=== joe is now known as Guest5790 [00:41] I'm using ebtables on Ubuntu 20.04 with KVM VMs but I'm finding that the more ebtables rules I have, the slower the UDP single threaded network performance is for the VPS that has its ebtables rule at the bottom of the ebtables rule list. This same issue is no where near as visible on Redhat based OS's. I've already tested HWE and 5.14 kernels with no change. Anyone any idea of a potential solution for this? [00:41] Hello, I came here reporting about one core being highly used on my laptop even with nothing open. It's still happening seemingly randomly, but I was able to get some screenshots from htop and ps about CPU usage. However I still can't see what's causing it. https://drive.google.com/drive/folders/1mvu7ClCOKziSCF3niXU-nCIzdfn83Tu6?usp=sharing Here are the screenshots. Thoughts on what could be causing it? [00:42] cuddylier: pastebin the ubuntu rules and rhel rules and maybe someone will have suggestions? [00:43] i would use nftables, it is in the kernel [00:45] computerjoe314: try running sudo perf top -- it might help spot whatever the machine is doing [00:45] computerjoe314, kworker ? [00:47] I use software that has ebtables functionality built into it, I would love to use nftables instead though. It's just odd the issue only appears with Ubuntu, at least in such an extreme way. [00:48] Here is an excerpt from the ebtables rules: https://pastebin.ubuntu.com/p/Tm6M6GMRnN/plain/ [00:48] The interface viifv9452 having the slowest speeds and viifv9446 the fastest. [00:49] maybe it is not the rules, but the order of rules.. [00:49] Single threaded UDP throughput is 1.79Gb/s to a remote location with 1 set of ebtables rules for one VM and when you scale that to 60 VMs worth of ebtables rules, the network performance drops to 500Mb/s for the last interface in the rule list. [00:50] no idea with ebtables [00:50] computerjoe314: try running sudo perf top -- it might help spot whatever the machine is doing [00:50] I know that nftables uses a map so avoids this issue of going down ebtables rules sequentially. === ur5us_ is now known as ur5us [00:57] can 2 nvidia gpu cards of different model be run and configured with 2 different nvidia drivers? [00:57] ball: with ubuntu live boot - all 4 monitors coming out of these 2 gpu's work perfectly fine - but after installing ubuntu, only 2 works [00:58] https://www.irccloud.com/pastebin/vpsgzov7/ [00:59] I think I've heard two different nvidia drivers at once just doesn't work [01:01] Do I have to install AMDGPU pro and how can I install the open source drivers? [01:02] vqueiroz: What is your Ubuntu version? [01:02] VMGuy23: Ubuntu 20.04 - I'm new on Ubuntu world =) [01:02] vqueiroz: 'ubuntu-drivers install' will probably do the right thing [01:02] I'm coming from debian [01:02] 11 [01:03] open amdgpu is installed, i guess. get the pro from their site? [01:03] sarnold: i don't think ubuntu-drivers will install the open source amdgpu nor the proprietary amdgpu-pro overlay. [01:03] vqueiroz: download the latest drivers from AMD website and install them with the script [01:03] Maybe this >> https://amdgpu-install.readthedocs.io/en/latest/install-installing.html#installing-the-all-open-variant [01:03] But I'm getting some glitches. I've installed the pro version. Might be related? [01:03] tomreyn: yeah the closed stuff will probably require going to their website, but I haven't heard what their closed stuff offers [01:04] vqueiroz: no, you do not need to install amdgpu-pro most of the time. and the default and fine amdgpu will already be installed and working [01:04] Also I had some OpenGL code that was working on my Debian 11 install with OS amdgpu driver and now glGenTextures returns 0 =( [01:04] install the All-Open variant ??? weird, you should have it already [01:05] Why is focal on archive? e_e [01:05] br.archive.... ? [01:07] sarnold: i'm not sure exactly either about the features. i think -pro is primarily targetted at 'professional use' (and there primarily with their 'professinal' hardware models. but all i've read about it is that -pro causes problems whenever people miss to read the compatibility notes (which usually rule out HWE kernels, or only provide compatibility for those with much delay) [01:09] vqueiroz: i think we can say that unless you *know* that you need -pro, you don't need, and moreover don't want it. [01:11] tomreyn: or perhaps some way to make sure they get a few bucks for their efforts from people who Really Care.. [01:13] sarnold: -pro is also free (as in beer) for all i know, if that's what you mean? [01:13] tomreyn: oh :) I figured it was expensive $$$ for commercial users [01:14] the 'professional' hardware is much more expensive (while not actually adding *that* much), that would be the 'supporter ticket' [01:29] `btrfs fi show` - could someone clarify? I added a 2T disk to a RAID1 and I think I should have > 1T free space, but ... [01:29] $ sudo btrfs fi show /home/dev/Data [01:29] Label: 'data' uuid: 091af054-b597-4215-b7bc-dacf1158a417 [01:29] Total devices 3 FS bytes used 1.62TiB [01:29] devid 1 size 1.82TiB used 1.08TiB path /dev/mapper/luks.data.1 [01:29] devid 2 size 1.82TiB used 1.08TiB path /dev/mapper/luks.data.2 [01:29] !paste [01:29] 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. [01:37] How can I be sure amdgpu is the driver being used? I was surprised there is no wiki on Ubuntu to tell me to mess with xorg config files [01:38] I uninstalled amdgpu and now I can render my stuff with OpenGL. [01:38] vqueiroz: xorg configuration is usually generated on the fly as xorg starts up. glxinfo should tell which renderer is active. [01:38] spci -nn | grep -E 'VGA|Display' or sudo lshw -c video [01:39] lspci ^^ [01:41] https://termbin.com/aiyd [01:42] tomreyn: Does that make any sense? ^ amdgpu [01:43] Ok so, trying again ... `btrfs fi show` - could someone clarify? I added a 2T disk to a RAID1 and I think I should have > 1T free space, but ... [01:44] http://sprunge.us/61hB0s [01:45] vqueiroz: i think so. glxinfo -B is what you want, i think, it should include the "Accelerated:" info also, without outputting too much. but this looks liek you're using amdgpu now. [01:46] i have a desktop with ubuntu 20.04 i cannot update any packages because the system was interrupted while updating a package. Every time I try to run dpkg --configure -a to fix it, my pc freezes up [01:46] tomreyn: It says accelerated, yes. [01:47] Dedicated video memory: 8192 MB [01:47] Total available memory: 24148 MB [01:47] How come? [01:47] some ram is dedicated to the gpu [01:47] 8gb in your case [01:47] devslash: system freezes upon configuring package? that's weird. i'd say boot to recovery and do a file system check [01:47] how do i boot to recovery [01:48] !recovery [01:48] If your system fails to boot normally, it may be useful to boot it into recovery mode. For instructions, see https://wiki.ubuntu.com/RecoveryMode [01:48] apt install -f would fix such break [01:49] vqueiroz: first line is video ram (actually located on the graphics card), second line is system memory (on the mainboard). [01:49] Ok, LL I want `btrfs fi usage` rather than show... [01:50] But that's strange too, because it has this line: [01:51] Data,RAID1: Size:1.63TiB, Used:1.63TiB (99.96%) [01:51] How do I interpret that, given that the esimated free space is 1.09T? [01:51] Free (estimated): 1.09TiB (min: 1.09TiB) [01:54] Diagon: i'm pretty clueless about btrfs, but here's a guess: "Data,RAID1: Size:1.63TiB" is the size of the RAID-1 device, "Used:1.63TiB (99.96%)" is the partitioned space on it, "Free (estimated): 1.09TiB (min: 1.09TiB)" is the available unallocated space on the existing file system. [01:56] i cannot do fsck in recovery mode [01:57] try; apt install -f [01:57] ok [01:58] it wont let me [01:58] says i have to run dpkg --configure -a [01:58] Then run dpkg --configure -a [01:58] so I tried that command and even in recovery mode, my pc locks up [01:59] pc locks up .. for how long? [01:59] permanently [02:00] i mean i waited a few minutes and caps lock doesnt light up on the kb [02:00] so it seems to be deads [02:00] dead [02:00] ouch [02:00] yes, oke, no caps light is an indicator for that [02:00] if i dont update i think it will continue to work [02:00] backup data and reinstall [02:00] but im permanetly stuck [02:01] i just reinstalled recently... [02:01] this is so fcking annoying... [02:01] do a memtest86 run, to exclude bad ram [02:02] i already replaced the ram with known good ram [02:02] same issue [02:02] bought a new cpu cooler [02:02] because i assumed that it was bad... [02:03] if you're at a text console now, you might get better diagnostics when it locks up [02:03] no [02:03] theres no output [02:03] its just frozen [02:03] and i have to do a hard reset [02:03] what package did it try to configure immediately before it locked? [02:03] it said building modulews [02:04] modules [02:04] which module? [02:04] do you *need* it? [02:04] i dont know [02:05] hang on ill send a screenshot [02:05] devslash: is ubuntu freshly installed on this system? has the system been locking up before? [02:06] fairly recently i dont think its been more than 1-2 months [02:06] it started recently [02:07] i'd agree on the memtest, though if you already replaced ram, it's more likely something else. but the memtest might still show errors, telling you it's definitely a hardware problem [02:07] or it might just lock up there, too, which would also say that [02:07] imgur.com/a/9ldrWBW [02:08] 404 [02:08] imgur.com/a/9IdrWBW [02:08] sorry I and l look the same when I is in lowercase [02:09] a common problem, unless you use a proper monotype font [02:09] its not my issue [02:09] its imgur [02:09] they generated the link [02:10] or copy-paste the text [02:10] cant [02:10] its way too difficult to type out [02:10] thats why i screenshotted it [02:11] I meant the imgur url ;) [02:11] i did type it [02:11] thats why I misspelled it [02:11] wish they wouldnt use l and lowercase I [02:12] is this ubuntu 21.10? [02:12] no [02:12] 21.04 [02:13] oh 20.04 LTS you said [02:13] sorry typo [02:13] i cant do anything without running into this issue where it wants me to run dpkg --configure -a [02:14] i have virtualbox installed. i dont know for sure that its the cause but i do know it installs kernel modules [02:14] if i could uninstall it, MAYBE that would fix it but i dont know.... [02:15] what about sudo dpkg --configure -a [02:15] im in a recovery [02:15] so i dont need to use sudo and as i mentioned htat locks up my pc [02:16] devslash: hah, if you've got the virtualbox *guest* stuff installed that is seriously unhappy. I can't remember now what it does, but this sounds familiar [02:16] im not sure that thats the cause.... [02:17] yeah. I wish the dkms output said what stupid module it was trying to build. it shouldn't hang when building modules, that feels like bad hardware territory. but still. [02:17] im only speculating... because when you install it, it compiles modules for your specific kernel version [02:17] i dunno im close to dumping this thing in the trash [02:17] so, next boot -- mv /etc/kernel/postinst.d/dkms to /root/dkms or something, and then try the dpkg --configure -a again [02:17] b==thats the command [02:17] oh you mean boot normally [02:18] then try that ? [02:18] or in recovery [02:18] i mounted the os rw in recovery [02:18] tomreyn - thanks, but ... I'm not sure I understand that. Here's the full output. [02:18] http://sprunge.us/oclnDC [02:18] recovery is nice, it makes it easy to get these photos of errors :) [02:18] so id assume i can do this from recovery too right sarnold [02:18] devslash: that's my hope, yeah [02:19] ok i moved dkms and im trying again [02:19] crossing my fingers [02:19] holy shitballs [02:20] it finished this time [02:20] hey all [02:20] question, hhow to remove a faultiy repository ? [02:20] find it in /etc/apt/source.list.d === Syntax-0 is now known as Syntax [02:20] i think that thats the correct path or very close to it [02:21] is it possible to run touch apps inside a vanilla install of Ubuntu? If someone could point me to a URL, I'd be grateful [02:21] devslash: sweet [02:22] you're a freaking genius [02:22] Diagon: sorry, i'm really not familiar enough with btrfs to comment. i do recommend that you read up much about btrfs and which scenarios are known safe and reliable with it. [02:22] devslash: so.. I'm not sure what the next steps are -- but it'd be worthwhile to try to figure it out [02:22] names in that derectory doesnt correspond devslash [02:22] well at least now i know what to do in this scenario [02:22] K, thanks... [02:22] devslash: you'll either need those dkms modules for something important -- you weren't doing it for fun, right? :) -- so maybe you need to troubleshoot why it fails.. or maybe it's enough to remove whatever module you're building [02:22] its for my graphics drivers [02:22] right now it boots in low res [02:23] but w/e [02:23] bummer :( [02:23] better than locking up [02:23] i can hopefully fix it [02:23] i might update to 21.10 [02:24] or maybe install the HWE kernel on your system, if the newer kernels may have support for your cards in it already, and not require the dkms packages [02:25] i dunno. my card is an nvidia gtx750ti [02:25] hmm, lomiri might be what I'm looking for [02:26] devslash is already using linux-image-generic-hwe-20.04, but not the latest version [02:29] question, hhow to remove a faultiy repository ? did you add a PPA? [02:29] no idea what a PPA is [02:29] VIA: Personal Package Archive [02:29] ok so spoke too soon... [02:30] Someone maintains it by themself [02:30] oh [02:30] normally [02:30] hi guys [02:30] sarnold, i did apt upgrade and it got stuck at 80% [02:30] VIA give us some output of faulty repository? in a pastebin [02:30] devslash: aw man :( [02:30] when I say stuck I mean the entire pc freezes [02:31] =( [02:31] I think memtest86 .. or memtest86+, depending [02:32] I'm going to try and see if i can luck out and borrow a processor from someone at my work. I really hope that someone has a processor that matches my socket [02:40] Hello [02:43] Beginner level to Linux. General experience and basically know my way around a terminal [02:44] jay20: just ask your question :0 [02:45] I got two raspberry pis and two thermal printers that connect on USB. [02:46] going to use print node from woo commerce, how do I print two printers at once. Same document [02:47] Any ideas on how to print the same document to two printers [02:53] Anyone here got an idea? [02:53] jay20: print it to one, then to the other [02:55] It’ll have to be an automatic type of thing where anything printed is printed on both A and B automatically === VMGuy23 is now known as hh === hh is now known as VMGuy23 [03:07] why did canonical make multipass [03:07] what does it achieve? [03:07] easy ubuntu vm [03:07] VMGuy23: isn't qemu/kvm easy vm? [03:08] i'm trying to see why a whole virtualisation system was created, at first I thought it was primarily for snapcraft [03:08] i'm still struggling to understand the way that snapcraft really works [03:08] quazimodo: true but like single command, no download, ubuntu vm in a few mins [03:08] VMGuy23: who's the target? [03:09] i would expect devops engineers & sysadmns to be able to handle server virtualisation using qemu/kvm as business as usual [03:09] so i presume it's not for them? [03:09] quazimodo: snap devs who compile their apps in the container so it will work in the snap no matter what the ubuntu ver [03:09] ok so it /is/ primarily for snapcraft? [03:09] i think so [03:10] ok understood [03:10] i tried snapcraft once [03:10] do you know if the snap itself launches a vm? [03:10] i think the snap does [03:10] but a very light one [03:10] ok so the build image could be some much heavier image? [03:10] this sounds like a multipart docker build [03:11] build in heavy image, copy assets into the light deployable [03:11] i just wish there were c4 diagrams of these things :( [03:12] i dont know that much but snaps use core ubuntu images with added dependencies for building and running and i think multipass might use the same very light vm [03:12] that sentence confuses me [03:13] snaps use core ubuntu images with added deps, they're running those images within multipass right? [03:14] i think so [04:31] Hello!! I have ubuntu 20.04 and my swap memory is at a certain partition, not in a file (> swapon -s ===OUTPUTS==> /dev/dm-2 partition 1003516 1003516 -2); How can I increase the size of swap memory? Can I create an additional swap file + swap partition OR can I use either a partition OR a file (not both) for swap? [04:34] Hi [04:36] My computer froze on an upgrade and I can no longer start it, unfortunately it is encrypted, is there a way for me to reinstall it with a usb (I'm on a live now) and repair it? [04:38] the answer is yeah almost always, but if you have important data on there you should consider making a disk image [04:39] there's a sizable risk that in trying to fix things you lose your current things [04:42] I need a few files from it the hdd, so perhaps I could just get those and then wipe it [04:42] ah that ought to be straight forward [04:42] back those up first, then decide if you want to fix or wipe [04:42] SupremeKai: what on earth [04:43] why you don't just have a sufficiently large swap partition? [04:43] is there some functional requirement? [04:43] depending on your partition scheme you could use a tool like gparted to increase your swap space size [04:44] i'm not sure if you can ever have multiple swap spaces/files combination [04:44] add a swap file, ought to work :) [04:45] quazimodo, I tried but I don't have unallocated free space available to add to the partition [04:45] rk4 i dont really get how to mount it and access it [04:45] rk4, can I have both? [04:45] SupremeKai: your ssd is too full? [04:45] how can yo uhave space for a swap file? [04:46] o my bad, perhaps you can unallocated some space [04:46] gparted lets you resize partitions [04:46] quazimodo, it is a bit, but the main concern is that I don't know exactly how to do it with gparted [04:46] SupremeKai: i believe so yeah [04:46] I already tried that path [04:46] right [04:46] you can even give them priorities [04:47] How can I assign priorities? [04:47] rk4 i used the gui "disks" thing, entered the password for the partition, but i don't get how to get the files [04:47] quazimodo, to tell you the truth I'm kind of fearful of messing up the partitions nbeucase I need to get some unallocated space [04:48] SupremeKai: if your data isn't backed up you're in a worse situation tbh [04:48] ssd & hdd fail all the time [04:48] get your important stuff backed up man [04:48] Friendlyfox: unsure how to use the guis, i use cryptosetup luksOpen, then mount [04:49] rk4: lol i had to resize /everything/ using that [04:49] what a pain in the dick [04:49] but actually worked really well [04:49] Friendlyfox: possibly the automagic has already done the luksOpen for you, and even the mount [04:49] quazimodo, yup that's my problem, I don't have any backup of my stuff [04:49] had to figure out how big the encryted space is, then how big it should be, then this then that [04:49] I need to get an extra ssd [04:49] SupremeKai: dropbox it! [04:53] gosh i will never ever not do a backup again [04:53] yup :) I'll start by backin' up all this stuff :P [04:54] thank you all quazimodo rk4 Friendlyfox [05:03] rk4 yeah i dont get it to work, not sure if i messed it up [05:04] https://twitter.com/kuschku/status/1450247425852125185 ouch [05:06] "probably corrupted filesystem" [05:06] Friendlyfox: pastebing `lsblk` [05:06] [also blkid] [05:18] rk4 at part sda3 i got luks, it got root (lvm) and swap_1 (lvm) [05:19] what do I do with that in the terminal just to access the files? [05:22] `ls`, `cd` if it's mounted, if it's not you'll have to mount [05:22] I think I've done this before and just used disk and entered the password and it just "popped" up in gui [05:25] Says its unlocked on Disks [05:26] /dev/sda3 [05:26] Where do I find it? [05:29] ok so I see, the mountpoint with lsblk [05:29] but not on sda3 [05:30] what do you use to do it you said rk4 to unlock and unmount a luks encrypted part? [05:30] and mount* [05:36] normally for me it's something like cryptsetup luksOpen /dev/sdaN unencrypted; mkdir -p /mnt/unencrypted; mount /dev/mapper/unencrypted /mnt/unencrypted; cd /mnt/unencrypted; ls, approximately [06:07] fk4 thank you for your help [06:18] i guess it worked[?] [06:43] I noticed that there is no firefox process right now in my ubuntu even though firefox is running. Instead its geckomain. When did this happen? google returns very few results on geckomain [06:44] is it like that for everybody (Geckomain and not firefox)? [06:45] Which ubuntu version, and are you using .deb or .snap firefox? [06:45] For me (deb/22.04), no, it's not [06:45] alkisg 61796 4.4 5.1 5270192 833452 ? Sl Οκτ23 249:21 | _ /usr/lib/firefox/firefox [06:47] *20.04, sorry [06:50] alkisg, 18.04 using deb [06:50] bunsen: and the output of `ps faux | grep firefox` is nothing? [06:52] alkisg, it returns a bunch of lines. not sure what it is. i just ps -A | grep firefox which returned nothing [06:53] These bunch of lines are the firefox processes [06:54] You can check their pid vs the `ps -A` pid, to see how the same processes show up in ps -A [06:55] E.g. /usr/lib/firefox/firefox -contentproc.... shows up in `ps -A` as `file:// Content` [06:58] alkisg, seems like the root process i GeckoMain not firefox (also in gnome system monitor) [06:59] lemme restart firefox. maybe it crashed [07:00] nope. the command is firefox but the processname is GeckoMain. [07:01] i was asking because i was mainly curious and suspicious. used to be a firefox process but now its called GeckoMain? And google doesnt return much either [07:07] bunsen, don't google "geckomain", try "gecko mozilla" or "gecko engine". How "GeckoMain" got to be the $0 of some Firefox process I dunno but any Mozilla program is too complicated for me [07:09] rfm i know that firefox uses the gecko engine. i dont know when the process name changed from 'firefox' to 'GeckoMain'. It gave me system32.exe flashbacks thats why i was investigating [07:14] bunsen: I don't know what `ps -A` shows, but `cat /proc/61796/cmdline` still gives /usr/lib/firefox/firefox for me [07:16] It's going to depend on exactly which firefox version he has, it's just somewhere there's an exec* call with args[0] "GeckoMain" for some name reason. [07:16] quazimodo, VMGuy23-: no, snaps don't run in Multipass VMs, they may be _built_ in one, though [07:17] alkisg, i think that is the command. for me the command is still firefox. The process name is GeckoMain. Like if i run top there is GeckoMain instead of firefox. That was not how i remembered it being before. [07:17] rfm: `ps -A` shows GeckoMain here too, but I can't find GeckoMain anywhere in /proc/pid/* [07:17] I'm not sure what it maps to [07:18] quazimodo: as for "why Multipass?" qemu/kvm is used under the hood, but you can't compare `multipass launch` with the steps you need to do to download / prepare / run an image. Multipass is also cross platform [07:18] quazimodo: see https://multipass.run/ for more info [07:19] alkisg, in gnome-system-monitor the process detail shows commandline as firefox [07:19] for geckoMain [07:19] *GeckoMain [07:19] alkisg, it's just whatever was in arg[0] when the exec call was done, I think. [07:20] OK found it, /proc/pid/comm [07:20] And indeed in 12.04 it's firefox [07:20] So at some point it did change [07:22] alkisg, does ps -A|grep firefox return anything for you? [07:23] bunsen: no [07:24] alkisg, in latest 20.04? So its not just me. I guess it is a change then not some virus lol thats a relief [07:25] bunsen: yeah [07:25] thanks for your time === klausfiend1 is now known as klausfiend [10:24] Saviq: so in fact multipass is an orchestration layer around qemu/kvm [10:25] quazimodo: correct, and around other hypervisors appropriate for a given platform [10:25] perhaps ~ sort of akin to docker engine & dockerd ? [10:25] right [10:25] i _wish_ it's website just said that my god [10:25] i was thinking that it's some sort of nightmare [10:26] and then... why on earth is it called multipass? [10:26] the name implies some sort of semantic 'thing' like multiverse [10:26] confused the heeck out of me [10:28] quazimodo: qemu/kvm is quite a technical detail, most people that go to the website will have no idea what that is and get confused even more [10:28] especially if they're on Windows / macOS [10:29] As for the name, it's a pun on The Fifth Element movie, no real semantic meaning [10:29] !discuss [10:29] Want to talk about Ubuntu, but don't have a support question? /join #ubuntu-discuss for non-support Ubuntu discussion, or try #ubuntu-offtopic for general chat. Thanks! [10:30] Saviq: lol is that what it's from [10:31] i didn't catch it [10:31] corben dalas multipass [10:31] you seem authoritative, are you on the multipass project? [10:31] Yes, come along to #multipass if you're interested :) [10:32] also yeah I've run qemu/kvm macos on 20.04 laptop [10:32] it's not trivial is it === pah is now known as pa [11:09] so I see that snapcraft can run on macos [11:10] does that mean that .snap files can run on macos, or that snapcraft can build a .snap on a macos machine via multipass [11:10] ? === Rhvs is now known as Rhys === Rhys is now known as 074AAB04Z [11:35] Which of these packages http://ix.io/3D3x is the actual kernel? I want to `apt-get source` for it. === lous1F is now known as lousiF === hasdf3 is now known as hasdf === manuels2 is now known as manuels === notepad7 is now known as notepad === SmokenatorZ65 is now known as SmokenatorZ6 === toozej86 is now known as toozej8 === scjosh4 is now known as scjosh === Madatnek- is now known as Madatnek === mnepton is now known as mneptok === rvgate2 is now known as rvgate === wille1 is now known as wille === mikesmoniker_ is now known as mikesmoniker === pitninja_ is now known as pitninja === coetzeer__ is now known as coetzeer === jmcgnh_ is now known as jmcgnh [11:56] answer: linux-image-unsigned-4.4.0-148-generic === saribro is now known as swinnenb === sal9000 is now known as Guest9572 [12:17] Hey guys, can I ask for help here or do I need to join a sub-channel? [12:34] Anyone have any idea why I have globally terrible performance in proton w/ steam? I've tried tons of different Nvidia Drivers--I have dual graphics w/ a GTX 1050 and intel igpu, but I'm in NVIDIA Prime mode w/ Nvidia-SMI output reading that it is processing games:( I had the same issue in Pop and Ubuntu w/ 21.04. Weird thing is that sometimes the games run perfectly fine if they aren't in fullscreen? I've tried [12:34] lots of stuff I can list below if it helps. DXVK Full output shows like 100% GPU usage regardless of settings so something weird is definitely going on [12:42] milesthewizard, runs fine when not in fullscreen? sounds puzzling. I dont know much but i have faced the opposite issue. Had to make sure compositing was off. [12:43] Some games run fine when in windowed/Borderless, on some drivers. Really weird. [12:44] I have heard people have that issue on windows. it seems not uncommon. [12:44] Even games w/ native proton support are running badly. Thought maybe could be a WM issue will make sure I've toggled compositing [12:45] milesthewizard, only proton games ? no issue with native? [12:46] Tbf, I have tried the new satisfactory beta which has native vulkan support and had same issue [12:47] Not an expert dont know if its a deadend but may want to look into if window manager compositing is an issue. Compositing should be off when you are in fullscreen. If you can switch instantly between desktop and game (instead of 'glitching' to black for a second) that usually means compositing is ON [12:48] Oh really? Yeah that transition is seamless. [12:48] it could also mean that your pc is fast :D [12:50] lol if only that was reflected in the gaming performance :'( [12:50] but when i tried games , i could get a sense of how fast it would take to tab in and out of a game. [12:51] so for example. csgo. if compositing is off alt+tabbing in and out of the game would take a second or so. but if it is ON it would be instant (but fps would be low) [12:52] i am not sure if this will work but it is quick to try out anway. [12:54] 1. sudo apt install icewm 2. logout and login to icewm session (in ubuntu clicking the cog should reveal wm options) 3. launch steam and try a game (dont open chrome though because icewm doesnt launch gnome-keyring and your cookies will get deleted) [12:55] milesthewizard, i dont know if it will work but icewm is supposed to be a tiny non-compositing wm. [12:58] I'll reinstall 470 driver and then toggle compositing and let you know what happens! [12:58] Do I need to do via terminal or is there place in gui? idk if you remember [12:59] milesthewizard, yeah again i am not a expert. just an idea. may want to find someone more knowledgeable . [12:59] is it plain vanilla ubuntu (GNOME)? [12:59] also, are you running on X or wayland? [13:00] Yeah it's plain vanilla [13:00] X I didn't think Nvidia played nice w/ wayland [13:00] why are you reinstalling the driver? which driver is it now? [13:00] did you try with wayland? [13:01] i'm a sad panda [13:02] i thought snapcraft is installable on osx, therefore .snaps /could/ be built to work there [13:02] boy was I wrong [13:02] and now i'm a sad panda [13:02] have you tried eating a bamboo stick [13:03] ahh pandas, how are they not extinct [13:04] sudo make --happy sad_panda [13:04] Hi folks [13:18] i've had a few serious problems with snaps on ubuntu re fonts but other than that... wowee [13:18] what a breeze to build & distribute a thing eh [13:19] i'm at a loss on how to build & distribute this app in such a nice way for the osx users, the windows dudes have wsl2 ubuntu so they're covered [13:20] i dont like snaps at all. always face issues [13:21] here we come [13:21] quazimodo: feel free to discuss in #ubuntu-offtopic [13:21] appimage seems to be the simplest way of sharing apps on linux from user pov . just click to run [13:21] bunsen: ^ [13:22] leftyfb: yep my bad, kinda got away from me === Anton is now known as chatting === chatting is now known as anton [13:40] Hello === sal9000 is now known as Guest7394 === genii-core is now known as genii === pong is now known as beaver === Andrew_ is now known as AndrewYu [14:00] does do-release-upgrade also upgrade say... python3? === sal9000 is now known as Guest5751 [14:01] or do I need to do an apt command to upgrade python afterwards? [14:01] bancroft: do-release-upgrade upgrades all packages that have been updated since the version you are running to the version you are upgrading to [14:01] leftyfb: so it would update python [14:06] leftyfb: thanks [14:11] https://hexchat.github.io/ HexChat 2.16.0 is the actual version and needs an update to it [14:11] dust: ? [14:12] leftyfb, whats unclear? [14:12] dust: what is your ubuntu support question exactly? [14:12] [16:11:20] https://hexchat.github.io/ HexChat 2.16.0 is the actual version and needs an update to it... didnt u read? [14:13] !latest | dust [14:13] dust: Packages in Ubuntu may not be the latest. Ubuntu aims for stability, so "latest" may not be a good idea. Post-release updates are only considered if they are fixes for security vulnerabilities, high impact bug fixes, or unintrusive bug fixes with substantial benefit. See also !backports, !sru, and !ppa. [14:13] so update it... [14:14] dust: that's not how this works === LjL^ is now known as LjL === sal9000 is now known as Guest7068 [14:33] dust Ubuntu and Debian do not use a rolling release system. Maintainers cannot "just update" packages in Stable/LTS version. [14:39] dust, https://wiki.ubuntu.com/StableReleaseUpdates === lotuspsychje_ is now known as lotuspsychje === pi1 is now known as johnjay === beaver is now known as pong === pasiz7 is now known as pasiz === leaftype is now known as guest2 === guest2 is now known as leaftype [16:07] Hi [16:08] welcome Guest21 [16:08] thank you [16:08] i cant connect to internet by cabel [16:08] how can i solve this problem [16:09] Hello, since I upgraded to 21.10, Blender can't find my GPU [16:09] It also happened when I upgraded to 21.04 [16:10] Is there any one to anwser? [16:10] I couldn't fix it so I'm looking around and noticed I'm on X11, it didn't switch to Wayland when I upgraded [16:10] Bardon do you know how i can connect to internet by lan cabel? [16:10] On the login screen, the little gear at the bottom right corner offers two choice : Ubuntu or Kodi. What does Kodi has to do here?? [16:11] Guest21: I just plug it in and it works [16:11] For me doesnt work [16:12] i cant connet by Wifi too [16:13] hey guys...I know you can shorten the terminal prompt out put when showing the current directory by editing the bash script, but how do you do it with the xfce-4 terminal? [16:14] opening up ~/.bashrc shows no PS1 line to edit? [16:14] Guest21: Open a terminal and type "ip a". You will get several lines, look for one that looks line "eno2" [16:15] ah...thanks! [16:15] oh wait [16:15] Guest21: See if there's something that says "UP" or "DOWN" [16:15] lol [16:15] OK Thank you [16:17] i see en02 but there no up or down [16:17] i see this : Open a terminal and type "ip a". You will get several lines, look for one that looks line "eno2 [16:17] sorry i send wrong === pasiz7 is now known as pasiz === pasiz1 is now known as pasiz === dob1_ is now known as dob1 [17:46] I have a binary that I'd like to ship from ubuntu's apt repos, what do I need to do to get the ball rolling in that direction? [17:49] or maybe a better question is how to release a package on ubuntu? [17:59] reedrichards: you have a closed source application that you want to make available to all ubuntu users? [18:06] alkisg, [18:06] i'm reedrichards now [18:07] https://github.com/brevdev/brev-cli [18:07] I'm packaging this, I think I just need to package it for debian [18:07] MrFantastik: yeah I saw your chat in #debian:matrix.org [18:07] Sure, if you get it to debian it'll come to ubuntu [18:08] You'll need to do some reading and it'll take some time and you'll only be able to send new versions every two years when debian releases, but it's certainly worth it [18:11] 2 year release cycle wow [18:11] where would be the best place to look for why my ubuntu install woke from sleep and then sat on a black screen unresponsive? [18:12] had to force reboot and im not seeing much by way of logs [18:13] MrFantastik: other options include: your own apt repository or (Launchpad) PPA or opensuse OBS project, or snap [18:15] we do have statically compiled binaries we can ship too [18:15] (also, i don't think it's 2 years through debian - ubuntu takes versions from debian unstable (which then go into by-yearly ubuntu releases), and those news version can be uploaded to debian unstable whenever you like to basically. [18:15] but I think it would be a better user experience for them to be able to install with apt install brev [18:15] oh that's good then [18:25] Although most ubuntu users use the lts editions which also have a 2 year release cycle [18:25] That's not a bad thing though, it makes systems stable [18:52] software as binairy only, is beyond the open source ideology.. === Erisa43 is now known as Erisa4 [18:53] but binairies are so light and fluffy [18:54] you can eat a dozen of them and not even notice [18:54] security by obscurity ,.. [19:23] hi. is someone here with knowlege in linux forensics? [19:24] !ask | w3gi [19:24] w3gi: Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience [19:25] i am curious if someone can recommend a good tool for forensics investigations of linux systems [19:26] osforensics or belkasoft seem to be quite limited when it come to linux [19:26] autopsy get at least a timeline togehter but back triage functions [19:27] lack bot back :) [19:27] *not [19:33] w3gi, do you know of 'forensics-all' pkg ? [19:34] A person who has a Ubuntu vm wants me to help and remote via ssh. They have users set up with public/private key. What do I tell them how to give me access - I think I know but I would like confirmation. They are novices I think [19:36] this guide gives CLIET to SERVER auth, https://help.ubuntu.com/community/SSH/OpenSSH/Keys#Transfer_Client_Key_to_Host [19:36] but you want it the other way [19:37] ioria: yes but I search a tool which make my job easier [19:37] GrandPa-G: you may like this https://tmate.io/ [19:38] i can read all logs manually and do all fs stuff with TSK and so on [19:38] but that is ineffitient as hell and take forever [19:38] sarnold, I doubt they want something different than ssh [19:39] have you ever tryed a windows forensics case with a prof. tool like OSForensics or EnCase? [19:39] GrandPa-G: this is a wrapper around ssh and tmux so that they can see what you do, or give you read-only view, etc. it's worth a minute of looking :) [19:39] sarnold, not my choice, they said ssh [19:40] upload your keys to their 'vm', by letting them enable password for 1 login [19:40] oerheks, I have been trying to tell them that but they don't know how to get just one user to have a password. [19:42] oerheks, is there a simple document I can send them that tells how to just enably my password? [19:42] GrandPa-G: do you have a key on a launchpad or github account? ssh-import-id can retrieve a key from either one, it's pretty easy [19:56] take for example a case of child abuse - i don't want to dig trough 56.853 JPG files by hand. I want to apply a filter and get 300 of the 500 childporn-images with maybe 800-1000 fals positive ones [19:56] and i can find the needed evidence in minutes and not in a couple of days [19:58] i search a tool which run TSK, Photorec, and many other commands, which parse logs and collect and crossreference all files, add the browser history and many other artefacts to the pool of data and lay me all out in a convinient form that i can search and filter [19:58] a classification like that sounds like something that would take an expensive training process to prepare [19:59] I believe microsoft has such a tool, or an api, but I'd be very surprised if it's easily packaged for end use [19:59] that's why belkasoft x cost arround 5.000 USD / year and the tool is great [19:59] i have no problem to buy a commertial tool [20:00] but X for example don't get even along with LVM partition and that's why it's a great tool for windows but totally suck for linux [20:01] OSForensics has the same flaw and offer me a undelete-function, search for files with wrong extention and a apache log parser - that's maybe 1 or 2% and 98-99% i still need to do by hand [20:01] can you dd the filesystem inside to local storage? rather than work with the lvm? [20:01] yes but i can' alter evidence [20:01] can't [20:03] guess what they will say in court if the MD5 sum of the file i analyzed will not match up with the MD5 sum of the file the police gathered... [20:03] all finding will be worthless [20:04] sure [20:04] md5 the drives, md5 the contents of the containers, md5 the files inside, etc, all the way down [20:04] and i am fully aware that i can do all by hand - it just will not take 40-60h as usual it will take 10x longer [20:05] it don't help when I have a image file created with a ditto or something like that [20:06] then i have just one log along with that which tell me the MD5 of the imaged file, the officer, the time and date, the case ans some other details [20:06] but I will not have MD5 sums of every file in the drive [20:07] because they image the drive and the tool just calculate the MD5 of the drive and the image and compare them if they match [20:07] according to that report i can check the MD5 of the image as well and ensure noone tampered with the image before i get it [20:08] but i need to work with that what i have [20:09] i am sure there are tools which are spezialized in linux - it's just a question which ones are good [20:13] and even if i could extract the partitions to images then I have still the problem the tools which are focussed on windows will not help me that much === Erisa47 is now known as Erisa4 === Milencho_ is now known as Milencho [20:14] ok X would list the images and i can use the AI porn recognition but if it's a hacking case I would do all log analysis and a ton of other things by hand [20:16] it's rediculous - linux is 10x more open and i don't have to deal with decoding cryptic values from the registry or some other artefacts but still i need 10x longer to investigate due to the lack of good tools [20:16] there could be a business opportunity here :) afaik autopsy/sleuthkit is about the only thing pre-made for this to run on windows, and it's not nearly as full-featured as eg encase.. [20:17] I wouldn't be surprised if it's a small enough market that there's just no demand for making a linux-based tool [20:20] Hi. Emptying trash (~/.local/share/Trash and others) with 'Dolphin' doesn't really work, because some of the files are owned by root. OK, so I found out that in the terminal I can use `trash-empty`, but it has similar problems. [20:20] When I run it with sudo though it apparently doesn't want to empty my '/mnt/elements/.Trash-1000' directory (why? because it works only on dirs belonging to root in this case?). So what is the proper solution? I have heard that simple `rm -rf /mnt/elements/.Trash-1000/*`. is not the right way, so what is? [20:20] take even the commands - i try w, who and last [20:21] none of them showed a login of www-data on my testsystem [20:21] konrados: chown -R user:user ~ [20:21] but the auth.log show that a session was opend and ended fot that user [20:21] konrados: if you don't have a really good reason to have root-owned files under your home dir... [20:22] Then just use the normal GUI tools to empty the trash [20:24] i would be happy if there is even a tool which parse and lay the logs out in some way [20:25] alkisg, ok, but there are other 'trash' dirs, on various mounted disks, I wanted to create a simple bash script, like 'empty-trash-all', a general one, working on all the mounted devices... I don't know them ahead, so I was thinking maybe about calling 'trash-empty' with sudo privileges but with 'normal' user as ... 'user' so trash-empty... ehhh... seems like I'm overcomplicating :) OK, then I'll do chown on the 'trash' directories, [20:25] everywhere? [20:25] w3gi: try utmpdump(1) for that [20:25] and i can't belive that there is no demand or no tools - how many badly maintained servers become a hosting-platform for malware, wansomware, phishing sites, etc. [20:26] sarnold: can I run that on a ro mounted image? [20:27] it don't help me if utmpdump run on a live machine [20:27] konrados: if you delete files on disk "element", it can't move them to ~/.local, because it would take a lot of time and space to move them between disks [20:27] w3gi: yes, you can tell it what files to read [20:27] good i add that to my list of tools to test [20:28] konrados: so it names them with the UID, /.Trash-. They should already be with the correct UID there, because if root had deleted them, they'd be in .Trash-0 [20:28] konrados: so a plain sudo chown, followed by a normal trash deletion without sudo, either from the console or the GUI, should do it. No chowns elsewhere [20:28] *sudo chown -R user:user ~ [20:29] alkisg - thank you! OK, will do that. [20:29] konrados: and even that first chown is only needed because of bad sudo use before [20:29] You shouldn't ever had root owned files under your home dir (without a very good reason), so when you do get them, you need to see what caused them, and avoid that command [20:29] *have [20:30] ok, got that, thanks again! Will investigate. [20:30] 👍️ [20:31] sarnold, using ssh-import-id is going to put my key in their heiarchy, not what is needed, yes? [20:35] GrandPa-G: ssh-import-id will put your key into the ~/.ssh/authorized_keys file for the current user, or into an authorized_keys file that they choose, via ssh-import-id -o /home/whatever/.ssh/authorized_keys etc [20:36] sarnold, then they still have to move it to my structure, correct? I could have emailed them the public file in the first place, right? [20:38] GrandPa-G: I don't know what a structure or a hierarchy are.. [20:39] GrandPa-G: yes, you could just as easily mail them your fingerprint, but this is a nice simple single command that they could run, if you've got a key published in launchpad or github [20:39] sarnold, /home/user/whatever is a hierachry [20:41] any more tool suggestions which make my work easier? [20:41] Hi all - a bit of a new linux user here. I installed Ubuntu on a computer with a second hard drive (/media/username/DATA). This PC used to be a Windows machine, and I'd rather not lose the data on this drive. However everything is Read Only - can't delete, can't create new folders, etc. I've been looking for a way to change the permissions, but the suggestions I found online aren't working. Would you have any suggestions? [20:41] (I've got a terminal window open and I'm on that drive now.) [20:43] sudo chmod -Rv a+w * replies back listing items saying "failed to change mode" for each item. [20:44] trevorgib: NTFS filesystems have different permission rules [20:44] sarnold aha [20:45] trevorgib: check the mount.ntfs(8) manpage, it has some uid= and gid= mount options that you can use to force a user and group to own the files [20:45] thanks - much appreciated [20:58] I just remembered I used to use squid(proxy) back in the day. Is that something people use now? [21:02] Aavar: yup. it's still fantastic. I've got a few running around here, via the sduid-deb-proxy package [21:03] sarnold - I got it! Yes! Thanks so much. [21:03] trevorgib: sweet :) [21:03] starting to feel like I might be able to do this stuff after all ;) [21:10] sarnold: Cool... ill try to set it up just for fun :) [21:11] aw woe is me. my nickname is no longer registered :( === gallomimia is now known as Gallomimia [21:13] sarnold: hmm... quick two questions. What port is the default port for squid and how do I set it up in my browser (edge, brave)? [21:14] (If it has some quick answers. If not i'm sure i'll find it om google ;)) [21:14] follow a proper squid guide? [21:15] found it :D [21:15] tnx again. [21:17] yoyoyo [21:17] nuggets! shalom. [21:18] is there a way that allows me to send SMD from an non sim/android devvice? eh would just need the parsing convertingg etc. but its a bit length and i cant say how much it hurts my eyes ans hands [21:19] Via how is that related to ubuntu support? [21:20] #android would be the place to be [21:21] ha fck i igive uo fml ims just wrie a mail and sen see from there fml xD [21:23] oerheks: apoologies the context was not apparent to you, and likely neer will since you lack creativity + imagination, thats for your help anywys ! [21:23] thanks** i lack proper typing ;}}} [21:25] VIA, totally offtopic [21:27] hi all, I am having trouble with setting a static IP with netplan on ubuntu 18.04 [21:27] kevinnn: ok? https://linuxconfig.org/how-to-configure-static-ip-address-on-ubuntu-18-04-bionic-beaver-linux [21:27] kevinnn: should be pretty straightforward [21:28] I know for sure my static IP config works because I have tried setting it in a vanilla debian 11 instance with /etc/networking/interfaces [21:28] netplan --debug generate and netplan --debug apply both show no errors [21:28] I *believe* the issue has to do with my gateway address [21:28] when I run the command "route" it doesn't look quite right [21:29] zamba: I have tried this many times over [21:30] pastebin your /etc/netplan/*.yaml and post the url? [21:30] kevinnn: maybe pastebin your config, the generated systemd configurations, ip a and ip r outputs, someone may have an idea [21:30] oh. i think i have my question thought of.... is it possible, or recommended, to duplicate an installation from an unpartitioned SSD onto a pair of SSDs with raid and crypto? i thought i might shrink the partition and use dd after setting it up [21:31] Gallomimia, depends, standard ubuntu would install in one partition [21:31] Aavar: the usual port is 3128 I think, and I don't know offhand how to configure a browser to use it, it shouldn't be too hard; there's some browser plugins that you select proxy per tab, that might be helpful [21:31] yes it just did [21:31] i would reinstall, with crypto [21:31] the installer had crypto greyed out? [21:33] Gallomimia: multi-boot setup? that that is why it was greyed out in the installer. [21:33] negative [21:34] bare naked SSD, fresh out of the box [21:34] or you didn't realize that to make the installer do full disk encryption, you also need to enable lvm [21:34] ah [21:35] or you did manual partitioning [21:35] can i get some more info on how that works? my previous setup had manually nested block devices... raid5, crypto, LVM [21:35] no it was the first one [21:35] didn't bother with LVM [21:35] this is a desktop, right? which ubuntu release? [21:36] at any rate, im still fairly sure i want to use the two smaller SSDs in a raid. new ubuntu straight off the website, downloaded today [21:36] desktop not server [21:36] sarnold: where do I find my generated systemd configurations? [21:37] Gallomimia: the desktop installer won't support setting up software raid with encryption [21:37] hence my question... could i just dd my install? [21:38] Gallomimia, you can, unencrypted most likely will work [21:38] make sure you reinstall grub with the proper UUID ( new) [21:38] kevinnn: I think /etc/systemd/network* -- I don't have one up and running anywhere handy.. [21:38] https://help.ubuntu.com/community/Full_Disk_Encryption_Howto_2019 is a guide for more complex encryption scenarios than the desktop installer supports [21:39] pretty sure i've done the reinstall grub thing like a thousand times troubleshooting my old setup [21:39] old rig, i was going to say. it's the same rig but with a few changes in storage [21:42] dd'ing the single file system woudld not work right away, not when you're newly introducing FDE or SW-RAID or LVM === fling_ is now known as fling [21:48] Hey there. Seems the upgrade from hirsute to impish broke my wifi. since rebooting my machine after the upgrade, my wifi-device is unavailable to networkmager: [21:48] `NetworkManager[1081]: [1635247689.9040] device (wlp3s0): Couldn't initialize supplicant interface: Failed to D-Bus activate wpa_supplicant service` [21:48] Is this issue familiar to anyone? I have installed and loaded `backport-iwlwifi-dkms` but no good. Didn't find any helpful cases via websearch either... [21:50] you should discuss which network chipset you have there [21:50] Okay, couldn't find the systemd files, but I've got everything else. Here is my netplan config: https://pastebin.com/tYJcisKW. Here is ip a: https://pastebin.com/PEb6dnEH. Here is ip r: https://pastebin.com/WdmGKkF2 [21:51] any help would be so greatly appreciated!! [21:52] lux[m]1: lspci -knn | grep -EA3 'Ethernet controller' [21:52] kevinnn: your default gateway is outside the subnet [21:53] tomreyn: okay, but this EXACT config works on other VM's on my hypervisor. on debian 11 specifically. Also confirmed with my colo who provided the IP's that this is legit [21:53] so... [21:53] just add a route for that gateway [21:53] so that the system will know which interface to use to get to the gateway [21:54] add a route? [21:54] something like this? [21:54] routes: [21:54] - to: 172.16.0.0/24 [21:54] via: 192.168.1.100 [21:55] yeah, something like that should do the trick [21:55] comment #3 https://bugs.launchpad.net/netplan/+bug/1767607 [21:55] Launchpad bug 1767607 in netplan.io (Ubuntu) "missing point-to-point option in netplan configuration" [High, Triaged] [21:57] First, thanks for your work in ubuntu. [21:57] Now, perhaps a subtle issue exists in ubuntu. During and for a period of time after running two copies of smallhog after compiling smallhog.c for an extended mount of time (accidentally) as well as sleeping the system during this time from https://docs.kernel.org/admin-guide/cpu-load.html, resizing resource monitor using the secondary click and [21:57] resize results in flickering issues including sometimes with icons from the desktop showing up above windows. Sometimes, keep on top was enable for a couple of windows. This issue was not reproducible after a period of time after "ctrl c"ing the two copies of smallhog. [21:57] Environment: Using Ubuntu 20.04.2 LTS in a live session with only updating a few things like gcc (Ubuntu 9.3.0-17ubuntu1~20.04) 9.3.0 [21:57] and firefox 93.0 (64 bit) (just updating this perhaps does not solve all security with browsers). Other installed/updated things probably don't matter. [21:57] tomreyn: yes, right.. this is my chipset: https://pastebin.com/rYFCXdyG [21:59] Unrelated to lux post, my processor is Intel Core-i3 3000 mobile series with integrated graphics. [21:59] Perhaps smallhob messed with some timing for refreshing windows or something. [21:59] smallhog* [22:00] hi folks, dunno if #ubuntu is the right place - after upgrading to 21.10 Firefox (which is now a snap) does not access my saved passwords anymore, although they seem to have been copied in the migration [22:00] Is this a known problem with a known solution? [22:01] have you tried firefox -p [22:01] to access the profiles [22:01] no, not yet [22:01] (guessing here not sure how snaps handle firefox profiles if firefox is indeed in snap) [22:02] the bookmarks have been moved over, so it's not that the profile has been lost completely, just the passwords [22:02] but I'll try [22:02] interesting, not sure about that [22:02] lux[m]1: there's https://bugs.launchpad.net/ubuntu/+source/linux-firmware/+bug/1933938 - which is supposedly fixed in in linux-firmware on focal and impish [22:02] Launchpad bug 1933938 in linux (Ubuntu Impish) "Missing CNVi firmware for Intel AX211/AX201/9560/9462 on ADL-S/ADL-P" [High, In Progress] [22:03] Guest201: same situation, it offers access to the default profile, but does not show the passwords [22:03] passwords are encrypted in firefox so not sure about that [22:04] lux[m]1: check which kernel variant (generic or oem) you're using and whether the intel ax200 loadable firmware in /lib/firmware/ is up to date. [22:07] Guest201: about:logins would list known accounts in firefox password manager [22:08] Guest201: and there is https://github.com/louisabraham/ffpass but i don't know whether ti would work with the snap without further customization (probably not) [22:09] Yeah, but the bookmarks are still visible. That suggestion, "about:logins" may still help. [22:09] * tomreyn -> bed [22:09] tomreyn: thanks, updating the firmware in /lib/firmware from https://www.intel.com/content/www/us/en/support/articles/000005511/wireless.html did the trick! [22:09] I've checked the logins.json in the old installation and the new snap config directory, they are identical [22:10] interesting [22:10] could there be a problem because I did not set a master password? === calwig_ is now known as calwig [22:11] Is it asking for one? [22:11] no, just not showing any saved passwords [22:11] I could try to add one and see how it behaves [22:13] Maybe that will work [22:13] Is this Firefox 93? [22:14] yup [22:14] I have an old laptop with an Intel Media Accelerator, is there any good driver for it? The one as shown in neofetch is generic for 2xx-4xx Intel Atom processors [22:15] weird, after adding a new password it shows in logins.json with the correct next id, but firefox still shows only that new one. [22:15] I'll try restarting firefix now [22:16] forefox. whatever. [22:22] phew. I've copied keys3.db and keys4.db over from the old installation, that fixed it. Looks like the migration does not handle that correctly. [22:23] ty for helping me think :-) [22:27] No problem, interesting. I created a test profile and noticed not setting a master password still resulted in an encrypted logins.json. I still need to try out Ubuntu 21. [22:28] So the key must be somewhere else [22:28] Good catch to you :) [22:30] I think if you don't set a master password firefox creates a random key to encrypt the logins (not really safe, I know). That key might be in keys3.db or keys4.db. Alas, I did not compare the contents before copying the files. [22:40] my neofetch says GPU: Intel Atom Processor D4xx/D5xx/N4xx/N5xx but i think thats too generalised. how to get intel gma driver? [22:41] tip if you are in HexChat you can go to window then send system info [22:41] im not [22:41] im in srain [22:41] oh [22:41] im on an old laptop [22:41] there is also sudo lshw [22:41] also i dont want people seeing my pc name and username [22:41] also assuming you are on linux [22:41] im on mint, so yes, linux [22:42] how long does it take [22:42] !mint | VMGuy23 [22:42] VMGuy23: The Ubuntu channels can only provide support for Ubuntu and its official flavors, since other distributions and derivatives have repository and software changes. So please use their dedicated support venues, for example: Linux Mint (#linuxmint-help on irc.spotchat.org), Kali Linux (#kali-linux), and LXLE (#lxle) [22:42] if you're on mint there is a driver manager app [22:42] shouldent take long [22:42] took about 3 seconds on my hardware [22:42] Client: HexChat 2.16.0 • OS: Ubuntu "impish" 21.10 • CPU: Intel(R) Core(TM) i5-4308U CPU @ 2.80GHz (3.06GHz) • Memory: Physical: 7.5 GiB Total (4.5 GiB Free) Swap: 13.0 GiB Total (12.9 GiB Free) • Storage: 1.4 TB / 1.7 TB (287.1 GB Free) • VGA: Intel Corporation Haswell-ULT Integrated Graphics Controller @ Intel Corporation Haswell-ULT DRAM Controller • Uptime: 30m 47s [22:42] VMGuy23: feel free to ask for help in #linuxmint-help. Not here [22:43] ok [23:09] o/ which package provides the systemcall manpages. I just tried `man 3 read` and got nothing. I have `manpages-dev` installed. [23:18] $ dpkg -L manpages-dev|grep man2/read [23:18] /usr/share/man/man2/read.2.gz [23:19] gh0s1: its section 2 not 3 [23:20] I got it, section 3 is in manpages-posix-dev [23:20] thanks folks :) === RaGE_Syria82 is now known as RaGE_Syria8 === genii is now known as genii-core