/srv/irclogs.ubuntu.com/2025/02/11/#ubuntu.txt

Guest69420i have an odd issue with mame for ubuntu... the mouse cursor movement is mirrored... if i roll my mouse up or left, it moves down or right.... how would i reverse this?00:08
enigma9o7turn monitor upside down?00:24
enigma9o7or maybe just mouse00:25
=== jgee8 is now known as jgee
=== antonispgs1 is now known as antonispgs
Guest69420enigma9o7, it's only in MAME, no other programs :p01:54
=== Square3 is now known as Square
=== gandalf72 is now known as gandalf7
=== robert3 is now known as brosna
=== foo is now known as foo-old
=== foo_ is now known as foo
k_szeIn Ubuntu (noble), is there a way to tell apt to install a meta package as a specific real version? I forget.05:51
Bashing-omk_sze: No - meta is a list of what to install - See: https://help.ubuntu.com/community/MetaPackages .05:58
k_szeRight, I forget the correct terms.06:00
=== pusher is now known as keypusher
C8-Comp-26hi07:18
C8-Comp-26hi07:19
Bashing-omC8-Comp-26: Hello - you have a support question ?07:19
C8-Comp-26na bruv just testing my college comps07:19
C8-Comp-26im sorry07:20
=== JanC is now known as Guest7443
hid3Good morning everyone.08:30
hid36.11 kernel is out for 24.04.208:30
=== coderandy is now known as Randy_
=== JanC is now known as Guest4906
=== JanC is now known as Guest5649
CorvusCoraxHi.  I have an issue with my ubuntu 20.04 LTS -- whenever I boot kernel 5.4.0-205-generic and load nvidia drivers - my network card stops working - the driver freezes in kernel space on any network access causing network using processes to also hang indefinitely.  in 5.4.0-172-generic that doesnt happen.   any suggestions what to do?11:49
CorvusCoraxin "recovery mode" i have network, but no graohics - it seems it only stops working once something changes the graphic resolution11:52
grywhat graphics card is this and what driver11:52
CorvusCorax Driver Version: 570.86.15,   VIDIA GeForce RTX 2080 Ti   (nvidia-driver-570-open [570.86.15-0ubuntu1])11:53
* mgedmin would try to upgrade to 22.04 lts, hoping the nvidia driver bug doesn't affect kernel version11:54
CorvusCoraxthats from nvidias PPA but i had the same issue earlier with older driver 470 from ubuntu repo11:54
CorvusCoraxcan't upgrade to 22 due to incompatibility with ROS1 - there is no ROS1 for 2211:54
CorvusCoraxassuming I stay with 172 -   how come apt upgrade keeps the 172 around on top of the 205 - and by what rules -- can i enforce to keep that with apt-mark or will that break security updates to the kernel?12:02
CorvusCoraxapt list --installed gives me linux-image-5.4.0-172-generic/focal-updates,focal-security,now 5.4.0-172.190 amd64 [installed,automatic] and linux-image-5.4.0-205-generic/focal-updates,focal-security,now 5.4.0-205.225 amd64 [installed,automatic] --  but no other versions atm12:03
mgedminthere's a linux-image-generic metapackage that depends on the latest linux-image-X.Y.Z-W-generic version12:04
CorvusCoraxI assume doing an "apt-mark hold linux-image-5.4.0-172-generic" would be safe to avoid losing my one working kernel version12:04
mgedminwhen a security update comes out, you get a new version of linux-image-generic that depends on the new kernel12:04
mgedminthe old one is no longer depended by anything and would normally be autoremoved12:04
mgedminbut apt autoremove has special logic for kernel packages where it keeps the last working version, just in case you need to revert to it12:04
CorvusCoraxah ok12:04
mgedminso you end up always having two kernels (for each flavor, like -generic)12:05
CorvusCoraxso if i update now with 172 running its gonnna keep 172 around since its the "working version" and replace the 205 ?12:05
mgedminI'm not sure, more likely it'll keep the two most recent versions12:05
CorvusCoraxwell apt-mark hold should avoid an autoremove i hope12:06
mgedminor maybe it tries to keep the version you're currently running, I don't remember the precise logic12:06
mgedminbut if you do not run the latest kernel, then you don't get security updates in any case12:06
mgedminit's best to report any regressions with `ubuntu-bug linux`12:07
CorvusCoraxthis is really weird, i have other PCs running which differ only by tiny details (same motherboard, same network card, slightly different GPU version ) where this issue doesnt happen12:07
mgedminmention the working kernel version and the failing one, so people have a chance of identifying which backported patch broke things12:07
CorvusCoraxhow do i report that best ? i can't really generate any bug report as theres no useful debug output (I can't produce even a dmesg since i can't sudo - as sudo for some weird reason wants to check the hostname which freezes the machine since network access which freezes the process12:08
ograCorvusCorax, try https://bugs.launchpad.net/ubuntu/+source/linux/+filebug12:16
CorvusCoraxhttps://bugs.launchpad.net/ubuntu/+source/linux/+bug/2097885 <-- tried my best12:34
-ubottu:#ubuntu- Launchpad bug 2097885 in linux (Ubuntu) "Network driver freezes when nvidia driver is loaded on 205-generic" [Undecided, New]12:34
centI can connect to a mail server using mutt, but mutt gets stuck after receiving "220 2.0.0 Ready to Start TLS"12:44
centUbuntu 22.0412:45
centmutt 2.1.412:45
centIt does NOT send login to the server and must be killed as it cannot be terminated.12:46
CorvusCoraxdoes it work with other STARTTLS enabled servers?12:47
pragmaticenigmacent: have you tried enabling debug mode to see what the trace looks like?12:48
centpragmaticenigma, yes -d 512:48
pragmaticenigmacan you post those to https://bpa.st and send us the link?12:48
centThat is why I can see where it got stuck12:48
centpragmaticenigma, I'll try .12:49
centpragmaticenigma, https://bpa.st/6DZA13:05
BluesKajHi all13:08
pragmaticenigmacent: Have you contact the host provider to see if they have any suggestions about what TLS encryption they support?13:08
pragmaticenigmacent: From what I gather from other bug reports, the stall is incompatibility of TLS ciphers used for encryption. I'm not familiar with mutt enough to know what to suggest.13:09
centpragmaticenigma, One more thing. My .muttrc has been taken from a host (Debian 12) where it works like a charm. No changes made to the config.13:09
centpragmaticenigma, Tha is fine. Thanks for hints anyways.13:11
tomreynpersonally, i'd prefer explicit TLS, do they provide that on 465?13:11
tomreynyou can use package "swaks" to analyze smtp encryption issues, or just openssl s_client.13:13
tomreynhttps://wordtothewise.com/2023/10/can-you-starttls/13:14
tomreynhttps://www.rfc-editor.org/rfc/rfc3207 is the relevant RFC13:14
tomreyn(for starttls)13:14
CorvusCoraxu could also run wireshark to analyze the stream - if it fails already in the handshake it should be able to decode that13:15
CorvusCoraxfor the payload stream u need to somehow get mutt to dumb the session keys, but as I unerstand it doesn#t get far enough13:15
Guest73hi13:17
nutellaGuys, where can I create a bug? https://bugs.launchpad.net/ I don"t see any possiblity here13:23
nutellaeven if logged in13:24
nutellaperhaps I"m blind13:24
pragmaticenigma!bug13:25
ubottuIf you find a bug in Ubuntu or any of its official !flavors, please report it using the command « ubuntu-bug <package> » - See https://help.ubuntu.com/community/ReportingBugs for other ways to report bugs.13:25
nutellapragmaticenigma, actually this is my bug: https://pastebin.com/raw/tKT5riJt what package name do i put here?13:37
nutellaNotebbok keeps freezing and the i have to hard reboot it. I'm on ubuntu 24.04 with kernel  6.8.0-52-generic13:37
pragmaticenigmaWithout any guidance on "what" in that log file I'm supposed to be looking at, I cannot help you13:39
CorvusCoraxi915 driver it seems13:39
CorvusCoraxbut i don't know what package that is in13:40
CorvusCoraxcould be linux  (the kernel itself)  or a separate driver package13:40
CorvusCoraxhttps://bugs.launchpad.net/ubuntu/+source/linux/+filebug13:41
pragmaticenigmaI don't think it is the i915... appamor is being triggered by something13:41
pragmaticenigmathere is a lot going on in there... and honestly, I'm not skilled at understanding most of it. I'd be looking for help before just going off and filing a bugreport. to make absolutely certain you have narrowed the bug to the kernel13:43
CorvusCoraxto be fair, the issue is definitely triggered in kernel, as theres drivers and IRQ handlers complaining about timeouts... the question is just -- is it the kernel package (in built driver) or a separate driver package that causes it - or broken hardware that just triggers the issue somehow...  if you have broken hardware that would explain the symptoms also and then thers no bug whatsoever13:45
nutellaI have the feeling it was introduced in one of the two last kernels. before it was fine. it happens mosttly, the freezing, when I run youtube or google meet13:46
pragmaticenigmanutella: what is "before"13:46
CorvusCoraxBIOS flaws could also trigger that ind of behaviour -- so the question is really what triggered it - and can you go "back" and see if its really absent in earlier versions13:47
nutellaWith one of the older kernels it was still fine. pragmaticenigma13:47
nutellaI mean yes I could install an older kernel version again see if it still happens13:47
CorvusCoraxif some hardware component broke and rendered the system unstable, then going back to older kernel version would not make it disappear, even though the 1st symptoms might have corelated with an update13:47
nutellacorrect13:48
CorvusCoraxso that would be a good test13:48
pragmaticenigmanutella: that wasn't the question... I'm looking for a specific13:48
CorvusCoraxthat would also narrow the package down. if going to older kernel fixes it, you can file a bug with the "linux" package (see link above) and provide as much info as possible, including output of lshw , the working and non-working versions, symptoms and how to reproduce - etc...13:49
pragmaticenigmaLike was it a different version of Ubuntu. Because kernel versions do not change within a release cycle. Patches are applied, but the version remains13:49
nutellano it was the same 24.04 version. but some 1 or two months ago this issue started. imo after a kernel update13:49
nutellathat is all I can say13:50
pragmaticenigmathere is also the issue that something is triggering apparmor in those logs, at the same time as the i915 timeout messages... ther is a lot going on at the same time... someone skilled in reading the log needs to help here. can't just be throwing random thoughts at it, and dumping a kernel bug without some legwork is going to get ignored13:50
nutellaalright thanks for the analyses. So you guys think I should not file a bug yet but try our perhaps an older kernel13:52
CorvusCoraxyeah, but tbh decyphering what exactly happens here and why (aka which bug is cause and which errors are effect and which are noise caused by unrelated issues) goes beyond my skill13:52
nutellaLet deepseek reason about it with internet search: https://markdownpastebin.com/?id=e85b35e1675349549edec78b835d3c5213:52
pragmaticenigmathere's a lot of time in 1 or 2 months ago... so no... to claim it was a kernel update is just something being fixated on at the moment, because that's the most prominatent thing in the log13:52
pragmaticenigmakernel dumps a lot of things into the log13:52
nutellapragmaticenigma, i noticed yes13:52
CorvusCoraxso my suggestion would be to file the bug, and at the same time ask elsewhere for help with analysis - and attach that on the bugreport if you find out more13:52
pragmaticenigmathat's why I say... someone that knows how to read that is going to be decipher what truely is a fault13:52
nutellai forgot to say one thing: GRUB_CMDLINE_LINUX_DEFAULT="quiet splash i915.enable_psr=0 i915.enable_guc=0" this was helping (no crash anymore) but the boot was really slow sometimes I nneded to go into the boot menu to select "manual" boot for it to come up again Computer then was also very slow13:55
nutellaso i disabled it again last week13:55
nutellaSo i guess that narrows the whole thing down to the grafic card/ i91513:57
CorvusCoraxmy hunch is theres an issue with the on board graphics card - but that includes its driver - or the GPU memory, which probably is system memory on a laptop13:57
pragmaticenigmanutella: have you done a memory check?13:57
CorvusCoraxtry older kernels which you knew worked and see if it still happens --- and yeah, memory check or other burnin-tests would help too to see if its a hardware issue13:58
nutellawill do. Thanks guys for your suggestions13:59
CorvusCoraxlinux has the annoying habit of running surprisingly stable even with faulty RAM ;)  - to the point where it becomes hard to notice that soething is wrong with the system except for random process crashes and very rare kernel oopses13:59
CorvusCoraxbut eventually you get telltale weird stuff like random filesystem corruption and the like ...14:00
pragmaticenigmaenable_psr should only be used if you experiencing screen flickering14:01
yes-ubuntuHi! I am playing around with tcgetattr and tcsetattr to change the terminal settings from a C program... and... somehow the changes persisted and messed up my terminal (no matter if I use zsh or bash), the command prompt displays far away on the line, plus, I redirected the CTRL+C and CTRL+\ handlers and after exiting from my program, these14:39
yes-ubuntuhandlers aren't restored? More, I logged out and logged back in and these are still persistent (not reset to default); any idea of how I could restore them (without having root access? and without restarting the machine?)14:39
pragmaticenigmayes-ubuntu: first, use libraries for doing those things. ncurses or similar libraries intended to perform actions to the active terminal will avoid issuing flags and commands that persist, and usually when the object is dereferenced it will restore settings it changed (check the documentation on the library to see how to make sure it does that.) As for how it persists across a reboot, I'm not sure. You can gain root access via rescue14:52
pragmaticenigmamode and if there's a help guide to resetting terminal, you should be able to follow it14:52
pragmaticenigmayes-ubuntu: ctrl + alt + F3 (or trying any F-key 1 through 7) should bring up a TTY terminal that may not experience the same problem.14:53
yes-ubuntupragmaticenigma: thank you. I have no root access (not my machine)... and, I had to use tcgetattr/tcsetattr as this is for an assignment :)14:53
yes-ubuntupragmaticenigma: ctrl + alt + F3 do not work on this machine14:54
=== justache is now known as justache2
leftyfbyes-ubuntu: I would contact your instructor on how to recover your "machine" (container? VM?)14:58
=== justache2 is now known as justache
yes-ubuntuleftyfb: they are willing that we learn... and got the reply "if you break something, you have to manage to fix it..."15:08
yes-ubuntu(which is a super cool approach... just that, I am frustrated...)15:09
leftyfbyes-ubuntu: good luc15:09
leftyfbluck*15:09
=== GJdan_ is now known as GJdan
pragmaticenigmayes-ubuntu: I would then reach out to your classmates that are also in the course... someone there may have an idea15:47
pragmaticenigmaor at least be familiar with the environment you are working in15:47
pradeepcan anyone give me mirror.list for all the distros to get mirror on a single repo server16:27
leftyfbpradeep: that likely doesn't exist and isn't an ubuntu support question16:28
=== rvalue- is now known as rvalue
=== neocharles is now known as Guest3002
UbuntuUserIt's insane that it's impossible to make the screen lock after some amount of time without the screen also going blank. Insane.17:36
leftyfbUbuntuUser: why would you need to do that?17:38
UbuntuUserI think that the only way is to change the screenlock program from Gnome to some other alternative. Does anyone have any suggestions?17:38
UbuntuUserI want to show off my new wallpaper to my coworkers17:38
UbuntuUserI'm planning to buy a Porsche Macan and I want everyone to see it on my desktop17:38
UbuntuUser*screenlock17:39
enigma9o7Have you considered taking a photo, putting it in a frame, and putting it on your desk?17:39
enigma9o7It doesn't even require electricity to remain on.17:39
enigma9o7But I tried forever to find a way to make the screen blank without locking, and couldn't figure that out either.17:40
hid3UbuntuUser: I fully understand your wish, seems valid. I had a very yound and sexy GF when I was working back in the days at a company where I used Ubuntu as a desktop. I added a few hot pics of her to be as a screensaver and switched off monitor going blank timout at all17:40
leftyfbhid3: lets not17:41
hid3point is, it should be customisable17:41
hid3but not 'why would you need to do that?'17:41
UbuntuUserapparently, not on Gnome it's not17:41
UbuntuUserMaybe I should try slock17:42
UbuntuUserhttps://tools.suckless.org/slock/17:42
=== neocharles is now known as Guest2972
UbuntuUserokay, xautolock may show promise, but I'm not sure whether it will work on Wayland. Back in my desktop-ricing days everyone used X17:44
pragmaticenigmathe key letter is in the name... X17:45
pragmaticenigmaso no17:45
leftyfbUbuntuUser: for wayland, try: dbus-send --session --dest=org.freedesktop.login1 --type=method_call /org/freedesktop/login1/session/self org.freedesktop.login1.Session.Lock17:45
UbuntuUserpragmaticenigma: I hoped there would be a backward-compatibility layer, I think there was for a time, but maybe no more17:47
UbuntuUserleftyb: didn't work, I'll look for xautolock alternatives for Wayland, there appears to be https://github.com/dukzcry/crap/tree/master/idler17:48
pragmaticenigmaUbuntuUser: XWayland is that compatibility layer... but what your suggesting is something that specifically relies on a feature of X... XWayland is just a virtual window for applications to become usable under a wayland session17:48
CorvusCorax@enigma9o7, a important usecase is when you need to remote connect to the desktop17:54
CorvusCoraxthat often doesnt work if the screen turns off completely17:54
CorvusCoraxso unless your xrdp (or whatever eayland has instead) can turn the screen back on, you will not get a remote session17:55
=== carlos_ is now known as carolvs
leftyfbCorvusCorax: sounds like something is broken on your end. That shouldn't be a thing17:55
=== carolvs is now known as Caronte
CorvusCoraxwwell i havent tried it with wayland, just with Xrdp on Xorg -- the issue is when the screen turns off, there is no more virtual framebuffer for xrdp to mirror on the remote connection.   or - there wasn't with the nvidia drivers, on my intel laptop the same thing worked (it apparentlz only turned the actual displaz off, not the rendering)17:57
CorvusCoraxso maybe thats a nvidia issue17:58
CorvusCoraxanyway that was Xorg, so the solution was to disable power management for the monitor so it would stay on when in screensaver17:58
UbuntuUser"xdg-screensaver lock" and "dbus-send --type=method_call --dest=org.gnome.ScreenSaver /org/gnome/ScreenSaver org.gnome.ScreenSaver.Lock" both work but they immediately blank the screen, maybe some Wayland xdotool alternative can unblank it immediately, I'll be looking into this17:58
pragmaticenigmaCorvusCorax: I have never experienced an issue with a screen powered down and connecting to a remote screen share.17:58
pragmaticenigmaSo something was wrong with your setup17:59
CorvusCoraxmaybe17:59
CorvusCoraxanyway preventing the screen from turning off fixed it for me17:59
CorvusCoraxso i just assumed that was required, maybe it wasnt17:59
CorvusCoraxthe other thing where its important is when i record sim experiments - but thats a different issue - here i not only want the screen to be on but also prevent the screensaver or lock screen from engaging (as that would ruin the recording)18:00
pragmaticenigmawhat more likely was the case is depending on the greeter. there used to be a hook in older versions of gdm, sddm, and others that allowed VNC and other remote desktop sharing services to allow sharing the active X session, insetad of just the user session. The only way I have had that continue to work is with lightdm, which still has the feature available.18:00
CorvusCoraxand again thats a different issue with wayland, since screen recording now depends on the compositor18:00
CorvusCoraxyeah. maybe that was it18:01
pragmaticenigmacorrect, X-server would allow clients to share a single framebuffer. Which was both nice and a bit of a vulnerability18:01
UbuntuUserwtype won't work on Gnome, brilliant18:02
CorvusCoraxits fine with sufficient levels of access control.  if i have a root process and sufficient access i can also directly inspect any process memory while its running - that also could be considered " a bit of a vulnerability" -- whether it becomes a vulnerability depends on whether theres a path for unauthorized users or processes to access that feature18:03
pragmaticenigmaI'm just going to put this out there... UbuntuUser, honetly, no one is going to give a rip about your background on your screen. The attempts to try and get this working aren't worth the risk as to way you need a screen lock in the first place. As someone else suggested, put a picture on your desk in a frame if it is that important to you18:03
CorvusCoraxhey would that mean i can't show my cool matrix style screensaver demo under ubuntu wayland anymore?18:04
CorvusCoraxwell i guess a smart picture frame running ubuntu could do the trick instead18:05
CorvusCoraxas long as i disable the lock screen18:05
CorvusCoraxbtw I once worked for a company building kiosk-computers for large clients - they would have these available for workers to browse inventory or safety manuals and the like - sometimes even SCADA interfaces -- one feature the client always asked for was a lock screen showing their company logo and branding18:06
CorvusCoraxso for that usecase just turning the screen off also wouldnt be an option18:07
pragmaticenigmalets stick to ubuntu support, not life stories please18:07
UbuntuUserYES18:07
UbuntuUserdbus-send --type=method_call --dest=org.gnome.ScreenSaver /org/gnome/ScreenSaver org.gnome.ScreenSaver.Lock ; sleep 3; sudo ydotool click left18:07
UbuntuUserworks18:07
UbuntuUserbut you need to run ydotoold also18:07
CorvusCoraxbut granted these kiosks ran their own flavour of the distribution in question anyway - so in that case they'd just patch ubuntu to disable to monitor power management, if need be as a hardcoded fix wayland18:07
UbuntuUser...but now I need a way to detect screen idling...18:09
UbuntuUserdbus-send --print-reply --dest=org.gnome.Mutter.IdleMonitor /org/gnome/Mutter/IdleMonitor/Core org.gnome.Mutter.IdleMonitor.GetIdletime18:10
UbuntuUserthis works as well18:10
pragmaticenigmaUbuntuUser: maybe this article can help: https://fostips.com/keep-lock-screen-display-linux-gnome/18:11
CorvusCorax@UbuntuUser, another more radical option, if all else fails, ditch wayland - switch back to a Xorg system18:11
pragmaticenigmaUbuntuUser: read the entire article, and make sure you note the Ubuntu versions in the steps18:11
=== SDGNelson2 is now known as SDGNelson
pragmaticenigmaCorvusCorax: that makes no difference as it isn't X or the compositor that is controlling that18:12
CorvusCoraxah18:12
UbuntuUserI think I already all the stuff I need to make a script to make it work, but it's just insane that I have to do all this just to make an automatic screenlock without the screen going blank18:12
leftyfbUbuntuUser: gdbus monitor -y -d org.freedesktop.login1 | grep LockedHint --line-buffered18:12
CorvusCoraxhave a nice day everyone and thanks for the help, gotta run :)18:13
Guest9471hi19:40
Guest9471whats up19:40
enigma9o7For my goal (blank without lock), under X its trivial.  issue, under x its easy to blank the screen (xset dpms force off), but haven't found a way under gnome wayland.19:46
pragmaticenigmaexcept they wanted the opposite... lock without blank19:51
enigma9o7Yep I know, I was here for the initial querry, when I also mentioned I wanted the opposite.19:51
=== sonOfRa_ is now known as sonOfRa
pragmaticenigmaenigma9o7: I remember figuring it out once before20:15
pragmaticenigmaI have my Fedora instance setup to blank the screen, but not lock20:16
pragmaticenigmabut nothing on demand20:16
=== jafarabbas33 is now known as Vani
=== uf89 is now known as uf
enigma9o7Yeah I was looking for how to turn off the screen now.  My other laptop has a "display off" button which I find quite useful, and of course desktop pcs have a monitor with an off switch.  but the laptop running oractular/wayland doesn't have such a button, and not being able to setup a keyboard shortcut instead is annoying.21:23
AimDesireeI'm an entrepreneur and developer working on a search engine called Scrollgen, designed with Gen Z in mind. It's currently in beta, and many features are still in development. If you're curious, you can check it out at [www.scrollgen.com](https://www.scrollgen.com) and share any feedback on what could be improved or added. There's no app yet, but21:35
AimDesireethe web version is available for testing. If you want to follow my progress, you can find updates on Instagram: [https://www.instagram.com/scrollgen_official/](https://www.instagram.com/scrollgen_official/).21:35
leftyfbAimDesiree: spam is offtopic here21:35
AimDesireeSorry, I just want people to know about a search engine I'm working on21:36
AimDesireeI should have ask first21:36
leftyfbAimDesiree: this is a support channel. Not the place for spam21:37
enigma9o7We get unsolicted ads everywhere else tho, it doesn't seem right for their to be no spam at all here, so at least thanks for that Desiree.21:38
leftyfbenigma9o7: lets not21:38
AimDesireeok, I won't do it anymore21:38
pragmaticenigmaThe whole libera network is no spam21:41
pragmaticenigmaand what zoomer is on IRC????21:42
=== seanh__ is now known as seanh
=== y0sh- is now known as y0sh_
Ruby64hi tkob23:17

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!