[00:01] F_Carson: Again - wher are you pulling from - as : ' apt list -a libedataserver-1.2-24 ' does not reflect that version as available in the repo. [00:01] where* [00:03] https://youtu.be/R3YJ0brUmb0 [00:06] sorry to hijack: can i use do-release-upgrade to go from 20.04 to 20.10? when i try it with -d, it tries to take me to 21.04, which i can't do yet (one of my snaps crashes on it) [00:08] signofzeta: -d is (D)evelopment version :D [00:08] ah, so it should gladly upgrade LTS to non-LTS without flags? [00:10] I hate you, systemd. [00:10] That is all. [00:10] signofzeta: you can get that via editing a config file [00:10] pulling straight from the main repository: libedataserver-1.2-24/focal-updates,now 3.36.4-0ubuntu1 amd64 [installed,automatic] [00:10] signofzeta: depends on what is set in the config file: /etc/update-manager/release-upgrades . [00:11] Bashing-om, sarnold: Thanks! I'll try it when I can. [00:11] signofzeta: /etc/update-manager/release-upgrades [00:11] oh hah sniped by Bashing-om [00:11] some days we snipe, some days we get sniped. happens to the best of us :-) [00:12] :) [00:16] F_Carson: What shows ' apt policy libedataserver-1.2-24 ' - as the installed version depends do not match what is in the repo. [00:23] Installed: 3.36.4-0ubuntu1 [00:23] Candidate: 3.36.4-0ubuntu1 [00:23] Version table: *** 3.36.4-0ubuntu1 500 [00:24] 500 http://us.archive.ubuntu.com/ubuntu focal-updates/main amd64 Packages [00:25] 100 /var/lib/dpkg/status [00:26] ?? [00:28] F_Carson: "Depends: libecal-2.0-1 (>= 3.36.5) but 3.36.4-0ubuntu1 is to be installed" - Unsure of what is not going on here. [00:38] <[Eli]> Welcome in warsoul :-) [00:38] thanks [Eli] :) [00:38] <[Eli]> Anytime [00:46] well, i'm stumped. it seems right buggered. === M4he is now known as mahe [01:37] PSA: neovim clobbers all timestamps on file edit, losing creation date, and when reporting it they say it's working right and to f off. DO NOT use neovim unless you're ok with losing data [02:44] miya === genii is now known as genii-core [03:20] Lenovo Thinkbook 14 IML with touchpad - having trouble with touchpad.......does not work. Only work around is using a patched 5.9 rc versioned kernel to get it working...the ernel unsigned [03:21] If anyone has found a way to make it work without using the patched 5.9 rc kernel, do let me know [03:21] the touhpad is elan [03:22] have tried using 20.04, 20.10 and even 21.04 - Elan touch pad does not work in any version uness patched 5.9 kernel is used [03:43] kedar_apte: I don't have that touchpad, so all I can recommend is visiting Ubuntu's Launchpad and searching for or submitting an issue. Sorry I can't be of more help. [03:49] dumb question: does Ubuntu have an emoji keyboard? On my Windows PC at work, I can hit Win+. to pull it up. [03:50] and I know macOS has its own emoji picker. [03:54] thanks @signofzeta [03:54] Ubuntu does have an emoji picker...it can be found by searching for emoji in the Dash [03:55] thanks, kedar_apte! didn't even think of that. [03:56] kedar_apte: as far as your trackpad issue, that sounds like an issue with the mainline kernel itself. i'm sure someone at kernel.org is aware of it. i'm not sure how you raise things to them, but if Ubuntu Launchpad is no help, look there. [03:56] BTW - the company I work for - its name is Zeta :) [03:56] @signofzeta - sure thanks [03:57] i state on the record that i am affiliated with no one, and my crackpot opinions are entirely my own. :-P [03:57] 😀 [03:57] picked it up from the emoji picker [03:58] I'm old school. I know my emoticons. But I occasionally need an emoji, so hey. [04:13] can I get a hand configuring an hp document feed scanner real quick? === Guest2107 is now known as toolz [05:24] Hi what is difference in lts and normal mode in do release upgrade ? [05:24] I am on lts , what will happen if i do normal [05:28] sanbot: lts means from 18.04 to 20.04 to 22.04. Normal means that 20.04 will go to 20.10 to 21.04 etc [05:29] Will i lose lts ? If i do normal on lts ? [05:29] Yes [05:33] alkisg: ok got it === bsmith093 is now known as Guest2249 [05:36] but we can do "normal" updates until the next LTS and then switch back to LTS, right? ~1y from now [05:38] Right [05:39] But note that if one is on 20.04, they can't update directly to 21.04. They need to go through 20.10 first. === wode is now known as wodes === djhankb0 is now known as djhankb [07:09] Hi. https://wiki.ubuntu.com/Kernel/OEMKernel reads: "OEM kernel rebases to the master kernel on every SRU cycle, so it gets the same fixes (including but not limited to CVE fixes) from master kernel." -- Does anyone know whether this includes Canonical Livepatch Service support as well? [07:10] generic, lowlatency, aws, azure, oem, gcp, gke, ... https://wiki.ubuntu.com/Kernel/Livepatch [07:10] for 20.04 lts, that is [07:11] oerheks: Thanks a lot! I've looked at the wrong places for this information… That's what I wanted to hear! \o/ [07:11] have fun! === nomad1 is now known as nomad_fr === klebers_ is now known as klebers [07:50] hi. i have trouble updating my firmware on my thinkpad t480s. im using kubuntu 21.04 and found this bug: https://github.com/rhboot/shim/pull/374 [07:50] Pull 374 in rhboot/shim "BUG/RFE: more set_second_stage() fixes and improvements" [Open] [07:51] i tried compiling the changes ... successfully but it still did not work... but im not so sure if updating the package actually worked... since dpkg -L doesnt show me any "package files"... do i need to recompile shim-signed as well since im using secureboot? trying recompile the shim-signed didnt work tho [07:51] https://stackoverflow.com/questions/67800819/changing-file-name-encoding-in-python-while-monitoring-directories [07:54] TheBigK: you can't build shim-signed yourself, that requires a very special private key [07:56] He can't sign it with the private microsoft key, but if he installs he own MOK key in his UEFI firmware, can't he sign it with that? [07:57] *his [07:58] I'm reading the bug report now trying to figure out why a shim update would be needed for fwupdmgr [08:00] ah, I see the last firmware update on my X390 happened a couple of days before I upgraded to 21.04 [08:01] if you disable secure boot, will fwupdmgr skip shim and use fwupd64.efi in the boot entry directly? you can then reenable secure boot once the firmware is updated [08:06] ah, but the intent is to see if that shim PR fixes the bug [08:07] so yes, signing the shim with the keys in /var/lib/shim-signed/mok is probably best (but I don't know how to do that) [08:21] mgedmin: i can just skip shim ? i tried disable secure boot... that didnt help either... [08:21] i thought efi is required to fwupdmgr to work [08:23] the usual boot order for fw updates is efi firmware -> shim -> fwupd, after which it does the work then reboots into the usual boot order of efi -> shim -> grub -> linux [08:23] I don't know if fwupdmgr update checks whether you need shim, or whether it always configures EFI boot variables to use shim + fwupd [08:24] I was mostly thinking out loud [08:24] secure boot means that the efi -> shim step needs shim to be signed with a private key that is trusted by the bios [08:25] I think I was wrong earlier with my idea of signing shim with the key from /var/lib/shim-signed/mok [08:25] it's not the efi firmware that trusts keys enrolled with mokutil, it's shim itself [08:26] * mgedmin 's brain hurts [08:26] may be im waiting for a fix from the ubuntu side... and do the firmware update with a usb ... [08:26] i thought it could be easy to apply to patch myself and fix it that way... but im failing miserably :D [08:27] oh hey with secureboot off you can do fwupdmgr upgrade, and then use efibootmgr to change the Linux-Firmware-Updater boot entry to boot into fwupdx64.efi directly, bypassing shim [08:27] before you reboot [08:27] this is probably the simplest workaround available right now [08:28] i thought i tried that... hmm... [08:28] well, simplest would be to wait for ubuntu to fix their shim, but that will take weeks/months [08:28] i try it one more time and see if that works [08:42] it worked... finally... thanks mgedmin [09:34] Anyone here with experience of joining an AD using Ubuntu workstations? The joining of AD works flawlessly (using sssd and adcli), however - login events aren't visible in the AD logging (wineventlog:security) which are required for external applications (FSSO). [09:58] kyentei: are you running desktop or -server? [10:00] kyentei: when you log in...does it spit you back out to the log in again? [10:00] kyentei: are your policies configured correctly? [10:01] lotuspsychje: desktop [10:02] tatertots: login works just fine. It's just that the login event doesn't appear in AD logging. We've been using the Ubuntu VDIs for well over a year now. :-) [10:02] tatertots: What policies are you referring to? Windows login events are logged fine. Though, that's a different team within the organisation so I have no insight in how that's all configured. [10:03] kyentei: ok..so you are able to log in successfully, you are wanting to see more verbose logging on the windows server side of the house [10:04] tatertots: indeed. Credential validation is logged, but the login events aren't. They're required to "connect" the username and the originating Ubuntu workstation IP adress in order to assign certain firewall rules. [10:05] kyentei: you aren't going to increase WIndows server side log verbosity from the "client" side...so not sure what your expectations are in this environment which is primarily for Ubuntu desktop scope [10:05] tatertots: The login events are, however, logged for Windows workstations. So the verbosity is already there, I'd think. [10:06] I am, perhaps falsely, assuming it has to do with the authentication/login method used. [10:07] kyentei: do you have access to one of the ubuntu client systems right now? [10:07] Yes [10:07] kyentei: the policies i mentioned above can be observed from the ubuntu client [10:08] kyentei: in terminal> realm list === cbreak_ is now known as cbreak [10:09] The only policy it returns is allow-realm-logins [10:09] kyentei: which would allow a domain administrator to validate the policies [10:11] kyentei: have you already consulted with the enterprise/domain admins about your query ? [10:12] tatertots: not the exact query. Though, using "realm join" - I'm not sure what the underlying query is. I have posted the sssd.conf in an askubuntu question, which I could link to if you're interested. [10:13] kyentei: if i were in your shoes i would consult with them on the detailed specifics, they can probably shed some light on it for you...your policy is not very granular [10:13] kyentei: i could show you my policies which are more granular in nature [10:14] kyentei: however this wouldn't justify changing ubuntu client policies across the entire infrastructure [10:15] kyentei: are you guys using pure FreeIPA/sssd or some "middle ware" like PBIS for getting ubuntu clients on the network? [10:17] We're connecting straight to AD using sssd and sssd-ad. [10:17] I've also contemplated using sssd-ldaps instead, but I'm not sure how well that would work [10:17] I'll consult some more with the admins on the windows side of things. Thanks for your feedback :-) [10:18] kyentei: no problem [10:18] .. and help! [10:24] kyentei: i just checked my logs and my ubuntu client log ins are logged ...so if your enterprise admins deem it justifiable they may alter the policies to get your desired results [10:24] or approve a change it policy [10:25] in/it [10:25] Hi all === fantomas1 is now known as OnkelTem [10:26] I was installing virtualbox on 20.04 and suddenly I received the following dialog: https://i.gyazo.com/79f12b1947f3b06e553fc682f2170648.png [10:27] Systemm UEFI and blah blah. What the heck? [10:28] Since when when one installs a user package they're also configuring the system secure boot? [10:28] As far as I remember, I don't use any secure boot on my laptop [10:29] OnkelTem: are you chatting from the computer right now? [10:29] tatertots: yep. Why [10:31] OnkelTem: check in terminal> mokutil --sb-state [10:31] OnkelTem: virtualbox needs additional kernel modules; kernel modules need to be signed in secure boot systems [10:31] OnkelTem: yeah do that mokutil command above [10:32] virtualbox's kernel modules use dkms, i.e. they're compiled (and signed) on the installing user's machine [10:32] Thanks, doing [10:32] OnkelTem: what was the output of that command? [10:32] this needs a signing key to be present in the filesystem and enroled into EFI variables, which is a one-time action [10:32] SecureBoot enabled -- ops [10:32] OnkelTem: mmmmmhhhmmm [10:33] it's basically next, next, next finish, write down a one-time password, reboot, enter the same password and next next next finish [10:33] Thanks mgedmin! I thought it's seomthing wrong :) [10:33] one of the reasons I hate virtualbox is this hassle with 3rd-party kernel modules [10:34] I try to use libvirt or gnome-boxes or something that's based on KVM and is supported natively [10:34] which makes using Vagrant a bit of a pain... [11:04] tatertots: Thank you! Sorry for the late reply, lunch break. I'll pass that message along! [11:05] tatertots: Could you send me which policies you are using? [11:11] kyentei: https://termbin.com/qbyv [11:12] tatertots: Are we talking domain controller GPO? [11:12] tatertots: thanks :-) [11:14] tatertots: The only difference I see is the login-policy is allow-permitted-logins on your end and allow-realm-logins on mine. And I don't have the permitted-logins and permitted-groups in there at all. [11:15] kyentei: i get logging when a AD member is a "permitted" member and also when they are NOT...appears in security logs which is the same place you mentioned desiring to see a more verbose logging [11:16] Do I configure the allow-permitted-logins in sssd.conf, or is it assigned by AD? [11:16] kyentei: this more granular approach of explicitly permitting or denying gives a increased level of logging [11:19] keypusher: sssd.conf and your AD infrastructure will need to be configured accordingly...so if you look at my output...there is a AD group called "Linux Admins" who have sudo powers since that is apart of their role in the organization and then regular old unprivileged "domain users" which is a default AD group that is already existing [11:20] kyentei: sssd.conf and your AD infrastructure will need to be configured accordingly...so if you look at my output...there is a AD group called "Linux Admins" who have sudo powers since that is apart of their role in the organization and then regular old unprivileged "domain users" which is a default AD group that is already existing [11:20] sorry typo [11:20] No worries, caught it [11:21] kyentei: so it will involve some planning with your enterprise admins [11:21] kyentei: if you are the Linux admin then you will make sssd.conf in accordance with the AD OU/membership structure [11:22] tatertots: I was playing around with the config earlier. Have restored it to what we have in salt and the output of the realm list command is a tad different [11:23] kyentei: show me your output...or just tell me what the "diff" is [11:23] tatertots: https://askubuntu.com/questions/1342185/joining-ad-works-but-how-to-appear-in-the-winevent-security-log [11:23] tatertots: Not sure if I'm allowed to link there. If not, sorry. [11:26] kyentei: permitted-groups: linvdi_allowed [11:26] kyentei: with that policy ...do you see logs when a non member attempts to log into ubuntu client PC's? [11:27] tatertots: That setting is just to configure whether or not login to the Ubuntu client is allowed, right? [11:27] kyentei: correct...it'll check AD membership and permit or deny based on it [11:27] tatertots: I'm not sure if we see logs if users without that group are attempting to log in. I just know they can't [11:28] tatertots: But the issue isn't logging in to the Ubuntu machine itself. that works [11:28] kyentei: i understand...you want to see more verbose logging [11:28] kyentei: a more granular policy should achieve this to some degree [11:30] tatertots: I'm not entirely sure what you mean with granular policy. A policy set on the linvdi_allowed group in AD? [11:32] Sorry. I'm "just" a linux admin - sort of doing the Ubuntu desktops on the side at this company, and have next to zero experience with windows related stuff. [11:33] kyentei: i understand....test it out by attempting to log in with a non member (it should fail as per your policy) and see if there are events [11:36] kyentei: also ..you mentioned seeing different results from salt...are you looking at the same Ubuntu PC (curious as the enterprise admins could have different policy for different Ubuntu systems by design) [11:38] kyentei: for example...in my organization any Ubuntu that is NOT a LTS/long term support has a more strict log in ..regular users aren't even allowed to touch non LTS systems [11:39] kyentei: the linux admins on the other hand can do what they will [11:51] tatertots: Narrowing it down ... Looks like it's "just not being logged" because of DC policies [11:52] tatertots: The thing about salt I mentioned is that we maintain sssd.conf in there, but I wanted to apply manual changes. [11:55] is there a way to completely remove systemd-resolved? [11:55] it breaks dns lookup [11:58] tatertots: So I overwrote mine - which resulted in the realm-joined policy - then I reverted to what we configure in salt [12:03] gordonjcp: you can replace the /etc/resolv.conf symlink with a plain file that lists your preferred dns servers === bsmith093 is now known as Guest5261 [12:09] kyentei: that makes sense...rare to see or justify non explicit policies in production or even test environments ...ideally it's scope is narrowed down to AD OU group memberships [12:10] tatertots: Seems I'm not the first person to run into this issue though: https://www.reddit.com/r/fortinet/comments/8qmj3q/fsso_with_linux_sssd/ [12:13] kyentei: i notice that reddit is fortinet related...and those guys can sometimes use the DHCP in the router/fortigate instead of Windows (in which case i'd expect them to run into those problems) [12:13] kyentei: are you a fortigate shop or environment? [12:14] tatertots: Yeah, FSSO is Fortigate SSO. [12:14] kyentei: is the fortigate handling your DHCP? [12:14] tatertots: I'm not entirely sure. It's a huge government agency and I'm not entirely sure what all teams have opted to use. [12:14] Does anyone know where actually the files in /etc/skel are located in the source code? [12:15] tatertots: I think infoblox handles DHCP. [12:15] tatertots: afaik, they're not part of fortinet [12:15] kyentei: i'm a fortigate shop too...but i don't allow fortigate to handle DHCP here [12:17] tatertots: Do you use FSSO? [12:17] kyentei: no [12:18] tatertots: Lucky :-p they offer very little documentation as well. [12:19] berkant: a quick dpkg -S reveals that the files in there are all installed by the bash package. So I reckon that'd be the bash source package. [12:19] kyentei: FSSO was under consideration for a bit but after the enterprise team combed over it with a fine tooth comb it was decided it will not be used [12:21] tatertots: I guess if it works, it works well.. [12:22] Aha, thanks, just found it under `~doko/+junk/pkg-bash-debian`. [12:23] tatertots: Maybe you've got some experience with this. I recently upgraded all VDIs from 18.04 to 20.04 and since then login/unlocking takes about 30 seconds due to group fetching. It's a lot of groups, but GDM wasn't *this* slow to unlock on 18.04 [12:26] tatertots: Did you see an increase in login/unlock times between 18.04 and 20.04 on your AD connected machines? [12:41] kyentei: i can't say i observed any increase in login/unlock times, I have a handful of Ubuntu's that went from 16.x>18>20.x and have had pretty consistent, predictable behaviors as far as the AD infrastructure log in are concerned [12:42] tatertots: Alright, that's good to hear :-) Perhaps it's the abundance of AD groups we're in... [12:43] update to yesterday's problem of mine: it seems culprit might be virtualbox guest additions + pulseaudio [13:58] how can i get better cpu perforamnce on ubuntu 2004? [13:59] i have a laptop with intel 6200u cpu, 2 core, 4 threads, and in idle i see 7 to 30% cpu usage across all cores [14:00] when i load youtube in firefox it often hits 100% and starts stalling and when playing a video cpu usage sits at 50% [14:01] is this normal? do i need a 4 core 8 thread cpu to run ubuntu desktop effectively on a laptop? [14:06] Guest16: It's my understanding that Firefox lacks hardware decode for videos, so a CPU spike while playing a video would be expected. I'm not sure what's happening with your idle CPU usage. [14:08] ok i will try chrome and see if it spikes too, although i think it too sits around 50% while playing a video. [14:09] 50 isn't 100 🙂 [14:11] now i open firefox with just a blank start page and click on the hamburger menu thing on the right and move my cursor down to get to "help" and "about firefox" and the selection highlight is having a hard time keeping up with my mouse movement as i'm hovering over the menu options. [14:12] it's the same experience as when your battery is dry, but my indicator says 52% left on battery [14:12] (not playing anything now, just 1 terminal window open after updating, and navigating in menus in firefox) [14:13] firefox 88.0.1 [14:13] well, this sounds more like the system is either swapping to disk because it is out of ram, your CPU being overloaded or your graphics driver having switched to full software rendering [14:14] Guest16: do you have the proper nvidia drivers installed? [14:14] true, ogra has a point: how much ram installed? [14:15] 4 gb + 8 gb so 12 gb ram installed, about 14% in use [14:16] is that enough to get by? [14:16] I wonder if that is nearly enough for firefox these days :-( === petris_ is now known as petris [14:18] Guest16, 4GB and 8GB ? do you mean you have two different RAM bars in your mainboard ? (are you sure they use the same speed and frequency ?) [14:19] (or did you mean 4GB ram and 8GB swap) [14:19] kuudes much do i need for firefox then? [14:20] sorry, I am not an expert, someone else will know it better [14:20] would #firefox work best for this? [14:21] yes 12 gb ram installed in total ( 4 gb soldered, 8 gb ram stick installed by me (corsair bla bla) when i swapped out the 4 gb that was in the slot from factory) [14:21] and you made sure they use the same speed specs ? [14:22] i am 96% sure they are 100% same spec, i remember chasing them down and ordering from amazon to match up [14:22] k [14:22] any nvidia or radeon card involved here ? [14:23] yes there is some nvidia gpu in here, plus the intel gpu [14:23] aha ... might be a graphics driver issue then [14:25] about system page says i have "NV118 / Mesa Intel HD Graphics 520" [14:25] is that of any help? [14:25] Mesa [14:26] that points to software rendering [14:26] ah ok [14:27] looking at display section now and i see fractional scaling is enabled, could that be an issue? [14:27] * ogra isnt a graphics expert but i'd start looking in the third party drivers tab in the software settings first [14:27] oh for sure [14:27] "may increase power usage, lower speed, or reduce display sharpness" [14:28] fractional scaling and a driver doing software rendering means it will eat CPU for *all* graphical operations [14:29] Guest16: open the "software & updates" application. Then go to the "Additional Drivers" tab [14:30] i disabled fractional scaling and my cpu idle dropped from 6 to 30%, to now 1 to 17% but now i can't read text from a distance [14:31] you could install gnome-tweaks and simply scale via font size [14:31] :/ [14:31] not ideal but it wont eat your CPU [14:31] Guest16: have you looked at which drivers you have installed???? [14:31] but first look for a proper driver like leftyfb suggested [14:33] ok i see "GM108M [GeForce 940M]" and under there are radio buttons and "Using x.Org x server" is selected [14:33] Guest16: step#1 put the scaling back to what it was. step #2 open the "software & updates" application. Then go to the "Additional Drivers" tab and either take a screenshot (preferred) or tell us what driver is selected and which is available [14:33] Guest16: you only have 2 options? [14:34] ok let me first go back and enable fractional scaling like you said (no there are like 5 or 6 different options in there) [14:35] hi [14:35] well, you want one of the proprietary nvidia drivers most likely [14:35] Guest16: a screenshot would be best. Though, whichever is "recommended" is .... recommended [14:37] if you want better battery life you can start tunning Linux Virtual Memory subsystem following oracle's hint at the end of /etc/sysctl.conf : [14:37] vm.dirty_background_bytes = 10485760 [14:37] vm.dirty_background_ratio = 5 [14:37] vm.dirty_ratio = 40 [14:37] vm.dirty_expire_centisecs=1000 [14:37] isapgswell: can we help you with something? [14:37] isapgswell: also, please use pastebin [14:37] i increased my battery life by almost 1 hour [14:38] leftyfb ok [14:38] this is great. you should blog about it [14:39] please help ubuntu testing these virtual memory conf [14:40] linux uses up to 20% of ram memory to virtual memory subsystem [14:41] with that hint battery discharge slowly [14:42] do not forget to run: [14:42] sudo sysctl -p [14:42] i have enabled fractional scaling and taken a screenshot of "additional drivers" tab here: https://i.imgur.com/sV8hdzy.png [14:43] rebooting is better [14:43] Guest16, click the top one [14:43] and then the "Apply Changes" button [14:43] (you will likely need to reboot after this) [14:44] ok (i need to reboot anyway) [14:45] applying changes... it's taking time... [14:45] it downloads the driver and compiles bits of ti on the fly [14:45] *it [14:45] ah ok i see the bar now [14:46] indeed that takes a little [14:46] that sounded weird... "i see the bar now" :) [14:46] hello world, may i know the topic of discussion here plz? [14:46] Guest16, have a beer for me too [14:46] 🙂 [14:47] :D 🍻 [14:47] akupedia: this is a support channel. Feel free to chat in #ubuntu-offtopic or #ubuntu-discuss [14:52] reading the uefi secure boot message... is it better to disable uefi before installing third party driver? [14:53] 🍺 [14:56] the install would more then likely fail if you do NOT disable secure boot [14:57] i entered a password for uefi, rebooted and now i have a super small blue window called "perform mok management" with menu options to enroll key, reboot and so on. looks like bios screen but covers only upper left 1/4 of my screen. is this expected? [14:57] if you didn't disable secure boot or can't figure out how to...that is expected [14:59] Guest16, https://wiki.ubuntu.com/UEFI/SecureBoot/DKMS [15:02] yes it looked something like that, i selected to enroll key, then it asked for password, i typed in, then the reboot option, and now i'm in ubuntu and it looks like it's enabled [15:02] https://i.imgur.com/DFcYk9b.png [15:03] awesome [15:03] your graphics behaviour should be a lot better now i'd expect [15:04] idling at 0 to 5% looking good already [15:04] yep [15:11] indeed, i'm wathcing a youtube video in firefox and cpu usage goes more up and down now instead of "sitting" at 50% point, it's more like 13 to 25% now and i can hear a small turbine now :) [15:12] that sounds about normal [15:12] i need more time to test everything but this is looking much better already, thank you! [15:12] welcome 🙂 [15:16] one odd thing about this now is that fractional scaling is disabled, and if i enable it and set it to 125% and apply, far right and bottom part of desktop goes out of view so the dock and menu button become invisible but i can see the trash bin [15:18] should i not be able to set scaling with this new driver now? i noticed there is a new app called "nvidia x server settings" maybe i need to set it in there instead? i seem to recall a similar problem previously when enabling fractional scaling. maybe i just need one more reboot+ [15:24] it scales normally at 200% (with or without fractional scaling enabled) [15:27] Guest16, https://carmelosantana.com/fix-fractional-scaling-4279/ === genii-core is now known as genii [15:35] so i can set the PRIME profile to on-demand for a temp fix? and for a permanent fix i need to install more upstream drivers from that ppa? [15:37] Guest16, you can try , but if you do it, first purge the ones you installed [15:40] Is this the new channel? Also, I am unable to connect to the Internet with my QEMU guest system (curl, w3m, Firefox, etc. is not working, while the Terminal displays that the Internet is turned on). [15:43] maxFlexGuest: yes it is [15:44] I am also unable to pinpoint what exactly is causing network trouble (I am on Linux 5.8.0-49 lowlatency to try to connect, since 5.8.0-50 lowlatency couldn't work) [15:48] dpkg -l | grep -i nvidia-driver reveals it has been upgraded from 460.73 to 460.80 and yes the "on-demand" mode did fix scaling [15:48] maxFlexGuest: you'd have to analyze your network configuration, a kernel by itself isn't going to alter and or fix a misconfigured network [15:49] I tried ifconfig, but the network is still misconfigured. [15:51] The installation ISO is also unable to connect to the internet (Ubuntu Studio version 20.04 LTS) [15:54] ioria and ogra thank you both for your help! cpu/gpu performance improved and fractional scaling is now fixed. beer is on me :D 🍺🍺🍺 [15:54] Guest16, glad to hear [15:54] 👍 [15:55] maxFlexGuest: do you have access to the system right now? [15:56] leftyfb is welcome too ;) 🍺 [16:03] Yes, I have access to the FULL system. [16:59] what is the right way to && sudo-required commands, eg. sudo apt update && sudo apt upgrade [17:02] explodes: this will work, yes === ano is now known as Guest9963 [17:31] coucou [17:32] test [17:32] failed [17:33] quel est le nom du service qui tourne en arrière plan pour xchat ? [17:36] Guest9963, there is no service associated with xchat as far as i know [17:36] Thanks ioria ! [17:36] ok === _erlon_ is now known as erlon === thessag is now known as thegass === Guest9963 is now known as ano === ano is now known as Guest9205 === Guest9205 is now known as ano788 [18:03] clear [18:09] xchat? That isn't developed anymore for about 9 years now. I wonder why people still bother with it when it's dead. [18:09] it was revived recently i think? [18:12] probably fir the same reason mamy ppl even is not developed for years i know some ppl still use bitchX [18:14] for example hexchat is not developed anymore but its kept upgraded i think same happen with xchat [18:16] Outside of security concerns, IRC doesn't change enough to really require much updates to a client once it is stable [18:16] Linux ano-Inspiron-5485 5.8.0-53-generic #60~20.04.1-Ubuntu SMP Thu May 6 09:52:46 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux [18:16] Bureau [18:16] Documents [18:16] Images [18:16] Modèles [18:16] Musique [18:16] Public [18:17] !paste | ano788 [18:17] ano788: For posting multi-line texts into the channel, please use https://paste.ubuntu.com | To post !screenshots use https://imgur.com/ !pastebinit to paste directly from command line | Make sure you give us the URL for your paste - see also the channel topic. [18:18] lotus|NUC: xchat? No. It's succesor is Hexchat and that one has been around for years now. [18:30] savez vous comment installer OTR pour xchat ? [18:36] ano788: though all languages are welcome we prefer to use english here [18:36] @ano788 - try this plugin [18:36] sorry, do you know how to install OTR for xchat [18:36] https://launchpad.net/ubuntu/precise/+package/xchat-otr [18:37] ano788: xchat is dead [18:37] use hexchat instead [18:37] or Kongregate [18:37] sorry- Konversation [18:37] as i said earlier, xchat hasn't been developed for 9 years now [18:37] spartanturtle: or quassel [18:39] Personally I found Konversation the easiest to just setup and get going, but hexchat, Konversation and Quassel are all valid alternatives [18:40] agreed [18:40] i have dependencies to install [18:41] but i don't know which dependencies i have to install [18:43] ano788: did you read the above? [18:43] if it's still about xchat..... [18:43] given that it hasnt been actively developed for 9 years.. those dependancies probably havent been tracked in YEARS. it would be better to move to one of the alternatives mentioned [18:44] what's the status of ubuntu channel on freenode? [18:44] @mekaneck yes i've seen it's some libraries who i haven't [18:45] according to the blog, the only official channel is here on Libera Chat [18:45] Guest16: that offtopic chatter [18:45] that/that's [18:46] !ot [18:46] #ubuntu is the Ubuntu support channel, for all Ubuntu-related support questions. Please register with NickServ (see /msg ubottu !register) and use #ubuntu-offtopic for other topics (though our !guidelines apply there too). Thanks! [20:27] hello [20:28] what is better supported on ubuntu zfs or btrfs? [20:28] also i forgot. where is channel for ubuntu on arm? if there is any ;) === ubot3 is now known as ubottu === ubottu is now known as ubot3 === ubot3 is now known as ubottu [20:29] i can't say for sure, heap, but I use an encrypted ZFS root and boot, and it's been great for me. haven't tried btrfs yet. [20:30] signofzeta: wou! but i assume not on arm? [20:30] heap: well, no, on Intel. but the ZFS part seems mature enough. [20:31] signofzeta: i see; bc on these arms u have to have /boot on vfat [20:31] but thats okay / i can have then rest on ZFS [20:31] for a nice backup/restore; question is how to manage MBR; [20:31] you need that for UEFI boot on PC's, too. the rest is encrypted. [20:31] the only cleartext stuff is the bare minimum needed to get a ZFS driver into memory. [20:33] signofzeta: is that zfs in kernel or userspace? [20:34] heap: good question. i'm not sure. with / on ZFS, though, I imagine it's in-kernel since it's in the initrd. [20:42] question is RAM; if i run ZFS on 16-32GB sd card. .. i asume not a big ram is needed? [20:43] ZFS doesn't *need* much memory, but it will cache aggressively if it can. the exception is if you use deduplication, which is incredibly memory-intensive. [20:43] but if you're getting into dedupe, ARC's, L2ARC's, and SLOG's, you might want to pop over to #zfs -- assuming that channel even exists, lol [21:10] hello an application installation script is failing, i suspect because COMPILER_PATH and LIBRARY_PATH are picking up the ~/../anaconda3/bin instead of the system bin even though I've changed my $PATH variable [21:18] Hello people. How can I know the upgrade to 21.04 is complete? Ubuntu server upgrade from 20.10. [21:18] I'm trying to upgrade my #mastodon instance but when I do `yarn install` it gives me this error: https://paste.ubuntu.com/p/GhRmFDvDpf/ [21:18] Mastodon people told me that the upgrade is incomplete. [21:19] And I should ask Ubuntu people to make the upgrade (distro upgrade) complete. [21:20] Can you please help? [21:21] Nevermind. I checked and the lxc is still on Ubuntu 20.04 [21:21] Damn it. [21:21] arh: heh, so you updated the host, but not the specific guest? [21:21] sarnold, yeah. [21:22] arh: d'oh. good thin gyou found the real situation quick ;) it wouldn't have been fast to find that through the usual debugging, hehe [21:22]