[00:47] <loswedseded> do you have your own AI in this channel?
[00:50] <arraybolt3> No, though there's a factoid bot.
[00:51] <arraybolt3> !ubottu
[00:56] <loswedseded> !ubottu, can you hear me?
[00:56] <loswedseded> !ubottu, when will 24.04 be released?
[00:58] <arraybolt3> You can direct message the bot to experiment with it without causing spam.
[01:01] <leftyfb> loswedseded: 24.04 will be released in April
[01:36] <dgutov> Well, it didn't last long: I got a crash again with a bracktrace I've seen previously: https://pastebin.com/cB1csEV2
[01:36] <dgutov> Googling shows previous reports shows pages like https://bugzilla.redhat.com/show_bug.cgi?id=1836090 which went nowhere.
[01:36] <dgutov> But what do you know, "minimize.svg" (the only useful constant findable in the backtrace) does get referenced in many files inside ~/.local/share/gnome-shell/extensions/unite@hardpixel.eu/themes/ (there are many themes in there which all use this filename).
[01:36] -ubottu:#ubuntu- bugzilla.redhat.com bug 1836090 in Fedora "[abrt] gnome-shell: _st_theme_resolve_url(): gnome-shell killed by SIGABRT" [Unspecified, Closed: Eol]
[01:37] <dgutov> I wonder where apport sent my crash info -- it apparently did, but I can't find any corresponding bug on launchpad. Perhaps I should report to gnome-shell directly.
[01:39] <sarnold> try this:
[01:39] <sarnold> xdg-open https://errors.ubuntu.com/user/`sudo cat /var/lib/whoopsie/whoopsie-id`
[01:46] <dgutov> sarnold: Thanks! I got this: https://errors.ubuntu.com/user/e941f0eddf365fe3fc22dc2de28329c4ac207d5a2f44dc1882ddc0fd47dc985420ee38c216133b5a658351cb00842cecb739f6abe111b82cffe3f8a9c7a98991
[01:46] <dgutov> This indeed look like my log of crashes, but I'm not able to view the contents of each individual one because "Sorry, you are not a member of a group that is allowed ..."
[01:47] <dgutov> Any chance you *are* a member of that group and can see any actual bug report(s) that got filed as a result? I'm not sure requesting access to see everybody's reports is the best course of action for me personally
[01:51] <sarnold> hrm, this thing doesn't make it easy to walk from the bug reports to the error bucket that they got put in :(
[02:13] <dgutov> Oh well. Thanks for trying. I reported that to gnome-shell in the meantime, let's see how they react.
[02:13] <sarnold> dgutov: yeah, I can't find anything that looks quite like yours :(
[02:13] <sarnold> dgutov: maybe move your extensions out of the way for a bit and see if that helps? it's weird that you've had a year without crashes, and last january, gnome-shell crashes..
[02:14] <dgutov> I've upgraded to GNOME 45.1
[02:14] <sarnold> try this on the extensions: cd ~/.local/share/gnome-shell/ ; mv extensions extensions.old
[02:15] <dgutov> I'm not sure if I was without crashes, to be honest. But in the past, gnome-shell restarted safely, more or less, keeping applications running (the Xorg session). With a recent bug in GNOME 45.1 it just crashes on an attempt to restart, so I switches to wayland. That helped for a few days, but not too much.
[02:17] <dgutov> The backtrace mentions a file name which belongs to an extension (and is referenced inside .css file in it). So it's *probably* related, but it hardly seems like the extension's fault.
[02:21] <sarnold> dgutov: the new gnome-shell design gives extensions a crazy amount of power -- crashes from them are pretty common :(
[02:28] <tomreyn> dgutov: you could try something along the lines of    gdb -q -n -ex 'continue' $(which gnome-shell) $(pidof gnome-shell)
[02:28] <tomreyn> but you'd need to do this from an ssh session you can control from another computer, so that you can still analyze the process once it crashes. or at least from a tty
[02:29] <tomreyn> you could do a 'bt full' in there once it does.
[02:31] <tomreyn> or, possibly easier, have the core file written elsewhere than the apport pipe it normally goes to, but on disk, and then analyze the core file later one.
[02:31] <sarnold> heh, gdb gives me that same reaction, too!
[02:31] <tomreyn> looks like gnome-shell crashed
[02:32] <dgutov> Well, that made my current gnome-shell session freeze XD
[02:32] <tomreyn> hmm sorry, the "continue" should have made it continue, not sure why it wouldn't
[02:33] <tomreyn> you can always type that out on the gdb prompt, as long as you can type something there
[02:34] <tomreyn> (other computer, ssh)
[02:34] <dgutov> Anyway, 'apport -g' turns a .crash file into a gdb session rather successfully. That's where I got the stack trace from.
[02:34] <dgutov> Sorry, 'apport-retrace -g'
[02:35] <tomreyn> ah that's possibly the better way.
[02:37] <dgutov> Ok, so you're suggesting to attach gdb waiting for the next crash? That sounds like a fine approach, maybe I'll do that if the gnome folks really decide to investigate and ask for some more information.
[02:37] <tomreyn> yes, this was my plan
[02:47] <dgutov> Hmm, LP#1788429 has a very similar backtrace. But it's 3.5 years old and apparently "fixed".
[02:47] -ubottu:#ubuntu- Launchpad bug 1788429 in gnome-shell (Ubuntu) "gnome-shell crashed with SIGSEGV in g_file_get_parent → _st_theme_resolve_url → _st_theme_node_ensure_background → st_theme_node_paint_equal → st_widget_recompute_style → ... clutter_actor_show → ScreenShield::activate() → ScreenShield::lock()" [Low, Fix Released] https://launchpad.net/bugs/1788429
[02:51] <sarnold> dgutov: that certainly looks like it spends enough time in libst to match yours, even if the 'earlier' paths are pretty different -- but the error tracker's most recent entry there is from Wednesday, August 17, 2022 09:32 UTC
[02:52] <dgutov> Yup. And its related gnome-shell issue has a committed fix (https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/2757)
[02:52] -ubottu:#ubuntu- Issue 2757 in GNOME/gnome-shell "Gnome shell crashes when screen lock attempted" [Closed]
[02:56] <sarnold> nice
[06:56] <poweruser__> irc.openjoke.org
[07:42] <Yakov84> When i use search in folder GUI filed for it, it freezes and says Files not responding
[08:15] <en1gma> im running ubuntu 22.04.3 lts amd64-desktop. im trying to use nautilus browser to ssh into another device to the / location. it puts me in home and i can not go higher up a level. i can do it with terminal and ssh. why cant i with nautilus
[08:19] <ravage> https://p.haxxors.com/mwnlsb7z.webm
[08:28] <en1gma> 'mc' lets me do it
[08:29] <en1gma> cool. i love mc anyway
[08:48] <olle> Did something weird happen with firefox on ubuntu last week(s)? It's eating my cpu for breakfast right now.
[08:48] <olle> Using Xubuntu 20.04 LTS
[08:48] <olle> Or Ubuntu 20.04.6 LTS
[08:49] <olle> Or maybe some video encoding/decoding driver has failed recently
[10:43] <manwhowouldbekin> Hi all! While troubleshooting issues with a file syncing app, I was led to check the version of OpenSSL installed on my Ubuntu. I found this https://pastebin.com/ayzJbikY That is, the version that is used by default on my system is that provided by Anaconda/Miniconda. How can I change that? Is there a way to do it via Anaconda itself? That is, if I could upgrade from 1.x.x to something like 2.x.x there, it would
[10:43] <manwhowouldbekin> probably work for me.
[12:44] <Kronos> Hi
[12:45] <Kronos> Does UncomplicatedFirewall use nftables?
[12:46] <Kronosphere> Hi. Does UncomplicatedFirewall use nftables?
[12:49] <ravage> there is only nftables
[12:49] <ravage> everything else is a wrapper
[12:53] <Kronosphere> Ok, my understanding is ufw uses iptables
[12:54] <Kronosphere> ufw's man page talks of iptables, but not nftables.
[13:06] <Kronosphere> hmmmmmmmm
[13:14] <hygl> I have got two dell latitude 5491 running 22.04/23.10 . On both, when I connect the power supply, they do not charge immateriality, but it takes half an hour or longer before they change from 'not charging' to 'charging'. Is there an option, that they start charging right after connected to the power supply?
[13:29] <pagios>  hi all, if i have file1.7z.001 file2.7z.002 etc, how can i extract them ?
[13:37] <sc6502> pagios, do you have 7z installed?
[13:59] <Lcawte> I'm getting a lot of complaints about cgroups (not existing) in my server logs from systemd/systemctl. Any idea where I start
[14:00] <Lcawte> I suspect it's somewhat related that SSHd won't start (sshd -t complains about missing priv seperation)
[14:04] <leftyfb> Lcawte: you have to be leaving out some critical details. What release of ubuntu? How exactly did you install it? What sort of hardware/hypervisor/container is it installed on? What types of really deep customizations have you made that would cause these sorts of issues?
[14:05] <leftyfb> Lcawte: A default install of ubuntu server onto an x86 or even arm device can install sshd and will not have any of the issues you described
[14:28] <Enissay> Hmm, I just noticed I have a couple movies on my NAS with creation dates in the future `-rwxrwxrwx 1 eny eny  16G Jan  1  2098 Nice.Movie.REMASTERED.1080p.BluRay.x264-GAZER.mkv
[14:30] <leftyfb> Enissay: I don't see how this relates to Ubuntu. Any number of things could have happened or is still happening on an almost infinite number of OS's
[14:31] <Enissay> I was just curious how that can occur to avoid it... Only a couple files had such thing out of many... I updated that manually anyway, so, just for info xD
[14:36] <Lcawte> leftyfb: Yes, my apologies. Was 20.04 on a dedicated box, 64bit. Wasn't an issue at boot (as far as I could tell, provider rebooted post PSU replacement) - not really any specific customizations, one additional cgroup directory. Rebooted and it's playing nice so far.
[14:42] <dgutov> In continuation of yesterday's chat, what if gnome crashes, with an entry like this:
[14:42] <dgutov> Jan 05 14:16:27 potemkin gnome-shell[3533]: #0   555bb20aa9d8 i   resource:///org/gnome/shell/ui/init.js:21 (3d2e25e70ba0 @ 48)
[14:42] <dgutov> Jan 05 14:16:27 potemkin gnome-shell[3533]: == Stack trace for context 0x555bb1fe39d0 ==
[14:42] <dgutov> Jan 05 14:16:27 potemkin gnome-shell[3533]: GNOME Shell crashed with signal 11
[14:42] <dgutov> ...but there is nothing else in the journal, and apport doesn't pick up the crash log? Is that a problem to look into further?
[14:43] <leftyfb> !paste | dgutov
[14:44] <dgutov> Here's a fuller excerpt: https://dpaste.com/DYBA5XEGH But there are just 3 important lines here, and anywhere, that I could find.
[14:45] <dgutov> I'm guessing it's again related to the external hdd's problems (either the disk itself, or the enclosure), but it's odd that the crash is not recorder, nor is it sent for collection. This might affect many others, and we wouldn't know.
[14:53] <Solarium> Good Day everyone, I need some help tracking down why my WiFi is intermittently just freezing up. Sometimes it happens on a long suspend, sometimes it happens if the laptop is sitting there, sometimes while I'm actually using it.
[14:53] <Solarium> I'm using Ubuntu 22.04.03 LTS on a Thinkpad P1 Gen 4. The Wifi card seems to be a Intel AX210 using the kernel driver iwlwifi. Is there perhaps proprietary intel drivers I need to get? I was looking around and couldn't find any, maybe I missed it.
[14:53] <Solarium> If anyone could help I'd appreciate it
[14:54] <leftyfb> Solarium: start by disabling wifi power save
[14:55] <Solarium> leftyfb Sure, I'll lookup how to do that, thanks. Does that mean that when the laptop is suspended, it won't power off the wifi card too? or only during operation. I appreciate it
[14:55] <leftyfb> no
[14:56] <Solarium> Ah ok thanks
[14:56] <leftyfb> power save has little to do with suspending the laptop
[14:56] <leftyfb> though suspend on all computers in general is a whole other set of problems. I tend to stay clear of that
[14:57] <Solarium> got you
[14:57] <Solarium> $ cat /etc/NetworkManager/conf.d/*.conf
[14:57] <Solarium> [connection]
[14:57] <Solarium> wifi.powersave = 3
[14:57] <Solarium> I'll need to find what value to change that to I'm guessing
[14:57] <leftyfb> 2
[14:57] <Solarium> Ah ok, Thanks!
[14:59] <jeremy31> Solarium: also check results for iwconfig
[15:00] <Solarium> What specifically would I be looking for using iwconfig? link quality and retry short limit or something?
[15:00] <leftyfb> "Power MAnagement"
[15:00] <Solarium> It's set to on
[15:00] <leftyfb> A/a
[15:01] <leftyfb> chabnge your setting, reboot and see what it's set to
[15:01] <Solarium> Will do, Thanks a lot
[15:01] <Solarium> I'll change it and see if it happens more, If I'm not back have a great weekend!
[17:09] <dgutov> So... I set up a breakpoint with gdb in a VT (at Alt-F4), and when gnome-shell crashed this time, gdb aborted as well. No chance to print the backtrace or do anything: https://imgur.com/a/5G8Wnye (the "show logging" command is something I typed before the breakpoint was hit).
[17:09] <dgutov> Is that another symptom of faulty memory, or a misconfigured system? The external drive was disconnected this time.
[17:16] <leftyfb> dgutov: do you still have the defective drive plugged in?
[17:16] <dgutov> Unplugged since before the last crash (mentioned above)
[17:16] <leftyfb> dgutov: and do you have non-stock gnome extensions removed?
[17:22] <dgutov> No, those are still around. I'm trying to learn to produce useful backtraces for gnome-shell that would help fix any incompatibilities.
[17:23] <leftyfb> dgutov: step #1: find out what is causing the issue   step #2  report it as a bug   step #1.5 or #3 spend time trying to capture backtraces and dig into why thr cause is causing problems
[17:24] <leftyfb> the easiest method in this case would be to remove all added gnome extensions and see if the problem goes away. Then add them back in 1 at a time until the problem can be replicated
[17:24] <leftyfb> process of elimination
[17:24] <leftyfb> once you have your culprit, you can spend time looking further into the issue
[17:27] <dgutov> I only have ~3 non-trivial extensions, and one of them is Ubuntu's built-in (Indicators). The problem does not happen predictably, so eliminating the causes might take a while
[17:27] <dgutov> I tried to capture the backtrace - and it didn't work at all (gdb died on gnome-shell's sigsegv)
[17:29] <leftyfb> dgutov: you're already spending lots of time chasing cryptic messages. You can do both at the same time. Remove the non-stock extensions
[17:34] <Yakov> Pls guide me how to kill terminal 4241
[17:34] <Yakov> https://justpaste.it/9wikd
[17:35] <leftyfb> Yakov: do you mean PID 4241?
[17:35] <Yakov> yes
[17:35] <Yakov> tried kill, kill -9, pkill
[17:35] <leftyfb> reboot
[17:36] <Yakov> lol
[17:36] <arraybolt3> if `kill -9 4241` doesn't work, it's probably stuck on a syscall.
[17:36] <Yakov> cant reboot
[17:36] <arraybolt3> you can try `sudo kill -9 4241` as a last-ditch effort
[17:36] <arraybolt3> that's the ultimate sledgehammer, if that doesn't work you're out of options except to wait.
[17:37] <leftyfb> as I said, reboot :)
[17:37] <Yakov> case: I have a lot of gnome-terminal instances opened, I need to close all, exept one, thats why I need to kill
[17:37] <leftyfb> reboot or ignore it
[17:37] <Yakov> process is killed with sudo
[17:37] <Yakov> but still there are two terminals opened
[17:41] <dgutov> leftyfb: The difference is I'll be working in a less functional environment during that time. And reproduction might take a while: last time, the "good" period lasted several days, even with the extensions enabled.
[17:41] <dgutov> Anyway, switched them off now. I'll see what happens.
[17:45] <Yakov> guys, any suggestions upon my case?
[17:46] <leftyfb> Yakov: reboot or ignore it. Those are your only 2 options
[17:46] <Yakov> I need to do it every day
[17:46] <leftyfb> then you're doing something very wrong
[17:47] <Yakov> I have one app which is static, other secondary apps call their terminals which I need to kill
[17:47] <leftyfb> define "app" that "call their terminals"
[17:47] <Yakov> so I need to kill all terminal, excluding GUI and one terminal to do OS commands
[17:47] <Yakov> c++ application
[17:48] <leftyfb> custom?
[17:48] <Yakov> my app
[17:48] <leftyfb> ok, fix the bug causing the problem
[17:48] <leftyfb> it's making a syscall that is hanging forever
[17:48] <Yakov> well, it was just gnome-terminal I called with desktop file
[17:50] <leftyfb> Yakov: reboot, don't touch any custom apps. Does the stock gnome-terminal open and cloe on it's own using the built-in desktop file without problems?
[17:50] <leftyfb> close*
[17:50] <Yakov> yes
[17:51] <leftyfb> Yakov: then the problem is with your app
[17:51] <leftyfb> I recommend /join #c++ for help with tracking down the issue with it
[17:52] <Yakov> the thing is I did not called my cpp app in this boot, I'm just trying to find syntax
[17:53] <leftyfb> syntax for what?
[17:55] <Yakov> I've opened 4 terminals, I need to kill last 3 and keep first one
[17:55] <Yakov> https://justpaste.it/9wikd
[17:55] <Yakov> pgrep -P 5379 | xargs kill -9 removes tree of processes, but terminals themselves are still open
[17:56] <leftyfb> keep track of the PID's of the terminals you open and loop through that dict to kill the last 3
[17:56] <leftyfb> again, this is a #c++ issue, not ubuntu
[17:56] <Yakov> wait, I did not open my c++ app yet, its just stock gnome-terminal
[17:57] <leftyfb> then why are you trying to kill terminals using CLI? Just close the windows
[17:57] <leftyfb> with your mouse
[17:57] <leftyfb> or type "exit" in each terminal
[17:57] <Yakov> I need to kill 20 terminals, and keep two
[17:58] <leftyfb> ok, drop the 2 you want to keep over to a corner of your desktop and close the rest
[17:58] <leftyfb> why do you have 20 gnome-terminal windows open if you only want 2?
[18:00] <Yakov> I debug some shit :)
[18:13] <dgutov> Okay, I just had a crash again, with only two stock extensions (Ubuntu Dock and AppIndicators and another -- css-only -- extension called "Light Theme"):
[18:13] <dgutov> Jan 05 20:03:35 potemkin kernel: Code: 14 0f 1f 44 00 00 49 83 ed 08 4c 39 6c 24 10 0f 84 ee 02 00 00 48 8b 83 28 01 00 00 4c 89 e6 4a 8b 2c 28 48 8b 45 00 48 8b 00 <48> 8b 38 e8 bc>
[18:13] <dgutov> Jan 05 20:03:35 potemkin kernel: gnome-shell[3539]: segfault at 0 ip 00007f9730bb180c sp 00007fff422aff40 error 4 in libst-13.so[7f9730b86000+54000] likely on CPU 10 (core 4, socket>
[18:13] <dgutov> This one again wasn't captured by apport
[18:13] <leftyfb> damn, for Yakov: for PID in $(ps -ef |grep $(ps -ef |awk '/[g]nome-terminal/ {print $2}') |awk '/bash/ {print $2}'|head -n-3); do sudo kill -9 $PID ;done
[18:13] <leftyfb> but they left
[18:14] <leftyfb> dgutov: did you reboot after removing the extensions?
[18:14] <leftyfb> dgutov: also, "Light Theme" is not a stock extension. Remove it
[18:16] <dgutov> Yep. Although what I did was disable+reboot. Is that enough?
[18:16] <dgutov> Okay, I'll disable "light theme" as well
[18:17] <leftyfb> I would remove them entirely
[18:17] <leftyfb> you're trying to get back to as stock as you can get
[18:17] <leftyfb> eliminate anything non-default
[18:21] <dgutov> Okay, removed and rebooted
[18:36] <yes-ubuntu> hello... is it possible to upgrade from 22.10 straight to 23.10 ?
[18:37] <lotuspsychje> yes-ubuntu: 22.10 is end of life by now
[18:38] <yes-ubuntu> lotuspsychje: okay... my question still stands :)
[18:38] <leftyfb> yes-ubuntu: possible, yes. Likely to result in a stable system, no
[18:39] <yes-ubuntu> leftyfb: thank you. how can I invoke the upgrade procedure? :D
[18:39] <leftyfb> yes-ubuntu: non-LTS releases are only supported for 9 months
[18:39] <lotuspsychje> yes-ubuntu: i would not reccomend upgrading from eol versions, the longer you wait, the more security issues coul arise
[18:39] <leftyfb> yes-ubuntu: I might also point out, it's not supported and not recommended
[18:41] <yes-ubuntu> I can't really understand that... ok, it is how it is. However, if it was possible to upgrade from 22.10 to 23.04 and then to 23.10, why isn't it possible to do the upgrade now?
[18:42] <lotuspsychje> yes-ubuntu: leftyfb said its possible, we just dont reccomend it
[18:42] <leftyfb> yes-ubuntu: I would highly recommend just installing ubuntu 22.04 from scratch. Stick with LTS's unless you are going to be upgrading every 9 months
[18:42] <leftyfb> yes-ubuntu: or spend the time to upgrade manually through each version leading up to 23.10
[18:43] <yes-ubuntu> so, is it possible to revert from 22.10 to 22.04 and then upgrade to 23.10 ?
[18:43] <leftyfb> no
[18:43] <lotuspsychje> yes-ubuntu: and if you choose the non-lts road, please upgrade to the next release 'before' your current release is end of life
[18:43] <leftyfb> also, upgrading from 22.04 directly to 23.10 is also not officially supported
[19:09] <bernd> hallo
[19:31] <bernd> hello
[19:32] <onkel> moinsen
[19:32] <leftyfb> onkel: hello. What can we help you with?
[20:13] <akram> h
[20:15] <akram> can we use android in lubanto
[20:16] <tomreyn> akram: you can connect your computer running lubuntu to another runnnig android. and you can use virtualization or containment to run variants of android on (under) lubuntu.
[20:19] <NikNikovsky> does anyone know a way to get my wallpaper into Ubuntu?
[20:22] <tomreyn> NikNikovsky: you mean for everyone, or just for you?
[20:23] <NikNikovsky> for everyone to see
[20:25] <tomreyn> you could try to have it added to an existing package, by requesting addition to where the package source is hosted. or by talking to the package maintainer. or you could write and publish a blog post to explain how others can manually add it to their ubuntu installation.
[20:26] <NikNikovsky> i want to add it for EVERYONE to see, something like some people in the community did
[20:27] <yes-ubuntu> okay! I have installed 23.10 and I have edited /etg/default/grub with nano and before restarting the system, I would need to execute sudo update-grub, however, I get this error: "/usr/sbin/grub-probe: error: failed to get canonical path of `/cow'." anyone could point me in the right direction? (I am still in the try ubuntu session, after having
[20:27] <yes-ubuntu> installed the OS to the disk)
[20:27] <tomreyn> NikNikovsky: there is usually a wallpaper competition for ubuntu, but there you'd neeed to comply with existing style guides, probably not what you're looking for. https://discourse.ubuntu.com/t/mantic-minotaur-23-10-wallpaper-competition/37235
[20:30] <tomreyn> yes-ubuntu: you need to boot into the installed system first of all
[20:30] <tomreyn> there's another approach, but it's more complicated
[20:30] <yes-ubuntu> tomreyn: I thought so :)
[20:40] <peacefulman> for 22.04, anyone get auto screen resizing with virtualbox working? I installed the virtualbox guest modules, but it seems I have to manually set the resolution
[20:54] <yes-ubuntu> Hey! I have just installed Ubuntu 23.10 from a USB drive by selecting the Legacy USB at the one time bios boot menu. The screen was very pixel like (one pixel was huge on the screen)... nevertheless, the install went on (with nice graphics and good resolution) and after it finished, I removed the USB and restarted the system... and the same low
[20:54] <yes-ubuntu> resolution black and white text display appeared, saying, there is no boot drive found? I can't seem to be able to select the internal SSD not even from the one time bios boot menu?
[21:58] <mcc> Hello. I would like to upgrade the firmware on my laptop. There are several weird issues I am encountering.
[21:59] <mcc> One, the first time I tried to do it after getting a notification I had a firmware update available, I got an error indicating Ubuntu had apparmor rules preventing me from upgrading the firmware. That sounds smart, but is there not some way to escalate privs? https://files.mastodon.social/media_attachments/files/111/654/885/772/750/809/original/575d5ca6ca41f168.jpg
[22:01] <mcc> Two, when I went to try and test this again today, it says "Current version 0.1.30, Latest version 0.1.31" but hwen I click "Update to latest" it asks if I want to update Embedded Controller to "0.1.30". Why? https://usercontent.irccloud-cdn.com/file/OjPlnR2M/Screenshot%20from%202024-01-05%2016-57-39.png
[22:02] <mcc> Three, when I attempted to run the firmware updater tool in sudo, it just failed with a baffling error message. What does any of this mean? https://usercontent.irccloud-cdn.com/file/ImkF1AiX/image.png
[22:02] <tomreyn> about "one": did you receive another message stating what you said a message stated, or just the message which you posted a screenshot of (i don't think *this* message says what you said your message said).
[22:03] <mcc> This is on Ubuntu 23.10
[22:04] <mcc> tomreyn: I'm sorry, I don't understand the question. I just got this big error box and then a second error box telling me not to reboot the computer while the firmware is updating. My comment about apparmor was a paraphrase. What I mean is it did not install and the error popup mentioned apparmor.
[22:04] <tomreyn> about "two": if you clicked on "update to latest", this looks like a bug in the firmware updater snap to me.
[22:04] <mcc> OK. Do you expect it is only a GUI error or might telling it to update do the wrong thing? Should I file this?
[22:05] <mcc> (Incidentally, it also seems like a bug to me that after failing to update firmware and showing a error box saying it had failed to update firmware, it afterward showed the second "do not reboot while we install" box, apparently contradicting both the first error box and what the app was doing at that moment.)
[22:05] <tomreyn> about "three": you should never run GUI applications as root
[22:06] <tomreyn> (there are some applications parts of which will need to gain elevated permissions, but that's a different thing, and they will take care of it by themselves)
[22:06] <mcc> Okay. Well, at the moment it appears the firmware updater snap is not correctly obtaining its elevated permissions
[22:08] <causative> wtf the gnome rdp server doesn't work if the gui is on the lock screen????
[22:10] <tomreyn> regarding your response "OK. Do you expect it is only a GUI error or might telling it to update do the wrong thing? Should I file this?" - i assume this is in response to my response to topic "A"? I fos: I'm not ebtirely sure what the effect of the error message presented is. what the message states is, i think, that apparmor rules prevented the firmware-updater application from getting a list of certain dbus properties from the running gnome
[22:10] <tomreyn> session.
[22:11] <causative> paternalistic pseudo-security crap
[22:11] <tomreyn> causative: you are welcome to have opinions, but this is a support place. do you have a support question?
[22:12] <causative> I want to remote desktop into my desktop, from my phone, wherever I am. This was working before. How do I get it working again.
[22:12] <causative> or vnc into my desktop, whichever.
[22:13] <leftyfb> causative: sudo apt install gnome-shell-extension-manager
[22:13] <tomreyn> mcc: regarding your remark in parenthesis, i would also agree that this seems like a bug, too.
[22:13] <leftyfb> causative: this will allow you to login remotely via RDP if it's on the lock screen, but not if you haven't logged in locally yet
[22:15] <mcc> tomreyn: Sorry, I was unclear. I guess I have three questions: (1) How should I proceed (as I want to upgrade my firmware but I do not currently appear to be able to)? (2) Do the oddities in topic A or B seem to imply bugs which should be filed? (3) In (b), when it asks if I want to update the embedded controller to 0.1.30, do you think it *means* 0.1.31 (eg: should I expect Update to update to 0.1.31, despite the message)?
[22:15] <tomreyn> mcc: regarding your remark "at the moment it appears the firmware updater snap is not correctly obtaining its elevated permissions", I think that elevated permissions and apparmor are not the same thing. i assume the firmware updater was able to gain elevated permissions, but some of what it tried to do seems to have been prevented by apparmor.
[22:15] <mcc> ok, perhaps so
[22:16] <tomreyn> mcc: on your next reboot you will see whether or not the firmware updater succeeded in staging a firmware update, in case this is yet unclear.
[22:16] <mcc> tomreyn: i am going to try to file some of these items if I can reproduce them after an apt upgrade. do you think it would be more helpful to file one bug per oddity (0.1.30/0.1.31 confusion, apparmor/dbus error, spurious second dialog after the apparmor/dbus error) or one superbug?
[22:16] <mcc> tomreyn: re "on your next reboot..." Oh. That makes sense.
[22:17] <tomreyn> unless it actually stated that preparing firmware update failed, it may well have succeeded despite of the (misleading, and not helpful) apparmore related error message.
[22:18] <mcc> i will try a reboot, then.
[22:18] <tomreyn> mcc: always file separate bugs, only after checking which of those already have been filed
[22:18] <causative> thanks leftyfb
[22:19] <tomreyn> leftyfb: installing the gnome shell extension manager will allow connecting via RDP to a screen locked gnome shell session?
[22:19] <tomreyn> if so, could you explain why? i'm just trying to lean, didn't know this.
[22:20] <tomreyn> *leaRn
[22:20] <leftyfb> sorry, missed a step
[22:20] <leftyfb> after you you need to install the allow-locked-remote-desktop extension
[22:20] <leftyfb> https://extensions.gnome.org/extension/4338/allow-locked-remote-desktop/
[22:20] <JanC> that makes more sense  :)
[22:20] <leftyfb> yeah, sorry
[22:21] <tomreyn> causative: ^ i think this goes to you
[22:21] <leftyfb> googling for people and ansibling at the same time :)
[22:21] <causative> yeah I already got it, thanks
[22:22] <tomreyn> leftyfb: ah, glad we could clarify this one. :)
[22:22] <mcc> What is the package for the software updater? (this is about something separate from the firmware updater issue). I tried to search on the bug reporter and got "sorry, something went wrong with your search" https://usercontent.irccloud-cdn.com/file/4gCYsFKH/image.png
[22:22] <causative> if my left shift, but not my right shift, is failing to work because of some software reason, is there anything I can do to fix it short of logging out?
[22:22] <causative> anything I can restart that would help?
[22:24] <tomreyn> you could try restarting your gnome-shell session on Xorg (won't work on wayland), by pressing alt-f2, typing "r" and pressing enter.
[22:24] <leftyfb> causative: I don't think so
[22:24] <tomreyn> replacing, not restarting, sorry
[22:24] <leftyfb> tomreyn: yeah, they were asking about this yesterday. It's not possible on wayland
[22:25] <tomreyn> There's always the chance that causative is running Xorg, I guess, unless you already clarified this yesterday, too.
[22:26] <tomreyn> mcc: it'S the ubuntu-software snap, i think
[22:26] <mcc> thakns
[22:27] <tomreyn> hmm apparently this is wrong, at least therE's no such snap listed on snapcraft.
[22:27] <tomreyn> .io
[22:28] <mcc> It doesn't matter because I pressed the return key inside the component button and it posted my issue (to the SQLite3 component) :(
[22:29] <tomreyn> i think it's https://snapcraft.io/snap-store
[22:30] <mcc> So it's too late now, unless I can reassign the bug component myself, or you can do it https://bugs.launchpad.net/ubuntu/+source/sqlite3/+bug/2048405
[22:30] -ubottu:#ubuntu- Launchpad bug 2048405 in sqlite3 (Ubuntu) "multiple packages failed to install/upgrade: unable to create '/usr/lib/x86_64-linux-gnu/libsqlite3.so.0.8.6.dpkg-new' (while processing './usr/lib/x86_64-linux-gnu/libsqlite3.so.0.8.6'): Read-only file system" [Undecided, New]
[22:30] <mcc> I have just rebooted. As you predicted, the firmware DID update during the reboot. So that's great.
[22:31] <tomreyn> did it upgrade to the same version you already had, though, or to the newer version?
[22:31] <tomreyn> journalctl -b | grep DMI:
[22:31] <mcc> It upgraded to the newer version. 0.1.31.
[22:32] <mcc> It is possible(?) that the weird "upgrade to 0.1.30"? was caused by the fact the update had already run and was pending install.
[22:32] <tomreyn> good, so i guess you just have a misleading error message and a misbehaving firmware update (trying some restricted action which it probably should not be trying) and an incorrectly reported target version.
[22:32] <mcc> Yes. The "do not interrupt me while I'm installing this!" message was the correct one, not the error message.
[22:33] <tomreyn> mcc: which package should 2048405 be reassigned to?
[22:34] <tomreyn> (it does look quite sqlite related)
[22:34] <mcc> tomreyn: really? if you think it's sqlite related it can stay. as noted tho it happened with 3 different packages.
[22:35] <mcc> tomreyn: it should be assigned to whichever package owns the program that runs when I search in the "show apps" menu for "software updater" and click "software updater". above, you speculated it was snapcraft.io/snap-store.
[22:35] <tomreyn> mcc: i guess the real question is why you have a read-onl file system at /usr
[22:35] <mcc> tomreyn: yes, that's an interesting question. actually, it is persisting after a reboot
[22:36] <mcc> Or rather, after a reboot, attempts to run software updater results in this: https://usercontent.irccloud-cdn.com/file/mgIUbY13/image.png
[22:38] <tomreyn> hmm your dmesg (posted on this bug report) is full of suspends + resumes
[22:38] <mcc> over what time period?
[22:39] <mcc> I would expect several suspend+resume cycles a day because this is a laptop which spends most of its time closed and stowed under a desk.
[22:40] <mcc> Do you know what the component is for `/snap/bin/firmware-updater`, or how I would find out? the launchpad "select a package" search is not very helpful… :(
[22:40] <tomreyn> ~300k seconds
[22:41] <mcc> That's 83 hours. I have definitely suspend/resumed many times over the last 83 hours.
[22:42] <mcc> I attempted a sudo apt-get update && sudo apt-get upgrade and got the same read-only file ystem error.
[22:42] <tomreyn> actually 451545 s
[22:42] <tomreyn> 125h
[22:42] <mcc> well, same expectations.
[22:43] <tomreyn> well, try to find out why your file system is read-only
[22:43] <tomreyn> use journalctl
[22:43] <tomreyn> use mount
[22:43] <tomreyn> or   cat /proc/mounts
[22:43] <tomreyn> verify /etc/fstab
[22:43] <tomreyn> use lsblk and / or blkid
[22:44] <tomreyn> there are no bugs with those packages, your file system is not mounted, which is not related to these packages
[22:45] <mcc> I agree, which is why I wanted to file against software updater :( however from what you say maybe it is not the software updater either.
[22:45] <tomreyn> the software updater should not be causing your /usr file system to be mounted ro
[22:46] <tomreyn> and so far we have no indication that it did
[22:46] <mcc> might the firmware updater have done this?
[22:46] <tomreyn> we could speculate on random packages having caused it, or you could check your logs on what happened
[22:46] <mcc> like, locked the filesystem in anticipation of a reboot, then failed to set things right after the reboot. seems a little unlikely maybe...
[22:47] <tomreyn> no, this doesn't make any sense
[22:47] <mcc> ok.
[22:47] <tomreyn> it would prevent a clean shutdown
[22:48] <tomreyn> the firmware updater application just puts the firmware update in place where the firmware can then find it and update itself on reboot.
[22:48] <mcc> i have gathered all those things you asked for above, except for journalctl, which without further refinement goes back to oct 17.
[22:48] <mcc> I can post a zip here— there is no security/privacy concern with posting them in the irc channel is there?
[22:48] <mcc> /etc/fstab looks normal as anything to me , by which I mean I don't see any weird options.
[22:48] <bprompt> mcc:  there are places you can post a zip file for hosting btw
[22:49] <tomreyn> journalctl -b    get you the log of the active session,  -b -1   gets the one from before last reboot
[22:49] <tomreyn> ...and so on
[22:49] <mcc> bprompt: i believe irccloud does this for me, i just want to get a confirmation these drive GUIDs aren't sensitive information somehow
[22:49] <tomreyn> you can filter logs messages to only see warnings (-p4) or just errors (-p3)
[22:50] <tomreyn> --list-boots shows which session matches which time period
[22:51] <bprompt> mcc:  in essence, you can /dcc send NICK to anyone, doable with an irc client menu too, however the NICK must consent to receive it, and you'd want to first ask for allowing it, many may say no
[22:51] <tomreyn> there can be unique hardware IDs on the journalctl output (just liek on dmesg)
[22:52] <mcc> This is from current session journalctl, today: Jan 05 17:29:12 Anthy kernel: EXT4-fs (nvme0n1p3): mounted filesystem ecb1730b-cbe1-457c-a4f8-23e2eaf89020 ro with ordered data mode. Quota mode: none.
[22:52] <mcc> It doesn't say why it did that.
[22:52] <tomreyn> there could also be information you have entered / assigned, such as wlan access point names or your user accounts full name.
[22:53] <mcc> ok well hopefully that ecb thing i just posted isn't sensitive :P
[22:53] <tomreyn> where did it mount this one to, though / where does it usually get mounted to? it's not necessarily wrong to have ro mounts
[22:54] <tomreyn> filesystem and partition UUIDs are automatically generated, there is not much sensitive info in there, other than when they were generated.
[22:56] <mcc> https://data.runhello.com/j/ubuntu/readonly/readonly-fs-mystery.zip
[22:57] <mcc> those are the mountpoints and such
[23:00] <tomreyn> can you show the output of:  findmnt --target /usr
[23:00] <alcosta> Hi everybody! Is here leftyfb!
[23:01] <tomreyn> mcc: i'm guessing nvme0n1p3 is your /
[23:01] <mcc> I'd expect so
[23:01] <mcc> /usr sysext overla ro,relatime,lowerdir=/run/systemd/sysext/meta/usr:/run/systemd/sysext/extensions/cosmic-sysext/
[23:01] <tomreyn> though the lsblk output is weird
[23:01] <mcc> Hmm so this is interesting
[23:01] <mcc> I did install a systemd sysext
[23:01] <mcc> I forgot this until now because I thought it was insignificant
[23:02] <mcc> (sorry)
[23:02] <mcc> My expectation from the directions I followed was that this would result in an overlay of additional files visible in the /usr directory , not that it would supplant the /usr mountpoint entire with a readonly one
[23:03] <tomreyn> i'm not really familiar with systemd sysext, but a quick glance on the man page seems to suggest it does change things quite a bit. i suggest you undo it, if that's possible, or reinstall
[23:04] <mcc> yes, from the manpage "System extension images are strictly read-only, and the host /usr/ and /opt/ hierarchies become read-only too while they are activated." :(
[23:04] <mcc> well, that's easy to fix then.
[23:05] <tomreyn> possibly, but undoing the harm that the fact that /usr has been unwritable for a while may not be as easy to fix
[23:05] <tomreyn> you will see
[23:05] <tomreyn> maybe someone else here is more familiar with what you did.
[23:05] <alcosta> Eventually I solved the problem I discussed with leftyfb! Thanks to his suggestions!
[23:06] <Cahata> hello
[23:06] <alcosta> Now I have another problem, in some way similar to the previous one, but different!
[23:06] <Cahata> is ubuntu on rpi 5 usable as home pc ?
[23:06] <tomreyn> alcosta: thanks for reporting back (i don't know what  it was about). if you think it is useful knowledge, you could share how you solved it.
[23:07] <alcosta> Hi tomreyn! In the exact way leftyfb suggested me to work with it!
[23:07] <tomreyn> alcosta: nice. :)
[23:08] <tomreyn> Cahata: possibly, with limitations.
[23:09] <bprompt> Cahata:  last I bother checking, is just a headless version, streamlined, but is just headless, is all
[23:10] <alcosta> Hi have two desktop PC, PC A and PC B, running ubutnu 20.04. On PC A I have a Virtual Machine VMA, on PC B I have a Virtual Machine VMB. I wanted that all the machines can communicate therebeteen. All the Machines are running ubuntu 20.04.
[23:10] <tomreyn> Cahata: a SoC is not really comparable to a 'proper' desktop or laptop computer. but you should be able to run ubuntu on it in a usable way.
[23:10] <tomreyn> Cahata: i do assume that you can run a graphical desktop on it.
[23:10] <mcc> After removing the extension and rebooting I can now run software updater correctly. Tomreyn thanks for the help
[23:11] <tomreyn> mcc: you're welcome. i suggest closing you bug report(s) then, at least those you cannot reproduce now.
[23:12] <alcosta> Eventually, with the suggestions of leftyfb about the definition of a bridge network, I was able to connect all the machines! :)
[23:13] <mcc> I added a note to https://bugs.launchpad.net/ubuntu/+source/sqlite3/+bug/2048405 requesting it to be closed, but if there is a button to close it myself I do not see it. Will changing status to "Invalid" do it?
[23:13] -ubottu:#ubuntu- Launchpad bug 2048405 in sqlite3 (Ubuntu) "multiple packages failed to install/upgrade: unable to create '/usr/lib/x86_64-linux-gnu/libsqlite3.so.0.8.6.dpkg-new' (while processing './usr/lib/x86_64-linux-gnu/libsqlite3.so.0.8.6'): Read-only file system" [Undecided, New]
[23:13] <Cahata> tomreyn that for sure... just intrested in small pc as a sandbox for torrents ...
[23:13] <Cahata> thanks for answers :)
[23:14] <tomreyn> mcc: yes, that's the only option there is
[23:15] <tomreyn> Cahata: it should work for a small torrent seed node as long as you don't expect to use too much bandwidth.
[23:15] <alcosta> Now, I would like to solve a similar problem, but a litle different: I have a PC A with a VMA, both running ubuntu 20.04 and a PC B, running windows 10, with a VMB running ubuntu 20.04. How can I enable the four machines to communicate therebetween?
[23:16] <tomreyn> whats a VMA and a VMB?
[23:17] <mcc> One more question. Is there a convenient way to get a window back into usable range when the title bar moves above the top of the screen? https://usercontent.irccloud-cdn.com/file/1BkSsNtf/image.png
[23:18] <tomreyn> ah i guess you mean "VM A" (as in "virtual machine one") and "VM B" (as in "virtual machine two")
[23:18] <bprompt> tomreyn: in essence he said, VM running on PC "A" and VM running on PC "B"
[23:18] <mcc> Ubuntu is constantly popping up this on-screen keyboard (I do not want this, ever, but I looked into it and there appears to be no way to disable it), when it pops up all the windows move upward to make room for it, and sometimes they don't move back down afterward
[23:18] <bprompt> alcosta: pardon my meddling, let's nevermind VM for a second, what are you trying to achieve at all?
[23:18] <tomreyn> mcc: hold down meta and move the window around anywhere on the window
[23:18] <leftyfb> alcosta: I would recommend you join #kvm to learn how to configure your VM's to utilize the bridges you created. Once you do that, they should all be on the same network as the hosts
[23:19] <alcosta> VMA is a Virtual Machine running ubuntu 20.04 on the PC A, whereas VMB is a Virtual Machine running ubutnu 20.04 on the PC B. In this case, PC B is running under Windows 10. Have I to define a similar bridge network under Windows 10? How have I to do that?
[23:19] <tomreyn> mcc: that's drag + drop with your mouse
[23:19] <mcc> tomreyn: sorry, i don't think I know what meta is. alt, ctrl, and compose did not do anything.
[23:19] <leftyfb> bprompt: 4 machines on the same network. 2 are hosts, 2 are VM's. We already created bridge interfaces on the hosts. Now the VM's need to be configured to use the bridge interfaces
[23:19] <tomreyn> mcc: "windows key"
[23:20] <mcc> Oh. I reassigned that to compose. Um. Hm.
[23:20] <bprompt> hmm ok
[23:21] <dgutov> mcc: There's this: https://extensions.gnome.org/extension/3222/block-caribou-36/
[23:21] <mcc> dgutov: oh, that is great. thank you so much.
[23:21] <mcc> i have a laptop with a touchscreen so i literally never want or need the onscreen keyboard.
[23:21] <dgutov> np! same here
[23:22] <alcosta> Hi leftyfb, thank for your suggestions on bridge network. Eventually, it worked! All the machines can communicate therebetween! Now, a bit different situation: what if when PC B is running Windows 10 and the other three Machines are running ubuntu 20.04?
[23:22] <leftyfb> alcosta: ok?
[23:23] <leftyfb> alcosta: if all machines are on the same network (check the ip addresses, they should all be the same except for the last octet) then they are all "connected".
[23:23] <alcosta> Under ubuntu 20.04 the four Machines are correctly working! They communicate all together!
[23:24] <mcc> I think I am not going to file my issues with firmware-updater because the chances they are due to the readonly filesystem issue are so high.
[23:25] <tomreyn> mcc: i'd make the same choice, and keep an eye on whether they re-occur next time a firmware upgrade is available
[23:25] <mcc> makes sense. thanks
[23:26] <alcosta> Now I'm trying to work with another condition, you can also consider a new PC C, running Windows 10, with a Virtual Machine running ubuntu 20.04.
[23:26] <leftyfb> alcosta: you need to /join #windows for that
[23:26] <alcosta> In the real case, PC C is the same computer B in condition of dual booting with Windows 10.
[23:27] <alcosta> Ah, there is a channel windows too! Good to know!
[23:27] <leftyfb> alcosta: if Windows is the host, then you need to go to #windows for support with creating a bridge interface for that host
[23:29] <alcosta> Ah, OK! Thank you leftyfb! I will do this way! I will ask in windows channel! Thank you again for the suggestions about bridge network! This was a good week for me! Great job well done! :)
[23:35] <alcosta> And a last (maybe but.. maybe not last too!) question: I developed a new App that partially is coded in nodejs and partially in python (among some other things). The way I used for communicating data between nodejs and python is through pipe files available under ubuntu 20.04 (and linux in general). I want the same App can work under windows 10 (or 11, is good too).
[23:36] <leftyfb> alcosta: that's beyond the scope of this channel. Try #python or #nodejs
[23:36] <leftyfb> !alis | alcosta
[23:38] <alcosta> AH, ok! Thank you leftyfb and ubottu! I'm new in this chat, I'm using it from less than two weeks! Thank you again in any way! :)
[23:40] <bprompt> alcosta: https://netsplit.de/channels/?net=freenode   <---- you can do a quick channels search there as well
[23:41] <bprompt> alcosta: some of the guys at #javascript also do nodejs btw, since it's really just js for the server
[23:45] <alcosta> ubottu - I write as you said: «/msg Alis help list» but... nothing happens!
[23:46] <alcosta> here as in #libera!
[23:47] <alcosta> Any way, I will ask in #windows - #python and #nodejs! Thanks again!
[23:50] <Bashing-om> alcosta: Alis' output will show in a new window.
[23:52] <bprompt> alcosta: https://netsplit.de/channels/?net=Libera.Chat    <---- updated  link, to search channels
[23:52] <bprompt> gave you an old one before
[23:52] <alcosta> Bashing-om, I don't see any new window!
[23:54] <Bashing-om> alcosta: running the command like:  /msg Alis help list ?
[23:55] <Kangarooo> For triaging bug report. To check a bug, i would like to know is this good strategy- to find all files containing shortcut meta, so i can find shortcuts configuration files is it good to search all files? what search would be best?
[23:56] <alcosta> Bashing-om - I cut and paste /msg Alis help list and nothing happens!