[00:00] I understan it's not ubuntu fault [00:00] debian abandoned it for 'seucirty' reqasons. [00:00] So I guess it's time to move away from debian crap after 20 years. Sigh. [00:01] Not only does it take out the driver, it makes it harder to even put a binary driver in there. [00:02] works fine with ubuntu [00:02] I use binary drivers from nvidia all the time on my servers [00:03] because I care about performance, and I don't want to use the outdated drivers that sometimesship with ubuntu [00:08] Hash: i don't understand, what you are on about, first of all it's nvidia who puts you in this position, and then even blaming people who care about security. (which most linux users do). so if you want to build yourself an insecure one, find out how to install the driver. and I expect the same with other distros [00:08] additionally, nouveau works perfectly fine without any performance loss [00:09] It does not. [00:09] This is provable using games. [00:09] If you don't understand what I'm on about, the kindly take some time, think and ponder it over, and then get back with me. [00:09] I read what you typed [00:09] Nvidia isn't the one that removed the driver in favor of nouveau/security. [00:09] Hash: but ubuntu removed it because nvidia doesn't support it anymore [00:09] Debian did. Ubuntu sources from Debian. [00:10] "i need 340" <- what kernel ? [00:10] build it.. https://launchpad.net/ubuntu/jammy/+source/nvidia-graphics-drivers-340 [00:10] Ubuntu didn't actively remove it. It followed Debian. Debian removed it. [00:10] nvidia-340 doesnt work with jammy or hwe kernel on old os [00:10] and make a dkms [00:10] you needto use the weird version [00:10] patches version i mean [00:10] i use it, lemee find the ppa [00:10] I might have to go back down to 20.04 [00:10] driver ppa does not have it. [00:10] enigma9o7[m]: thanks [00:11] its needs a patch from kerbek 333 or sometone [00:11] Hash: everybody tells you that nvidia is at fault, and seems like you still don't believe it [00:11] sorry i drunk but ifind it just a min [00:11] murmel: everyone could be wrong. [00:11] And I could be right. [00:11] What exactly is nvidia's fault? [00:12] enigma9o7[m]: 🍻 [00:12] Can you quantify? All you said was, it's theri fault, it's their fault, without quantifying anything.. [00:12] Hash: file a bug and move on. Or don't and move on. Good luck [00:12] some hardware get phased out.. [00:12] enigma9o7[m]: https://launchpad.net/~kelebek333/+archive/ubuntu/nvidia-legacy [00:12] Hash: nvidia doesn't support newer kernel versions, so how can you imagine should debian or ubuntu support that version? yes there are hacks to get it working [00:13] https://launchpad.net/%7Ekelebek333/+archive/ubuntu/nvidia-legacy [00:13] but as it implies, you don't know how long that will work (as a new kernel releases) [00:13] tomreyn: that's got "trust" written all over it :) [00:13] Yeah tom beat me damnit. [00:14] *sigh* [00:14] leftyfb: i did not mean to recommend anything ;) [00:14] For those of us with old gpu like that tho, its either use ppa, or use noveau. [00:14] I mean it'll be great if ubuntu were to support it but they're not. [00:14] You know what's even more infuriating or addingg insult to injury, when stuff like this happens, and the whole of the population jumps down a person's throat. [00:14] Use a new GPU! Nope. Out of the question. [00:15] Use nouveau, I can't because the performance objectively is terrible. [00:15] --my-next-gpu-wont-be-nvidia [00:15] It works for somethings but not for others. [00:15] xorg-modulepath-fix ... oh, you are bound to xorg, no wayland [00:15] enigma9o7[m]: amd has same issues. [00:15] These are not solutions to problems [00:15] Hash: file a bug and move on. Or don't and move on. Good luck [00:15] Oh, I'll just shun it. Or, I'll just do something else. [00:15] Yeah noveau sucks for me too btw. [00:15] i run an amd 5450, fine... [00:15] lolz [00:15] I use nvidia-340 [00:15] Client: HexChat 2.16.0 • OS: Ubuntu "jammy" 22.04 • CPU: Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz (800MHz) • Memory: Physical: 22,9 GiB Total (19,7 GiB Free) Swap: 2,0 GiB Total (2,0 GiB Free) • Storage: 1,0 TB / 4,1 TB (3,1 TB Free) • VGA: Advanced Micro Devices, Inc. [AMD/ATI] Cedar [Radeon HD 5000/6000/7350/8350 Series] @ Intel Corporation 4th Gen Core Processor DRAM Controller • Uptime: 2d 8h 38m 58s [00:16] That's nouveau in 22 [00:16] I might have to downgrade to 20 [00:16] install a 22 chroot/container for application [00:16] Hash: there is no downgrade path, only a fresh install of a previous version [00:16] I don't know what else to do. [00:16] Hash: I told you what to do [00:16] For 20 years in Debian I used stable, but with testing/sid chroots for package buldilng and latest software using [00:16] get the drivers from nvidia [00:17] i thought you didn't want to use any debian based distro anymore as they don't support 340 anymore [00:17] man it's the firs ttim eI sen your nickname! [00:17] wow. [00:17] they provide binary drivers in all supported versions [00:17] dont do that [00:17] yeah don't do that Hash [00:17] You already did that, Hash, and it didn't work. [00:17] Hash: file a bug and move on. Or don't and move on. Good luck [00:17] i my laptop is Nvidia GT216M [GeForce Gt 330m} so 340.108 works well for me. [00:17] 20 years... but that card is 13 years old [00:17] dont install from nividia script no matter what you do, you'll cause yourself lotsa hassle [00:18] What I'll do is install ubuntu 20 in a diff disk [00:18] boot that, use nvidia, and record you a vide of game performance [00:18] and it wont work anyway [00:18] nivdida ofoficial only supports up to linux 5.4 [00:18] Oh man, dude, homie bbro [00:18] diggity dude man sir home slice. I need nvidia or else my gaming suffers. [00:18] !ot | Hash [00:18] Hash: #ubuntu is the Ubuntu support channel, for all Ubuntu-related support questions. Please register with NickServ (see /msg ubottu !register) and use #ubuntu-offtopic for other topics (though our !guidelines apply there too). Thanks! [00:19] noise in the channel [00:19] I'm going to ignore th ebot [00:19] nvidia binary drivers are binary drivers. Only nvidia can maintain them properly [00:19] Whoever did that, you're on ignore. Okay> [00:19] The bot has useful info.... [00:19] You don't get to tuse the bot to harass me as I ignored you. Dont' do that again. [00:19] Hash: Pretty sure you just ignored an operator. [00:19] I understand. Don't do it again. [00:19] if nvidia doesn't support newer kernels, then no one else can properly do it either. [00:19] I don't care. I don't appreciate random harassment. [00:19] Please stop. [00:20] I really wonder what you do if he does it again [00:20] I'm already frustrated with an problem and I don't need people adding insult to injury and antagoniging me even more. [00:20] I told you, I will just ignore the bot. [00:20] anyway, this is not the dicussion I am trying to have. [00:20] What is the problem? [00:20] Nevermind, I was wrong. Sorry. [00:20] You've already been provided with info on the PPA to use nvidia-340 with current ubuntu, if you want to. [00:20] Yes, its not officially supportecd by canoncical, but it works, and your other option is noveau [00:20] I need to figure out the nvidia driver for 22 or else downgrade to 20 [00:20] Hash: so you got all the options laid out. decide for yourself [00:20] It was linked earlier by me and tom. [00:21] enigma9o7[m]: so much traffic going fast I did not see the ppa [00:22] I can't find it in the log up there. [00:22] 3~ [00:22] This is what happens when instead of helping, 120 people jump to criticize you. [00:22] https://launchpad.net/%7Ekelebek333/+archive/ubuntu/nvidia-legacy [00:22] I have no idea where that ppa link is anymore, my IRC buffer isn't that large. [00:22] but it won't matter, they have me on ignore :) [00:23] leftyfb: Hash: PPA [00:23] https://launchpad.net/%7Ekelebek333/+archive/ubuntu/nvidia-legacy [00:23] Tyank you [00:24] !logs > Hash [00:24] Hash: Please see my private message [00:24] Ty [00:24] enigma9o7[m] also has me on ignore :) [00:24] Oh that's useful! [00:27] I'm now confused, isn't ubottu a bot? [00:27] yes, officially [00:28] it is, but you have me on ignore so you didn't see me trigger it. Or this message :) [00:28] enigma9o7: I'm guessing tomreyn used a special variant of the bot command to make the info go to a private message. [00:28] https://new.ubottu.com/factoids.cgi?db=ubuntu3&search=&order=added%7CASC&page=24 [00:29] Yes it is. But if I ignore someone, and then they use the bot to harass me in any manner, it's harassment. They were ignored and I don't want to speak to them or hear from them. If they circumvent the /ignore by using the bot to talk to me, I will happily ignore the bot. If this results in a ban, I'll find another distro. I will simply not allow anyone to cross my boundaries. Remain civil and [00:29] helpful or else no need to try to support anyone on IRC. [00:30] your attitude triggered a volunteer to give you a hint. if you feel harrassed, join #ubuntu-ops [00:30] oerheks: careful, he'll put you on ignore too [00:58] hello, is this the support/ask dumb questions channel, or is there another channel for this [00:59] feel free to ask dumb questions here :) [01:00] I'm running Ubuntu 18.04 on a headless media server. have a static IP and duckDNS set up so I can access it remotely. Always works fine, until I go to reboot. After I reboot, I can't access it remotely using my custom domain, I can only access it using the local IP address. Once I ssh in, I type "sudo netplan apply" and things go back to normal [01:00] until the next reboot. [01:01] My guess is I looked up multiple ways to configure static IP / remote access when I first set it up forever ago, and there's something that conflicts with Netplan that I don't use anymore, but I can't remember what it is! [01:01] any ideas you might have about what conf files to look at, etc would be much appreciated [01:02] Guest92: how many files do you have in /etc/netplan/ ? [01:03] just one! [01:04] Guest92: I don't want to distract from leftyfb's efforts here, so just store this url aside for a bit, and come back to it if you're still stuck and out of ideas later on https://systemd.io/NETWORK_ONLINE/ [01:05] (it's guessing at a cause way before a cause has been determined.. bad sarnold :) [01:05] Guest92: can you post the contents of the file in /etc/netplan/ ? [01:06] Guest92: is/was this machine running a desktop environment? [01:07] sarnold thanks! [01:07] leftyfb yes give me a moment, there isn't much there. Also no, it's headless so just installed 18.04 server [01:08] network: [01:08] version: 2 [01:08] renderer: networkd [01:08] ethernets: [01:08] enp7s0f1: [01:08] dhcp4: true [01:08] !paste | Guest92 [01:08] Guest92: 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. [01:10] Guest92: cat /etc/netplan/* | nc termbin.com 9999 === raczka4 is now known as raczka [01:14] https://paste.ubuntu.com/p/nBTkcpkscF/ [01:15] leftyfb there you go! [01:15] Guest92: that's not a static ip [01:15] oh I know, it still has the same ip every time I reboot though [01:16] the ip address isn't the issue, but I just have no idea what is or why netplan apply fixes it [01:16] define "access it" [01:16] sorry, ssh [01:16] after reboot, ssh user@192.xxx.xxx.xx works but ssh user@domain.duckdns.org does not [01:17] Guest92: do you have port forwarding on your router? [01:17] after "sudo netplan apply", back to normal [01:17] yes port 22 is forwarded properly [01:18] Guest92: This may be a hacky solution, and hopefully you'll find something better, but if all else fails, you could write a custom systemd unit that will run a Bash script that runs "netplan apply" for you. [01:19] That would make it work on every boot. And if it doesn't work, just add a "sleep " line before the "netplan apply" to delay execution until the right time. [01:19] that is only masking the problem, not fixing it [01:19] something doesn't add up here [01:19] leftyfb: True. Like I said, hopefully they'll find a better solution. Just throwing out an emergency "make things work if nothing else does" out there. [01:19] yah I thought of that too! but I figured I'd ask about it first in case it's a really simple thing that I conf'd wrong way back when and someone would just say "look here over here dummy" [01:20] Guest92: can you verify right now by looking at the ip of the machine in question, verify the ip your router is port forwarding to, reboot and verify the ip of the machine in question again and see if you can ssh via the dns hostname [01:22] yep! so currently, local IP is set to xxx.xxx.xxx.99, as it should be. I'll reboot and report back ' [01:23] meantime, can also confirm that the router has a IP reservation for xxx.xxx.xxx.99 set, with ports 22 + 2 others forwarded. [01:25] now, after reboot, "cannot connect to host domain.duckdns.org port 22: operation timed out" [01:25] This might be silly, but maybe this is a race condition? Is it possible some component of networking isn't getting initialized until after Netplan tries to do its magic, thus requiring a "sudo netplan apply" to try again after the interface comes up? (I don't actually know how Netplan works, so this might be totally off, but I ran into a sorta similar problem some time ago.) [01:25] "ssh user@xxx.xxx.xxx.99" does work and let me log on [01:25] arraybolt3[m]: ssh works fine, I do not think this is an ubuntu issue at all [01:25] I type "sudo netplan apply", then logout, and when I log back on, I can ssh using the domain again [01:26] Guest92: I'm pretty sure this is an issue with your router and some weird state it's getting in with your machine going "offline" and doing a soft reboot [01:27] Guest92: netplan apply is just re-initializing the networking [01:27] leftyfb: Right, but my thinking is, that maybe this is happening: 1: System boots. 2: Netplan tries to apply too early. 3: Interface comes up. 4: User tries to SSH over DNS, fails. 5: Users tries to SSH with IP, works. 6: User runs "sudo netplan apply" *after* the interface is up. 7: Now everything works. [01:27] basically renewing your dhcp lease [01:27] router keeps a lease.. [01:28] arraybolt3[m]: it's not really "SSH over DNS", it's "SSH over port forward" that's the issue [01:28] SSH is working perfectly [01:28] locally [01:28] right [01:28] other than that, that's not an ubuntu issue, that's a router issue [01:28] leftyfb: I know, sorry for bad terminology. oerheks: Oh, that's right, I didn't think about that. [01:29] leftyfb: I get it. [01:29] again, it's the router getting into a weird state of "losing" the client when it reboots, but only "finds" it again after you renew your lease after the link is up again [01:29] I've not been following along.. is the dns giving an externally-reachable IP address, and the router won't let an internal host use that IP / port to forward to the machine in question? [01:29] I know I'm not the expert here, but I'm pretty confident the router's not the issue. I can only say that because I've had this problem for almost a year (I've just been ignoring it), and in that time I've had 2 modems and 2 routers. and the router sees the server just fine after the reboot. [01:29] put that ip in a mac filter, if your router gives such option [01:30] Guest92: the fact that ssh works perfectly means it's not an ubuntu issue. Your only issue is port forwarding over your router [01:31] Guest92: ubuntu has nothing to do with ssh coming over you router [01:31] Guest92: Were you using two different routers of different brands? [01:31] Guest92: Or the same brand? [01:31] sorry I should have clarified that, yes [01:32] Guest92: you know another test you can do..... reboot your machine, confirm you cannot ssh in via the port forward, then reboor your router. I bet it works [01:32] what other services has ubuntu used, currently or in the past, to do/configure what netplan does? I'm wondering if I have a conflicting config somewhere that I set up but don't know about, which is only overridden when netplan is applied again [01:33] leftyfb that's a good idea, I'll try that now [01:33] Guest92: netplan and ssh are working perfectly [01:33] there is nothing wrong with netplan, ssh or anything to do with the network (software-wise) on your machine [01:38] Guest92: After you finish leftyfb's suggestion, there's one other thing I'd be curious to see. Pastebin /etc/resolv.conf after a fresh reboot, then do the netplan apply and pastebin that file again. [01:39] Guest92: I notice in there you're specifying a particular nameserver, but dhcp would probably also be giving you a nameserver. I assume these will be different. [01:39] InPhase: that won't make a difference when port forwarding from the internet through the router to a local ip [01:40] InPhase: resolv.conf is only for the router to resolve hostnames/DNS, not for incoming traffic from the router [01:40] leftyfb: I believe sshd does a reverse dns lookup. [01:41] leftyfb: And connecting from a localhost vs an external ip is functionally a different behavior under that, with potentially two different resolvers. [01:41] s/localhost/192.168 subnet/ [01:41] I'm back! still having the same issue; router sees the server on the network, IP forwarding is still correct etc [01:41] Only for logging purposes or if you have hostname restrictions in the public key [01:42] contents of /etc/resolv.conf is exactly the same before/after [01:42] leftyfb: Yeah, there would be other digging needed. But seems that's not a difference then. [01:44] Guest92: test it while booted to a live usb (install ssh and setup your public ssh key) [01:44] I definitely remember messing around with multiple conf's when I was setting this thing up originally. So I know my netplan conf is correct, I know ssh isn't the issue etc etc I'm just pretty sure that there's another config somewhere that I set up, didn't need, and have forgotten about, that's conflicting somehow [01:44] @left [01:44] how does your netplan looks like now? [01:44] static ip and dhcp4 no ? [01:45] leftyfb I did set up another server to xxx.xxx.xxx.88, and set it up otherwise identically, and it's never had these issues. [01:45] oerheks https://paste.ubuntu.com/p/nBTkcpkscF/ [01:45] Same physical machine? [01:46] yes, just different boot drive [01:49] Guest92: If you're feeling adventurous and confident in "otherwise identically", you could do a diff -r between the /etc directories of the two drives and go on a hunting expedition for what is different that has anything to do with networking or ssh. [01:51] well it's obviously not "identical" otherwise, because of this thing I'm almost certain I'm missing / have forgotten about, but yes I take your point! that would be a tedious but probably effective way of finding it [02:01] intel micro code update.. - CVE-2022-21151, INTEL-SA-00617 - CVE-2021-0146, INTEL-SA-00528 - CVE-2021-0127, INTEL-SA-00532 [02:01] Processor optimization removal or modification of security-critical code for some Intel(R) Processors may allow an authenticated user to potentially enable information disclosure via local access. [02:01] Hardware allows activation of test or debug logic at runtime for some Intel(R) processors which may allow an unauthenticated user to potentially enable escalation of privilege via physical access. [02:02] Insufficient control flow management in some Intel(R) Processors may allow an authenticated user to potentially enable a denial of service via local access. [02:02] thanks for everything, folks! off to do some diff-ing === ForrestGumppp is now known as MeghanT === Kuraokami is now known as Guest5120 [04:07] is there any way to launch tasksel without it launching the ncurses gui? [04:28] yeah, command line [04:28] pass params too it [04:28] negative. it always launches ncurses [04:28] and there is no flag to stop it [04:28] i.e. tasksel install lxde [04:28] yeah yeah i know [04:28] that does nothing [04:29] it still launches an ncurses gui with a progress bar [04:29] i see i see [04:30] so just sudo apt install stuff [04:30] instead of tasksel [04:30] it does the same thing [04:30] hmmm, i guess i could get the package list from list-packages and pass that [04:30] huh. i like it. thanks! [04:30] theres a meta package [04:30] for the full ubuntu desktop? [04:30] I have sda5 456G Used 145G - I want to partition the drive to install another ubuntu boot partition (22.04) and a data partiiton. I fear that my data is spread out all over the drive and will be overwritten. Help me understand hot to do it. [04:31] yeah, that one is tricky, its called ubuntu-desktop [04:31] yeah, that's why i was messing with tasksel install ubuntu-desktop [04:31] mate had been a mess and fail on 2 different computers, one is an HP laptop. [04:32] tripleb, i dunno for sure, i only know what i'd try if i wanted to do that; i'd boot a live iso and run gparted [04:32] That's like making a left turn without looking. I need more advice from those who may know. [04:33] shrink the partition, make a new one. [04:33] or two [04:33] enigma, I am not going to just try. I heard that ext4 puts important things at the high end of the partition. [04:33] trial & error is fun [04:33] thats how I learn [04:33] go right ahead. [04:34] or just go left without looking ;) [04:35] But thanks your giving me your point of view. enigma9o7[m] - I shat sleep on it. [04:40] well, funny enough `tasksel --task-packages ubuntu-desktop` just outputs ubuntu-desktop^ [04:40] either way, `apt install -y $(tasksel --task-packages ubuntu-desktop)` seems good enough === Guest3189_ is now known as theciscokid [04:45] anyone in here familiar with the foot terminal? [04:47] I have question about a package for Ubuntu. I wrote a puzzle game engine called Free Hero Mesh, and someone might want to include it in the Ubuntu package manager (and possibly also Debian, Devuan, etc). It requires Node.js to build only if the code has been modified; if you are compiling unmodified code then it is not needed. [04:49] Additionally, it does not use Unicode (I will not fix this; however, I do intend to add support for more code pages, including multibyte code pages (e.g. EUC-JP) in future), and the man page uses the .so and .ds commands. [04:50] If you want package yourself, a good place to publish is launchpad. Helps you figur eout how to make the source package and stuff. [04:50] Its not trivial either, packaging is pita. [04:50] Until you understand it. [04:52] Dependencies for run time are SDL1.x and SQLite. (However, ensure that the version of SQLite in the package manager is not too old; on my computer it is, so I had to compile SQLite myself.) [04:52] After that try to get it in debian. [04:52] Is it acceptable for a package to have man pages with the .so command and/or .ds command? [04:55] (If necessary, I believe it is possible to convert any man page that uses these commands to one that does not use them.) [04:58] Note also that I do not use Bazaar; I am using Fossil instead. (I am not sure how easy it is to make them interoperable; I will read the Bazaar documentation and try to figure out if I can.) [05:00] zzo38: you would best be going to debian, getting it there "upstreamed" and then it will be automatically pulled into ubuntu. Problem being, if you don't package it yourself, it's very unlikely that somebody will do it [05:02] Fossil identifies files by their SHA-1 hash (new versions also support SHA3-256). If Bazaar can also identify files by their SHA-1 hash (without any extra header added like that which Git adds), then it should be possible to transfer files between them, I hope. [05:02] I did ask in Debian, and may try again later to ask, but it is unsure that I would package it by myself. (I asked in Debian more than a year ago though, I think; not recently) [05:03] zzo38: if you want to formally apply for a package, you would need to send a RFP (request for package) to their mailing list [05:08] Hi all, silly question; I've never set up my server to back up before. going to try upgrading to the new LTS, but wanted to have a backup before I do. I looked at various backup utilities, but restoring from them doesn't look very straightforward if I end up needing to. I'd rather just have a working clone of the boot volume. Is this possible? [05:08] Would it really just be as simple as setting up rsync to copy everything from / ? and then in the event that the primary boot volume is hosed, just swap the drives and carry on? [05:11] Guest11: if you want something gui like, you can also use clonezilla to clone the whole disk (so you can restore the whole disk if something goes wrong) [05:11] I'm not running a gui [05:12] Guest11: I am talking about clonezilla (it's a live iso) which means you boot it from a usb stick best case [05:12] What I had done is using "tar --anchored -c -v -S --numeric-owner --one-file-system -p" and pipe that to gzip and store the file on a USB flash drive, and then record the files onto one or more DVDs. [05:12] which gives you a gui to clone the whole disk [05:12] zzo38: idk about you, but my data is bigger than a usb stick ;) [05:13] That is how I did backups on my computer. I was able to restore the files onto a new hard drive after that; after that the only thing I needed to do is to restore the boot loader and fix the UUID of the file system in the boot loader configuration file. [05:14] murmel: OK. I suppose a lot of people have more files than I have on my computer. (The files on my computer will fit on three or four DVDs after they are compressed.) [05:14] zzo38: uh you burn dvds? for backups? ^^ [05:15] (Also, you can use more than one USB stick; you do not need to use only one. However, you must figure out how to best split up the directories so that they will fit.) [05:15] zzo38 so this is kind of what I mean... rather than "restore the files onto a new hard drive" as you said... I'd rather just ... already have them on a drive? then in the event of an issue, I just install that drive, rip out the old one, and carry on. putting it onto a stick, and compressing it, etc just means more work and more downtime in the [05:15] event of an issue, no? [05:15] murmel: Yes; I insist to use WORM (write once, read many) media for backups. (Or at least, I would do, but my DVD recorder does not seem to be working. I still ahve the backup on USB, though.) === theciscokid is now known as polo [05:16] If you can connect two hard drives to your computer at once, then yes you can do that, but I still like to keep separate backups (on WORM media, as I said), which is why I did not do that. [05:17] it's a media server with several drive bays and usb ports. I don't need frequent backups, I probably wouldn't even update the clone more than once every few weeks. minimal downtime is the key factor in my case. [05:18] Guest11: eh sounds like then, that you do want to look into rsync, and backup your config files extra (so you don't have to backup the system itself) [05:19] murmel can you elaborate on that? what do you mean by not having to backup the system itself === drkokandy_ is now known as drkokandy [05:19] Guest11: nowadays most people just backup their config files and have an automated install or just reinstall manually. so they don't have to fiddle with efi and/or UUID for disks etc [05:20] (basically you would need to fix booting the system) [05:20] building your own automated install, should take you max 1 hr for a basic one [05:22] I always backup the entire system, so it is not necessary to reinstall anything other than the boot loader. [05:23] zzo38: not everybody wants to fiddle with the bootloader (as the install is easy to do) [05:23] (The boot loader configuration is included in the backup, like everything else, but the boot loader itself isn't.) === polo is now known as Guest6607 [05:24] so this is kind of what I'm getting at.... what you're describing sounds like.. more work? so I'm wondering what I'm missing. from what I can tell, your method is: 1) reinstall OS onto new drive first, then 2) use restore utility or scripts to restore config files to where they were, and presumably reinstall programs, etc. . Whereas my thinking is [05:24] 1) install drive clone into same drive bay as original disk 2) boot up, already good to go since it has *everything* [05:25] in my case the system is an old Mac Pro tower. so I think boot loader stuff etc is all handled by Mac efi anyways. it just looks for a bootable volume in the front drive bay, and works its way down from there [05:26] Guest11: you would still need to edit UUIDs, the only exception afaik is if you use dd to clone the drive (which also copies over zero space) [05:26] so it may take longer, but you can boot from it (afair) [05:30] I do not reinstall the OS at all; the OS is part of the backup and can be restored. Only the boot loader needs to be reinstalled; the OS is already there. [05:30] thanks for the info. sorry for the dumb question but why would I need to edit the uuid? I do get that each drive will have a different one, I'm just not sure what its role is. my thinking was just that the system, on boot, would look for a bootable volume in drive bay 1, find it, and boot to it. why would it care that the volume has a different [05:30] uuid if it's bootable? [05:31] If you are cloning the disk, then you do not need to change the UUID. However, if you are making tape archive files like I do, then the new UUID will be different and you will need to fix it. [05:31] (Maybe it is different for UEFI; I don't know. I do not use UEFI and I do not like UEFI.) [05:32] this ^, as I said, if you dd the disk no need for changing UUIDs [05:32] (You also do not need to reinstall the boot loader if you are cloning the disk, since that will copy the boot loader too) [05:32] zzo38: I guess you have a nice future then === themill_ is now known as themill [05:35] zzo38: just pointing out, as intel already announced to disable features if you choose legacy boot (which won't be supported forever, even by bios companies) [05:48] Another thing about Free Hero Mesh: One of the files included (misc/har.c) possibly should be in a separate package. It is not required by Free Hero Mesh, but is useful with it; it is also useful with OHRRPGCE and possibly other packages (none of which require it). Additionally, an implementation of farbfeld is required to use the picture import/export function of Free Hero Mesh (it is not required if you do not use this feature). [05:48] How do you deal with the package requirements that I mentioned (including some that I mentioned earlier, about Node.js, man pages, etc)? [05:57] zzo38: well idk about manpages, but the rest shouldn't be a huge problem. (as for example node.js shouldn't be a huge issue, as nobody would change the game) [05:58] sqlite is kinda a bummer, but I guess as long as you don't push the versions too fast, it's not that of a huge problem as the sid/testing branch would reletively fast change to the new verison [06:44] hi, any advice how I could increasemy disk space in EFI? I have unallocated space available [06:46] is the unallocated space available directly next to the efi boot partition? [06:46] if not forget about increase it [06:46] how can I check it? [06:47] do you have gparted installed? check in the menu [06:50] gpkumar it looks like this, i have dualboot https://imgur.com/a/HYUddPW [06:51] direct link: https://i.imgur.com/Rhbp4O7.png [06:51] no hope? [06:51] it is not to the next of the first partition [06:52] you can do it by moving every partition, but I highly discourage you to do that [06:52] it will most likely result in errors? [06:53] what it EFI partition moved to unallocated one? [06:53] not errors, but takes lot of time and there are chances of data loss [06:53] or doesnt work like that [06:53] nah it doesn't work like that [06:53] damn thought ssd nvmes are different [07:33] how do i resize the default swap file on 22.04? is it just swapoff -a; truncate -s 10G /swap.img; swapon -a; or is there something more to it? [07:36] hans https://linuxhint.com/change_swap_size_ubuntu/ [07:44] One stupid question: where could I buy them (the hardcopy)? https://pasteboard.co/NzxIILRNB8le.jpg [07:45] asarch: that's offtopic here [07:46] #ubuntu-offtopic would be a better place to ask this :) === atol-71 is now known as atol71 [08:05] Thank you [08:05] And sorry for ask [08:09] no problem [09:06] hey [09:17] hans_: i think your approach should work fine (assuming "/swap.img" is in /etc/fstab as such). [09:18] you may need to "mkswap /swap.img" also (before mounting it), i'm unsure whether this will happen automatically. [09:44] is there an Ubuntu alternative to Debian's "python-certbot-nginx" package ? [09:45] lel yeah, it's called "python3-certbot-nginx" [09:45] !info python3-certbot-nginx [09:45] python3-certbot-nginx (1.21.0-1, jammy): Nginx plugin for Certbot. In component universe, is optional. Built by python-certbot-nginx. Size 35 kB / 158 kB [10:48] Hi, all. Just rebooted my Ubuntu and discovered that no sites load in browsers. I get a "check your proxy settings" error message. As far as I know, I do not use a proxy. At the same time, I get a successful ping to Google from command line, IRC working and messengers working. How can I troubleshoot this? [10:48] manwhowouldbekin: other network related things still work? can you sudo apt update for a test? [10:49] lotuspsychje, Yes, they do. I have actually tried that very command for a test myself. It worked. [10:55] manwhowouldbekin: wich ubuntu release and browser(s) version did you try? [10:56] lotuspsychje, I am on Ubuntu 22.04. I have tried Vivaldi (Chromium) and Firefox. [10:57] pretty weird, are you behind a router or firewall manwhowouldbekin [10:57] lotuspsychje, I am behind a router. I cannot even get to that router's IP address from the browser. It gives the same error. Rebooting didn't help either. [10:58] manwhowouldbekin: did you do anything special before this started to happen? recent updates or anything else? [10:59] lotuspsychje, Not that I remember. I typically suspend the machine, so a week or two can pass between reboots. I was messing with OpenVPN at some point (trying to setup client). I do think there were some updates during this time period as well. [11:00] messing with vpn could result in unwanted configs perhaps? [11:01] lotuspsychje, Possible. I was going off tutorials mostly. [11:02] try to undo what you did on vpn, see if you can revert things [11:02] if it was updates/network related, i dont think apt would still work right [11:05] lotuspsychje, Sounds reasonable. I honestly do not know where to begin. Will google around. Need to sign off now, but might return later with same question. Thanks! [11:05] !vpn [11:05] For more information on vpn please refer to https://wiki.ubuntu.com/VPN [11:06] manwhowouldbekin: you could also trace your /var/log/dpkg logs to see what recently happened to your system [11:06] perhaps also leave a; journalctl -f running while you seek your problem [11:07] turning off network/ turn back on could give error clues too [11:38] Hi all [11:41] Hi BluesKaj [11:43] hi tomreyn [11:43] .o/ [11:52] Does the ubuntu 20.04 on WSL2 have somehow fewer packages than the proper version? [11:53] no [11:55] So you can install libtinfo5 on WSL2 focal? [11:55] !info libtinfo5 [11:55] libtinfo5 (6.3-2, jammy): shared low-level terminfo library (legacy version). In component universe, is optional. Built by ncurses. Size 97 kB / 540 kB [11:57] !info libtinfo5 focal [11:57] libtinfo5 (6.2-0ubuntu2, focal): shared low-level terminfo library (legacy version). In component universe, is optional. Built by ncurses. Size 82 kB / 522 kB [11:57] yes you can [11:59] Fullscreen YouTube-videos and others now no longer prevent my screen from sleeping. This happened after updating from 20.04 to 21.10 - I'm using i3. How do I go about fixing this? [11:59] (Screen falls asleep after 10 minutes) [12:01] ravage: then I wonder why users report they can't https://gitlab.haskell.org/haskell/ghcup-hs/-/issues/379 [12:01] Issue 379 in haskell/ghcup-hs "ghcup asked me to install libtinfo5 but it's not available (WSL2, Ubuntu 20.04)" [Opened] [12:04] ravage: https://gitlab.haskell.org/haskell/ghcup-hs/-/issues/379 [12:04] Issue 379 in haskell/ghcup-hs "ghcup asked me to install libtinfo5 but it's not available (WSL2, Ubuntu 20.04)" [Opened] [12:04]