[00:33] Just to let you know the shutdown button on 19.10 when you are on login screen after booting doesn't work [00:39] Poo [00:39] Oii* [00:39] Hello guys I am installing Xubuntu and i came accross a error [00:40] The grub-efi-amd64-signed failed to install into /target/ [00:41] I did some research and my gparted in live session said I have a EFI partition already [00:43] how do you mean you already have an efi partitioon, do you have two now? [00:43] Only one [00:43] with boot and esp flag [00:44] so the installer will have created it before it tried to run grub.install [00:44] Yes I think so [00:44] so this does not yet explain why installing grub failed. [00:44] are you still on the live system? [00:45] Yes [00:45] what does this, when run in a temrinal, output? journalctl -b | grep 'DMI:' [00:45] ctrl-alt-t to start a terminal [00:46] oh, xubuntu, this may be different, not sure [00:46] but you can surely find it on the menu [00:47] Xubuntu kernel DMI: Hp Hp Pavilion x360 Convertible 11-ad0xx/830D, BIOS f.08 06/13/2017 [00:49] XubuntuNub: so, what i recommend: disable "(UEFI) Secure Boot" in BIOS (if possible), upgrade bios version to the latest non beta version, install again. [00:49] How do you update the BIOS? [00:49] Sorry My name is self-explanatory on my skill on OS [00:50] bios upgrades are providsed by mainboard vendors. in your case by hp [00:51] i'm trying to get you more info, hold on [00:52] XubuntuNub: https://support.hp.com/us-en/drivers/selfservice/swdetails/hp-pavilion-11-ad000-x360-convertible-pc/15551379/swItemId/ob-234561-1 [00:53] F.45 Rev.A is current. yours is F.08 [00:57] XubuntuNub: got to leave it there for now, good luck! [00:57] Alirght thank you === zbenjamin is now known as Guest48989 === zbenjamin_ is now known as zbenjamin === Leion1 is now known as Leion [02:01] if I type ` nmcli radio wifi off` [02:02] the next line appears, as though the command has executed [02:02] however, when I then type `nmcli device wifi` [02:02] it lists all the local wifi SSID's [02:02] how can I turn the radio off? [02:03] unloading the module may do it [02:03] After writing a Ubuntu image to a USB stick (to create a live USB stick), typically there's lots of free space left, if the USB stick is big enough (e.g. 32 GB). Has anybody succeeded in using the remaining unpartitioned free space? [02:03] isn't that a bit extreme? [02:04] k_sze: Just use gparted to expand the partition to encompass the rest of the space [02:04] I tried to create ext4 in LUKS in the free space, but gnome-disks ends up being really confused and loses track of the original Ubuntu partition (it now thinks the original ubuntu partition is just free space) [02:04] is there no other straightforward way to disable the wifi radio? [02:05] bumblefuzz: Did you try rfkill on it yet? [02:06] how do I do that exactly? [02:08] bumblefuzz: rfkill ... then see the number on the left beside the one you want to disable. Then:rfkill block [02:15] I think I'm running into this: https://askubuntu.com/questions/441016/deleted-windows-e-drive-now-how-do-i-give-that-partition-to-ubuntu#441254 [02:15] which is utterly stupid. [02:15] why doesn't gnome-disks automatically align (or even ask me to?) [02:18] Seems like an unfixed bug from at least 5 years ago. [02:18] https://askubuntu.com/questions/377253/unable-to-format-usb-drive-with-disks-udisks-error-quark-0 [02:27] Guys [02:29] I dont know what to do anymore. I have "The grub-efi-amd64-signed failed to install into /target/" and My Partition is all good (one EFI System with the flags boot and esp and one ext4) [02:29] is the filesystem full? are there more errors in the logs? how about dmesg? [02:31] dmesg? [02:31] dmesg prints the kernel messages [02:31] if you're not familiar with it, it might be better to pastebin the whole thing and ask for advice -- 99% of what's in there is unlikely to matter much one way or another [02:32] FHold one [02:32] I am doing the pastebin [02:36] https://pastebin.com/GxqNG8TD [02:39] [ 579.735250] [Firmware Bug]: Page fault caused by firmware at PA: 0x679de770 [02:39] ewwwww [02:39] Not good? [02:39] [ 579.735470] efi: Froze efi_rts_wq and disabled EFI Runtime Services [02:39] How many month to live [02:39] I wonder if that line is a consequence of the first problem or just some random unrelated thing :/ [02:39] XubuntuNub: I suggest looking for bios updates from hp [02:40] ahh [02:40] Welp... It doesn't have any OS installed anymore... [02:41] So... Yeahhh.... [02:41] @sar [02:41] oh that isn't going to be fun :( [02:41] Want me to tell you my BIOS option? [02:42] sometimes they'll have bootable media, sometimes you can do bios updates from within the bios itself [02:42] I don't know HP gear, it wouldn't do me much good, sorry [02:42] (not that I know anything real well, I don't really deal with anything other than booted running systems..) [03:17] If I'm on Bionic can I add the Eoan Source repository to make my own backports for certain userland applications? [03:19] Well, not just userland. But apps that loads of other things don't rely on [03:22] Kon-: Sure, but while you "can", will likely be a good way to learn how to UN-break your system :) When the libs break will be a painful experience. [03:26] Heh, fair enough [03:39] Hey I use WSL2 in windows 10, how do i enable a service? I have been doing it like this in my bashrc and it seems slightly ghetto https://termbin.com/uv4d [03:45] !wsl | adamantium [03:45] adamantium: Windows 10 has a feature called Windows Subsystem for Linux, which allows it to run Ubuntu (and other Linux distro) userspace programs without porting/recompliation. For discussion and support, see #ubuntu-on-windows or ##windows. For installation instructions, see https://msdn.microsoft.com/en-us/commandline/wsl/install_guide [03:45] thanx [03:45] adamantium: :) [03:48] Is there a way to easily determine if a HWE kernel is currently set as the default kernel or otherwise the currently loaded kernel? [03:51] einichi: ' ls -al / '. The vmlinuz and initrd.img will tell that tale :) === not_phunyguy is now known as phunyguy [03:52] cat /proc/cmdline #check the current boot-up kernel [03:52] From what I can tell, when HWE is installed, it doesn't affect the kernel image names [03:52] Unless I'm mistaken? [03:53] !hwe | einichi [03:53] einichi: The Ubuntu LTS enablement stacks provide newer kernel and X support for existing LTS releases, see https://wiki.ubuntu.com/Kernel/LTSEnablementStack [03:53] When you install the HWE kernel you will continue to recieve updates for both the HWE kernel and the "original" kernel version [03:53] whats the best layout for storing images from a webapp on my server? [03:54] ... this isn't a channel for development. [03:54] o.O [03:54] *my ubuntu server [03:54] still applies. that's not a support question. [03:56] hypercore: might be more appropriate for #php [03:56] depending on what you're using. [03:56] hypercore: ##webdev as well perhaps === HsiaoTing_ is now known as HsiaoTing [03:56] ill try there, thanks [03:57] I still don't have a clear way to determine if the currently loaded kernel is a kernel provided via HWE [03:57] einichi: check your installed packages. HWE kernels have hwe in their package name. [03:57] einichi: then compare to your running kernel version. === not_phunyguy is now known as phunyguy [03:59] or at least it'll tell you what kernel version series is considered HWE [03:59] i forgot, the actual kernel packages don't include it, only the meta package does. [04:01] Right now on a freshly installed 18.04 the kernel version does not change at all when I remove the HWE metapackage and install the linux-generic package, and it again stays the same when I remove linux-generic and reinstall the HWE metapackage. [04:02] !info linux-image-generic bionic [04:02] linux-image-generic (source: linux-meta): Generic Linux kernel image. In component main, is optional. Version 4.15.0.66.68 (bionic), package size 2 kB, installed size 15 kB [04:02] !info linux-generic-hwe-18.04 bionic [04:02] linux-generic-hwe-18.04 (source: linux-meta-hwe): Complete Generic Linux kernel and headers. In component main, is optional. Version 5.0.0.32.89 (bionic), package size 1 kB, installed size 12 kB (Only available for i386; amd64; armhf; arm64; ppc64el; s390x) [04:03] ... weird [04:06] Hmm, so uninstalling the metapackage doesn't get rid of the 5. kernel, nor does it set the prior kernel to the default kernel... [04:06] I guess that makes sense since it's just a metapackage [04:10] einichi: linux-generic: It is only a "meta-package" depending on linux-image-generic and linux-headers-generic. Those two are themselves meta-packages depending on the respective latest image/headers packages. [04:11] einichi: You can see this for yourself by issuing apt-cache show linux-generic, apt-cache show linux-image-generic and apt-cache show linux-headers-generic. [04:42] does the boot-repair tool not exist in ubuntu anymore? [04:42] i searched for it with apt but it doesn't appear. i'm having some EFI issues [04:43] EriC^^: one for you ^ [04:43] ah wait my bad, I didn't read closely enough. it's actually in a ppa [04:45] my problem is i had grub booting with mbr and i'm trying to get it to be efi now [04:52] Bashing-om: Thanks [04:53] What is the correct way to set the GA kernel as the default kernel? My first thought is to edit /etc/default/grub, but is there a better way using apt? [04:57] einichi: 'sudo update-grub' should set to latest installed kernel - if other is desired, then yes editing etc/default/grub one can set the booting kernel. === jje_ is now known as jje === Napsterbater is now known as Guest11735 === Napsterbater_ is now known as Napsterbater [05:10] I just love outdated tech articles on the internet, especially ones from after the trend started of trying to hide when a post/article was written so that people would continue to bother reading it. [05:12] On that topic, anyone have recommendations on MTAs for Ubuntu server? (I'm still on 16LTS.) Minimal setup effort would be good. I just need it for sending messages from PHP scripts--already set up for actual email hosting on this domain with a third-party. [05:21] JackOLantern: Is sendmail not available? [05:31] ok luckly i think i somehow managed to fix my EFI boot partition [05:31] now i'm booting into ubuntu without problems [05:32] i think the key is the --boot-directory=/EFI/ubuntu line? not sure [05:39] Is 19.10 not doing something right with BTRFS snapshots or not coexisting right with apt-btrfs-snapshot? I am getting these messages on every package operation, since 19.10 update: "Invalid snapshot ''IO Error (open failed path:/mnt/@ errno:2 (No such file or directory))." https://pastebin.com/HdgcyVPd [05:42] I would love to jump to ZFS (were using it for years on illumos) but I wonder would Ubuntu be happy to live inside ZFS Boot Environment. [05:43] (beadm tool) [05:44] is btrfs only in 19 and up? [05:45] i'm reading on wiki it has a lot of features but is speed one of them? [05:45] btrfs is available for years [05:45] it's usually faster then ext4 [05:45] beside other things [05:45] ok. i don't remember if i saw it during my ubuntu 18 install [05:45] wait so other distros have it too? [05:46] one cpuld for a long time do manual install and just create partition that mounts to "/" with whatever FS one wants/included [05:46] I bet they do [05:47] snapshot integration is.. light. it just makes snapshots when doing apt operations, that might be enough [05:47] well i don't want to reinstall ubuntu yet again [05:47] maybe i'll try btrfs some other time [05:48] there's "update" of FS from ext4 to btrfs , it just do it in-place [05:48] sounds scary lol [05:49] I've done that 7 years ago or something , and also moving configuration from single disk to 2-way mirror and back, all in-place [05:49] it's not actually scary, because it works and it's fast. ext4 is sad history for a long time.. since 2006 and ZFS in production. [05:50] can i just look on askubuntu for a guide to do it? [05:50] yeah zfs and btrfs have similar ideas right [05:50] does ubuntu have a problem with nvidia? [05:50] BTRFS only edge over ZFS that I know is that it can set RAID level per-dataset, across multiple devices. ZFS do that for naow only per-pool. [05:51] spinningcat, nvidia makes good but proprietary drivers, someone might say that is at the same time blessing and curse. [05:52] yeah i have to use the nvidia driver if i want to clone my VGA and my HDMI monitors sensibly [05:52] johnjay, well, Btrfs tried to imitate and partially inovate. [05:52] hmm dont know either my gpu is broken or ubuntuhave a problemö with nvidia [05:53] nikolam: i feel like i'm behind the times now if i don't switch to btrfs. but at the same time if i can't measure the improvement in speed it feels like chasing a fad [05:53] spinningcat, or if using right driver for your, maybe, older card. Use driver shipped with Ubuntu. [05:53] johnjay, All Linux is behind times for a long time, but it did catch up. [05:53] einichi: yes, sendmail is an option. Will it be difficult to "harden" and just use for sending from PHP? [05:54] Will it cause conflicts w/ existing setup (w/ full mail send/rcv set up on this domain as being hosted by a 3rd party)? [05:55] johnjay, speed is not the matter, snapshots, software RAID, I have recovered form misconfiguration or when not being able to boot, multiple tims, by bringing back old state of all the system. See Phoronix for speed measurements. [05:55] times [05:56] nikolam, there are two drivers available [05:56] snapshot meaning your system didn't boot for some reason, and you reverted to an earlier state? [05:58] johnjay, well yes. I were doing god knows what and then calmly booted to older state of things. On illumso distros and ZFS that's a standard for 13 years or something with 'beadm' and Boot environments. Linux is all new to that. [06:01] Ubuntu have some specifics, system is inside "@" named dataset. so one boot in recovery environment and 'mv' operation to @ (or better btrfs snapshot form previous snapshot, then mv to @) moves you to previous state of system (providing apt-btrfs-snapshot created it) [06:02] 19.10 prints some weird new messages doing apt operations, that's why I came to ask anyone else seeing this. https://pastebin.com/HdgcyVPd [06:33] i want to sort a set of files in a folder based on a increment number inside the filename.. how can i do that? [06:34] i have the regexp to identify the number inside the filename [06:35] zamba: can you give an example [06:36] nevermind, got it working :) [06:36] besides, it was the wrong channel.. i meant to ask in #python :) [07:03] zamba: ah cool :) === akem__ is now known as akem [07:22] hello [07:22] the login password issue is still not fixed in 19.10 [07:23] I select the keyboard as Portuguese but the special characters are stored as US [07:42] Anybody having copy/paste issues in 19.10? Many times, today, I've tried to paste but instead of getting the most recent text that I copied, I get the previous text. [07:43] Maybe I just need to reboot and stop keeping so many windows open at once. ;) [07:46] JackOLantern, I have never seen copy paste problem [07:47] oh wait, sorry I'm not on 19.10 [07:47] Is it possible to change the default connection mode within Blueman bluetooth manager? Have Sony bluetooth headphones that always auto-connect as headset, even when I choose handsfree or audio sink during setup. [07:48] doug16k, ok no problem. [07:50] my machine can't save core dumps. my /var/log/apport.log has lines like this: FileNotFoundError: [Errno 2] No such file or directory: '/var/crash/_usr_bin_gdb.1000.crash' [07:50] I want to look at that core dump [07:52] /var/crash exists [07:59] to be sure I checked `sudo systemctl status apport`, looked fine [08:00] http://paste.ubuntu.com/p/Vh5jgjSsCk/ [08:05] doug16k: maybe the dir /var/crash isn't writable for some reason? ls -ld /var/crash [08:20] if i put 'xset xxx' in .bash_profile, the X would hang there probably due to lack of terminal at that time. but if i put it in .bashrc, it slows down bash startup noticeably, with about 3-5 secs among other things. how to deal with it? [08:33] i am unable to add ppa to install unetbootin ... https://paste.ubuntu.com/p/Fz6dbyYCT9/ === nikolam_ is now known as nikolam [08:42] ls -ld /var/crash -> drwxrwsrwt 2 root whoopsie 4096 Oct 25 03:56 /var/crash [08:43] EriC^^, ^ [08:45] and /var is drwxr-xr-x 14 root root [08:46] seems it is trying to chmod a .crash file that never got created [09:18] hrrm weird ubuntu 19.10 issue... [09:18] Thinkpad X201 worknig beautifully, but booting ubuntu soemitimes swiches the machine off at boot! [09:19] intermittently, someimes does sometimes doesn't [09:19] early in boot sequence in any case [09:19] ubunt logo and red dots... [09:19] and off ewe go [09:19] no signs of cpu overheating or any runtime issues once booted [09:20] any $clue here ?!?!?! [09:22] booting with noapic acpi=off then works ... [09:25] Hi,I have a 18.04 server that *started rebooting every night at about 2am* [09:26] That happened since when I reverted the kernel to the stock 4.something series; I had been running vanilla 5.x before, and no reboots happened. [09:27] Every night in the logs I see this: Oct 24 02:17:01 stingray CRON[24637]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly) [09:27] ^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@ [09:28] The previous message is several minutes earlier, and I see nothing in the logs anyway. The directory /etc/cron.hourly *only* contains the .placeholder file. [09:28] ^@ is NUL bytes [09:28] which tend to appear around the end of logfiles on hard crashes [09:29] Habbie, yes, seems like a sudden reboot and the log is truncated, but it's curious, that the last line is always the cron thing [09:29] well [09:29] for any random time of day, wouldn't the last log line often be a cron thing? [09:29] Habbie, not at all on this machine :) [09:29] ok [09:29] but you're right, it's cron.hourly anyway [09:30] You're right twice, since thinking about that, there are other cron.hourly entries just fine along the day. [09:31] What could cause a nightly reboot? [09:32] if you weren't so sure it was related to switching kernels, i would blame something outside of your computer [09:33] vincenzoml: have you tried apt-get remove windowsupdate? (Just a joke.) [09:33] Habbie, I'm not sure at all (will retry the 5.x series). But it happened coincidentally. One thing could be that thermal control for that chipset only reached maturity in kernel 5.x, so maybe there's some nightly process that makes the CPU very warm and then the system is halted. [09:33] But what could be such a process in ubuntu, that is not advertised in the log? [09:34] vincenzoml, maybe it's only advertised in the log if it runs to completion - run 5.x for a night :) [09:34] Hmm, yes, good suggestion. Also now the 5.x series is in bionic-updates so I don't need vanilla it seems. [09:36] is there a way to tell my system to install the 5.x series and keep it updated, without specifying explicitly the kernel version (e.g. 5.0.0-15)? [09:36] i'd hope there is a metapackage for that, but i don't know [09:36] well, thanks for your help! [09:38] Hey, i got an IO error while trying to mount an SD on 1 computer but it works just fine on 2 other computer and even on a Raspberry, all running Ubuntu 18.04, also i have the voltage signal warning only on the computer with IO error: https://pastebin.com/dwzYft6g [09:39] I think it must be the SD card reader driver on the computer with the problem, what you think? At first i tought the SD card was broken or something :/ [09:40] What strange is that i can mount the first partition on that SD which is FAT, but not the 2nd and 3rd which are ext4. [09:41] 2 seconds away from getting rid of ubuntu and putting windows back on [09:41] How do I even diagnose my constant crashes? [09:42] It's crashed like 4 times in the past 2 hours. It's an issue over multiple versions of ubuntu over multiple re-installs [09:49] hrrm [09:49] logs suggest kernel getting wrong idea cpu temp 100 degC and so shutting itself down! [!] [09:50] well syslog, anyhow, something in there [09:51] some suggesitosn oniline this can be wrong information problem [09:51] though problem intermittent so not alwasy able to debug! [09:51] trying a 5.0.0-32 kernel manually instalred fom debs, bootable, but then problem not recur os kerel 5.3.x so not yet clear what excatly triggers the failure! [09:52] help ;p [10:20] When I try to connect to wifi, it gets stuck at configuring interface. How do I fix this? [10:21] R13ose: hrrm, have you had same machine working okay with same wifi network, on ubuntu before? [10:21] Yes [10:21] R13ose: change in ubuntu version ? kernel update? [10:22] I'm not expert on that particular area but giving a bit more info on circumstances that do and don't trigger the fault... others may recognize issue ... [10:23] Yes but was doing the same before version update. Kernel update, I don't remember before not working. [10:23] R13ose: so you've never had it working? intermittent? works eventually? etc [10:23] more to the point, does it work fine on other wifi networks withouth this failure? [10:23] Works on and off now for some reason. [10:24] you need a comprehensive fault report ... [10:24] I haven't tested other networks. [10:26] R13ose: given you some hints on 'narrowing down' issue [10:26] R13ose: do try other networks, be patient, do ask about another time etc... [10:27] When I am out, I will test other networks [10:29] hello. after update to kubuntu 19.10 can't find php7.2 [10:30] php7.2-fpm isn't found [10:30] can't install for some reason [10:30] yuradoc, php-fpm [10:30] yuradoc, this will give you 7.3 [10:31] but my development is on 7.2 [10:31] i can't upgrade quickly [10:32] https://www.php.net/manual/de/migration73.incompatible.php is pretty small. [10:32] i see https://launchpad.net/ubuntu/eoan/+package/php7.2-fpm is avalable [10:32] but can't install it [10:34] "apt install php7.2-fpm" not found package [10:35] https://packages.ubuntu.com/search?suite=eoan&searchon=names&keywords=php7.2 I recommend using the official package search rather than launchpad (that can mean it may have existed at the start of eoan development, but was later dropped in favor of the current stable version 7.3) [10:36] so there are only way to build from source? [10:38] There's a relatively popular php-ppa, but I can't recall its name right now. It will probably still have 7.2. But I'd strongly recommend migrating, the 7.2->7.3 migration is not a big deal, I think it took us about half a week for our largest application [10:38] i use drupal 7 [10:38] it hasn't migrated yet [10:39] ok, thanks. but possibly there is incompatibility with mysql 8 also [10:39] need to check [10:41] Ah yes, derpal and worstpress, the bane of ever upgrading PHP to anything stable. Condolences [10:43] Howdy all [11:05] Hi, does someone use a HP Thunderbolt Dock 230 G2 with Ubuntu? [11:06] HP Thunderbolt Dock 230W G2 [12:53] In ubuntu (>= 18.04), is it generally possible to install all of the different (semi-)officially supported desktop environments and switch between them at any time? [12:54] I mean when I install one desktop environment, it doesn't force me to uninstall packages required by another desktop environment. [12:56] k_sze: i think it's anybody's guess, mine is that it could work, but you might need to tweak a few stuff, iirc installing lubuntu with unity on 16.04 messed up the notifications in unity becoming lubuntu style and some stuff like that [12:56] in general it's doable i guess, i'd say if you dont want to risk your install, try it in a vm first [12:58] EriC^^, right, I remember trying KDE Plasma and it switched my splash screen and mouse cursor even when I went back to GNOME. [12:58] It was quite annoying. [12:58] yeah, they're really dont have any foresight for dual desktop's installed and stuff [12:59] :/ [12:59] maybe somebody should make some tool that let's you choose stuff easily, or switches between them based on what you load [12:59] that'd be pretty cool and ubuntu-specific, might get more people to use ubuntu too [13:00] Maybe put each desktop environment into its own partition, lol. [13:02] :D [13:02] The thing is, I really want to try the different desktop environments to decide which one I like best. But it's sounds like soooo much trouble. [13:03] k_sze: that's really where a vm can be very useful [13:03] make 1 ubuntu, clone it, install different desktop, clone the original, install another, try and see then barebone install what you want [13:07] There are things that can't be properly tested with a VM though. [13:07] such as GPU driver stability. [13:07] hello everyone :) [13:07] I have upgraded to the latest version and it's looking pretty good so far [13:08] the ui changes and the improved performance and everything else [13:08] thanks! [13:08] except that Chromium has begun acting weird since it has been packaged as a Snap app [13:09] k_sze: i guess [13:09] I can no longer open a pinned Chrome app in its own window [13:09] it opens the generic Chromium window instead [13:10] is that because of how snap manages things or is it an issue with Chromium itself? [13:10] sentiment: report a bug with the 'snap' tag and mention it in this thread: https://discourse.ubuntu.com/t/call-for-testing-chromium-browser-deb-to-snap-transition/11179/174 [13:10] ah ok :) [13:11] I just noticed something else btw [13:11] clicking that link just opened the Chromium window and didn't start the navigation [13:13] could be you're suffering from the "snapd upgraded your chromium in the background and now everything's wacky until you'll close all the windows and restart it" situation [13:13] note: this also might mean your browsing session might get lost [13:13] as in, the new chromium will restore the state from when snapd decided to do the silent background upgrade, not from when you close the window [13:13] (this includes tabs, bookmarks and history) [13:14] mgedmin: hmm upgraded when? like now? I don't think so. it did upgrade the apt package though [13:14] you can check with 'snap changes' in a terminal [13:14] I didn't know it does silent auto upgrade [13:14] "Auto-refresh snap" is the silent auto upgrade in the background [13:15] nice command :-) [13:15] though it says Chromium was updated yesterday [13:17] the relevant part is whether you've restarted Chromium after it was updated or not [13:17] mgedmin: can you pin a website to the launcher and see if it show its own icon or the generic one when you launch it? [13:17] you can detect this by looking at the blue dot in the dock [13:17] mgedmin: I have [13:17] k [13:17] how does one pin a website to the launcher? [13:18] click the three dots menu and -> more tools > create shortcut [13:18] do I check the "open as window" (tr?) checkbox? [13:18] yeah [13:19] (create shortcut was translated as "create hot key" in lt_LT, which made no sense -- chrome's translations are pretty poor quality overall!) [13:19] My wifi is still down. What do I do? [13:19] sentiment: I get no launcher? just a new window with an orange border [13:20] I close the window, still no icon in the launcher [13:21] mgedmin: it creates a shortcut in the launcher app list [13:21] greetings, how can i obtain the kernel headers for old kernel, like 4.1.20 [13:21] you have to open it from there and then add it to favorites [13:21] (to pin it to the dock) [13:22] R13ose: what does iwconfig show? [13:22] sentiment: I don't see one [13:23] there's one "Chromium We..." with a totally blank icon [13:23] weird. type chrome://apps in Chromium [13:23] it just opens a new window [13:23] it should list your shortcuts (apps) [13:23] chrome://apps shows the thing [13:23] when it's open launcher shows two orange dots next to the chromium icon [13:24] mgedmin: yeah, that's the problem [13:24] it used to open with its own icon in 19.04 [13:24] so your bug is confirmed [13:24] sentiment: I have wlp3s0 which shows IEEE 802.11 and lo which says no wireless extensions. [13:24] I'm gonna post it to the site you mentioned [13:25] R13ose: can you paste the output? [13:31] sentiment: https://pastebin.com/UuC2CHVA [13:44] Any further thoughts? [13:44] R13ose: try "iwlist wlp3s0 scanning" [13:45] you should see your acess point in the output [13:46] How can I download a package and install on some other machine which doesn't have internet access? [13:48] mra90: You can try looking for the deb files in the repositories, however, there are usually dependencies that apt automatically fetches on update/install. [13:48] I need hd progream [13:48] program [13:49] mra90: try "apt-get download" [13:49] I try to download sources by apt-get source hd [13:49] E: Unable to locate package hd [13:49] what's hd? are you sure of the name? [13:49] R13ose: do you? [13:50] yes hexdump, [13:50] sentiment: you can't use apt on one machine to download packages for another [13:50] hd is hexdum [13:50] hexdump [13:50] sentiment: You won't get all the dependencies the other machine will need [13:50] pragmaticenigma is right [13:50] pragmaticenigma: so what I can do? [13:51] sentiment: yes I see that. When I try to connect to that in the GUI, it get stuck on configuring interface [13:51] can he/she use snap instead? [13:51] mra90: You connect the machine you need the package on, to the network and use the tools available. Your request is a use case that isn't supportable with the tools available in Ubuntu [13:51] pragmaticenigma: I cannot connect it the network card is broken ;/ [13:51] HW failure [13:52] mra90: You can purchase an USB network adapater for really cheap these days... sounds a lot easier to do that to me [13:52] first of all I need to find out in what package the HD app is located? [13:53] I think it would be easier to fetch that package on other machine and install offline [13:53] mra90: There is no package called hd... [13:53] R13ose: try "iwconfig wlp3s0 ap " [13:53] pragmaticenigma: Yes so how to locate package in which HD is? [13:53] mra90: As I said, the software and tools Ubuntu ships with are not intended or capable of accommodating your use case [13:54] !info hd | mra90 [13:54] mra90: Package hd does not exist in eoan [13:54] !info hd bionic | mra90 [13:54] mra90: Package hd does not exist in bionic [13:55] so where is it? :( [13:55] !offline [13:55] If you need to download Ubuntu packages using another machine or OS, check the desired packages in Synaptic and select File > Generate package download script. See also !APTonCD [13:55] /usr/bin/hd comes from bsdmainutils on eoan [13:56] Hi. Does anyone know a way to stabilize videos in ubuntu? I mean a simple way, like opening a video, press stabilize, and save the result. [13:56] I currently upload my video to google photos, use the stabilizer on the app, and download the video again. [13:56] but it's not very... efficient. [13:57] okay I found it on my machine it just has different name [13:57] hexdump [13:58] however my question how to locate this utility and mainly its package is still vaklid [13:58] there must be a way [13:58] mra90, see !offline [13:58] oerheks: offline? [13:58] easy to download and transfer to a non-network machine [13:59] what is it? [13:59] read back please? [13:59] why repeating, such a waste of bits [13:59] !offline | mra90 [13:59] mra90: If you need to download Ubuntu packages using another machine or OS, check the desired packages in Synaptic and select File > Generate package download script. See also !APTonCD [14:00] ok thanks [14:00] !APTonCD [14:00] APTonCD is a tool with a graphical interface which allows you to create one or more CDs or DVDs with all of the packages you've downloaded via apt-get or aptitude, creating a removable repository that you can use on other computers - See also !offline [14:01] ./ffprobe: error while loading shared libraries: libavdevice.so.58: cannot open shared object file: No such file or directory [14:01] where do you find libavdevice.so.58? [14:02] realies: try https://packages.ubuntu.com/, there's a filename-based search box a bit down that page [14:03] realies, more details please, what are you doing, what ubuntu version etc [14:03] but, uh, it would be better to find a properly packaged version of this thing, whatever it is [14:04] sentiment: I did that [14:04] oerheks, compiling ffmpeg from source, 18.04 [14:04] mgedmin, is not available in bionic unfortunately :( [14:05] libavdevice57 is available but not 58 [14:05] if you're compiling from source, then it should pick the version that is available (the one for which you have the -dev package installed) [14:06] weird... [14:06] where did you get the source? [14:06] oerheks, was following https://seanthegeek.net/455/how-to-compile-and-install-ffmpeg-4-0-on-debian-ubuntu/amp/ [14:07] so that would be wget https://ffmpeg.org/releases/ffmpeg-4.2.1.tar.bz2 [14:07] which is the latest version [14:08] libavdevice* packages are built from the ffmpeg source package [14:08] read the comments, lolz [14:08] so you shouldn't need to install new ones if you're building from source? [14:08] good luck realies [14:09] if you've built and are now trying to run things without installing, you may need to set the LD_LIBRARY_PATH environment variable so that your locally built binaries can find your locally built libraries [14:09] is there a repo for the newly released python 3.8? [14:09] if you're trying to install, well, careful you don't break the OS [14:09] isomari: ppa:deadsnakes has it [14:10] mgedmin: that repo does not work? [14:10] disco and up gives 3.8 https://launchpad.net/ubuntu/+source/python3.8 [14:10] isomari: besides, eoan has python 3.8.0 in the main archive (as an "available" version, but not a "supported" version) [14:10] sentiment: what further should I do? [14:10] mgedmin: how do I install it? [14:11] isomari, install on what version? [14:11] eoan [14:11] sudo apt install python3.8 [14:11] Err:7 http://ppa.launchpad.net/deadsnakes/ppa/ubuntu eoan Release [14:11] 404 Not Found [IP: 91.189.95.83 80] [14:11] then you already have 3.8 [14:11] ah, deadsnakes only supports LTS releases [14:11] I C. ok [14:12] python3.8 is not found on u eaon installation. [14:13] https://packages.ubuntu.com/eoan/python3.8 [14:13] do you have universe enabled? [14:13] yes [14:16] isomari: can you run this and post the output? sudo /bin/true && cat &>/tmp/aptlog < <(sudo apt-get -y update 2>&1; apt-cache policy 2>&1; sudo apt-get -syV full-upgrade 2>&1; apt-cache policy python3.8 2>&1;); nc termbin.com 9999 it wont change your system [14:17] I just ran install python3.8 and it's installing. [14:17] What do I do about my wifi? [14:18] R13ose, ask a question with full details about it? [14:19] When I try to connect to my wifi, this gets stuck with configuring interface and doesn't fully connect. How do I fix this? [14:23] that's a good start but not a lot of details, yet, R13ose. do you know which ubuntu version and flavor your have there, or is it a headless server, which kernel version you run, do you know which wireless chipset you have and which driver is loaded? [14:23] 19.10, kubuntu [14:24] I don't know any of the other answers [14:24] ok, that's fine, those can be worked out. [14:24] R13ose, for kernel, paste the output of "uname -r" [14:26] 5.3.0-19-generic [14:26] about the wireless NIC: lspci -knn | grep -EA3 '(Network|Ethernet) controller' | nc termbin.com 9999 [14:27] There is also an error when I try to connect to the wifi saying "secrets were required but not provided" [14:29] The wireless NIC: https://termbin.com/zpbo [14:30] R13ose, it lacks authentication probably; delete and create again the connection in NetworkManager [14:33] Still the same issue [14:33] R13ose: if this doesn't fix it, yet: did this wireless work previously on any linux version, on any ubuntu version, on any other OS? what changed before it stopped working? [14:34] R13ose, restart the router [14:34] I don't know what changed but worked before. The previous version this worked but was going in and out last night even before I upgraded to 19.10. [14:34] Let me restart router. [14:36] here are some module options to try: https://askubuntu.com/questions/675352/wireless-disconnects-intermittently-with-intel-corporation-centrino-advanced-n-6 [14:38] this is a fairly old chipset, BT coexistence and power saving may not work that well. and maybe 802.11n, too, as this post suggests. [14:43] I want to report a bug against Chromium using ubuntu-bug, but it doesn't let me write details of the bug. [14:43] any clues? [14:44] chromium is a snap now, right? [14:44] it just proceeds with a generated report [14:44] the apt package name is chromium-browser, the snap is called chromium [14:44] oerheks: yes, it's a problem with the new snap version [14:45] is that how I am supposed to report a bug? [14:45] ubuntu-bug is a symlink to apport, i think, it probbaly has no understanding of snaps [14:45] snap info chromium [14:45] what about this then: https://forum.snapcraft.io/t/call-for-testing-chromium-browser-deb-to-snap-transition/11772/29 [14:46] that's point to launchpad for bug reporting, right === JanC_ is now known as JanC [14:46] publisher: Canonical✓contact: https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bugs?field.tag=snap [14:46] do you have your launchpad account logged in? [14:46] nope, I'm gonna try iy [14:46] it [14:47] weird how that link refers to ubuntu-bug [14:48] i see "the apport hook has been updated to include relevant information about the snap package and its dependencies" [14:49] tomreyn: so is it possible at last or not? [14:49] so according to this post you shoulöd indeed be able to use ubuntu-bug chromium-browser [14:49] yeah, but it just sent a generated report [14:49] i have no first hand experience [14:49] looks like you now need to report two bugs instead of one [14:49] and didn't let me write details of the bug or symptoms [14:49] okay thx [14:49] haha yeah [14:50] tomreyn: that worked out, thanks [14:50] R13ose: what is "that"? [14:51] restarting the router (this was ior1as idea, not mine), or adding some module parameters (which?)? [14:53] tomreyn: the answer in the link you provided. [14:53] R13ose: there are multiple suggestions there, which is why i'm wondering which one you used [14:54] sentiment: see these bugs which have already been filed against the snap: https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bugs?field.tag=snap [14:54] apport-cli worked [14:54] I'm about to browse that same page [14:55] it doesn't open automatically after clicking a link in the terminal [14:55] it really needs updates and fixes [14:56] ouch, wonder why that is [14:56] ctrl-clicking links in gnome-terminal works for me [14:57] (but I have done some bananas things with default browser associations and mime types for reasons) [14:57] (my default browser is a shell script that dispatches the url either to firefox or to chromium-browser, depending on the link's domain) [14:57] tomreyn: the commands under this text "To restart the services without rebooting to test alternative configurations:" [14:57] sentiment: try middle mouse button on link [14:58] R13ose: ok, thanks! [14:58] No problem [14:58] always good to know which fix worked. [14:58] rud0lf: middle button would paste, no? [15:02] How do I make sure libffmpeg.so is updated all the time when I use the opera browser? [15:03] i'm not sure now, i may use another terminal app [15:05] R13ose: which libffmpeg.so do you mean? the one which comes with opera? is it not self-explanatory then? or one which comes with another packages? [15:06] found it [15:06] https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/1732482 [15:06] Launchpad bug 1732482 in chromium-browser (Ubuntu) "[snap] doesn't properly save desktop files for "create shortcuts" action" [Medium,Triaged] [15:06] dpkg -S libffmpeg.so will help you determine which of these files you have are from debian packages (installed via apt), and which package they belong to [15:06] R13ose: ^ [15:07] belongs to 2017! [15:07] !who | sentiment [15:07] sentiment: As you can see, this is a large channel. If you're speaking to someone in particular, please put their nickname in what you say (use !tab), or else messages get lost and it becomes confusing :) [15:07] tomreyn: the one here: https://github.com/iteufel/nwjs-ffmpeg-prebuilt/releases/ [15:08] :) [15:09] R13ose: well, looks like you identified one of the major disadvantages of not using ubuntu packages. [15:09] Oh [15:09] i should say *supported* ubuntu packages [15:10] no support, no upgrade path, no timely patches for critical security vulnerabilities [15:11] (nor bug fixes) [15:16] Does Opera Browser support Ubuntu packages? [15:17] it's proprietary, so you'd need to look for third party packages. but they seem to provide apt repositories [15:22] So there is a way to update libffmpeg.so when I update other packages? [15:24] not for the one you ninstalled from some (randomly chosen?) github repository, no. === beaver is now known as morebaby [15:26] Okay thanks [15:29] I get this error in my dmesg https://paste.ubuntu.com/p/fSGt3vWXNH/ any idea what it is and if its harmless [15:29] it's a skylake cpu [15:31] aiena: which ubuntu version, which kernel version? [15:31] it's not harmless [15:32] it's most likely also not going to break your hardware, but something you should try to fix. there are data transport problems on your PCIe bus. [15:34] this can be due to bad / insufficient power supply, broken PCIe card, broken mainboard / PCIe slot, broken riser card [15:34] tomreyn https://paste.ubuntu.com/p/YC3jJYWY7K/ should give you the basics [15:34] tomreyn its a laptop [15:35] nc termbin.com 9999 < <(lsb_release -ds;cat /proc/{version,cmdline};) [15:35] aiena: ^ please show the kernel version as well [15:35] ah sorry [15:36] aiena: or you can actually just post a full kernel log: journalctl -b | nc termbin.com 9999 [15:36] https://termbin.com/mf7s [15:38] aiena: did this happen only recently? do you actually have notable effects? [15:38] no notable effects [15:38] I havent been monitoring this laptop full time as i had installed linux foro my cousin [15:39] he's just concerned about the error but ubuntu bots fine, the wifi works and audio works too [15:39] aiena: can you run journalctl | grep 'Multiple Corrected error received' | head | nc termbin.com 9999 [15:39] It seems to be in the journal for pretty log [15:41] pretty long [15:42] i also see thi son your log - did we discuss this the other day? mce: [Hardware Error]: Machine check events logged [15:43] tomreyn no I see this in the log too and was curious about it [15:43] it seems like a CPU error but since all apps and ubuntu work well i dont know it's meaning [15:44] have you looked for a firmware upgrade? [15:44] i havent discussed this before these errors are things I've seen for the first time myself [15:44] byfirmware upgrade you mean BIOS upgrade [15:45] (okay, there was someone else here with a dell system and mce log records here recently, so i was wondering whether it was you.) [15:45] yes, "bios upgrade" [15:48] aiena: https://www.dell.com/support/home/uk/en/ukbsdt1/drivers/driversdetails?driverid=fr32j&oscode=biosa&productcode=inspiron-15-3567-laptop [15:48] tomreyn yes there seems to be a bios upgrade [15:48] it's currently at BIOS 01.02.00 11/30/2016 [15:49] which is probably the 1.2.0 one listed on the "Version" history you can expand on the bottom of this page [15:50] tomreyn problem is the bios upgrade tool is a .exe file lol [15:50] "can be used to install from any Windows or MS DOS environment." [15:50] you can create a bootable freedos usb stick and place it there [15:57] aiena: your other option would be contacting dell support, since this computer is (or was?) supported on ubuntu 16.04. [15:58] so they should make some efforts to provide bios updates for linux, too, they actually provided some via lvfs https://fwupd.org/lvfs/search?value=Inspiron which you can install from "ubuntu software", but this is a rather old one. [15:58] it's still newer than the one you have now, though [15:59] so this one specifically https://fwupd.org/lvfs/device/com.dell.uefi1d4362ca.firmware [16:00] if you'll contact support, maybe remind them that their ubuntu customers would generally appreciate them upgrading their firmwares via LVFS again [16:02] tomreyn apparently the bios upgrade ends in exe but the bios recognises it [16:03] so i dont really need freedos but I'll sort this out lalter [16:03] thanks for the tip lets see if it dissapears [16:03] ah you can just put it on a fat32 formatted usb stick? nice. [16:04] aww gone. anyone what device [8086:9d14] is? [16:05] CarlFK: vendor id 8086 is .. inside joke... intel [16:06] you can also search /usr/share/misc/pci.ids for it [16:08] tomreyn: It's not listed in that file [16:08] it is in mine, after running update-pciids -> Sunrise Point-LP PCI Express Root Port #5 [16:09] also on aienas, see line 45: https://paste.ubuntu.com/p/YC3jJYWY7K/ [16:09] not seeing it in mine tomreyn ... odd [16:09] 8086 9d60 100 Series PCH/Sunrise Point PCH I2C0 [Skylake/Kaby Lake LPSS I2C] [16:10] CarlFK: 9d14 != 9d60 [16:10] close enough ;) [16:12] pragmaticenigma: so if you run update-pciids and then run grep 'Sunrise Point-LP PCI Express Root Port #5' /usr/share/misc/pci.ids you don't see it listed? [16:12] nm... figured it out... greping '8086.9d14' doesn't match the data form [16:12] *format [16:13] right, the format of the file differs from lspci (-nn) output [16:14] hi latest samba build requirements package is looking for python-gpgme this is no longer the package name but a package hasnt been updated to reflect this one out of about 30 packages can anyone tell me how to get around this problem [16:15] steps to reproduce problem: https://paste.debian.net/1110639/ [16:16] this is on bionic [16:17] use punctuation, it does serve a purpose. [16:17] transhuman: where did you find that list of packages? [16:17] s/reflect this/ reflect this. This is [16:17] on sambas site [16:18] url? [16:18] https://wiki.samba.org/index.php/Package_Dependencies_Required_to_Build_Samba  [16:18] and your ubuntu version is? [16:19] its necessary to build samba from scratch in order to get samba to work with windows 10 shares [16:19] ubuntu bionic [16:19] so lsb_release -ds returns what? [16:20] um never heard of that command how do i use [16:20] you type or copy and paste it to your terminal [16:20] ah ok 19.04 [16:20] no [16:20] transhuman: the ubuntu 18.4 script just has python3-gpg [16:21] Does Ubuntu create proprietary software? [16:21] transhuman: I bet that just works [16:21] which scirpt CarlFK [16:21] transhuman: the one on the page you sent me [16:21] Ubuntu 18.04 bootstrap/generated-dists/ubuntu1804/bootstrap.sh [16:22] uio, Canonical? [16:22] + community or however it works... [16:22] uio: I use my ubuntu box to write software that is pretty proprietary [16:23] CarlFK, Right, that's what I was asking. lol [16:25] CarlFK, I didn't ask 'can Ubuntu be used to create proprietary software'... Does Ubuntu [as an entity, organisation, Canonical + community] create proprietary software? [16:25] transhuman: it is not necassary to compile from source Samba. You have to learn how to configure it, and verify which protocol version your computers are trying to use. [16:25] uio: this is not a support question. you can discuss ubuntu in #ubuntu-discuss, other topics in #ubuntu-offtopic [16:26] tomreyn, Depends on how narrow one's definition of support is... but sure, I'll head there. [16:26] thanks pragmaticenigma that works [16:27] uio: I suspect the answer is still yes. but your question probbly ins't well defined enough. what are you getting iat? [16:27] except has to be updated to perl-modules / perl-modules-5.26 [16:27] CarlFK, [I've moved over to offtopic :)] [16:29] pragmaticenigma: i have this already on my ubuntu 18.04 system it allows me to connect to windows shares without back dating windows to smb-1.0 which is dangerous but is the way everyone thinks is the way to fix it [16:29] going to write up an article about it or do a youtube video on it if i get time [16:30] transhuman: If you are using the command line to mount, you cannot use type "cifs" as that is version 1.x [16:30] windows requires smb 3.1 for security reasons, using smb 1.0 is dangerous [16:32] hello [16:32] about 10 million people just go ahead and install smb 1.0 to fix the problem ... a big no no [16:32] transhuman: please keep this to support topics only. your commentary is not helpful or beneficial === decadent is now known as depraved [16:33] i have some bad experience with trying to install an earlier version of ubuntu on a simple, low budget ACER laptop where the problem lay with the BIOS ... Now i am considering a dual boot on a Medion laptop ... anyone can tell me if it will work? [16:33] ubuntu inter-operation with windows isn't important OK [16:34] transhuman: Stop, I'm trying to help you, but your continued disregard for this being a support channel is making me less interested [16:35] ok have to go anyway [16:35] thanks for the help its appreciated [16:35] transhuman: here's the solution: https://askubuntu.com/a/1009485 [16:35] and I am telling you from the other side that that is dangerous [16:36] no worries have to go anway [16:36] transhuman: that tells you how to set it so it uses version 3 [16:36] you didn't read it [16:37] yes i did W10 uses 3.1 [16:37] anyway thank you you solved my problem [16:45] Hi. I've got a Ubuntu server connected to an AD domain, where users log on with their AD username. Works great. On this server, I want users to be able to mount a SAMBA share, without entering their password again. How do I do that? [16:45] I use PBIS for the AD integration (formerly known as likewise-open) [16:46] reask in #ubuntu-server ? [16:47] oerheks: oh, thanks, will try there [16:48] mgee: search about automountfs [16:48] lupulo: thanks, will do [16:48] mgee man automount [16:50] lupulo: I know automount a bit, but got a bit lost on how to re-use the already active credentials. Will search further [16:59] I'm having lots of trouble with an upgrade from 16.04 to 18.04, anyone around to help? [16:59] I recently did two systems, and then upgraded the kernel to 5.x [17:00] but it went great [17:00] for there is some weird dependency thing I cannot figure out [17:01] some package you need wants it? === CDuv is now known as CDuv_afk [17:02] no, do-release-upgrade fails no matter what [17:02] I have run distro-upgrade first, unfrozen some packages, etc, nothing works [17:03] claviger-pc: Pastes to https://paste.ubuntu.com/ what fails looks like [17:03] can you paste the output when you try, to pastebin.com? [17:03] ok, main.log ? [17:04] you using a term window? the text from that [17:07] https://paste.ubuntu.com/p/S8Q3hNTjZq/ [17:07] that is the main.log [17:07] the screen log is a hundred MB and is useless [17:08] how do check battery life via cli? [17:08] CarlFK: https://paste.ubuntu.com/p/S8Q3hNTjZq/ [17:10] claviger-pc: such packages and the dependencies they introduce in apt, can make upgrades fail: Foreign: weewx, Obsolete: libicu64 libssl1.1 libwebp6 [17:11] tomreyn, okay, but I tried to unistall the libraries, they seem essential [17:11] weewx I suppose I could unistall and reinstall [17:12] there's no libssl1.1 on xenial [17:12] so it can't be essential [17:13] the same goes for th eother two libs [17:13] ok, I've got issues then, even worse: https://paste.ubuntu.com/p/znyKC6fXwc/ [17:14] if i try to uninstall them, it basically wants to remove a huge plethora of packages [17:14] try them one by one [17:14] wait, are you on 16.04 or already on 18.04? [17:15] it's not clear how far your upgrade got [17:15] on 16.04 [17:15] upgrade fails [17:16] the log you posted suggests you did not start the upgrade, yet [17:16] this is true [17:16] if fails out each time [17:16] ubuntu 19.10 battery icon no longer available [17:16] surface pro 6 :( waaaaaah [17:18] on asus laptop the icon is visable [17:19] claviger-pc: so clean your xenial up first of all, then try the upgrade again. apt-forktracer can help you identify (maybe more) packages which should not be installed. here's a wrapper script around apt-show-versions i wrote for the purpose of identifying packages which should be downgraded or removed: https://github.com/tomreyn/scripts/#foreign_packages [17:20] claviger-pc: you can also apt install packagetokeepinstalled packagetoremove- (note the trailing dash) to help apt find a more deirable solution [17:24] tomreyn ok, I am trying this second one first [17:28] Hello Guys, I am trying to install Xubuntu and I have a error message : grub-efi-amd64-signed package failed to install into /target/ [17:29] Here are the dump file : Yesterday https://pastebin.com/GxqNG8TD [17:29] https://pastebin.com/QtXiyWqk [17:30] tomreyn [17:30] sorry, is there an easy way to downgrade? [17:30] claviger-pc: apt install package=targetversion [17:33] Anybody can help me? [17:33] LoneBadger: which ubuntu version are you trying to install there? [17:33] Xubuntu 19.10 [17:34] Since Ubunut 18 LTS is too demanding for my PC [17:34] hmm this kernel is from july [17:34] i dont think that's how 19.10 installers released [17:35] is this a pre-release installer? [17:35] How can I tell that? [17:35] what's the file you downloaded called? [17:36] I don't know and I just want to provide you with anything that can help me [17:36] you don't know which ISO file you downloaded? [17:36] Xubuntu-19.10-desktop-amd64.iso [17:37] hmm this looks like a release [17:37] LoneBadger, is lots of disk space free on /target? [17:40] /dev/sda2 Ext4 Mount point is /target and unsused space is 449.42 GiB [17:40] unrelated but I suggest staying with ext3 for a while, so you have more variety in supporting recovery programs. :) [17:41] I'm confused about what the current problem is. a dump file means sth. crashed? or is it a problem with installing a package? or does apt/dpkg crash? [17:41] LoneBadger: do you still have the iso file? can you run sha256sum Xubuntu-19.10-desktop-amd64.iso [17:43] if you have a weak machine but lots of eagerness, you could try starting with a minimal install (Ubuntu mini iso, or if you're deep into it, debootstrap) and build step by step from there. [17:44] I like the debootstrap approach because it assures me I'll have the least amount of unwanted stuff. [17:44] Tom [17:45] No Such file or directory [17:45] oh I see now, the crash dump seems to be about broken wifi drivers. try using ethernet and disable wifi if possible. [17:45] LoneBadger: you need to run ithis in the very directory the iso file is in [17:45] at least until you get the other stuff stable [17:47] LoneBadger: if you know the path to the iso file, post it here and i can tell you the command to run from an directory. [17:48] sha256sum -b /full/path/to/whatever.iso [17:48] It is in /dev/sdb1 [17:49] "in" meaning inside a filesystem, or raw-copied onto sdb1? [17:49] Sbd1 is my usb key where the ISO is [17:50] what program did you use to write it onto that USB key? [17:50] LoneBadger: do you still have the .iso file oyu downloaded, though? [17:52] Yes In this PC (not the one I am installing Linux on [17:52] LoneBadger: and which operating system does this PC where you still have the iso run? [17:52] Win 10 [17:52] Filepath is "C:\Users\Nicolas\Downloads\xubuntu-19.10-desktop-amd64.iso" [17:53] can you find the exact file size in bytes? then we can verify the bytes on the USB key [17:54] 1,609,344 [17:54] LoneBadger: please open powershell and run: Get-FileHash C:\Users\Nicolas\Downloads\xubuntu-19.10-desktop-amd64.iso | Format-List [17:55] LoneBadger: then post the output to https://paste.ubuntu.com and tell us the resulting url [17:55] https://paste.ubuntu.com/p/PDcf3R6Qcs/ [17:56] LoneBadger, meanwhile, on the Ubuntu, run "sudo head --bytes=1609344 /dev/sdb | sha256sum -b" [17:57] this hash matches http://cdimage.ubuntu.com/xubuntu/releases/19.10/release/SHA256SUMS [17:58] i'm checking the xubuntu 19.10 installer kernel locally in a VM now [17:59] https://i.imgur.com/ka6Aw4A.png [18:00] LoneBadger: you must have accidentially written something else onto your installer [18:00] What is it? [18:00] How do I get rid of it? [18:02] LoneBadger: the file you posted here insitally, https://pastebin.com/raw/GxqNG8TD , shows that you were running an installer with this linux kernel version: Linux version 5.0.0-23-generic (buildd@lgw01-amd64-030) (gcc version 7.4.0 (Ubuntu 7.4.0-1ubuntu1~18.04.1)) #24~18.04.1-Ubuntu SMP Mon Jul 29 16:12:28 UTC 2019 (Ubuntu 5.0.0-23.24~18.04.1-generic 5.0.15) [18:03] LoneBadger: however, the xubuntu 19.10 installer actually comes with kernel version 5.3.0-18 #19, as seen on my screenshot at https://i.imgur.com/ka6Aw4A.png and also confirmed by http://cdimage.ubuntu.com/xubuntu/releases/19.10/release/xubuntu-19.10-desktop-amd64.manifest [18:03] I tested Ubuntu Yesterday on a live session but it was too demanding [18:03] LoneBadger: what you need to do is to again write the .iso file you have on your windows computer to the installation media. [18:04] LoneBadger: you mean ubuntu gnome? [18:04] Ubuntu 18.04 [18:04] LTD [18:04] LTS* [18:04] okay, but how is this related to the issue at hand? [18:05] you want to install xubuntu 19.10, right? [18:06] I doN,t know my bad [18:06] oh, maybe you mean to say that the log you posted at https://pastebin.com/raw/GxqNG8TD is an ubuntu 18.04 LTS installer log? [18:06] DL the Xubuntu ISO [18:06] Yes [18:07] so you don't actually have a log of the xubuntu 19.10 installer? [18:07] what is this second output you posted? https://pastebin.com/raw/QtXiyWqk [18:07] this is some linux kernel 5.3.0 kernel log. maybe this is from the xubuntu 19.04 installer? [18:08] This one is today [18:08] today doesn't help me much [18:09] Do you think the two kernel are conflicting each other6 [18:10] ? [18:11] hmm, we're having communication problems. ok, forget everything i said so far, i will just work based on assumptions. you're saying that you ran into the error message "error message : grub-efi-amd64-signed package failed to install into /target/" while installing xubuntu 19.10 [18:11] Yes [18:11] can you disable "secure boot" on your bios and try installing again? [18:11] Wilco [18:14] LoneBadger: also if the installation fails again afterwards, boot the installer again, then open a terminal (press ctrl-alt-t) and share your systemd journal with the volunteers helping here: journalctl -b | nc termbin.com 9999 [18:15] is there any way to limit gparted's disk scanning to the disk(s) that I want to work with? I have some other (very slow) disks and it takes forever to pointlessly read their partition tables each time gparted refreshes. [18:15] if uploading it to termbin.com (what this command does) does not seem to work (i.e. the command wont return a http address) then get help here with transferring the log to the other computer. [18:15] LoneBadger:^ [18:15] Formating the USB atm and placing the ISO [18:15] I will do that tom [18:16] LoneBadger: okay, good luck. i hope you will have the "(UEFI) secure boot" option on your bios. [18:16] It is disabled [18:17] LoneBadger, you can skip formatting. the ISO contains all the partition and filesystem info required. [18:17] Sven_vB: i think gparted always reads all partition tables when it starts [18:17] tomreyn, even worse, it reads all of them whenever it refreshes, e.g. because I created a file system [18:18] it makes work feel like batch computing in the olden days, submit your request and come back in a few hours… well, minutes, but still. [18:19] Sven_vB: you could physically disconnect the other storages, or use a different software. [18:19] tomreyn, nope, one of the slowest is the live USB where gparted and the casper-rw resides. [18:20] yeah the 2nd would probably be the better approach. what other nice GUI-y partition managers do we have for xenial and/or bionic? [18:21] Sven_vB: there is also gnome-disks, but i'm not convinced it is nicer. i prefer the working on a terminal. [18:21] or maybe a CLI PM that displays similar details in its partition list, esp. GPT partition label, FS type, FS label, human-readable size? and ideally print the labels in full even if they're longer than 20 characters [18:21] with an option to include free space in the same list as the partitions [18:22] i guess parted and fdisk or sgdisk do this [18:22] maybe not all of it, i'm not sure about these specific features [18:22] really? nice [18:22] oh ok [18:23] but they all limit themselves to working on the given storage [18:23] for a moment I hoped they'd suddenly have them and I just didn't notice [18:23] Installing [18:23] yeah but managing partitions w/o proper overview is kinda yolo. [18:24] "risky" is the better word. [18:26] yes, that's why i prefer to worked on specirfic devices, well targetted, not missing potential warnings, using proven-to-work utilities on a terminal [18:26] *work [18:27] Alright [18:28] Option of Installation : Reinstall Ubuntu 19.10 / Erase Ubuntu 19.10 and reinstall / Erase disk and install Xubuntu / Experimental :erase disk and use ZFS / Something else [18:30] What do I chose tomreyn [18:31] Anyone ever have the issue where the GRUB only loads every other boot? [18:31] LoneBadger: can you actually post the log first of all, it may be worth reviewing this [18:31] If I don't have Any USB? YEs [18:32] LoneBadger: is the installer connected to the internet? [18:33] !smart | k1tsun3: check your disks' health [18:33] k1tsun3: check your disks' health: smart is Self-Monitoring, Analysis and Reporting Technology, a monitoring system for hard drives. See https://help.ubuntu.com/community/Smartmontools [18:33] https://termbin.com/e2rz [18:34] LoneBadger: rgeat. so... it says "secureboot: Secure boot enabled", did you try disabling it on the bios? [18:34] Will disable again [18:35] k1tsun3, do you mean its menu won't show, or what observation makes you think it didn't load? [18:35] LoneBadger: again (so you did it before?)? please post another log afterwards [18:36] k1tsun3, also, if GRUB didn't load, what else loaded instead of GRUB? [18:37] tomreyn Disabled again [18:37] The health is fine, its a new-ish ssd and none of the values are over threshold. Essentially whats happening is that every other boot it will go through diagnostic startup and then boot to a completely black blank screen. When i restart it boots to GRUB, but if i restart again its back to black screen [18:40] https://termbin.com/y50b [18:40] k1tsun3, might the black screen be GRUB doing stuff in secret? configure it to always show its menu, with a reasonable delay in case your graphics systems need to wake up first. you can also do the next step in the same re-config, configuring it to make your default OS boot verbosely. [18:40] LoneBadger: good, now it's off! "secureboot: Secure boot could not be determined (mode 0)" [18:40] *Celebration* [18:41] Want me to installl? [18:41] LoneBadger: i'll read on on the log for a while. you can install again in the meantime. [18:41] LoneBadger: if you don't need to preserve any data then just do a fresh install [18:41] Fresh install [18:41] "Erase disk and install Xubuntu" [18:42] LoneBadger: ^ [18:42] Done [18:42] Thank tomreyn You have been very helpful :D [18:42] Installing.. [18:44] LoneBadger: you're welcome. i just hope it succeeds this time. make sure you update all packages after installation + reboot, since there's a problem with the iwlwifi (intel wireless) driver loading its firmware. [18:48] Installing Libre Office [18:48] Installing system now [18:48] there's no need to post your progress ;) [18:48] oh indeed, failure due to network problems was another issue why I switched to debootstrap. or rather, Ubiquity's choice of installing the boot loader very late in the process, so a minor failure e.g. in installing a game would stop it from installing GRUB. priorities. [18:50] k1tsun3: maybe that's a firmware issue? maybe you want a "bios upgrade"? [18:51] k1tsun3: journalctl -b | grep 'DMI:' should report computer model and current mainboard firmware version [18:52] tomreyn grub-efi-amd64-signed failed to install on /target/ [18:52] LoneBadger: :-/ can you post the log from /var/log/installer ? [18:53] good thing it used LoneBadger's time to install LibreOffice then. :> [18:53] LoneBadger, have you selected 'Download updates while intalling' ? [18:53] Yes [18:53] On Ubuntu where does an electron app try to read its certificates from? [18:54] LoneBadger, I'd really try a minimum install to get some Ubuntu working somehow, best with ethernet, and deal with stuff like wifi drivers and LibreOffice later once the base system works. [18:55] kur1j, probably the system-global openssl certs. if you want to know for sure, you can spy on its file system access using strace -fyxo. [18:57] LoneBadger, third party software too ? [18:58] Disable [18:58] LoneBadger: you recently did a firmware upgrade on this system, right? if so, did you load vendor defaults afterwards? [18:58] LoneBadger: and did you make any customizations to the settings there other than the "secure boot" thing you just changed? [18:59] Upgarded the BIOS since someone suggested it yesterday [18:59] Changed the boot order so the USB would boot first [19:00] LoneBadger: probably i suggested the bios upgrade. ;) can you look for a "IPC" option on the bios? [19:01] LoneBadger, and you tried also 'without connection' ? [19:01] Sven_vB: unfortunately it isn't because I have it in my global cert store and it isn't working [19:01] How do I see the /var/log/installer logs [19:01] I am in the directory [19:01] LoneBadger: i'm asking because your log shows some "Invalid pltconfig, ensure IPC1 device is enabled in BIOS" errors [19:01] LoneBadger: yto view directory contents, type "ls" [19:01] I want to send them to you and I will look for IPC option [19:02] kur1j, I'd strace it then. [19:02] Sven_vB: will do. thanks for the tips [19:02] LoneBadger: i *think* you can also use trhe graphical file browser to access this location [19:03] LoneBadger, you can send the logs with (ls -l; head --bytes=4G * ) | nc termbin.com 9999 [19:03] Can someone help me with my problem on ubuntu 19 my DNS just stops working randomly i can't even ping my router it says: "ping: sendmsg: No Buffer space availabe" [19:04] when i turn wlan off and on it works for like 5 minutes and then the same thing happens [19:04] osw, does that happen on a virtual machine or real hardware? what's your memory use? ("free -m" tells you) [19:04] on real hardware, to be specific a dell xps 13 [19:05] total ~8000 used ~1000 free 3500 [19:05] cache ~3k [19:05] so there seems to be some availabe [19:06] try monitoring network interface stats like RX/TX queue length whithin the few minutes while it works [19:06] what tool do you suggeste to do so? [19:06] sry im unix noob [19:06] dunno. last time I did so I used ifconfig but nowadays we have ss [19:06] iotop? [19:06] https://termbin.com/5csa [19:07] tomreyn You helped me yesterday? :O Thank you for that too [19:08] wow. Ubiquity scoring again. maybe when that project grows a bit more mature they'll find someone who knows how to import Gtk properly. [19:09] so just so i understand wht is RX/TX? i'm not sure where to find it on iotop [19:10] osw, I think it's radio amateur's abbreviations for receive and transmit (send) [19:11] Sven_vB: ok I found it...its the same as chrome in the /home//.pki folder...which sucks because of this missing feature in Ubuntu...and I don't know how to get around it...none of the work arounds work https://bugs.launchpad.net/ubuntu/+source/ca-certificates/+bug/1647285 [19:11] Launchpad bug 1647285 in nss (Ubuntu) "SSL trust not system-wide" [Wishlist,Confirmed] [19:12] LoneBadger: i don't rmemeber whether it was me, i'm just notirious at recomending firmware upgrades. and i have recently seen a bios version string of y computer like yours, so it's possible we talked ysterday [19:12] I remember when I patched a Ubiquity error similar to LoneBadger's "Could not translate page (prepare): 'NoneType' object has no attribute 'replace'" a few years ago. lots of users affected but nobody found the time to comment on the patch. I suggest using an installer that is actually maintained. [19:12] Sven_vB [19:12] kur1j, what feature is missing? [19:12] LoneBadger: hmm this doesn't seem like the right log. how did you post this? [19:12] i'm not sure what i am looking for right now === Etua_ is now known as Etua [19:13] ah I see the link now [19:13] ls -l;head --bytes=4G * |nc termbin.com 9999 [19:13] on /var/log/installer [19:13] Sven_vB: I linked it. Basically Ubuntu isn't allowing system wide certs not work [19:14] bad phrasing [19:14] osw, any recognizable pattern might help. especially if one of the queue lengths rises constantly, and errors start occurring at a certain Q length reproducibly. [19:14] but chrome/firefox and such won't load the system wide certs [19:14] LoneBadger, if you omit the () you can also omit the ls :) only the last command is termbinned [19:14] LoneBadger: what did this output other than the url? [19:15] kur1j, would it help to put a symlink to the global certs? [19:15] -rw-r--r-- 1 root root 5897 oct 25 14:53 debug [19:15] Sven_vB: well the issue is its loading it in a certs DB and every single user would have to do it [19:15] -rw-r--r l root root 18 oct 25 14:37 version [19:15] Sven_vB: im trying to have it where I can do it globally for each machine instead of per each user [19:16] kur1j, maybe a startup wrapper for the electron app can prepare the symlinks. [19:16] (… and then start the original app) [19:16] LoneBadger: so just those two files? hmm, i had expected to see more. are you still on the live installer where the installation failed now? [19:16] Sven_vB: well its annoying in general as firefox/chrome also have the same issues [19:17] tomreyn, the other Ubiquity debug output goes to syslog [19:17] It closed but I can re-open. I am on a live sessiona tm [19:17] Sven_vB: oh thanks. [19:17] atm* [19:17] Sven_vB: If I could the pk11 tool to load it like it saysit works in the bug report that would be a good workaround but it doesn't work for some reason [19:18] tomreyn, really, I'd try an installer that's maintained with enough peoplepower. how about curtin? [19:18] Why is apt-add-repository creating all those .save-files when adding a repo. [19:18] q9, backups in case it messes up [19:19] Sven_vB: on 'ss' i get only zeros on send/recieve besides on "icmp6", but i'm only pingig google [19:19] Sven_vB: feel free to lead if you are happy to explain properly for a user who is used to GUIs? [19:19] Sven_vB: Why does it touch the existing files at all? [19:19] Sven_vB: Also it fixes itself after reconnecting automaticly after some time in case this information helps you [19:20] q9, I've no idea. probably the authors' expression of their confidence in how safe their code is. :> [19:21] tomreyn Do you need anything else? [19:21] LoneBadger: in the meantime, can you post journalctl -b | grep -v iwlwifi | tail -n500 | nc termbin.com 9999 [19:21] Wilco [19:22] https://termbin.com/hvm9 [19:22] LoneBadger, you know you can install without grub and install it later clicking on 'Continue testing' ? [19:22] tomreyn, you're right, we'd need an installer with good support not just for the code base but also novice users. I wonder if there is any. [19:23] Wait what ioria [19:23] LoneBadger, [Firmware Bug]: Page fault caused by firmware at PA: 0x679cf7f0 [19:24] LoneBadger, you can use 'sudo ubiquity -b ' and then install grub [19:24] ioria, did Ubiquity create a user account and login already at that stage? [19:25] Sven_vB, what ? [19:25] LoneBadger: did you try to install using the experimental ZFS option by chance? [19:25] Never [19:25] ok [19:25] Sven_vB, 'Continue testing' [19:25] !who | LoneBadger [19:25] LoneBadger: As you can see, this is a large channel. If you're speaking to someone in particular, please put their nickname in what you say (use !tab), or else messages get lost and it becomes confusing :) [19:25] ioria, your suggestion of -b might supersede my question, I'll look it up [19:26] tomreyn Should I [19:26] Sven_vB, you run sudo ubiquity -b from the live installer (try ubuntu without install) [19:26] Sven_vB: also after it fixed itself it won't "crash" again, maybe it's something about when connecting to the wlan something gets set wrong? [19:27] meh, on bionic, there's no -b in man ubiquity [19:27] ah, in --help it has " -b, --no-bootloader Do not install a bootloader. [19:27] " [19:27] LoneBadger: you're asking whetzher you should install using the experimental zfs option? no, no, i just noticed somethign on the log you last posted which looks as if you had, but it will be a red herring. [19:28] yeah let's try whether --no-bootloader skips the crashing page [19:28] I doubt it though [19:28] tomreyn Ok [19:29] LoneBadger: so linux and your firmware still somewhat agree to disagree, they don'T get along too well. maybe you can try installing xubuntu 18.04.3 and upgrade to 20.04 LTS when it's released? [19:29] no one? :-( [19:29] hmm [19:29] I will ! :D DL Xubuntu 18.04 LTS [19:31] for a moment I was tempted to actually write a patch for LoneBadger's Ubiquity bug, as it seems really easy to fix. [19:31] :3 Please [19:31] LoneBadger, please, try again without a connection active; if it fails again you start the live session and from terminal run 'sudo ubiquity -b' after you follow this : https://askubuntu.com/questions/1028703/the-grub-efi-amd64-signed-package-failed-to-install-target [19:32] I just want to be part of the Kool Kids klub and honesly is so tired of WIN10 bullshit [19:32] maybe you need to set a password on your UEFI bios, seen that before [19:32] :P [19:32] nah sorry. by the time I patch this, other Ubiquity devs have invented 20 new bugs [19:33] if you want to get in the kool kids klub, better invest your time in learning how to use "curtin" (the curt installer) :) [19:33] What is that? [19:33] or try the Ubuntu mini ISO [19:33] not sure mini-iso can install in efi mode [19:33] ioria +1 [19:33] Sven_vB: afaik mini.iso doesn't even support efi [19:34] yeah, good trolling on friday [19:34] curtin is an alternative linux installer that aims to get your linux up and running as soon as possible, without asking too many questions. [19:34] maybe supergrubdisk can chainload the mini iso. haven't tried though. [19:34] he should check for a bios password, else it should not be trusted on some uefi machines [19:35] Sven_vB: what's the root cause of the failed installation you identified on LoneBadger's logs? [19:37] tomreyn, glad you asked. looking again made me realize I was chasing the wrong bug. we'd need to see the syslog to determine the actual problem. [19:37] tomreyn, Oct 25 14:51:13 xubuntu kernel: [Firmware Bug]: Page fault caused by firmware at PA: 0x679cf7f0 and below ... [19:38] shouldn't that be just a networking problem? so when we disable networking it should not occurr [19:39] Installing Xubuntu 18 LTS [19:39] ioria: this certainly appears to be a critical one, yes. there is also 'Volume group "sda" not found' before this, any idea if this is a problem? [19:39] tomreyn, ailed to register the EFI boot entry: Interrupted system call [19:39] *f [19:39] * Sven_vB not asking for the syslog because he's not motivated to debug it. [19:39] even a Call Trace [19:39] oh nice [19:40] ioria: we don't need to discuss errors below the "Page fault caused by firmware at PA: 0x679cf7f0" one. i'm just wondering whether there are any on top of it [19:40] what log is that in? I seem to have missed the link [19:41] https://termbin.com/hvm9 [19:41] tomreyn, nope, i don't think so ... [19:41] ah nice [19:41] "E: Can not write log (Is /dev/pts mounted?) - posix_openpt (19: No such device)" also seems wrong [19:41] I guess this is the core problem: finish-install[32289]: Temporary failure resolving 'ca.archive.ubuntu.com' [19:42] Critical part is Installing Grub2 [19:42] nonsense, that's to the very bottom [19:43] LoneBadger, consider also to install in bios mode ... it's not so bad after all [19:44] IMO failing to update EFIvars as a result of the firmware inhibited page fault is the root cause. [19:44] Ubiquity with pid 31490 survives even after failure to install grub-efi-amd64-signed so I guess it's not vital [19:44] so what ioria pointed out first [19:45] as in, could be fixed later [19:45] well how do you boot a system without a boot loader? [19:45] it would require chrooting. [19:45] and installing another kernel probably [19:46] let's see how 18.04 LTS goes [19:46] 19.04 in BIOS mode is indeed another option, if the firmware actualyl supports bios booting [19:47] also downgrading the firmware or TJ's acpi_osi workaround may help [19:47] ewwww that kind of scripting bug really shouldn't be in GRUB. xubuntu ubiquity[31490]: /usr/share/grub/grub-check-signatures: 22: [: Illegal number: [19:50] LoneBadger, would it be acceptable to install Ubuntu in a way that it cannot boot itself, but then use a USB key with a boot helper to kickstart Ubuntu? once it starts you could withdraw the USB key. [19:51] Yes Why not [19:51] LoneBadger, do you have a spare USB key that you don't need? [19:51] BTW the installer is still stalling at Grub2 package [19:51] No error message [19:53] LoneBadger: you mean the 18.04.3 LTS installer now? [19:53] Yeah [19:54] just dont install grub, do it after the install from a chroot, use standard efi paths if you have to if efibootmgr isnt working to add it [19:54] ah crap i had forgotten that you had actualyl tried the 18.04.3 installer befire [19:54] tomreyn: https://paste.ubuntu.com/p/mJmPKqwYMG/ a new try, the only thing left is weewx, you think it could really be the cause? [19:56] One of the main message I see is /usr/lib/ubiquity/ubiquity/gtk)components/nmwidgets.py:133 Warning ; Source ID 7971 was not found when attempting to remove it Glib.source_remove(self.rows_changed_id) [19:56] that's not a relevant messge about why your installation fails, LoneBadger [19:56] LoneBadger, when you found a spare USB key to use as the kickstarter, download this SuperGrub and let's try if it works for you: https://sourceforge.net/projects/supergrub2/files/2.04s2-beta2/super_grub2_disk_2.04s2-beta2/supergrub2-2.04s2-beta2-multiarch-USB.img.zip/download [19:57] LoneBadger's old 18.04.3 kernel log on this system https://pastebin.com/raw/GxqNG8TD [19:57] claviger-pc: looking [19:59] claviger-pc: can you remind me which ubuntu release you're trying to upgrade to which? [19:59] tomreyn 16.04 -->18.04 [19:59] claviger-pc: is there an error message? [20:00] claviger-pc: do you use this as a headless server or as a graphical desktop? [20:00] its a headless server [20:00] LoneBadger, you could also use the xenial USB key but doing so will remove the xenial from it. [20:00] claviger-pc: is "ubuntu-server" installed? [20:01] Anyone know how to check and change the system time zone from terminal? [20:01] TabMasher: timedatectrl [20:01] TabMasher: timedatectl [20:01] tomreyn yes, ubuntu-server is installed (ubuntu-server is already the newest version (1.361.4)) [20:02] claviger-pc: is there an error message? [20:02] tomreyn: there is no error message, just the two log files in /var/log/dist-upgradee === kostkon_ is now known as kostkon [20:03] tomreyn main.log and apt.log [20:03] tomreyn: That's just what I needed. Thank you. [20:03] tomreyn: https://paste.ubuntu.com/p/KK2CHPFKWd/ [20:04] LoneBadger, sorry, you can cancel the download. I found a better version [20:04] wOT? [20:04] claviger-pc: so, i don't see any error messages at https://paste.ubuntu.com/p/mJmPKqwYMG/ - it just looks like you had chosen not to carry out the upgrade. but i may be missing something. [20:05] LoneBadger, yeah I just remembered how big that file is but you only need the first 14 MB of it [20:05] tomreyn: no, there is no error messages, no prompts, nothing, just fails [20:06] claviger-pc: and foreign_packages lists nothing but weex in both sections? [20:06] I have another laptop running linux and I'm trying to test first to see if I can establish a TCP connection between that laptop and my desktop by using netcat. On my desktop I run "netcat -lp 4444" and on my laptop I run "netcat 192.168.0.101 4444". When I do that, I try to send a message from my laptop to my desktop, but nothing happens. This only occurs with TCP connections, not with UDP (if I do it with -ulp), [20:06] does anyone have some pointers on how I can go about fixing this? [20:06] Launchpad bug 4444 in vlc (Ubuntu Breezy) "vlc streaming output segmentation fault" [Medium,Fix released] https://launchpad.net/bugs/4444 [20:06] firewalls are disabled on both machines [20:07] tomreyn does not evn list weewx [20:07] oh right, because you have a repository setup for it [20:08] claviger-pc: i can't think of anything else really. maybe try picking a different mirror server just to introduce some variance [20:09] k, interesting, sufficent to just change /etc/apt/sources.list ? [20:09] HI guys. Does anyone know a cli tool that can monitor a process and start another based on performance of the system you can specify ? [20:09] tomreyn: or is there a way to purge the distupgrade files? [20:11] claviger-pc: nice idea, i don't actually know. you could maybe purge ubuntu-release-upgrader-core and reinstall it [20:11] Guys [20:11] I think we got through the ad64 signed bug [20:11] LoneBadger, actually you could use the Xenial live session to download just the start: supergrub2-2.04s2-beta2-multiarch-USB.img.zip wget -O https://sourceforge.net/projects/supergrub2/files/2.04s2-beta2/super_grub2_disk_2.04s2-beta2/supergrub2-2.04s2-beta2-multiarch-USB.img.zip/download [20:11] LoneBadger, and when it shows it has downloaded 14 MB or more, press Ctrl+C [20:12] It is installing the x86_64-efii platform [20:12] oh ok [20:12] I pasted the command wrong anyway. =) [20:12] wow, grats! [20:13] Giving me a lot of warning message but it is still going [20:13] claviger-pc: it seems to have a lot of reverse dependencies (read: many other packages depend on it). so you may want to just dpkg --purge it, bypassing some of the dependency checks. [20:15] @tomreyn interestingly, apt let me do it, but also uninstalled ubuntu-server, which I then reinstalled [20:15] Sven_vB tomreyn My PC now https://i.pinimg.com/originals/50/35/3e/50353e5b86de77d720baf7416c3ab3f7.jpg [20:15] claviger-pc: nice [20:15] LoneBadger: hehehe. not really, but at least you got it installed, nice! [20:16] It is not installed completely [20:16] but it got further [20:16] oh ok [20:17] but what's missing if the boot loader is done? [20:17] i think after this it's just cleaning up and unmounting [20:18] Started cleanup of temp directory [20:18] sounds good [20:18] :3 is it happenning?? [20:19] seems like it [20:21] claviger-pc: did it improve what ubuntu-release-upgrade does, though? [20:21] tomreyn: no, seems exactly the same [20:21] :-/ can you show the full terminal output? [20:22] tomreyn: here's the end of it: https://paste.ubuntu.com/p/2C54xgrpHk/ [20:23] claviger-pc: hmm, is this an i386 system? [20:23] nope [20:24] Linux claviger 4.4.0-167-generic #196-Ubuntu SMP Mon Oct 21 19:50:52 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux [20:24] seems fine [20:25] and there's no log file showing what you on the temrinal output in /var/log/ or /tmp/ ? [20:25] or /var/tmp ? [20:27] Still Installing [20:27] tomreyn: there's a whole folder, not sure what's useful: https://paste.ubuntu.com/p/WyCh68gNTG/ [20:27] thats in /tmp [20:27] thats in /tmp/ubuntu-release-upgrader-zez1jt0y [20:28] I wonder what takes LoneBadger's Ubiquity that long [20:28] but there is also /var/log/dist-upgrade, where there is both main.log and apt.log and the terminal output [20:28] claviger-pc: yes that's just the unpacked release upgrader code, not any of the logs [20:29] claviger-pc: can you compress /var/log/dist-upgrade and make them available somehow? [20:29] tomreyn: yeah, let me see [20:29] problem is its 150mb [20:30] the screenlog.0 [20:30] claviger-pc: 150 mb?! that's unusually much. you mean uncompressed, though, right? [20:30] yes [20:31] it should compress well [20:31] tomreyn Is it normal that it take so much time? [20:31] claviger-pc: maybe you can actually skip the screenlog for now if it's an issue [20:31] ...for uploading [20:31] how can I check whether my partitions align? do I need to divide their start sectors manually or can a tool do it for me? could there be other reasons why gparted shows 1 MB gaps between partitions? [20:32] yeah, but I could compress it, problem is, I don't think there is much there [20:32] LoneBadger: no, but you have an unstable system there. the wireless module will be failing constantly, you can watch this if you open a terminal window with ctrl-alt-t and type: dmesg -w [20:32] LoneBadger, when it shows "Reticulating splines…", half an hour was still within regular in older Ubuntus [20:33] Alirght [20:33] claviger-pc: i couldn't tell :) [20:33] Gonna take a shower while this grandma PC do its thing [20:33] Brb [20:34] LoneBadger: you could try unloading the modules failing there, or disable the wireless connection at least. [20:34] Wireless connection is off [20:34] How to unload some module? [20:34] ok [20:35] WAIUT [20:35] Starting Message of the Day! [20:35] did it reboot spntaneously? [20:35] then.. back to Source ID not found when attempting to remove it [20:36] it started at id 10 and now it is in the 10000 [20:36] LoneBadger: i'm not sure what you're looking at [20:37] are those BIOS messages or is it inside graphical Ubuntu? [20:37] It is the installer prompt of Xubuntu [20:37] /usr/lib/ubiquity/ubiquity/gtk)components/nmwidgets.py:133 Warning ; Source ID 7971 was not found when attempting to remove it Glib.source_remove(self.rows_changed_id) [20:37] can you send a screenshot? [20:37] oh ok [20:38] those Warnings with "Glib." in them are not important usually [20:38] it just means the programmers did a sloppy job [20:38] oof [20:39] and I'm not even sure whether it's the Ubiquity or Glib programmers. [20:42] so i tried to install ubuntu from a live cd, but because i dock my laptop, it goes into sleep mode if it boots >_> [20:42] i know you can disable that from the settings but is it possible to disable before booting? [20:43] maybe just don't dock until installation is finished? [20:44] tomreyn: I figured out that the screenlog does not get purged, so I removed it, ran again, here it is the first part: https://paste.ubuntu.com/p/fX5jGvrX8B/ [20:46] tomreyn: then it lists every package in the repos, i think, with "no candidate ver: ", then here is the end : https://paste.ubuntu.com/p/msWtC8W4bW/ [20:46] oerheks, you can use systemd-inhibitor [20:46] erry, ^ [20:46] can i boot with that? [20:46] erry, no I assumed you already booted it [20:46] i have to dig out the power cable to not dock cause inm lazy [20:46] erry, what's the current system state? powered down? [20:46] yeah [20:46] erry, so what will happen when you put it in the dock? [20:47] if i boot while docked it'll sleep [20:47] when it gets to x11 [20:47] erry, why would it do that? is that a known bug? [20:48] because the lid is closed,.. [20:48] it's a feature. [20:48] not one i like [20:48] claviger-pc: hmm, most peculiar. can you file a bug about this? [20:48] tomreyn: then there is also this lspci.txt: https://paste.ubuntu.com/p/6ns7Yy86Rt/ [20:48] erry, oh ok. you could probably boot with lid actions disabled. [20:48] tomreyn: yes, I suppose [20:48] how do [20:48] DISPLAY= ubuntu-bug ubuntu-release-upgrader [20:49] claviger-pc: ^ [20:49] erry, #systemd should know [20:49] ok ty [20:49] erry, you can also boot into recovery mode and reconfigure systemd [20:49] claviger-pc: this should gather the logs, then print a url you can visit from a desktop to complete the bug report [20:49] tomreyn: on it [20:50] Sven_vB: live cd :D i'm still trying to get a not-totally-borked install of some os >_> [20:50] i can undock if i dig the power cable ouit of where it is, [20:50] otherwise the power lasts like 5 minutes \o/ [20:50] erry, you're trying to boot a live CD (which?) while the lid is closed? [20:51] Sven_vB: ubuntu, latest lts, and yes [20:51] erry, do you have casper-rw persistence already? it's really useful. [20:51] ohh hmm not really [20:52] erry: had you considered opening the lid? [20:52] erry, oh ok so you need a solution that will work within 5 min? then it's probably easiest to just open the lid, and prepare stuff better next time. ;) [20:52] fiine i'll untanglle the power cable [20:52] :p [20:52] :D [20:53] if you're yolo, try kernel boot option systemd.HandleLidSwitch=ignore [20:53] but I've probably guessed the name wrong [20:53] oooh [20:53] that sounds like it could work, [20:53] let me try [20:55] or HandleLidSwitchDocked [20:55] logind.conf(5) [20:57] if you're another kind of yolo, you could try reading the docking station manual on which of the pins are power, and mask the other ones with tape. :D [20:58] electrical advice given on IRC is usually totally safe ;) [21:04] tomreyn: there is no ubuntu-release-upgrader package, so I am doing update-manager-core [21:05] tomreyn: is this weird? https://paste.ubuntu.com/p/gNPZ8qbwr7/ [21:07] yo, im trying to download tails to a usb. i used wget to download the os. im following a tutorial on a website that says to run the following command and i am getting the following error. Can somebody help me? [21:07] claviger-pc: ubuntu-release-upgrader should be the source package, but we can still reassign later [21:07] isohybrid /home/doz/Desktop/tails/tails-amd64-4.0.img --entry 4 --type 0x1c [21:07] isohybrid: /home/doz/Desktop/tails/tails-amd64-4.0.img: could not find boot record [21:07] tomreyn: the bug: https://bugs.launchpad.net/ubuntu/+source/ubuntu-release-upgrader/+bug/1849902 [21:07] Launchpad bug 1849902 in ubuntu-release-upgrader (Ubuntu) "upgrader from xenial to bionic fails without any clear error message" [Undecided,New] [21:08] tomreyn: there is no package: ubuntu-release-upgrader [21:08] claviger-pc: the apt priorities are weird, have you played withpinning? [21:08] tomreyn: sudo apt-get install ubuntu-release-upgrader [21:08] not knowingly or recently [21:08] seems like the i386 should not be htere [21:08] Polybius-, to an Ubuntu live iso? do you have some sort of storage/persistant part setup? [21:08] in case you upgrade that remotely, make sure you have a backdoor that upgrader doesn't know how to stop. don't rely on any SSH not even the backup SSH. [21:09] oerheks, what do you mean? [21:09] also make sure that backdoor won't be defeated by accidentially restored firewall rules [21:10] claviger-pc: you cannot install source packages, those are the packages which binary packages are builkt from. and you can install binary packages. [21:10] oerheks, the install was from their website so it should be the complete iso [21:10] Polybius-, with the "persistence" kernel boot option you can use casper-rw overlay to have changes to files persist accross live sessions. [21:10] ok, well, I have put the bug in [21:10] oh tails [21:11] is tails based on ubuntu? [21:11] claviger-pc: is there anything in /etc/apt/preferences* [21:11] Sven_vB: i think it is debian [21:11] !debian [21:11] Ubuntu and Debian are closely related. Ubuntu builds on the foundations of Debian architecture and infrastructure, with a different community and release process. See https://www.ubuntu.com/about/about-ubuntu/ubuntu-and-debian - !Repositories meant for Debian should NOT be used on Ubuntu! [21:11] nope [21:12] claviger-pc: can you post sudo /bin/true && cat &>/tmp/aptlog < <(sudo apt-get -y update 2>&1; apt-cache policy 2>&1; sudo apt-get -syV full-upgrade 2>&1;); nc termbin.com 9999 bingo!! [21:12] claviger-pc: so you do have apt pinning in place [21:12] tomreyn: https://termbin.com/k0l9o [21:12] Polybius-, are you sure it's an official tutorial? with tails there should be no need for hybrid [21:13] Sven_vB, okay hold on, before i figure out how to check, i just remembered. this usb was preformatted for luks and w/e filesystem format we chose at the time but it is empty. could that be the problem? or is the problem in the iso file iteself [21:13] here's /etc/apt/preferences: https://paste.ubuntu.com/p/hnFXg6zV9d/ [21:13] seems like it could be the problem [21:13] all i see are iso images https://tails.boum.org/install/ [21:13] Sven_vB, its not an official tutorial its from cyberciti [21:13] oh, not an ubuntu issue then [21:14] claviger-pc: well, yes, nice home made problem you have there [21:14] so, I should just delete it, no? [21:14] Polybius-, you can check the checksum of the ISO to tell whether it's valid or corrupted. if you use tails with intent to protect your privacy, trust no-one except tails devs, [21:15] claviger-pc: maybe move it out of the way then identify the intelligent person who wasted our past hour. [21:15] Polybius-, that hybrid tutorial may be well-intended but could still contain subtle mistakes that have sneaky effects [21:15] Polybius-, also you should put Tails on a medium that will be read-only later, to avoid malicious drivers or firmware from modifying it. [21:15] tomreyn: I confess and apologize, this was me, learning ubuntu and linux, perhaps 5 or 6 years ago [21:16] Sven_vB the install came from the offical website. and do you mean verifying the download? everything im downloading is innocuous [21:16] harmless [21:17] claviger-pc: hehe, looks like you learnt yet more today... please update the bug report with our findings ("apt-pinning was in use in /etc/apt/preferences*" or similar) so that others can find it should they run into the same situation. [21:17] and then we can probably close it [21:18] ("invalid") [21:18] Polybius-, then you can just copy the ISO to the USB thumb drive (readonly would be better see above). Ubuntu has tools like "cp" or "pv" that can do that copying. of course your old data on the USB drive will be hidden or even erased. [21:18] k, done, sorry... ;( [21:19] claviger-pc: that's ok, happens, and apt pinning is easily overlooked ;) [21:19] Sven_vB, thats fine. once the file is on the drive, can i boot from that as is? [21:20] Polybius-, yes that should work. make sure to copy it onto the drive itself, not onto a partition. if it isn't able to boot, you could use SuperGrubDisk to chainload it. [21:20] tomreyn there are things that are not smart, that you do because they are suggested, then later you find they come back to bite you [21:20] Sven_vB, the whole drive is one partition. ill try now, thanks man [21:20] (that is, if tails support GRUB loopback) [21:21] Polybius-, copying it onto that partition is useless. [21:21] Sven_vB, no the usb is only one partition [21:21] Sven_vB, and empty atm [21:22] Polybius-, having a partition there means your drive is a partition table followed by a partition. so the partition is not the entrire drive. [21:22] Polybius-, thus the boot loader address will be wrong by the size of your partition table [21:24] Polybius-, check in /dev/disk/by-id/ you will probably have a symlink named "usb-VendorName_ProductName" (your disk) and ("usb-VendorName_ProductName-part1" (1st partition). [21:24] so make sure to copy the ISO to the disk [21:25] using the named symlinks instead of a single letter as disk identifier reduces the risk of catastrophic one-letter typo. :) [21:25] Sven_vB, i've never changed anything about the file system after formatting the drive so no partitions have been created on it. when i installed another linux os (the same way) it had its own file structer with install by default. so i never had to dynamically set up any storage. i figured this would be similer [21:27] Polybius-, actually it just came to my mind that maybe that ISO isn't even meant to be booted from USB thumbdrives. in that case maybe you'll need sth. like UNetBootin or (my favorite) SuperGrub [21:28] Polybius-, what tool did you use to format the drive? [21:29] Sven_vB, just the "disks" program on ubuntu. the iso is ment for USB. i didnt origanally format the drive for this though. i was just erasing it. so it just so happened that i had a empty thumbdrive and decided to install tails on it [21:30] originally [21:30] tomreyn: thanks anyway and for understanding, ironically, i never even used bacula... [21:30] oh and no it's not similar. the ISO contains a structure similar to a partition table, which cannot by default co-exist with partition( table)s [21:31] Polybius-, ok then have a look in /dev/disk/by-id/ and see if you can find links that sound like your USB drive [21:31] Polybius-, that way you will know whether you have partitions on it [21:32] Sven_vB, dm-name-luks-cb3205bc-54f1-4f5d-912b-2ee06084d47e [21:32] dm-uuid-CRYPT-LUKS1-cb3205bc54f14f5d912b2ee06084d47e-luks-cb3205bc-54f1-4f5d-912b-2ee06084d47e [21:32] usb-Generic_Flash_Disk_06277CC0-0:0 [21:32] usb-Generic_Flash_Disk_06277CC0-0:0-part1 [21:32] all could be relevant [21:32] it shouldnt have any [21:33] wait [21:33] does luks create two partitions for anything? [21:33] there's only one partition, -part1 [21:33] but that wouldnt hurt anything... i dont hink [21:33] think [21:34] Sven_vB, okay then i should be good [21:34] Sven_vB, im about to just try to boot from it and see what happens [21:34] Polybius-, good luck! [21:34] ill be bock [21:35] I'm confident about that :) [21:49] Is there anyway to "test" this p11-kit-trust.so module? [21:50] what kind of test? [21:51] https://superuser.com/a/1312419/411058 I'm following those instructions [21:51] but its not loading [21:51] do you get an error message? [21:51] nope [21:51] im simply linking it like he says which should be a "drop in replacement" [21:51] but the browsers don't load it [21:52] there is nothing in the cert store at all any more not even the default certs [21:53] so more of a "drop replacement" then? === sjohnson is now known as krokula === krokula is now known as sjohnson [21:54] tomreyn: yeah it should be [21:54] I'm simlinking it like it says [21:55] this entire "cert store" thing is so damn cryptic (excuse the pun) but its miserable [21:56] so what are you actually trying to do? [21:56] add a company root CA to firefox? [21:56] yeah [21:57] and just adding it via the system certificate store it doesn't end up there? [21:57] I think mozilla enterprise config might help [21:57] i was thinking there used to be a mechanism that did so [21:58] https://developer.mozilla.org/en-US/docs/Mozilla/Firefox/Enterprise_deployment [21:59] "As of Firefox 64, an enterprise policy can be used to add CA certificates to Firefox. This is now the method recommended for organizations to install private trust anchors." – https://wiki.mozilla.org/CA/AddRootToFirefox [22:03] looks like i am wrong and there never was a mechanism to import those system certificates into firefox' nss db [22:04] or at least no more [22:05] last time I checked there were only ugly hacks. I'm still looking for a nice browser that also has git-friendly configs. [22:06] how about https://thomas-leister.de/en/how-to-import-ca-root-certificate/ [22:07] tomreyn no it doesn't [22:07] tomreyn: they have this issue https://bugs.launchpad.net/ubuntu/+source/ca-certificates/+bug/1647285 [22:07] Launchpad bug 1647285 in nss (Ubuntu) "SSL trust not system-wide" [Wishlist,Confirmed] [22:08] tomreyn: yeah thats the issue its in the ca-certificates file, but its not being loaded because libnssckbi.so is a static r/o file from my understanding [22:09] using the p11-kit-trust.so _should_ make firefox/chrome read from the certificates-ca.crt file [22:09] but its not [22:11] i think the firefox apparmor profile is disabled by default, otherwise this could be related [22:12] the advantage of the enterprise policy method is that you get official support from mozilla. [22:12] Sven_vB: that doesn't work on linux, only on mac and windows [22:12] oh :( [22:13] maybe ask on irc.mozilla.org, too [22:15] https://support.mozilla.org/en-US/kb/setting-certificate-authorities-firefox suggests it also works on Linux [22:16] Anyone else having problems logging into GMail with the new Firefox 70? It always asks me for my credentials. Doesn't happen on other sites, even YouTube works. [22:17] It does, I agree but I couldn't get it to work either [22:17] have you added the virtual security device? [22:18] Sven_vB: the pk11? [22:18] yes [22:18] the pk11-kit-trust.so? if so yeah thats what I was trying to do and it won't work [22:19] https://superuser.com/a/1312419/411058 that was what I was originally linking [22:19] tomreyn's link sounds to me like you need the .so to make the pk11 virtual sec-dev available, and then need to add it and authorize it. [22:20] Sven_vB: yeah "Certificates can be programmatically imported by using p11-kit-trust.so from p11-kit (add the module using the “Security Devices” manager in Preferences or using the modutil utility). " [22:21] exactl [22:21] y [22:21] did that, won't load them [22:21] i can view them its "loaded" but p11-kit-trust.so won't load the certs [22:21] then I'd try file system redirect approaches [22:21] its empty [22:21] Sven_vB: did that as well, empty [22:22] its like the p11-kit-trust.so is broken [22:22] that could be [22:22] but I don't know how to tell [22:22] or why [22:22] I meant fs redirect as in not using the p11.so [22:23] not sure what a fs redirect is then? I assumed a sym link [22:23] yes. was the launcher wrapper idea not feasible? [22:24] sorry, remind me what that was? for zulip you mean? [22:24] nah, making a script that prepares config in the homedir and then launches the real browser [22:25] I wanted to avoid having the user do anything [22:25] yes, thus a script that does it [22:25] I can add the cert manually to firefox with "certutil -d sql:$HOME/.pki/nssdb -A -t "C,," -n \-i " [22:26] adrian_1908: could be a side (or intentional) effect of the improved tracking protection https://www.mozilla.org/en-US/firefox/70.0/releasenotes/ [22:26] kur1j, how would your users launch the browser? [22:26] Sven_vB: double clicking on the Firefox icon [22:29] kur1j, if that's the only method, you can just replace that icon's target with your script. you'll probably want to register it for some URL schemas "a link has been clicked", too. [22:29] tomreyn: yes, that's what I'm thinking but it seems I'm the only one who ran into this. Can't quite seem to fix it. I would exempt e.g. mail.google.com from the stricter rules if possible. [22:30] well sure, that might fix firefox, but other apps chrome/zulip and such won't be resolved [22:30] i want to try and solve this at the system level [22:31] its one of my biggest gripes with Ubuntu is in an enterprise setting its not very robust [22:31] CentOS/Fedora have had this for like 5 years [22:31] adrian_1908: if you'Re the only one then it rather suggests firefox addons or local firewall / blocking rulesets etc. are getting in the way. or antivirus, if you run such. [22:31] problem is if misbehaving apps don't adhere to the system-wide config. [22:32] kur1j, you could check how fedora does it and replicate that. [22:32] p11- lol [22:33] but I can't get it to work haha [22:33] maybe you can use their p11.so? [22:33] easy fix: pay canonical to fix it [22:33] haha [22:33] yeah that would be another possibility. :) [22:33] if it works in other linux distros it dshouldn't be so tough [22:34] its got an open ticket for 4 years [22:34] https://bugs.launchpad.net/ubuntu/+source/ca-certificates/+bug/1647285 [22:34] Launchpad bug 1647285 in nss (Ubuntu) "SSL trust not system-wide" [Wishlist,Confirmed] [22:34] 3 [22:34] and if your hostname gives away what it seems to, then i, as a non canonical involved boting ubuntu user recommend your employer puts some peanuts into this box [22:34] * boRing [22:36] now I feel bad having volonteered my time on this. [22:36] could also find someone to do the patches needed to make it work properly and submit them upstream. [22:36] Sven_vB: ? [22:37] OTOH, I'll have the same problem in a few months in a private project, so I guess it was worth a try. [22:37] tomreyn: well was just trying to figure out how to resolve it... [22:38] i mean if its as just some simple configuration why not [22:38] oh actually maybe he's just on their network as a customer, not employee. [22:38] *they're [22:38] sorry for assuming gender [22:39] lol its fine [22:39] well nowadays you can't be as sure as back in the days. :) [22:40] shall I use 1 GB or 2 GB for my new /boot? [22:40] EFI is separate [22:40] the ESP I mean [22:40] 512MB [22:40] tomreyn: Any idea how I could track this issue down? It's only GMail, not YouTube, not Reddit etc. I tried disabling addons and changing the default protection profile, but whenever I close and reopen the browser, GMail asks me for my credentials again. [22:40] Only since FF70. [22:40] adrian_1908, watch the developer console for errors and network traffic [22:41] Sven_vB: tomreyn: regardless appreciate the assistance [22:41] adrian_1908: web developer toolkit [22:41] Ok, will try that! Alas I immediately get redirected to a different page [22:41] kur1j: good luck there. [22:41] adrian_1908, there should be a (maybe secret) option to avoid auto-redirect [22:42] its late, I'm over this haha. Yall have a good weekend [22:42] thanks, you too! [22:44] ic, I'll probably need about 70 MB per x86_64 kernel version [22:45] Well, if this doesn't get turn out to be a shared problem that gets fixed, then maybe I can use this as an opportunity to ween myself off GMail. I only use it for financial/work related stuff anyway. [22:45] *non-financial/work related [22:49] good morning all [22:50] hello manjaro user! [22:51] apt continues to update older kernels I still have installed? [22:53] doug16k: what are you seeing? [22:54] doug16k, do you mean they're included into the menu when you update your grub config? [22:54] well, what I did was take all the packages that dist-upgrade was going to install and constructed a huge `apt changelog ` command line [22:54] *when apt updates your grub config [22:55] I was wading through a huge list of kernel changes without realizing it was for the 4.15 kernel. I'm running 5.0.0 HWE though [22:55] are the old kernels still installed in the current system? [22:55] is it updating the 4.15 I'm keeping around too? [22:56] yes still installed [22:56] then it's intended to try and preserve them, i.e. preserve all that's possible to preserve. [22:56] if you later find your upgraded system works without them, you can still uninstall them. [22:57] or of course you can uninstall them before :) [22:57] how do I ensure it never autoremoves 4.15? [22:58] there are people with severe USB problems on 5.0.0, I get the dmesg logs they get. my system doesn't screw up though. I am afraid it might start screwing up someday though [22:58] oh sorry I think I misunderstood your question. [22:59] I get this bug with no issues. other people there get severe problems -> https://bugzilla.kernel.org/show_bug.cgi?id=202541 [22:59] bugzilla.kernel.org bug 202541 in USB "xhci_hcd 0000:15:00.0: WARN Set TR Deq Ptr cmd failed due to incorrect slot or ep state" [High,New] [22:59] so you want to make double-sure your system has that kernel after the update. in that case I'd upgrade in a chroot from a live session, then after that chroot in again and make sure that exact kernel is installed. [22:59] I want to ensure I can go back if I start getting the severe problems [23:01] sorry if I am being confusing. I don't mean 18.04 -> 19.10 upgrade. I mean just regular dist-upgrade [23:01] "go back" together with "ensure" sounds like you need a backup.;) [23:02] or how about you simulate your plan in a virtual machine? [23:02] ok nah that probably won't properly emulate all xhci madness [23:03] isn't there a way to pin a package or something [23:03] I heard that autoremove after installing hwe rips out everything before 5.x [23:04] before hwe* (kernel 4.15.x) [23:04] shouldn't autoremove only remove auto-installed packages? in that case the fix would be as easy as installing that kernel manually, explicitly. that should make apt change the "user wants it" flag and then find that's all it had to do. [23:05] ah [23:05] in an ideal world, at leas [23:09] I see "linux-headers-4.15.0-67-generic/bionic-proposed,now 4.15.0-67.76 amd64 [installed,automatic]"... that "automatic" means eligible for autoremove then? [23:09] I think yes [23:09] thanks [23:10] iirc that 'automatic' means it comes from the linux-headers-generic dependency [23:10] so if I explicitly apt install that and the -modules and -headers etc, it should not get autoremove'd? [23:11] doug16k, you could simulate the autoremove to find out [23:12] the autoremove logic is pretty complicated; it's along the lines of "keep the most recent two kernels and the kernel that we're currently booted into" or something. somehow it'll keep upwards of four kernels. [23:16] I don't see a --dry-run option or anything. simulate how? [23:17] apt doesn't always ask Y/n, I don't trust it [23:17] -s [23:21] neat! it is not saying it will remove 4.15.0-67 (the latest 4.15 one), but it will remove a bunch of older 4.15's [23:25] autoremove did what I wanted perfectly. thanks again [23:27] er, thanks first time :P [23:32] glad we could help :) [23:32] another question: `reboot` sometimes says something like: Operation inhibited by "doug" (PID 4078 "gnome-session-b", user doug), reason is "user session inhibited" [23:32] how do I list the inhibitors [23:33] I tried searching and I get 1000 results saying how to force it [23:35] I think I figured out the inhibitor - had a virt-manager vm running. but what if I didn't know. when there are "inhibitors" it will do that dreaded 1m 30s delay thing? [23:36] nope it wasn't the vm. still says that if I `systemctl reboot --dry-run` [23:36] gnome-session-b? what the? [23:38] someday I want to figure out that dreaded "waiting for session to stop" 1m 30s crap [23:38] it doesn't always occur [23:39] one time it was blocked on a virt-manager machine to shutdown. it counted up to 1m 30s then suddenly the timeout was 3m. counted up to that and it increased it again. pressed reset [23:39] systemd was trolling me [23:41] rickrollctl? [23:42] something like that [23:42] only rick was missing [23:42] if it played that song I *might* have let it go a bit longer :P [23:45] if ffmpeg rendered the video too in ascii, I *definitely* would have let it go longer [23:45] pgrep normally cuts process names after 11 characters. but you can use ps 4078 to see the details [23:46] .. in response to Operation inhibited by "doug" (PID 4078 "gnome-session-b", user doug), reason is "user session inhibited" [23:46] $ echo -n gnome-session-b | wc -c [23:46] 15 [23:46] tty2 Sl+ 0:00 /usr/lib/gnome-session/gnome-session-binary --session=ubuntu [23:47] the 15 comes from the kernel "comm" field for processes [23:48] oh thanks sarnold [23:49] https://elixir.bootlin.com/linux/latest/source/include/linux/sched.h#L210 [23:49] doug16k, libaa should be supported [23:50] if not, try avconv [23:52] welp. `systemctl reboot --dry-run` doesn't do a dry run at all. [23:52] I was closing one program at a time and retrying that command [23:52] once I closed chromium (which was playing music on youtube), poof, rebooted instead of dry running [23:53] I guess that youtube had an inhibitor [23:53] yes [23:53] but it blamed the gnome session instead of chromium [23:53] oh I see your question now [23:54] did you have a pending reboot from before you started issuing the systemctl reboot --dry-run commands? [23:54] then probably it used another, session-based mechanism, and gnome translated it to an inhibitor [23:54] I didn't. it started the actual reboot when I ran systemctl reboot --dry-run [23:55] while I agree that the reboot action should support --dry-run, meanwhile the work-around would be to close apps until the list of inhibitors goes empty. [23:55] man page specifically says reboot supports it [23:56] having a queued reboot take action immediately after my CD burn finishes would be rather surprising btw. [23:56] maybe I want to see the report first [23:56] "Just print what would be done. Currently supported by verbs halt, poweroff, reboot, kexec, suspend, hibernate, hybrid-sleep, default, rescue, emergency, and exit." [23:56] so I doubt there's really a queue for that [23:56] doug16k: I wonder... the manpage says it's systemctl [options] [command] --- how about systemctl --dry-run reboot ? [23:57] yeah maybe that option would need to go first [23:57] and no validation :( oh well, at least I know why now. thanks [23:57] actually it looks like it has to [23:58] reboot should really fail when run with unsupported args.