init as init process" [21:37] colimit: i see. maybe you can look through your past boots at jorunalctl --list-boots and identify the one where it froze like this and post this log as well (by passing the ID from the first column to journalctl -b IDGOESHERE | nc termbin.com 9999 ) [21:38] yes, there should be one [21:42] tomreyn: don't seem to have any... Is it possible to log would not get saved [21:43] I have one with boot options acpi=off and one with the option acpi=noirq (both of these "work") [21:43] colimit: yes its possible the log was not saved if it was during the early boot. [21:44] Do you want the logs with acpi=off and acpi=noirq? [21:45] colimit: is the acpi=noirq log with a 5.0 kernel, too? [21:45] colimit: it might be worth trying "acpi_osi=" [21:45] Yes [21:46] jeremy31: I tried that at some point, I think it also did not boot [21:46] colimit: okay, if you can stay around another 20 minutes fir me to review those while eating please post it, too [21:48] tomreyn: https://termbin.com/phw9 thanks [21:50] thanks, i'm not sure i can help much, really, this is mostly a learning experience for myself. but i'll try. and i agree about trying http://iam.tj/prototype/enhancements/Windows-acpi_osi.html if you haven't [21:50] (but drop both pci= options then) [21:52] tomreyn: That website you link to suggests I use acpi_osi=! "acpi_osi=Windows 2015" [21:53] is there reliable whatsapp client for ubuntu? [21:53] colimit: it suggests you should determine, on your computer, which values to use, and then to try those several values one by one [21:54] tomreyn: Yes, I mean, the script tells me to start with that [21:55] (it just happens to be the same as the example on the website) [21:55] colimit: give it a try if you like, i'll be around for another hour [21:57] tomreyn: Ok, will test this and reboot [22:04] so, there's a significant improvement: the trackpad works (as a generic mouse device) [22:04] but, I still had to add pci=noacpi to actually boot [22:09] tomreyn: https://termbin.com/6c6f for the new log [22:12] thanks, still diffing the previous two ;) [22:12] colimit: did you notice the nouveau traces in your log? journalctl -b and search for: ------------[ cut here ]------------ [22:13] Yes, it's either a different problem (graphics card) or also related to the irq stuff [22:14] for now, I don't need the graphics card to work (am using intel graphics, this is one of those hybrid devices) [22:14] ok [22:15] So my strings /sys/firmware/acpi/tables/DSDT output is actually quite long... [22:16] (it contains 16 entries) [22:18] the goal should be to find a string where it boots without the "can't find IRQ for PCI INT A" messages (which i assume requires also booting without pci=noacpi, if you can) [22:19] I guess I can go through them and check when boot works [22:19] also, you may wantr to boot without splash quiet but with... [22:19] !bootlog [22:19] To get a more verbose log of the boot process, add the following !kernelparm: debug systemd.log_level=info [22:20] will do that [22:20] i guess i'd only try the last 3 or so [22:21] the windows version dell had preinstalled on these should work [22:21] yes, it does [22:22] There were some duplicates among the 16, it's basically 01, 01.1, 01 SP1, 01 SP2, 06, 09, 12, 13, 15 [22:24] oh and use !sysrq for a chance to have logs written to disk [22:24] !sysrq | colimit [22:24] colimit: In an emergency, you may be able to shutdown cleanly and reboot by holding down Alt+PrintScreen and typing slowly, in succession, R, E, I, S, U and B. For an explanation, see http://en.wikipedia.org/wiki/Magic_SysRq_key [22:24] just SUB is enough, the others are ignored. use H to test whether it works [22:30] Ok, so I add !sysrq and !bootlog [22:31] colimit: those aren't kernel parameters, it's just how me make ubottu tell you about it here [22:31] !ubottu [22:31] Hi! I'm #ubuntu's favorite infobot. You can search my brain at https://ubottu.com/factoids.cgi | General info and channels at https://wiki.ubuntu.com/IRC/Bots | Make a clone of me, see !botclone [22:31] oh, lol [22:32] That explains why I didn't find anything in the kernel documentation :) [22:32] it's getting late [22:33] so debug systemd.log_level=info and I'll try the sysrq thing [22:33] I'll get rid of splash as well [22:33] and get rid of "quiet", too, yes [22:34] oh you already did [22:39] pnp.debug=1 *may* help getting suggestions on better workarounds for these resource conflicts you see on your log ("pci 0000:00:1c.0: can't claim BAR 14 [mem 0xa4000000-0xba0fffff]: address conflict with PNP0A08:00 [mem 0x8f800000-0xdfffffff window]" and following lines) [22:40] I tried the 12-15 versions, none of them worked without pci=noacpi === jmcgnh_ is now known as jmcgnh === jmcgnh is now known as jmcgnh_ === jmcgnh_ is now known as jmcgnh [22:44] I'm wondering if, in the OSI options, why I can't just add all of them [22:55] colimit: you could, but the idea there is to identify the one which works best and use that [22:59] colimit: so i finished reading those logs, don't have any more suggestions, i'm afraid. only to try again to boot without any extra options and somehow get a boot log (serial console, netconsole https://wiki.ubuntu.com/Kernel/KernelDebuggingTricks ) to get a better idea of the what the root issue is (and better search terms for the web which are not just side effects of kernel parameters you set). [23:00] if this doesn't succeed you could also try booting with variations of the pci parameter set to (=) one of: noearly nobar nobios biosirq [23:01] tomreyn: will try that later, thanks in any case [23:01] but this is not based on any deeper understanding but only based on reading https://www.kernel.org/doc/html/v5.0/admin-guide/kernel-parameters.html [23:02] good luck. [23:20] does someone know of a "caffeine" alternative that actually keeps ubuntu awak? [23:20] caffeine doesn't work, at all [23:26] wtf [23:26] is it just me? [23:26] for me it works on gnome3 out of the box [23:29] just turning off suspend isn't enough? why "caffeine" program? [23:29] my machine never sleeps [23:34] wow, playing with nvidia-smi. The power limit is bang on, my UPS shows it changing power (watts) exactly as commanded [23:35] I expected it to be approximate. no way, bang on [23:35] nautilus is no longer using my ssh keys to log into sftp [23:36] in ubuntu 19.10 [23:36] how do i fix this? [23:37] its kinda pissing me off as it breaks my workflow [23:39] scientes, how well does sftp usually work ? I usuually use CIFS/samba [23:39] if you already have ssh settup it is seamless [23:40] scientes: does it still work on the CLI? [23:40] and sambais not case sensitive [23:40] * scientes is just installing sshfs to work around it..... [23:41] scientes, can it handle streaming? [23:41] cluelessperson, it can do everything fuse can [23:43] gio mount sftp://servername_or_alias_from_ssh_config [23:44] yeah doesn't work [23:44] it asks for password when keys are set up correctly [23:44] both rsa and ed25519 [23:45] wellif it doesn't work on the CLI either then i wouldn't expect it to work in nautilus [23:46] you probably dont have ssh-agent running or no SSH_AUTH_SOCK / SSH_AGENT_PID in env [23:47] that didn't fix anything [23:47] too imprecise. [23:47] ssh-agent [23:48] then the above command [23:48] exported SSH_AUTH_SOCK for me [23:49] and the socket exists? and ssh-add -l lists the loaded keys? [23:50] it didn't list the keys, but i added it, and then it still doesn't work [23:50] still asks for password [23:51] but GPL3 (if enforced) actually prevents that anti-social non-sense [23:55] oh sorry, wrong channel for that last line [23:59] ahh, I think it isn't honouring the port settings in ~/.ssh/config