[00:03] Not sure if this is the right place to ask this but running virtual box with ubuntu, trying to ssh into it but not working [00:03] I forwarded port 22 from it to 3022 and its still not working [00:04] what error message do you get? [00:06] sarnold just cant connect [00:06] you can use -vvv to get more information to see what might actually be happening [00:07] be sure to check logs on the server side, too, sometimes additional information is logged there than is returned to the client [00:08] funhouse: is the VM using a bridged or NAT'd interface? Can you ping the VM from your host? Can you ping the host from the VM? === Bencraft145 is now known as Bencraft14 [00:11] um, looks like ssh isnt installed by default in ubuntu? [00:12] correct [00:12] if you mean the server... [00:12] ± seeded-in-ubuntu -b openssh-server [00:12] openssh-server is seeded in: [00:12] ubuntu-server: daily-live, daily-preinstalled [00:12] oerheks, ok got it thank you oerheks sarnold and dilyn [00:15] :) === esv_ is now known as esv [01:01] hello, i'm trying to backup important folders for specific programs such as firefox (extnesions and configurations.. etc ) but i don't know what folders should i actually back up.. im willing to install a new distro so things that's is system based will be generated on new system perhaps i don't wanna back up those? may someone help please? [01:02] backup your profile? [01:02] https://support.mozilla.org/en-US/kb/back-and-restore-information-firefox-profiles [01:03] krowH: usually people back up their entire home directory, maybe skipping ~/.cache/ [01:03] /.mozilla [01:03] but i would find out what the new ff 100 gives [01:05] i want to back up firefox data yes but also other things, like an appimage configuration files.. but for not all applications i have on system [01:05] in ubuntu 22.04 where is php-fpm? When you apt-cache show php-cgi it says you probably want php-fpm if using nginx [01:05] appimage.. interesting [01:05] sarnold: is /home will suffice ? [01:06] krowH: or /home/krowH, etc, whatever's appropriate [01:08] for php6 7 8? [01:08] ok thank you sarnold, one last thing please, i've read on some article someone recommends backing up /var /bin /etc /opt ... and so on... these things won't be important since i'm going fresh install or do i have to back up some of them? [01:09] you would only be interested in backing these up if you had modified something in them, usually [01:09] for instance, custom systemd unit files [01:10] snappy: https://packages.ubuntu.com/search?searchon=contents&keywords=php-fpm&mode=exactfilename&suite=jammy&arch=any ? [01:10] there's not a lot of point in backing up /bin -- there might be value in the others. eg, the user cronjobs are stored in /var/spool/cron/crontabs [01:10] but a lot of /var/ doesn't need backing up. it's hard to be precise there ;( [01:10] otherwise there's a literal package php-fpm [01:10] dilyn: nope i haven't modified anything.. i'm pretty new to linux to be honest -- i'm just going to install a new distro but there's some configurations that took long time to set up for some applications and i dont' wanna fo through them again in my new distro that's it [01:11] dilyn: oh interesting [01:11] ah i didn't do apt update, that'll explain things [01:11] :-) [01:11] krowH: if by customization you just mean things like editing the settings within applications, you can probably get away with just backing up ~, as it includes ~/.config where most of those things are likely to be kept [01:11] https://packages.ubuntu.com/jammy/php-fpm [01:12] otherwise as sarnold said, it's hard to know in general; but my rule of thumb has always been "you'd know if you did". for instance, if you setup a mail server you'd probably want to backup some directory in /var... [01:13] okay [01:13] that's a good rule :) [01:13] thank you sarnold and dilyn [01:13] also oerheks :) [01:15] have fun === Scotty_Trees is now known as earl_grey === earl_grey is now known as therealharvey === therealharvey is now known as Scotty_Trees [03:26] so i tried that weird snap thingy, installed a program and now? how do i start things from that snap? [03:27] no launcher in gnome, trying to start it from terminal crashes it [03:28] "illegal instruction (core dumped)" [03:28] how nice and now? [03:29] how get i rid of that crap now? [03:36] 0000 [03:36] de-factco, snap offers remove command too. [03:38] yeah i removed it now with snap remove --purge [03:38] thanks [03:54] Does anyone use mobile broadband? [03:57] madsam: Internally or through a hotspot? [03:57] Via a usb modem by AT Commands [03:58] madsam: Sorry, I've not done that. Maybe others will know. [03:58] arraybolt3: why haven't you? [03:59] madsam: I've just USB tethered to a cellphone or used mobile hotspot Wi-Fi. [03:59] arraybolt3: it's for lames man [04:03] arraybolt3: if you didn't know this, normally you can connect via AT Commands using the LTE modem, so the syntax is like [04:03] AT+CGDCONT=1,"IPV6","your-apn-ipv6-here" === ubuntu is now known as Guest1473 === ubuntu is now known as Guest654 === veegee_ is now known as veegee === Mibixy is now known as Mibix [05:30] so i tried to update to ubuntu 22.04 lol look at all these broken packages :o https://paste.ubuntu.com/p/7YgXYqp7z7/ [05:32] but i dont see any with apt or dpkg :/ [05:38] nvm found the error in /var/log/dist-upgrade/main.log === diskin_ is now known as diskin [07:08] man this update to 22.04 taking so long [07:08] been an hour and its only like 1/8th the way installing haha [07:30] The new social network for cupcake https://w.metty.us/invite5674 === Starmina_ is now known as Starmina [07:31] Guest459: Is this directly related to Ubuntu support? [07:33] Warning, all, this site linked to by Guest459 is not coming up in Google searches. I think this is probably a malicious link. [07:34] Hi, I have a problem with ubuntu becoming unresponsive with some io, like with rsync [07:34] nothing even very heavy [07:34] rsyncing from nvme m.2 to usb nvme m.2 [07:35] system becomes simply unusable, takes >30 seconds in some cases to see the mouse move and menus to update [07:35] I think in the past I had to use low latency to escape it being unusable? [07:36] Hi, I tried install Ubuntu 22.04 7 times and it doesn't find the boot after restart. I really don't know what to do. There is a single new ssd in my laptop. [07:37] Guest085: How much RAM is in your system? This sounds like thrashing, which can be caused by low RAM. [07:37] arraybolt3 8GB [07:37] arraybolt3 it may very well be thrashing, how do I manage it? [07:37] you think kernel is buffering io in ram? [07:38] webchat22   you probably need to try out the grub install option. [07:38] Windows 8.1 is better than your fucking Ubuntu I fucking install all the games on my computer and what the fuck do you want? What the fuck can you do on it bitch fuck you stupid fucking faggot [07:38] Guest085: Compress your RAM. [07:38] !ops [07:38] Help! Channel emergency! (ONLY use this trigger in emergencies) - CarlFK, DJones, el, Flannel, genii, hggdh, ikonia, krytarik, mneptok, mwsb, nhandler, ogra, Pici, popey, sarnold, tomreyn, Unit193, wgrant [07:38] Guest459  windows sucks. [07:38] Operators: Guest459 is casuing problems. Please remove them from the channel immediately. [07:38] Guest085 also, nvidia just released open source graphics drivers, so linux gaming is about to get a LOT better. [07:38] so you're also an idiot. [07:38] Guest085 like with the boot repair disc [07:38] ? [07:39] Guest085 sorry that was for Guest459 [07:39] I don't cause trouble and what the fuck do you want to tell me, motherfucker? [07:39] lol [07:39] Guest085: To compress RAM, open a terminal with Ctrl+Alt+T. [07:39] arraybolt3 compress??? [07:40] I'm in late on it, but yes you can compress ram or swap. zram, zswap. Interesting really. [07:40] !info zram-config [07:40] Guest085: Yes. You can actually make Ubuntu compress RAM contents rather than swapping them to disk. This makes your system behave like you have more RAM than it does. [07:40] zram-config (0.7build1, jammy): System service to enable zram support. In component universe, is extra. Built by zram-config. Size 4 kB / 18 kB [07:41] OK, Guest459 is out. Here's the command sequence for compressing RAM: [07:41] sudo su - [07:41] modprobe zram [07:41] arraybolt3   how do I simply configure the kernel not to buffer so much io to ram? [07:41] the disk io speed is the issue, I don't need to buffer it much [07:42] Guest085: I don't know. Really, with 8 GB of RAM, I doubt this is your RAM's fault. [07:42] Also, how do I decrypt user home directory in ubuntu? [07:42] arraybolt3 the system becomes so unusable I can't even troubleshoot it. [07:43] Guest085: Decrypt user home directory? Like from within a live ISO? [07:44] arraybolt3 yes, I have a usb nvme m.2 256GB to throw my home directories on before installing ubuntu's newest lts fresh. [07:45] Guest085: This looks like what you're looking for: https://askubuntu.com/questions/63594/mount-encrypted-volumes-from-command-line [07:46] arraybolt3  yes, I'm familiar with luks, but ubuntu home directory encryption uses some weird standard [07:48] Oh, I see, it's eCryptfs. [07:49] it's not in the liveiso apparently [07:50] I'm looking for how to deal with it... [07:51] Guest085: Bingo. https://askubuntu.com/questions/238047/how-do-i-mount-an-encrypted-home-directory-on-another-ubuntu-machine Looks like it takes one command. [07:53] arraybolt3 that's not available in the liveiso or as an install candidate [07:55] It still exists. It's in the package ecryptfs-utils. [08:10] oh wow [08:10] this home directory encryption stuff is bad [08:11] What's it doing? Error messages? Can't find where it put the files? [08:12] arraybolt3  it just sucks. [08:15] Guest085: Really, encrypting just the home directory is helpful, but it's not really all that secure against someone with the right know-how and some determination (due to the fact that the swapfile still isn't encrypted). If you want to use encryption, maybe try LUKS. [08:16] arraybolt3   my swap file is encrypted in luks yeah [08:16] arraybolt3  the ecryptfs stuff sucks because it's difficult to use and unwrap manually, frankly it's shoddy as hell [08:16] I didn't realize it was doing file encryption [08:17] ls [08:20] I had to do this bs: keyctl link @u @s [08:20] now this works: ecryptfs-recover-private .Private [08:20] take that back, it's not working [08:22] Ubuntu was recommending ecryptfs a few years ago, but then they switched to recommending full-disk encryption; so I switched my ecryptfs private folder to the newer and better supported fscrypt [08:23] There's also cryfs/gocryptfs and libpam-mount. Very handy. [08:24] alkisg  yeah, I'm abandoning ecryptfs and user directory encryption [08:24] it's stupid [08:25] might as well just luks everything [08:31] User directory or even single directory encryption still has its uses; but if FDE suits you, it's much easier [08:31] E.g. I wouldn't have anyone at work unlock my disk if it rebooted when I'm not there; so FDE is useless to me [08:32] get weird tinny audio here and there, is there a quick fix for it? have the codecs installed. [08:37] is there a window manager where I can resize/move my windows in gnome via keyboard shortcuts? [08:37] I still wanna use Gnome but have that feature [08:40] Hello, I have a question regarding keyboard configuration in ubuntu 22.04. My laptop is a chromebook, which has a bit special layout, for example the F keys are mapped to controlling the audio and similar. How can I convince gnome to use a different keyboard model? xkeyboard-config has support for this model and i have tried setting it both with [08:40] localectl and by editing /etc/default/keyboard, but gnome/ubuntu seems to ignore this [08:48] I did a bit more research, and it seems like gnome does not support configuring different keyboard models if wayland is used :-( [08:52] is there a way to display rsync progress occassionally? [08:52] I care about seeing the distance it's gotten [08:52] but don't want to slow it down for every file [09:02] ubuntu seems to slow down when I'm doing a bunch of disk io [09:07] vladoski: for tweaks in your system, try find dconf-editor values [09:35] Hi Everyone, I tried to install the new lts version the whole hight. I'm getting crazy now. I didn't sleep. After installation there is no grub on the system. I'm alone with this problem. Is new version faulty? [09:35] Hight is night [09:37] nemtuja: You did an upgrade, not a fresh install, right? [09:38] It is afresh instal. New ssd. Not dual boot [09:38] nemtuja: there are some changes on uefi & biosboot at grub level, see the releasenotes; https://discourse.ubuntu.com/t/jammy-jellyfish-release-notes/24668 [09:42] I tried uefi and legacy mode in bios. None of them work. Is it work for everyone? [09:44] nemtuja: sata ssd or nvme? did you disable secureboot? [09:44] Ssd. Secureboot disabled [09:44] nemtuja: ubuntu desktop or another flavour? [09:45] Desktop gnome [09:45] nemtuja: Hi, no it's not faulty [09:45] nemtuja: so at wich point exactly do you get an error? boot issue? [09:45] New version is working very well. [09:46] Yes. After installation boot device not found [09:47] I'm running actually new LTS version Kubuntu 22.04 with Xfce upgraded from 21.10 and surprisingly everything went nice and smooth, thanks to Cannonical. [09:47] IMHO they have done great job. [09:48] How did you install it nemtuja ? [09:48] Are you using the "Erase disk" option, or are you doing manual partitioning? [09:49] Try again and choose the option with automatic partitioning whatever it's called. [09:50] Erase disk [09:50] I use auto [09:50] you had to done something wrong perhaps with boot menu [09:51] is it your only OS on disk or are you sharing with another operating system nemtuja ? [09:51] vimart: New ssd, not dual boot. [09:51] The was no option doing anything with boot menu [09:52] nemtuja: What model is your system? By "new SSD", do you mean that you wiped an existing one, or you installed a brand-new one? [09:52] I see I got mixed up with nicks [09:52] Adata su800 [09:52] Brand new [09:52] nemtuja: Wow, that's really weird. What manufacturer and model is your system? Something like a Dell Optiplex something-or-other, for instance. [09:53] Hp probook 640 g1 [09:53] Hmm. What error message do you get when the system fails to boot? Is it just "No bootable image found, notebook will be shut down"? [09:53] nemtuja: were you doing manual paritioning or left the job to do by installator? [09:54] vimart: They told us that they used the "erase disk" option. [09:54] ok [09:55] Boot device not found Pkease Install an operating system on your hard disk hard disk 3F0 [09:56] nemtuja: Well, that's very odd, since it sounds like everything should work just fine. My best guess is either you got a bad brand-new SSD (unlikely), or you got a bad ISO download (way more likely), or you're using a bad flash drive (also somewhat likely). [09:56] after erase disk what happened? [09:56] nemtuja: Try downloading the Ubuntu 22.04 ISO again, and flashing it to a different flash drive than you've been using. Then try the install again and see if that works. A bad ISO or flash drive can cause all sorts of weird problems. [09:57] nemtuja: Wait a minute. When you were messing with UEFI and Secure Boot, did you accidentally disable SATA boot? [09:57] nemtuja: If you accidentally told your system to not boot from the internal SSD, that would cause your problem. [09:57] I use the same iso but tried on different flash drives [09:58] nemtuja: Try looking through your BIOS settings and see if you've disabled your ability to boot from a SATA drive. [09:58] nemtuja: You might have done this trying to get your system to boot from the USB in the first place, forcing it to not boot from SATA so it would boot from USB. That would cause exactly these symptoms AFAIK. [10:00] There are cd-rom boot, usb device boot and upgrade bay hard drive boot checked. No sata option in the list [10:01] nemtuja: OK. Put the system in UEFI mode, then press Esc over and over during early bootup, and try to enter a boot menu. Then select "ubuntu" in the menu. Some HP laptops have a problem getting Ubuntu to boot after a fresh install, since they're trying to boot Windows still. [10:02] (If you did a BIOS installation, this won't work, but if you did a UEFI installation, then this should do the trick.) [10:02] nemtuja: If this works, then I believe I know the last step to getting your system functioning, since I've had exactly this problem on an HP Elitebook. [10:02] (Or almost this exact problem, anyway.) [10:03] There are twu uefi modes uefi hybrid (with csm) and uefi native (without csm) till this moment it was in legacy mode [10:03] nemtuja: Do uefi native (without csm), then install Ubuntu again. [10:05] nemtuja: Once the install is done, when the system is first booting, do the same thing you do to get into BIOS setup, but go to the boot menu rather than BIOS setup. Then select "ubuntu". If that works, then I'm almost certain I know what the problem is. [10:05] Installation started [10:06] It's working [10:06] I thought when ichoose ubuntu it will start the intaller flash drive but it booted in the fresh system [10:07] Thanks [10:07] Hi there, I have a question. [10:07] nemtuja: Fantastic. OK, so let me find the instructions that got things working automatically for me. It might take me a bit. [10:09] I'd like to install Ubuntu on a laptop whose BIOS setup panel is locked with a password. Do you think it could be a problem ? I have no idea how secure boot flag and the like are set. I can boot on a flash drive and play with the live system though. Any pointers welcome, thx! [10:10] nemtuja: Bah, I can't find it. What you do is, get into BIOS options, and set Customized Boot as the highest priority. Then set Customized Boot to boot up EFI/ubuntu/grubx64.efi. I think that will work - if it doesn't, try setting the customized boot to EFI/ubuntu/shimx64.efi. [10:11] i did apt install quake3 and it is stuck trying to download a file from id software's ftp server and failing to connect.... it's been stuck there retrying to connect. how can i quit safely? [10:11] marcus-d: If you can boot the live flash drive, you can probably install Ubuntu without problems as long as you don't need any proprietary drivers. If your Wi-Fi is working in the live environment, and you don't need proprietary graphics card stuff, you should be good to go AFAIK. [10:12] marcus-d: You might get messed up by Intel RST, but it's likely that Ubuntu will warn you about that and not let you install if that's the case. [10:13] Thanks arraybolt3: wifi works with the live system and I was able to play HD videos smoothly. So I should be OK, that's encouraging. [10:13] Intel RST: I don't know that beast, I'll check that. Thx. [10:14] <_maddy> I am upgrading xubuntu to 22.04 in terminal usuing 'sudo do-release-upgrade', it hangs forever on 'Waiting for automatic snapd restart'... how to fix without messing up the upgrade? [10:14] arraybolt3 Thanks I'll have a look at that [10:17] it's stuck trying to download a file from id software's ftp server... i want to cancel the installation [10:19] ctrl-c does nothing [10:20] it's retrying 10th time now [10:25] xkill it>? [10:25] is it safe to killall -9 apt? [10:25] open a terminal window type xkill then click on the ftp client window [10:26] sams: it's doing it as part of apt install quake3 [10:26] oh [10:27] ctrl z might work [10:29] but will it break apt? [10:30] loganlee[gnu]: you could try SIGTERM with htop if you can locate the apt process. [10:30] loganlee[gnu]: you may want to read this https://askubuntu.com/questions/253925/how-to-safely-abort-apt-get-install before trying "kill -9". [10:30] marcus-d: yeh i know the process id [10:31] loganlee[gnu]: SIGTEM is safe. You may try that. But maybe the process won't be a [10:31] able to process it. [10:35] marcus-d: killed the processes holding the lock files and the main apt process is killed now [10:36] the package is marked as installed even though it was interrupted this way [10:36] can i safely purge it? [10:39] loganlee[gnu]: so were you able to kill it with SIGTERM? [10:39] I don't know about the purge. Why not try to remove it first? [10:43] marcus-d: from your link sudo fuser -vki -TERM /var/lib/dpkg/lock /var/lib/dpkg/lock-frontend [10:44] i guess this sends SIGTERM [10:44] to lock files [10:45] omg `sudo dpkg --configure -a` tries to download from ftp again.... [10:49] loganlee[gnu]: so probably the FTP thing happens during configuration. You may stop the process safely as you just did. And instead of running `dpkg --configure` just remove the package with APT. I am not an APT expert but that looks harmless. [10:55] arraybolt3 Where did you find that HP EFI bios settings? [10:58] marcus-d: holy crap i had to do `sudo fuser -v /var/cache/debconf/config.dat` to kill the process on config.dat then i could purge quake3-data. now both quake3 and quake3-data are fully purged [10:59] well `sudo fuser -v /var/cache/debconf/config.dat` then kill -9 [process id] [11:00] apt was locked on config.dat and preventing me from purging quake3-data [11:00] so i had to kill that [11:00] i think my system is ok now [11:00] apt-wise [11:02] loganlee[gnu]: glad to head that. Though you won't be able to play quake3 thanks to apt, that's a shame. Good luck. [11:04] marcus-d: i could have done `sudo fuser -vki -TERM /var/cache/debconf/config.dat` instead [11:04] ` [11:05] that will send SIGTERM to config.dat [11:07] well the process locked on config.dat file [11:08] arraybolt3 success [11:09] \EFI\boot\bootx64.efi instead of EFI/boot/bootx64.efi solved the problem [11:09] so 22.04 is not workin in legacy mode anymore? [11:11] apt is pretty stupid.... it keeps trying to download a file from ftp server that doesn't exist and i can't gracefully quit it [11:12] i have to do `sudo fuser -vki -TERM /var/lib/dpkg/lock /var/lib/dpkg/lock-frontend /var/cache/debconf/config.dat` to kill all the locked files then try to purge [11:28] Hi all [11:37] hi blu [11:38] hi BluesKaj [11:38] is there people playing in public servers on quake 3? [11:40] that would be an incentive to make this thing work [11:47] would a binary compiled on 20.04 run on 18.04? [11:47] SteelRose, sometimes, yes [11:49] I guess I better recompile the stuff then... [11:49] thanks oerheks [11:53] hi loganlee[gnu] [12:38] hi [12:40] hi [12:42] hi [12:44] How are we all? [12:44] alright [12:45] great news [12:45] How you tried out 22.04 yet? [13:08] How to register an account here ? https://w.metty.us/invite5674 [13:14] j [13:15] What happened? [13:15] porn links [13:15] I didn't even click on it [13:16] nice catch [13:16] that person was nasty before ... but had been qiet for the last few hours [13:17] wow [13:32] Hello! My 21.10 refuses to upgrade, help? [13:32] Upgrades to the development release are only [13:32] available from the latest supported release. [13:33] nina, there are still some open bugs with do-release-upgrade, you need to use the -d option if you want to force it [13:33] !ltsupgrade [13:33] Regular upgrades from the last but one LTS release to the latest LTS release, 22.04 "Jammy Jellyfish", are enabled days or weeks after 22.04.1 is released. This delay helps to ensure that any lingering issues are resolved before people upgrade production systems. If you'd prefer to upgrade now, use sudo do-release-upgrade -d [13:35] Same result [13:36] and you did update your system properly to all latest package versions first ? [13:37] sudo apt update; sudo apt full-upgrade [13:37] (make sure to have the -updates and security repos enabled) [13:39] Ah, just upgrade [13:40] But ye, says that i dont have anything to upgrade [13:40] so same result still [13:40] ill check the repos [13:44] Any repos in particular i should look for? They all seem default to me, never changed them [13:45] well, then it should work ... [13:47] Is there a sources list generator like for debian somewhere? [13:47] nina, https://discourse.ubuntu.com/t/jammy-jellyfish-release-notes/24668 has the instructions, perhaps they help [13:51] Thanks! I need to go, in case it doesnt work ill come back [13:51] <|\n> hello, trying to build python 3.10 from source on ubuntu focal vm, whom should i blame for openssl version compat-issues? [14:00] <|\n> should've been re-phrased probably: which version should i pick to be able to accomplish such a space grade task? [14:08] probably way later... [14:08] like 1.1.1o or something... [14:19] <|\n> dilyn, i mean ubuntu release itself [14:21] impish might have a recent enough openssl, otherwise jammy (as jammy comes with 3.10.4) [14:21] so if you want a known-working dependency stack, go with jammy [14:21] <|\n> oh looks good! thanks dilyn [14:21] np [15:03] I'm seeing crashes on 20.04 that I'm trying to track down a cause for. One of the things in the logs I'm wondering about is a corrected error for "[Hardware Error]: Unified Memory Controller Ext. Error Code: 49". Anyone know what that is and if that is actually an issue? Searching isn't pulling up much. [15:04] what kind of processor [15:06] AMD Ryzen 7 5800X 8-Core Processor [15:14] Delemas: Did you try just pasting the whole message into a search engine? [15:15] Yes multiple search engines and variations on sub strings. [15:17] I did find something on mce: Hardware Error microcode a201016 which suggest it might be some kind of power management issue but that doesn't make sense when the crashes happen during heavy use. [15:17] are you using XMP or any kind of overclock? [15:19] I'm an engineer. I deliberately avoid overclocking. [15:21] The only possibility wrt/ overclocking I'm aware of would be an independent computer store setting that up by default but I doubt that. [15:22] well might'nt hurt to check the bios to make sure XMP is disabled [15:22] otherwise could be bad ram or a faulty chip [15:23] the memory controller is on the cpu [15:23] or try a bios update [15:23] Hi guys [15:23] hey [15:24] I just switched back to Ubuntu from Debian, is there something that I should do as first thing? [15:26] dn_gian5x[m], well, first enjoy %) second configure for your work flow and/or fun %) [15:33] I know they updated the BIOS when I bought it like a month ago as the MB didn't support the CPU without it. Once I can reboot and go into the BIOS I'll confirm both XMP and "Global C-State Control" are disabled. Thanks. [15:43] Delemas: you can check your bios version with DMIDecode as well. if you can upgrade, put the update file on a usb stick formatted to fat32, and you *should* be able to flash from within the bios, depending on vendor [15:45] how can I reliably trigger the OOM killer artificially? so I can figure out why it's killing something it should have with the -1000 oom adjust [15:45] *shouldn't [15:47] " tail /dev/zero" is fast, but first take a read https://askubuntu.com/questions/1188024/how-to-test-oom-killer-from-command-line === vlm_ is now known as vlm === Starmina_ is now known as Starmina [15:55] Thanks yes, latest BIOS confirmed. Manual says "Global C-state Control" is on by default and XMP is disabled by default. I'm hoping one is ON and a turn off will end the crashes. === vlm_ is now known as vlm === vlm_ is now known as vlm [16:12] omg! after 3 weeks from initial 04-20-2022 release of 22.04 there is finally a "daily" available... i wonder if anyone will tell us what was going on? i bet something super big. [16:13] i downloading it right now [16:15] as yesterday we had a kernel update, probably [16:18] oh crap. it was a daily which is kinetic canary. its not ubuntu 22.04 official [16:18] google took me to wrong repo [16:19] its still 04-19-2022 as the last update [16:20] but at least there is a "daily" of 05-13-2022 [16:20] ah yes, google [16:20] they have old cache, that points now to 22.10 [16:21] what are we talking about exactly? [16:22] leftyfb, even with bing; when i search ' ubuntu 22.04 daily ' then i get old results, with urls that point now to 22.10 [16:23] https://cdimage.ubuntu.com/daily-live/current/ [16:23] i know. what the heck is going on with that [16:23] same here [16:24] we're complaining about search engine results? [16:24] no, just an old cache from beta daily builds [16:24] i havent got to a 22.04 daily yet. still trying [16:24] im downloading a 20.04 daily [16:25] only 22.10 dailys or 20.04 daily. been like this for a long time [16:26] the 22.04 final release does not work with my setup while in live usb mode [16:26] i have an nvidia 3060ti and screen is all white so i check for 22.04 daily all the time so i can use ubuntu 22.04 somehow [16:34] en11gma, use 21.10 and upgrade ? [16:35] cant upgrade with usb "try ubuntu". i mean you can if you make it persistent but when i do that it does something to my windows drive so i dont like to make it persistent [16:35] server and add a desktop [16:36] on the ubuntu 20.04 dailys do they add new chipset and gpu's when they come out so it will work when it finishes booting up in the "try ubuntu" mode? [16:37] those newer GPU' s might need a newer kernel [16:37] so the dailys provide that when they become available? [16:38] or is that only in point releases? [16:38] hwe will be added at point release ... (i forgot if it is .1 or .2) [16:40] ahhh. crap. yep thats a problem for me then because 22.04 was released with a pretty big problem with ppl who own nvidia gpu's while using the "try ubuntu". so i been checking for new 22.04 dailys for new updates that will fix the gpu problem [16:40] so now im just learning it wont fix it until a new 22.04 point release? [16:41] that is bad news for alot of ppl with nvidia cards and wanting to try ubuntu [16:41] tried nomodeset? [16:41] is that the same thing as safe graphics mode? [16:41] it is an optio during booting live iso [16:41] !nomodeset [16:41] Systems with certain graphics chipsets may not boot properly out of the box. "Temporarily Add a Kernel Boot Parameter for Testing" as discussed at https://wiki.ubuntu.com/Kernel/KernelBootParameters to add the "nomodeset" parameter there. [16:42] yea but that is the same thing as "safe graphics mode" option isnt it? [16:42] no [16:42] ohhh [16:42] no i have not tried that [16:42] i did the safe graphics mode thing that was the same thing [16:43] ok now i just wrote a 20.04 image now i need to write the 22.04 image and try that. thanks [16:43] on the screen try/install, F6 for nomodeset, IIRC [16:44] gonna try that now as i just found a usb stick that already has 22.04 on it [16:44] thanks brb === beaver_ is now known as beaver [16:58] Thanks people. Both XMP and Global C State options were enabled in the BIOS. One of which was done by a questionable new tech... Once turned off system no longer has errors on boot. Likely fixed but will know in a few days if it doesn't crash. [17:13] helloihave a problem with a upgrade in ubuntu 18.04 [17:14] https://paste.ubuntu.com/p/NDxz7FnTnT/ [17:16] `sudo apt --fix-broken install` [17:16] If it doesnt solve it, share output. [17:16] i do it is not work [17:17] seems you dont have deafult bionic python stuff [17:17] so its ppa related [17:17] i get message error [17:17] dpkg-deb: error: paste subprocess was killed by signal (Broken pipe) [17:19] deadsnakes ppa? [17:19] time to do a fresh install then [17:20] Hmmm well if yer willing to fresh install, may as well start with jammy. [17:21] darksis, try ppa-purge to get rid of the PPA [17:21] Unless 32-bit I guess. [17:21] ogra, i bet he cannot install ppa-purge now.. [17:21] seen this many times before [17:21] I use 22.04 ;D [17:21] ppa-purge should be standard. [17:21] On both of my servers ;D [17:22] if not you can easily just dleete the file in /etc/apt/sources.list.d [17:22] that wont help [17:22] Hmmm thats how I remove ppas.... [17:22] ppa-purge cleans up the dependencies ... just remving the sources.list entry will just break it more [17:22] oh ok [17:22] deleting that file is worse, it does not reverse packages, and you get no updates [17:23] my bad, not familiar with that app, sounds useful. [17:23] the prob here is that there are newer versions of distro packages in the PPA [17:24] I've personally always had bad luck when I tried to install newer python version, never works the way I want. [17:26] Better to upgrade ubuntu than upgrade python, if you actually need a newer one. === duxco is now known as duxsco === duxsco is now known as duxsco_ === duxsco_ is now known as duxsco [18:02] I'm trying to install the latest LTS of ubuntu, but I'm having serious issues. [18:02] When I use rsync for any io at all, ubuntu becomes completely unresponsive. [18:02] top shows gnome-shell riding at 30% cpu, always, which seems insane. [18:03] I'm not sure what to look for next. [18:10] @Guest89 are you using 22.04 when running the rsync? ( asking questions that are in my head sfter reading your post [18:16] can't believe I'm saying this: after with ubuntu for 14 years I now switch to debian, major reason? the out-of-box installation with ubuntu 22.04 on virtualbox on windows failed again, was never able to get that working with ubuntu 20.04 either due to some cloud-init errors, debian was installed without issues. [18:17] was planning to upgrade desktop to 22.04, now it will be debian instead. the virtualbox-oob experience keeps surprising me, it's too essential to forgive [18:17] yolo: do you have a support question you want help troubleshooting? [18:17] @yolo. yikes. I do understand the decision. [18:17] i filed a couple in the past, it eventually did not help [18:18] yolo: ok, feel free to take rants or discussions to #ubuntu-offtopic or #ubuntu-discuss. This channel is for support [18:19] Can you come up with an "exit statement"? [18:19] armadefuego1: please take it to #ubuntu-offtopic [18:21] Hey, anyone knows why I cannot make https://yt-dl.org/downloads/latest/youtube-dl readable with a simple text editor? It should be an open-source Python script but it looks like its encoded or obfuscated. [18:21] I apologize. I thought the question was much _on topic_. How were you failed. [18:25] armadefuego1 on most recent windows10, install a standard 22.04 or 20.04 and then boot up(no special customization whatsoever), then it hangs right away. shutdown reboot again, it sometimes even give me a login screen then hangs again. when i tried 20.04 it was stuck at cloud-init or something and did not even finish installation. [18:26] it's dead-on-arrival [18:26] tried on different windows machines with different virtualbox version, the same. i can't believe this was not well tested before release. [18:26] Thanks for the feedback. [18:28] My external hard drive wont show up as a block device [18:29] armadefuego1 yes, currently running live iso of 22.04 [18:29] @mint. does it show up as _any_ device? [18:29] gnome shell is sitting at 30% cpu, right off that's bad. [18:29] mint: what version of Linux Mint are you running? [18:29] @Guest89. That is not good, but, maybe one at a time? [18:29] live latest [18:30] fair [18:30] !mint | mint [18:30] mint: 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) [18:30] 22.04 latest LTS liveiso is becoming unresponsive when simply using rsync from nvme m.2 to usb nvme m.2 [18:30] Hi, new to ubuntu. I'm having a problem cloning VMs from a template, and wondering if there's a way to set up an initial (first/one-time-only) script/target to rm /etc/machine-id and run systemd-machine-id-setup ? [18:30] How do I even troubleshoot this? [18:31] @Guest89 slow rsync? that might me something i can help with. [18:31] davetoo_. What's the vm software? [18:31] Guest89: cd/usb live performance isn't really something that is guaranteed. It's really meant to install the OS [18:31] Proxmox [18:31] Guest89: install ubuntu and you won't have the issue [18:31] mint is based on ubuntu whatever [18:31] armadefuego1. it's only doing 40MB/s I shouldn't have to slow it.  It's limited by USB. [18:31] Even then, ubuntu should not be bothered by rsync taking 100% of one of the cores/threads. [18:31] mint: please seek support from linux mint [18:32] thanks [18:32] leftyfb. why would installing ubuntu change anything? [18:32] the live iso is supposed to be a demo of the OS, it should be the exact kernel loaded. [18:32] why would installing it be different? [18:32] Guest89: the OS is running from a squashfs filesystem on the same, slow usb device you're trying to perform file transfers on [18:33] @Guest89. so then the CPU might be a direct co-relation. And the Running on, essentially, a CD could contribute. [18:33] Guest89: the live cd/usb isn't meant for performance [18:34] leftyfb. I'm not performing file transfers on the USB stick or in the squashfs.   I'm running rsync between an nvme m2 and a usb nvme m2.   The OS and squashfs should already be loaded into ram. [18:34] leftyfb that makes no sense. [18:34] @leftyb. Yes. several factors contribute to the noted effect. [18:34] Like yes, if you run out of ram, it might be slow to LOAD things, but once they're loaded it should have no issue at all using the processor and ram normally to run at full performance. [18:35] armadefuego1. No, running a live iso should not contribute at all. [18:35] Guest89: it's an ephemeral environment. There's no changes that can be made to fix anything [18:35] Guest89: that's not true [18:36] On one level that is correct. on another it can be caused by internal locks in various portions of the system. In effect serializing a multi-CPU machine. [18:36] leftyfb] That's not how this works. [18:36] Guest89: if you think that is true, boot it up and then remove the usb, see how well that works out for you [18:36] * davetoo_ goes back to searching for systemd initial boot techniques... [18:36] leftyfb. it was happening with previous ubuntu lts while installed, and I worked around that by installing lowlatency kernel. [18:37] leftyfb.  Running software from USB has nothing to do with it's actual performance while loaded and running. [18:37] stop arguing with me. [18:37] lol [18:38] Guest89: ok, then remove the usb while it's running [18:38] leftyfb.   You don't seem to understand what you're talking about. [18:39] There are multiple ways that you can load an iso.    You can copy the entire FS into RAM to query, and then you can rip out the USB.   Or you can load things *as needed* from USB into ram. [18:39] Either way, the only performance issue USB should cause is in transfering data to/from the USB stick itself. [18:39] the software isn't going to run slower/faster because you ran it from USB. [18:40] there's something wrong with the software or its configuration. [18:40] There's no swap file, there's not enough ram used to even engage the swap file. [18:42] Just as a point to that. the live CD _may_ be looking for additional hardware at an increased rate. On the USB bus. The ISO9660 FS may present a set of locks that delay FS access. And Gnome may be looking for something you don't have. [18:42] I don't know any of that for a _fact_ [18:43] Ubuntu won't detect my external hdd [18:44] corenull: sorry, we're not playing that gam. Seek support from Linux Mint [18:44] Wym [18:45] I'm in Ubuntu [18:45] corenull: please run this and post the output here: ( cat /etc/os-release ; uname -a ) | nc termbin.com 9999 [18:45] Ok. Did you see the device added in dmsg? [18:46] armadefuego1. None of that matters.  Even if it's encountering locks on the USB device, when rsync runs the processor will be running it's binary instructions at the same speeds. [18:47] and seeing as its a demo ISO, the kernel being used should be the same. [18:47] corenull: got that output? === diskin_ is now known as diskin [18:49] safe graphics is the same as nomodeset. it shows it on the grub command line menu where you can edit the startup process [18:53] leftyfb you got that output [18:53] corenull: please paste the URL here [18:53] After you [18:53] corenull: I'm not the one seeking help [18:54] Maybe you need help [18:54] I wouldn't mind helping, but i can't see what is on your machine. [18:54] corenull: trolling is offtopic here. Please go elsewhere to troll. This is a support channel. If you'd like help, please follow the instructions [18:55] Gay ass ubuntu [19:09] Hi [19:09] How does ppa repos works? [19:10] When I add them the repos seems to be non existent === tuxifreund_ is now known as tuxifreund [19:21] WOW. [19:23] I've opened firefox from the usb live iso [19:23] and it's practically unusable. [19:24] Guest89, okay? and? what do you need help with? [19:25] ubuntu is lagging slowing to a crawl and being unresponsive when I do any IO at all. [19:25] and I have no idea what I'm supposed to do to figure out why how to fix it [19:25] install it [19:26] performance is not to be expected from the live usb [19:27] Your PC could also be poo, too. just install it and see how it is after [19:30] Chunkyz. I'm not installing this yet because I need to backup/copy off the storage contents before installing over it [19:30] Chunkyz [19:30]  and it being USB has nothing to fucking do with the kernel/OS environment running now. [19:31] Guest89: then do your backup, forget about it's performance [19:31] also, false [19:31] You don't know what you're talking about. [19:31] and I've made live usb iso images before. [19:31] if it's screwing up, it's because the software has been designed to screw up which is stupid. [19:32] ok, file a bug because you aren't seeing the performance you would like from your usb live environment [19:32] !bug | Guest89 [19:32] Guest89: If you find a bug in Ubuntu or any of its official !flavors, please report it using the command « ubuntu-bug » - See https://help.ubuntu.com/community/ReportingBugs for other ways to report bugs. [19:32] Guest89: there's nothing to be fixed since it's a live, read-only environment [19:46] leftyfb. dude, you don't even know what's wrong, much less able to describe what needs to be fixed. [19:46] also, there's a tmpfs overlay used so I can install things and run things to see if it helps. [19:46] so far I don't have any useful metrics. [19:49] Guest89: no need to ping me *twice* [19:50] also, probably a crap HDD or something and you're blaming Ubuntu. [19:50] Could also be your USB, but no... it's Ubuntu...... [19:50] lol [20:00] it dont look like ubuntu 22.04 amd64desktop does not have support for AX201 wireless adapter while using "try ubuntu". is there something i can download to a storage device and install that after im at the desktop in 22.04 amd64 with "try ubuntu" in a live enviroment? [20:01] or a module i can dl to a storage device and manually load it without restarting? [20:09] trying to get my machine to use Wayland, but as I login the only that shows is the Ubuntu logo and nothing happens [20:09] s/the only/the only thing/ [20:10] this is a desktop with a nvidia gtx 1080, the nvidia driver is the 510 === b is now known as Guest9092 [20:18] @find eisenhorn [20:19] rasta, on devices with nvidia graphics wayland is disabled atm. the plan is to change it in 22.04.1 [20:20] ecryptfs is a total hack and should not be built into ubuntu at all. [20:20] I suggest everyone avoid user home directory encryption. [20:21] rasta, you can try to install backport-iwlwifi-dkms. your may have to load the module manually after installing it [20:22] ravage huh? is this for me? [20:22] why would I install an wifi dkms? [20:24] oops [20:24] no of course not [20:24] that was for en11gma [20:25] `ecryptfs-mount-private .Private`. returns "Encrypted private directory is not setup properly" [20:25] what the hell. [20:25] * Guest89 hates this crap software [20:27] Guest89, do you have an Ubuntu support question? if not please discuss in #ubuntu-discuss [20:28] Yes.  Why is ubuntu becoming unresponsive and unusable when I use rsync? [20:29] ( while running from a live usb ) [20:29] like, fine, I expect those storage devices io to become saturated, and maybe a thread to be used to 100%, but that leave 7 other threads for the Desktop environment to function fine. [20:29] leftyfb. that literally does not matter and if it does the software is built wrong. [20:29] or configured wrong. [20:29] that literally does matter [20:29] Guest89: this may help https://www.brendangregg.com/usemethod.html [20:29] i think that question has been answered multiple times. probably not even a live usb problem. try another live linux to compare or install Ubuntu [20:30] balls [20:30] ravage I can't install ubuntu because I have to copy the drive contents out first. [20:30] ravage and no, the fucking question has not been answered. [20:30] balls [20:30] balls [20:30] balls [20:30] balls [20:30] balls [20:30] Guest89: then copy the data and forget about performance [20:31] is the answer to the question "how do I view metrics that show me why the system is unresponsive". "install ubuntu overwriting all my data" actually an answer?  No the fuck it is not. [20:31] !language | Guest89 [20:31] Guest89: Please avoid any language that may be considered offensive, including acronyms and obfuscation of such - also see https://wiki.ubuntu.com/IRC/Guidelines || The main channels are English only, for other languages, see https://wiki.ubuntu.com/IRC/ChannelList [20:32] leftyfb. The last time you said that, I replied the system becomes unusable, slows down, uses power, shuts down, crashes. [20:32] leftyfb. Stop trolling and go away, you're not helping. [20:32] Guest89: actually, I do have an answer as I've run into a similar issue recently [20:32] You made a suggestion, I explained why you're wrong, and you're continuing to interfere with actual support.  Stop bothering me. [20:33] leftyfb. So you've encountered similar BS before and you didn't mention it for the last hour? [20:33] go away stupid troll. [20:33] Guest89: rsync is caching the entire filesystem you are backing up and is OOM'ing. rsync smaller sections [20:33] leftyfb. Hey look, that's actually a sane and reasonable suggestion that seems logically sound.     Is there a way I can limit/lower the kernel buffering rsync's file operations into ram? [20:34] Guest89: just rsync smaller sections [20:34] I don't need the kernel to buffer much rsync stuff, if I could limit that to like 500MB that issue should be solved. [20:34] Guest89  That's ludicrous to manually rsync sections out of thousands. [20:34] Guest89: for i in `ls /path/to/source/` ; do rsync -av $i/ /path/to/destination/$i/ ;done [20:35] ... [20:35] No, I'm not doing that. [20:35] leftyfb. There needs to be a way to measure ram build up / caching by rsync, and then limiting it so it doesn't happen. [20:35] you can just usw cp and run rsync afterwards [20:35] Not just guessing and doing random workarounds hoping something works. [20:35] *use === dostoevsky1 is now known as dostoevsky [20:45] I think ubuntu is mismanaging interupts or something === ssh0732 is now known as ssh073 [20:48] Hello, I have a problem with the network card on Ubuntu. I have a 1000Mbps internet line. On my notebook comes over lan 950Mbps complete. But on my computer only 300-400Mbps arrives, the computer has two network cards, but the power arrives on both cards. [20:49] The question is, how do measure that something is actually wrong? [20:50] rsync seems to be fine for awhile, and progressively gets worse, seemingly confirming possible OOM [20:50] but then OOM logs should be visible [20:50] ---- [20:50] separate question, what is "Window manager warning overwriting existing binding of keysym" [20:58] @Guest89. If you want to investigate the interupt performance there are tools, like sar, iostat. [21:00] even _cat /proc/interrupts_ === pong is now known as beaver [21:06] Guest89 It's completely locking up during massive file transfers yeah? [21:07] @mal [21:07] @mlag [21:07] omg [21:07] mlaga97. not massive, but many at least it seems reasonable to suspect oom, I need to check [21:08] You need to set vm.dirty_bytes and vm.dirty_background_bytes to something same using sysctl [21:08] Let me check my values [21:10] Try these commands to set them in the live environment: [21:10] sudo sysctl -w vm.dirty_bytes=214748365 [21:10] sudo sysctl -w vm.dirty_background_bytes = 107374182 [21:10] Guest89 [21:12] Can someone please help me. I am getting very confused. [21:13] Can anyone here guide me how to run ubuntu in server with only RAM? [21:13] Guest89 The issue is basically that those values are normally set as a percentage of system memory, and that leads to stupid numbers on modern systems with modern amounts of memory and result in memory filling up quickly with write cache data and then hitting a wall and stopping *literally everything* until the data is flushed to disk over the small write pipe. Manually setting them to much lower values prevents the problem. [21:14] N360, what network card in the desktop? [21:15] sealion, what underlying problem are you trying to solve by doing that [21:15] Broadcom BCM5751 [21:16] and Broadcom BCM5778 intern on Mainboard [21:16] I have server(co located) and connected to residential network. I want to run a true "no logs" VPN server. So whenever server start everything is reset. Each server has 128GB RAM [21:17] I also want to run tor relay [21:18] sealion I would probably do a normal install to disk, install all the needed software onto it, then set up fstab to mount / as ro and combine with an overlay mount to ramdisk [21:20] sealion keeping in mind that, depending on your threat model, that is not sufficiently protected from government agents busting your door down and freezing the ram with an upside-down can of compressed air [21:20] mlaga97 How do I combine / with an overlay mount to ramdisk [21:21] N360, this may be related: https://askubuntu.com/questions/1396344/very-slow-download-speed-on-ubuntu-20-04-with-a-wired-connection [21:21] mlaga97 Govt. busting through my door is not issue for me. I am not in USA. [21:22] I live in a place where Police is more corrupt than a harddisk attached to a magnet [21:22] sealion This isn't the only way, but it is a way: https://superuser.com/a/1445917 [21:23] And yeah, like I mentioned it definitely depends on your threat model, just be aware that not writing things to disk is a few leaps away from "totally secure" [21:23] Still an improvement, just want to ensure you know that there are still risks to consider [21:25] mlaga97 I will be grateful to you if you could suggest few other methods if you have one [21:27] Uh, so I don't have any particularly thought-out suggestions, just nitpicks lol [21:28] The answer uses a bunch of startup scripts, and I feel like it should be doable using nothing other than fstab configuration [21:29] But I don't actually have any evidence for that claim, just a vague and potentially incorrect memory of it being possible to do each of the components (ramdisk, read-only mount, and overlay mounts) in fstab [21:30] But I've never actually tried to combine them, so there may be some issue with the integration [21:31] mlaga97 OK I will try it [21:35] sealion.  The way linux works is halfway there.  the vmlinuz thing is basically a squashfs or similar that's loaded to ram.   you just need to instruct that another FS is copied to ram, and use that with the kernel [21:36] Guest89 Can you point me to any guide on how to do this? [21:38] sealion. this is basically how companies setup kiosks and set top boxes where users aren't supposed to change anything. [21:38] You have a partition that remains unchanged or a read only FS that bots [21:38] and you have something cycle the power every hour [21:38] or daily [21:39] sealion https://github.com/chesty/overlayroot perhaps [21:39] I was thinking of writing a cron job to restart every 6 hours [21:39] sealion.   You need to look at grub scripting and boot settings I think [21:39] or uboot [21:40] add a pir switch and put it in a closet [21:40] mlaga97.  I'm running into another problem making it difficult to test the OOM issue. [21:41] mlaga97. this damn user directory encryption. (ecryptfs) is terrible and now it's not working [21:41] "encrypted private directory is not setup properly" like wut [21:41] it was working last night. [21:41] overclucker, maybe sit a claymore pointing down onto the server case with a tripwire to the door for good luck === dostoevsky6 is now known as dostoevsky [21:43] Guest89, I'm missing a bit of context, but if you are trying to do an emergency backup I would suggest using ddrescue on the entire disk and sort the "getting usable data" part out later [21:43] mlaga97: thermite would be easier :) [21:44] overclucker depending on how much of either you might have available I suppose, supply-chain being as it is these days [21:44] mlaga97: just en case they get past the bamboo spike traps [21:54] Guest89, actually reading back to see what's going on, yes I believe that dd or ddrescue would be much more appropriate tools for what you are trying to do and you will need to use sysctl to set vm.dirty_bytes and vm.dirty_background_bytes in order to avoid the lockup due to the major mismatch in throughput between the source and target devices. === dostoevsky2 is now known as dostoevsky [22:08] mlaga97. not emergency backup, just copying files [22:08] for an upgrade [22:16] mlaga97. I'm worried it's not OOM, I would expect to see syslog entries for that. === dostoevsky8 is now known as dostoevsky [22:28] Because it isn't *technically* OOM [22:31] It's the disk write cache (dirty_bytes) filling up and causing issues. It's a known problem caused by legacy limit allocation algorithms on systems with fairly abundant ram making large transfers with a target disk that has much slower write speeds than the source disk has read speeds. [22:32] https://bbs.archlinux.org/viewtopic.php?pid=1832171#p1832171 [22:33] https://upcommons.upc.edu/bitstream/handle/2117/329616/main.pdf [22:33] https://archived.forum.manjaro.org/t/decrease-dirty-bytes-for-more-reliable-usb-transfer/62513 [22:33] https://lwn.net/Articles/572911/ [22:34] (link spam will now cease) [22:36] The last one (the lwn article) describes why the problem still exists 10 years after the issue was identified. === dostoevsky2 is now known as dostoevsky === dostoevsky3 is now known as dostoevsky [23:51] 22.04 looks to install the nvidia driver on setup and I see my gpu in ubuntu-drivers devices, iirc there used to be a command to query info from the gpu such as temperature and load, etc, does that still exist? [23:51] jiffe1, nvidia-settings ? [23:52] ah, no, "nvidia-smi" ... [23:53] yup that's what I'm looking for, thanks