[00:00] (I've gotten better at this over the years, thankfully.) [00:00] Experience is proportionate to data destroyed. [00:01] jhutchins: LOL If only that were a joke... (totaldelete C:\windows\symlink\to\sourcecode comes to mind... oh that was such a mess, careful with symlinks on Windows!) [00:01] Boomer4582: Still doing my test, but I'm making progress. [00:02] All right, I am waiting on the next commands. [00:03] Alright, simulated the mess in a VM, now I can make sure my fix works. [00:06] Almost done... [00:12] OK, now we're just waiting on my Internet, which should take about less than 4 minutes, then I'll be able to reboot and see if it works. [00:13] I disconnected,  sorry arraybolt3[m] [00:13] what's the next command please? [00:13] Boomer4582: For in the future, Ubuntu will automatically remove unused kernels as a normal part of the update process. You will notice two kernels in there - that's on purpose, so that if something goes wrong after an update, you have a fallback. [00:14] That's fine. I'm still working on the test (about a minute left on my download), if this works I'll tell you the commands, if not, then you'll be glad I didn't tell them to you! [00:14] All right [00:16] Is it possible to run my INSTALLED apps when I have chrooted into my OS using a live OS USB? [00:16] such as my Bitpay installed app from Snap [00:16] Boomer4582: Sorta, but not really. [00:16] Boomer4582: Also, don't worry about payment - we're all here because we love doing tech support. [00:16] The whole point of Ubuntu is that it's really good and free. [00:16] Good people still exist huh [00:17] OK, here we go! Looks like it's working! [00:17] !cookie | arraybolt3[m] [00:17] arraybolt3[m]: Wow! You're such a great helper, you deserve a cookie! [00:17] oerheks: LOL [00:17] oh boy, as long as it doesn't break my encryption and possibly losing my crypto resulting in divorce, it's fine :') [00:17] Oy. Well, it almost worked... [00:18] almost? [00:18] I can boot, but it's not giving me a prompt for the encryption key. I obviously missed something, but what, I'm not sure. Glad I didn't tell you what I did! [00:19] oh Goodness [00:19] can't I just decrypt the partition and do this using a decrypted partition? [00:19] wouldn't that be easier? [00:19] Boomer4582: I don't know how, but I doubt if that's possible. [00:21] do we try something else arraybolt3[m]? [00:23] Install the Linux kernel: apt-get install linux-image-generic (no sudo required as you are root after a chroot) [00:23] wouldn't this work? [00:24] That's what I did, and it's what got me here. [00:25] Boomer4582: Can you do "cat /boot/grub/grub.cfg | pbput", and then give me the URL? [00:25] This will let me see what your grub.cfg file looks like. [00:26] Boomer4582: arraybolt3[m]: Might I suggest ' apt install --reinstall linux-image-generic ' ? should workie IF the partition is now Un-encrypted in a change root environment. [00:26] Bashing-om: The problem is that is updates something in a weird way, I think. I don't get a passphrase prompt. [00:26] "cat /boot/grub/grub.cfg | pbput" [00:26] 1minute, sorry [00:28] https://paste.ubuntu.com/p/rps5Dz8Jgg/ [00:28] Writing from other account on the machine, I had to install pastebinit [00:29] Guest45: OK. [00:29] https://ibb.co/kQ3vdFn  this is what installing pastebinit did [00:29] strange [00:30] that normal? [00:31] Guest45: Weird, it looks like it reinstalled the kernel for you. [00:31] yes.. shall I reboot? [00:31] It doesn't look like it worked. [00:32] cryptsetup: WARNING: target 'luks-886b72e4-c13e-4998-8956-21122ca4bef2' not [00:32]     found in /etc/crypttab [00:32] it also gave this :( [00:32] Hmm... Yeah, that's what I got too. [00:32] I hope it didnt break something [00:32] Let me reboot [00:33] Well, you're backing up all your files, right? [00:33] I think I did.. [00:33] I'm still able to access my encrypted drive even after this, so worst case scenario, you should be able to back up everything and then reinstall. [00:35] Nope [00:35] wont boot [00:36] Are you at a BusyBox screen? [00:37] https://ibb.co/XWKkMdg [00:37] God Damn [00:37] Hey wait, boot up from your live USB again. [00:37] doing [00:41] USB boot up stuck [00:41] ok it works now [00:42] Boomer4582: OK, do the commands that got you into the encrypted drive earlier. [00:42] Boomer4582: You probably can't see those now... [00:42] Boomer4582: OK, open the encrypted drive again. [00:46] Decrypted it, worked. Did all the neccessary commands, now I am in chroot environment. [00:48] Boomer4582: OK, run "update-initramfs -u". [00:48] I did run that inside chroot and it showed nothing [00:49] no output [00:50] Hmm... odd. It should have said something about "Generating /boot/initrd.img" [00:50] nope, nothing [00:51] Boomer4582: "ls /boot" [00:52] https://ibb.co/Lk0P7kJ [00:52] Hey, I think I just figured it out! Hold on one moment... [00:54] Guest58: Can not work from an image. Can you not show us the command and its result as text ? [00:54] I bet I forgot to install a package in the encrypted system. If this works, I'll give you the commands. [00:54] Bashing-om: I can deal with images. [00:54] Here's the text: [00:55] arraybolt3[m]: I try to zoom here and nope no can do :( [00:55] config-5.15.0-33-generic memtest86+_multiboot.bin [00:55] dev proc [00:55] efi sys [00:55] grub System.map-5.15.0-33-generic [00:55] initrd.img-5.15.0-25-generic vmlinuz [00:56] lost+found vmlinux-5.15.0-33-generic [00:56] memtest86+.bin vmlinuz.old [00:56] memtest86+.elf [00:56] Bashing-omroot@ubuntu:/# ls /boot [00:56] config-5.15.0-33-generic      memtest86+_multiboot.bin [00:56] dev                           proc [00:56] efi                           sys [00:56] grub                          System.map-5.15.0-33-generic [00:56] initrd.img-5.15.0-35-generic  vmlinuz [00:56] Guest58: Working, give me just a bit. [00:58] OK sure [00:59] grrr... ok, still working on it [01:00] I think I got it! [01:01] Argh, why? Still giving me that warning. [01:07] Boomer4582: type "ls /boot | nc termbin.com 9999" [01:08] what does that do? [01:09] Boomer4582: That would probably give a link for people to see the text of what's in your /boot directory. [01:10] https://termbin.com/5kphk [01:12] Guest58: I'm still working on this. If I can just figure out what command to make my kernel work, I can give you the fix. [01:12] all right man [01:13] which ubuntu is this Guest58 ? 22.04? [01:13] yes 22.04 LTS [01:14] !info linux-image-generic jammy [01:14] linux-image-generic (5.15.0.35.38, jammy): Generic Linux kernel image. In component main, is optional. Built by linux-meta. Size 3 kB / 20 kB. (Only available for amd64, armhf, arm64, powerpc, ppc64el, s390x.) [01:15] Boomer4582: type apt-get install --reinstall linux-image-5.15.0-35-generic [01:15] EriC^ I know that looks like it will work, but I tried something like that. [01:15] right now you're missing the 35 kernel in /boot, you only have an initrd [01:15] It didn't do it. [01:15] arraybolt3[m]: look at his paste for heaven's sake :D [01:15] Something about being in the chroot messes up encryption support. [01:15] I did. [01:16] what [01:16] I made a VM, replicated the error, purged and reinstalled the kernel, and... [01:16] Now I get BusyBox on bootup in the VM. [01:16] arraybolt3[m]: look here https://termbin.com/5kphk [01:16] I think I know what the problem is (the chroot is messing with encryption), but I'm not sure how to fix it. That's what I'm working on. [01:16] https://pastebin.com/tcqDmykM [01:16] did it work? shall I reboot? [01:16] EriC^: Yes I know. [01:16] he has the 33 kernel and the 35 initrd right now [01:16] EriC^: I know. He's got encryption on top of everything. [01:17] It's throwing a wrench in the works. [01:17] so? [01:17] Memtest86+ needs a 16-bit boot, that is not available on EFI, exiting [01:17] Warning: os-prober will not be executed to detect other bootable partitions. [01:17] Systems on them will not be added to the GRUB boot configuration. [01:17] Check GRUB_DISABLE_OS_PROBER documentation entry. [01:17] done [01:17] did it work guys or not? [01:17] If you want to make a VM and try it yourself, go ahead, but when it doesn't give you a password prompt and drops you into BusyBox, Don't say I didn't warn you. === mwhudson_ is now known as mwhudson [01:17] Guest58: Do update-initramfs -u again. [01:17] If it gives a crypttab warning, it didn't work. [01:17] no [01:17] i think you guys are confused [01:18] EriC^: You've not dealt with this error with encryption. [01:18] I just did this. I know what I'm talking about. [01:18] yeah you did what he did i get it [01:18] If he gets the crypttab warning, he's gonna get dropped to BusyBox when he reboots. [01:19] www.pastebin.com/YFWu9czc [01:19] what's ls /boot looking like right now Boomer4582 ? [01:19] EriC^: Sorry if I'm snippy, I'm kinda tired. [01:19] no worries buddy [01:19] EriC^: Thanks. [01:19] Please check the latest pastebin after I run update-initramfs -u [01:19] it didnt work, right? [01:20] Boomer4582: try ls /boot/*35* [01:20] do you see both the initrd and the vmlinuz ? [01:21] root@ubuntu:/# ls /boot/*35* [01:21] Boomer4582: I think I got it! I GOT IT TO BOOT! [01:21] OK, let's see if this works. [01:21] Tell me Array! [01:22] https://ibb.co/tb1z0f3 [01:22] OK, can you send messages even if the computer isn't working? [01:22] using my laptop [01:23] Agh, still not working automatically. [01:24] But, I now know more about how to fix this. I think we're almost there. [01:24] Boomer4582: can you run 'cat /etc/crypttab | nc termbin.com 9999' please? [01:25] EriC^: You read my mind. [01:25] :D [01:25] https://termbin.com/l436 [01:26] sda3_crypt UUID=886b72e4-c13e-4998-8956-21122ca4bef2 none luks,discard [01:27] yes guys? [01:27] Guest58: also 'blkid | nc termbin.com 9999' [01:27] OK, reboot the machine with the live USB still in it, and when you reboot into the live environment, don't open the encrypted drive. [01:27] I got 5 more minutes, sadly [01:27] https://termbin.com/srhe [01:28] afterwards I need to go, unfortunately have a plane to catch. [01:28] what else do I do now? I dont have time to do the process all over again this very moment. [01:28] let's give it a shot without having to restart all over [01:28] OK, "sudo cryptsetup luksOpen /dev/sda3 luks-886b72e4-c13e-4998-8956-21122ca4bef2" [01:28] ah i think i know why this is happening [01:28] tell me [01:28] yeah i was about to say but no [01:28] arraybolt3[m]: that's not it [01:28] OK. Well, reboot into the main OS. You'll get dropped to BusyBox. [01:29] you want to decrypt it as sda3_crypt [01:29] Yeah, but this will at least get the system to boot. [01:29] When in BusyBox, do "cryptsetup lukeOpen /dev/vda3 my_encrypted_volume". [01:29] reboot the live usb and use sudo cryptsetup luksOpen /dev/sda3 sda3_crypt to decrypt it, so everything lines up [01:29] so which command do I type now guys? [01:29] Type the passphrase. [01:29] Do EriC^'s. [01:29] What he just said makes much more sense. [01:29] right now it was decrypted as luks-uuid by the gui and it's messing things up cause crypttab uses sda3_crypt [01:30] Then do mount /dev/mapper/sda3_crypt /mnt [01:30] Then do all the commands for getting into the encrypted system, and then run update-initramfs -u in the encrypted system. [01:30] We can resume this when you get off the plane. [01:31] I think that will get you booting, though. [01:34] https://ibb.co/gvwSygs [01:35] I got here, some progress. [01:35] Boomer4582: If it drops you to BusyBox, I think we can fix it. [01:35] Yep. You'll get an odd-looking shell. [01:35] yes it dropped me to the terminal shell now [01:35] what's next [01:35] OK, do "cryptsetup luksOpen /dev/sda3 sda3_crypt" [01:35] Type the passphrase. [01:36] Now do "mount /dev/mapper/vgubuntu-root /root" [01:36] Then do "exit". [01:36] That should boot you up. Once that happens, open a terminal with Ctrl+Alt+T, then do "sudo update-initramfs -u", and you should be done. [01:36] EriC^: You were exactly right. If it wasn't for you, I probably wouldn't have figured out how to fix it. Thank you SO MUCH! [01:37] EriC^ I got my VM working again. [01:37] https://ibb.co/php5zMX [01:37] did I do them right? [01:38] nothing happened.. [01:38] Boomer4582: Yep, now do "exit". [01:38] Boomer4582: This should boot your desktop. [01:38] Good Lord, it worked [01:38] wow [01:38] haha [01:38] Boomer4582: Once that's done, open a terminal and do "sudo update-initramfs -u" to make it stick. [01:39] done. all good now I suppose? [01:40] arraybolt3[m]: no problem buddy, pretty sure you would have gotten it anyway though [01:40] Boomer4582: Done being the update-initramfs, right? [01:40] ante@workstation:~$ sudo update-initramfs -u [01:40] [sudo] password for dante: [01:40] update-initramfs: Generating /boot/initrd.img-5.15.0-35-generic [01:40] I: The initramfs will attempt to resume from /dev/dm-2 [01:40] I: (/dev/mapper/vgubuntu-swap_1) [01:40] I: Set the RESUME variable to override this. [01:40] you did 99% of the work really [01:40] does that mean it worked? [01:41] safe to reboot? [01:41] Boomer4582: If you did the update-initramfs, you're done and everything will be working. And if it's not, you can always use that trick inside the BusyBox shell to boot. It looks like it worked, reboot and see. [01:41] Yeah, reboot. [01:41] !cookie | EriC^ [01:41] EriC^: Wow! You're such a great helper, you deserve a cookie! [01:41] :D [01:42] All great, rebooted. [01:42] Just amazing. You guys truly are amazing. [01:43] 👍 [01:43] Boomer4582: Glad we were able to help. And I'm really glad EriC^ was here. === five619 is now known as five61 === octav1a_ is now known as octav1a [04:14] I am on 22.04 Xubuntu and it is updating this morning and just showed "W: Possible missing firmware" and states multiple missing amdgpu firmwares and then it stops updating.. [04:15] its after depmod and update-initramfs [04:15] i always get that W: [possible missing warning; as its "W" it means warning, it shouldnt stop anything from updating. [04:15] I have RX480 inserted in HP 600 G1 Twr [04:16] in any case, its W; its warning. It shouldn't stop anything from proceeding. [04:16] here it just halts [04:16] I can only hit cancel [04:16] Hmmmmm that sounds different, in my case I'm not prompted. What is the prompt asking you to cancel? [04:17] Last line in update window is "W: Possible missing firmware /lib/firmware/amdgpu/navi10_mes.bin for module amdgpu" , it's Software updater window. [04:17] hi, I hate the default file browser 'Files' with a passion, any recommendations for something better? [04:17] Now I see 'zpool list also halts hmm' [04:18] also known as 'Nautilus' [04:18] dunno what that means [04:19] dsc, thunar is fine. [04:19] But I dunno what you hate about nautilus. [04:19] But I use thunar and works for me. [04:19] be sure to add the xarchiver plugin [04:19] Nautilus is just one of those GUIs designed by programmers, not with users in mind [04:19] no offense [04:19] ill try thunar, thanks [04:20] well that seems wrong, whereas gnome stuff like nautilus seems to be designed by programmers with average users in mind, trying to hide advanced options, simlpify stuff, use sane defaults, etc. [04:20] I personally like a zillion options. [04:20] But tthats not gnome. [04:20] I suppose that would require me to switch to xfce (which is not a problem, but oh well :P) [04:20] So I don't think you'll like thunar better, but be sure to give it a try. [04:21] No, of course not, just `sudo apt install thunar` [04:21] alright [04:21] you can install every file manager [04:21] I assumed so as it pulls in libxfce-* related packages [04:21] xfce is gtk based [04:21] so there really isnt any extra libs [04:21] cool [04:22] if you try the kde file manager, sure it'll pull in qt, but reall you havent installed any qt or kde apps yet? [04:22] vlc virtualbox qbitorrent etc all pull in qt libraries [04:23] once you install one app, then the rest are fine, so if you wanna try the kde file manager, go ahead [04:23] yeah but Qt is not a desktop environment while xfce is [04:23] anyway downloading those xfce is not an issue ofc. [04:23] those xfce libs* [04:24] Hi! Can someone help me with some data restoration. My Synology NAS died on me. Followed some guides and now I'm stuck with an error that says: "mount: /mnt/md127: wrong fs type, bad option, bad superblock on /dev/mapper/vg1-volume_1, missing codepage or helper program, or other error." [04:25] Zeek7098: maybe try to make a complete copy of the disk by using `dd` (or any other recovery software) before trying to mount it [04:26] so you at least have a copy around [04:27] I don't have a copy [04:27] Sigh [04:28] It's my work files that I need [04:28] wife, daughter, bills... Jesus [04:28] Try re-reading what I said [04:28] that sounds more personal [04:28] dsc_,I'm a total linux noob. [04:28] Not to nit pick. [04:28] I would create an image of the disk [04:29] Ok [04:29] by using `dd` or otherwise using recovery software [04:29] thats just what I would do [04:29] is there any recovery software on linux? [04:29] im using Pop OS [04:29] I'd just give up and start over. Or pay$$$$ to data recovery service. [04:29] Cuz they know what they're doing. [04:30] enigma9o7[m], I'm literally on my last line here. Please, empathy. [04:30] and next time, back up [04:30] No money. [04:30] Yeah, I get it. [04:30] I get it enigma9o7[m] [04:31] not really recovery [04:31] clonezilla is a cloning tool [04:31] suppose you could try clonezilla to clone the disk but you need to know what you are doing [04:31] sorry i dunno anything actually useful [04:31] right [04:31] @dcs, Ok... I don't know what I'm doing. I'll look for a tutorial on clonezilla. [04:31] Zeek7098: it just copies the disk 1:1 without mounting it, you seem to have an issue mounting so .. maybe good to copy it as-is [04:32] then later you can attempt techniques such as file carving [04:32] if you install clonezilla from repo, it wont add menu entry, just `sudo clonezilla` to get into interactive UI [04:32] Wait a sec [04:32] My Synology drives (two 2TB drives + 1 4TB drive) are btrfs / HFS / RAID 5 [04:32] is this relevant? [04:34] you need an expert, this is not something that is easily fixed I think [04:34] I can't afford an export at this moment. [04:34] Wife, 8mo baby, bills. Work due this friday. I either pay up or go homeless. [04:35] You might get lucky and catch someone here, but what you're dealing with isn't common knowledge. [04:35] Either I find a way or I'm fucked. [04:35] Hence why I said earlier pay expert. [04:35] @enigma [04:35] well in the past when I was poor I spend time figuring it out myself, thats how I learned linux essentially [04:35] spent* [04:35] but it will take time [04:35] enigma9o7[m], no money. I'm not being cheap. [04:36] if its hardware failure related tho, getting a clone of it now is a good idea. [04:36] cuz if its hardware failure, it will continue to degrade [04:36] hard drives should be fine [04:36] and at some point become completely inaccassible [04:36] the issue was the Synology NAS [04:42] gma@pop-os:~$ sudo clonezilla [04:42] Clonezilla mode is device-device [04:42] Setting the TERM as linux [04:42] Starting /usr/sbin/ocs-onthefly at 2022-06-08 00:39:07 AST... [04:42] Start preparing device name cache files in /tmp/ocs-cache//... [04:43] Zeek7098: read some tutorials first about how to use clonezilla [04:43] these type of programs can do harm too [04:43] when not used correctly [04:44] it's wednesday at 12:44pm. I have to do this on the fly if no one can't guide me through it. I will only have a couple of hours on thursday to actually do the work. [04:45] besides... it's not like it's doing anything. [04:45] it's just hung on the last line. [04:46] also... I'm assuming getting data out of a 4TB array is going to take a couple of hours [04:46] as far as I know, clonezilla is an operating system for the sole purpose of creating disk images [04:46] so, previously I would burn clonezilla.iso on an USB stick [04:46] and boot from that [04:46] its an app. `sudo apt install clonezilla` [04:46] it's both [04:47] right, I would use the live cd [04:47] I'm using the app [04:47] but I also downloaded the ISO and placed it on my ventoy [04:47] anyway should not matter .. I just dont have experience with the CLI tool [04:47] sigh... I mind as well reboot and go into the live version [04:47] since the CLI is not doing anything. [04:48] thx for the help [04:48] so this program can take a disk (for example /dev/sda) and create an image out of it [04:48] the reason why you want an image is so that you have a backup at least [04:48] honestly, if I don't get it working on the next 24h it won't matter at all. [04:48] because like enigma9o7[m] mentioned, when there is a hardware issue, the problems might become worse over time [04:49] he didnt let me finish ;/ [04:49] What's a good e-book or YT channel to learn more about general stuff on Ubuntu? === Wild_Man is now known as Guest7934 === gschanuel7 is now known as gschanuel [06:14] hi where is there some proper directions recommended by canonical for install? ing python3 [06:15] just sudo apt -y install python3 [06:18] transhumanist: python3 is already installed. it is a main component [06:19] !info python3 | transhumanist [06:19] transhumanist: python3 (3.10.4-0ubuntu2, jammy): interactive high-level object-oriented language (default python3 version). In component main, is important. Built by python3-defaults. Size 23 kB / 90 kB [06:19] ok thanks for the confirmation [06:19] so much spam out there on the topic [06:22] this package has broken my system https://bpa.st/DQKA any idea how to remove it? [06:32] yo mama [06:32] uwu [06:38] transhumanist: what happens when you run: sudo apt --fix-broken install [06:38] it just spits out garbage... [06:38] and dies [06:39] I'm guessing you were following some blog/tutorial on how to install dart [06:40] trying to install some AI software [06:40] I was just hoping they might have some new magic for Ubuntu 22.04 [06:41] this is the problem when you dont support pcie pass through, virtualization, excellent cpu, plenty of ram and your developing on the same machine as you use for a desktop [06:41] I think you're trying to install two versions of the same package, one from some PPA and one from Ubuntu. What's the output of: apt policy libdart6-external-odelcpsolver [06:41] While Ubuntu offers libdart6-external-odelcpsolver6.12 instead (different name) [06:47] alkisg anyway to generate a list of installed packages using apt-get that were not part of original install? [06:48] Not easily, the "original installation" state isn't kept in apt. It's possible to search for "packages not in ubuntu repositories" though [06:49] But if you used a ppa, the easiest way to remove it is ppa-purge [06:54] !apt-clone [06:54] !clone [06:54] To replicate your packages selection on another machine (or restore it if re-installing), you can use the !software package "apt-clone" - See also !automate [06:56] I will keep that in mind for next install. this one is too broken to get back it seems [06:56] transhumanist: Pretty much the only way to track the original package set is to do something like dpkg -l > [06:57] transhumanist: You can get the lists of packages that the installer uses for various options, but I'm not sure where you'd find that. [06:57] ya I got most of those, mainly big thing is my VM's list and kqemu setup with bridge networking that takes the time to get back. I can't fully use VM's on my system because VM's crash a lot because of a real shitty system [06:57] You can check your recently installed packages in /var/log/apt/history [06:57] fridge.ubuntu.com seems to be down ;-( [06:58] would be helpful to get a list of which system utility installed a particular package [06:58] and what ones attempted the install too [07:02] You can see something like that with `apt rdepends package` [07:02] But for "which ones attempted..." you'll need /var/log/apt === diskin is now known as Guest2191 === diskin_ is now known as diskin === hackerman34 is now known as hackerman3 [07:39] Is it possible to hide file extensions in nautilus? [07:49] Hmm.. I don't see an extension for it anyway. [08:07] why on libreoffice start screen, libreoffice base is disabled ? do I miss some packages ? [08:08] dob1: LibreOffice Base isn't installed by default. [08:08] dob1: Many users probably won't have a need for it, but if you want it, you can install it. [08:09] arraybolt3[m]: ok, thanks [08:09] Let me see if I can find the package name. [08:10] dob1: Run "sudo apt install libreoffice" in a terminal, and that will pull in anything that's not included by default, including LibreOffice Base. [08:10] ok I give a try [08:32] i have some ubuntu devices. they can connect via ethernet, wifi, &/or lte. if i have both ethernet and sim in the device, it always defaults to ethernet (which is good). however, i have no physical access to my devices once they have been put in place (only ssh connections). I am looking for a way to change a connected via ethernet device, to switch to using the lte (sim card) instead. how do I do that? [08:33] i have found that by turning eth0 off, it does not help me as I simply lose connection to the device (until it is rebooted). I do know both the ip of ethernet and the ip of ppp0 (if that helps). [08:35] with regard to switching if I had access, i simply reboot with the ethernet cable unplugged, but that costs me 50usd to have a technician do that. [08:35] any ideas or pointing to the right direction is appreciated. [09:00] Hello I'm trying to atomate the update of a server ( cli ), but the update causes a `dialogue` box to pop up and ask, if I want to keep the old config files. Can someone tell me how I can I disable dis dialogue completely, so that it never asks. thanks. [09:15] haroon - can you paste the command you are using? I do things like this, but am unsure if it helps you. "sudo apt install aptitude -y" the -y at the end will automatically answer the prompts yes - perhaps in the man page for your command it has such a switch. === ice is now known as madmax [09:21] I use `sudo apt upgrade -y` but I still get this dialogue box in the shell https://pictshare.net/rf2yz3.png [09:24] haroon: https://askubuntu.com/questions/104899/make-apt-get-or-aptitude-run-with-y-but-not-prompt-for-replacement-of-configu === sams_ is now known as sams [09:40] nice - dsc_ [09:41] re: my question - i thought of alternative - is there a way for me to reboot a computer, but tell it to connect via lte (and not ethernet) after the reboot? [09:43] weedmic: no need to reboot, the solution is to programmatically interact with your network-manager [09:43] im not sure what network manager ubuntu uses nowadays [09:44] but you'd create a script that switches to SIM when it detects ethernet is down [09:44] i have nmcli, but i'm pretty sure it does not have control over the modem [09:45] but really this sort of task should be handled by your network manager, sounds like a trivial case [09:45] ic [09:45] i cannot create a script of something i cannot do manually, but do plan to create one once I have the steps figured out. [09:46] I think your goal should be to do this through nmcli, yeah [09:46] i have machiens that have two connections, but they are using ethernet - we want the default to be lte b/c the connection would be maintained for 2 hours with a powerloss. [09:46] re reading the manpage for nmcli, but p[retty sure it cannot see/control the modem (which is mmcli) [09:48] https://unix.stackexchange.com/a/268285 [09:48] maybe this helps [09:49] will check it out - ty [09:51] i had it backward, nmcli does not see the ethernet - only the modem === gacuxz1 is now known as gacuxz [09:52] eth0 ethernet unmanaged [09:52] nmcli d s [09:53] weedmic: honestly the network management situation is confusing to me, im not sure if ubuntu uses nmcli, netplan, systemd-network, networkd [09:53] :D [09:53] >_> [09:53] i think what I need is a manager that can "manage" all three (eth, wifi, lte) - not just a separate manager for each [09:59] weedmic: yes === kaiseisei1 is now known as kaiseisei [10:17] weedmic: NetworkManager should be able to handle all three AFAIK. How exactly do you have things configured that makes it so that you can't see LTE in NetworkManager? Also, can you take a look at nmtui (not nmcli, nmtui) and see if everything pops up in there? [10:18] by "NetworkManager" do you mean "nmcli"? [10:18] So, NetworkManager is an entire system. nmcli is the NetworkManager Command Line Interface. [10:18] nmtui is the NetworkManager Text User Interface. [10:18] nmcli - sees all,but manages only gsm (lte/sim), it states clearly with nmcli d s that is does not manage "ethernet" [10:19] When you use Ubuntu on the desktop, NetworkManager is what lets you see Wi-Fi, change settings networks, etc. [10:19] there is no desktop [10:19] Type "nmtui" and see if you can fiddle with stuff in there. [10:19] reviewing nmtui [10:19] ty [10:23] weedmic: This AskUbuntu answer looks helpful: https://askubuntu.com/questions/71159/network-manager-says-device-not-managed [10:23] nmtui - looked familiar - nmtui does not deal with lte/sim/gsm - only wifi & ethernet - under "add" the drop down list does not show lte/sim/gsm [10:23] ty - reading above [10:24] basically, nmcli does lte & wifi, nmtui does ether & wifi - looking for something that does all three (or could write my own if I knew how to do the steps manually) in python [10:26] See, that doesn't make sense to me, since Red Hat documentation indicates that nmcli can be used for Ethernet configuration. [10:26] (And I get that Red Hat != Ubuntu, but they're both Linux, and a lot of stuff is the same between them.) [10:27] very nice link - i'm forwarding it to our r&d head to see if it will goof anything up if I do that [10:27] I also did not understand until that post about it being in the wrong place [10:28] so, if it is "ok", and it works, then anyone can connect to a machine, click a few things and be done with it. no cable need be touched. [10:29] Also, read down through the answers, there's important stuff further down, looks like. [10:29] In particular, Ubuntu config sets everything but WiFi, GSM, and CDMA to unmanaged if I'm understanding correctly. [10:29] And that sounds like exactly the problem you're having. === Nicklificent is now known as Millicent [10:31] yes - it's a very helpful page [10:32] Out of curiosity, why does ethernet need disconnected at all? Can't you leave both connected and have two points of entry? [10:33] You should be able to connect to ethernet and LTE simultaneously. At least, I know my systems haven't had problems connecting to ethernet and WiFi at the same time in the past. [10:37] they are both connected, but... it appears the first to load (which is eth) is how all the connectivity to our software and the internet of things connects, and stays that way. we want lte to be on the batterires. [10:38] to work while on battery - we have no control over an ethernet port at someone else's location. [10:38] Oh, OK. I guess if stuff inside the server needs to access stuff outside, then that makes sense. I wonder if you can set the connection priority. [10:40] i am supposed to see into that too, but i really wanted to just down ifdown on the eth0 and have it work :D silly me [10:40] Hmm, looks like what you're trying is going to be easier. [10:41] i don't have faith in the prioirty thing - just based on phones - do you ever connect tothe right one automatically [10:41] What Ubuntu Server version are you on? [10:41] will connect to one and see [10:42] Welcome to Ubuntu 20.04 LTS (GNU/Linux 5.10.60-olimex armv7l) [10:42] afk [10:45] weedmic: When you come back, if you're able to and allowed to install software on the server, you can use ifmetric to set network priorities, I believe. https://askubuntu.com/questions/1327484/ubuntu-20-04-is-it-possible-to-prioritize-network-connection [11:19] fyi - we're going to unplug all the ethernet cables - lte only :) [11:20] experienced fedora/centos user, new to ubuntu. Installed 22.04 in an ESX VM, networking only seems to reach local subnet, "ip -4 a" shows correct default gateway "ip route" shows correct routes, I can SSH into the ubuntu machine, it can ping within the subnet, but no traffic seems to reach the default gateway or the firewall beyond that ... suggestions? [11:24] Can you ping the gateway? [11:25] yes, and can resove hostnames using internal DNS servers [11:25] first 6 or 7 pings seem to go unanswered though, then it picks up [11:26] That would be an issue... [11:26] Other devices can ping the gateway with no dropped 6-7 pings? [11:28] yes, just noticed, if I ping another local subnet, via the default gateway, first 8 pings were answered, then no replies again until 42nd packet [11:29] stopped at packet 89 restarted after 134 ... definitely odd ... something weird with packet offload perhaps? === jesopo is now known as jess [11:33] have done an "ethtool --offload ens160 rx off tx off" which seems to have improved things [11:39] pings are now reliable, to local subnet, and statically connected private subnets via default gateway, apt can resolve gb.archive.ubuntu.com, but not connect to it, still nothing showing on the firewall corresponding to apt's http requests, can't traceroute as I only did minimal install [11:40] tools like traceroute are indeed useful for network testing [11:41] @rob0 if only I had it installed, or could install it via apt ... can probably remount the install DVD and find the package [11:42] this default gateway, it works otherwise? Just this new Ubuntu is having trouble? [11:42] oh ESXi [11:43] yes, it's a big corporate network, just this box is having issues [11:43] maybe your gateway is not doing NAT for the vmware network [11:45] hundreds of other VMs are happy, mostly windows, but a handful of centos, does ubunto generally "like" ESX, I'm using a VMXNET3 paravirtual NIC, would an emulated E1000 be better? [11:46] disabling the NIC ofload certianly made a difference, I'll give a different virtual NIC a shot .. === diskin is now known as Guest1461 === diskin_ is now known as diskin === diskin_ is now known as diskin [11:57] Hi all [11:59] yes, using an E1000E NIC, apt is now whizzing along ... [11:59] aha, interesting [12:02] jammy does seem to have a driver for the VMXNET3, I'll scout around for any other problem reports, log a bug if I don't find anything, thanks chaps even for giving me somewhere to bounce my thoughts off ... === diskin_ is now known as diskin [13:40] in my experience that's how things go though? [13:40] at multiple companies [13:40] ie, java, python, etc [13:40] <[twisti]> so do i need to install it and submit another paste ? [13:41] [twisti]: no, I wasj just trying to figure out what sort of install this is [13:41] <[twisti]> ah, sorry for not prefacing my question with it [13:42] <[twisti]> im not really sure what the error message even means; im trying to connect to an ftp via an openconnect VPN tunnel [13:42] [twisti]: no idea why the tnftp client is doing weird things for you. Though, are you even using that client? ls -l /etc/alternatives/ftp [13:42] <[twisti]> looks like i am, yes [13:42] jongsta, if you are on focal 20.04, you' ll be fine [13:42] !info graphviz version 2.42.2-3build2 focal [13:42] 'version' is not a valid release [13:43] !info graphviz 2.42.2-3build2 focal [13:43] '2.42.2-3build2' is not a valid release [13:43] !info graphviz focal [13:43] graphviz (2.42.2-3build2, focal): rich set of graph drawing tools. In component universe, is optional. Built by graphviz. Size 577 kB / 3,101 kB [13:43] i hope your ' devs' agree with a security update, when available [13:44] [twisti]: which ftp [13:45] [twisti]: I wonder if the error has to do with the networking you have in the docker/VPN that is preventing it [13:46] <[twisti]> tnftp, like you said. the vpn is probably part of the problem, but i dont even understand what the error message means, so its hard to diagnose [13:48] what is openconnect? [13:48] [twisti]: ^^ [13:48] oerheks: right but we need to pin that version according to my boss for the devs [13:49] at some point i imagine doing an "apt install graphviz" will install a newer version/build [13:50] !pinning [13:50] pinning is an advanced feature that APT can use to prefer particular packages over others. See https://help.ubuntu.com/community/PinningHowto [13:50] so, good luck [13:51] hey, stupid but hopefully quick question perhaps, but see when using hdparm as a benchmarking tool (`-tT` or similar), does it erase any data? can I point it at a new blank SSD and an existing one to work out which is faster without burning my current data? [13:51] <[twisti]> rob0: a VPN tunnel software id say, it lets you connect to things like fortivpn [13:52] oerheks: sudo apt-mark hold libxfont1 … seems easy enough [13:52] that's the method we're planning to use [13:53] jongsta: I hope this is short term with the long term being the devs learn how to develop properly and not rely too heavily on static version numbers [13:54] yah, i'm discussing with my boss how he plans to address that issue [13:58] [twisti]: I am actually well versed in VPN in general; I mean, "what VPN protocol is it?" [13:58] <[twisti]> oh, sorry. it supports about half a dozen protocols, but the one we are using it for is fortivpn [14:02] That's not a protocol, that is a provider. I think they are multiple protocols. [14:03] s/are/use/ [14:03] <[twisti]> i dont know then [14:03] <[twisti]> the parameter is `--protocol=fortinet` and we were given no other information that its fortinet [14:03] <[twisti]> and connecting works fine [14:06] oh. Anyway it seems that something is broken in the docker network stack. [14:29] <[twisti]> looks like [14:30] <[twisti]> this whole VM is messed up, im going to set a proper one up === gandalf0 is now known as gandalf [14:40] -That's one of the nice features of VMs, tossing them out and starting over. [14:59] can anyone help me reconfigure my wireless dong, please. [14:59] its a WIFI 5 AC12000 USB dongle [15:04] type lsusb see what kind it is then search for it in synaptic maybe [15:06] Chalky: cat /etc/os-release | nc termbin.com 9999 [15:11] when I run snap install multipass I get an error access denied [15:11] see snap help login but you shouldn't have to log in with snap to install a package right ? [15:12] devslash: the snap online store is having issues at the moment [15:12] theres no workaround [15:13] this app that I want to install is only available to install via snap [15:13] just tried searching synaptic, nothing found. Tried install following description in github but no joy their either. [15:13] devslash: the servers which host the snap are having issues at the moment. You need to wait till they resolve the issue before you'll be able to download packages from the snap store [15:14] Chalky: cat /etc/os-release | nc termbin.com 9999 === atol-71 is now known as atol71 [15:32] <[twisti]> omfg i might kill someone today [15:33] <[twisti]> alkisg, leftyfb, rob0: finally figured out the `servname not supported for ai_socktype` error: no ftp/ftpdata in /etc/services [15:33] <[twisti]> thats literally all there was to it [15:33] Heh, I thought about that from the first minute, but I considered it too unlikely to be the cause (since it was saying :ftp), and I didn't mention it :) [15:34] (also I couldn't think of a reason why /etc/services wouldn't be the stock one) [15:34] because it's a docker image [15:35] and nobody should be ftp'ing from a docker image in 2022 :) [15:35] <[twisti]> i have no idea why it wouldnt be the stock one even in docker, its not a super slim alpine image or anything, regular ubuntu 22.04 lts [15:35] [twisti]: it's docker. It's not an ubuntu image. It's as slim as it needs to be in order to run the applications you built into it [15:36] <[twisti]> you dont have to tell me, the whole reason we are putting it in a docker image is because we want to have neither ftp nor this fortinet vpn running on an actual host [15:36] at least use lxd then :D [15:36] ^ [15:36] <[twisti]> so we fire up a docker image, download the new files from the customers nightmare setup and nuke the whole container afterwards [15:37] but i guess for a one-off execution docker works fine [15:38] <[twisti]> lxd vs docker vs containerd vs podman or whatever decision was above my head anyways === gschanuel9 is now known as gschanuel [17:10] Hi [17:12] :-) === kgiii_ is now known as KGIII === diskin is now known as Guest4879 === diskin_ is now known as diskin [18:47] hi , how do i kill a script .sh running on terminal without closing ? i tried using pid or -INT but it doesn't work. [18:48] CTRL + C ? [18:48] ^\ might do it [18:48] nope , i can mouse over the terminal [18:49] CTRL+C usually cancels any onling script if it's not running in the background. [18:49] being able to move the mouse pointer over a terminal window is not unexpected [18:49] CTRL + C is only possible if the window on the terminal is active or mouse over the terminal. [18:50] You can also do CTRL+ALT+F1, login, and then kill the PID from there. [18:50] To get back to the graphical interface, do CTRL+ALT+F7 [18:50] If the script overlapping the the terminal or the terminal is minimize ? [18:51] Yes. Regardless of what you're doing, CTRL+ALT+F1 will knock you to a terminal shell void of all GUIs. [18:51] alucardromero-l1 : how do i know which pid my script is running ? === wildmanne39 is now known as Guest8561 [18:52] ps auxw | grep .sh [18:53] ^ that, yes [18:57] sarnold : but my script is not running , once it run i can't check the script pid [18:59] sarnold : when the script is running , i can't hover to the terminal anymore [18:59] nbusrone: if the script isn't running, what's the point? [19:01] sarnold : Simple to said , i run the script on terminal Desktop/test.sh but when the test.sh running , i can't hover to the terminal to check for the pid but I want to kill the pid without closing the terminal similar with ctrl+C [19:03] is it very hard to cancel a script or a command in GUI [19:04] I use ^C to cancel scripts all the time [19:04] sometimes that doesn't work and I have to use ^\ [19:04] that's very rare [19:04] sarnold : i can't hover to the terminal [19:04] what does that mean? [19:07] sarnold : it's looping [19:12] Perhaps the script has handed off to a binary program you called with it. [19:19] sarnold , jhutchins : sorry , went to looping again , i have to close it find the sh pid [19:19] Ubuntu 20.04.4 installer defines 1.5 GB for boot, while it was 1 GB as I remember in 20.04.3; Is that extra 0.5 GB necessary to be kept? [19:19] sarnold , jhutchins : anyway to kill a .sh with infinity loop ? [19:19] sarnold , jhutchins : when mouse can't be use ? [19:22] moha: by default, yes. you can build smaller initrd's by including less modules and save some space, but the goal of saving disk space would be a silly motivation to do this most of the time (it can make sense for other reasons). [19:23] sarnold , jhutchins : how to cancel a script when the script using mouse as hovering in GUI ? [19:24] nbusrone: I'm sorry, I do not understand you :( I hope someone else can udnerstand what you're asking about [19:25] sarnold , jhutchins : The script is https://pastebin.com/WiYME0fM. Open a terminal run the mouse1.sh and will take over the mouse control.You can't hover to the terminal as mouse in a looping [19:25] aha!! :D [19:25] sarnold , jhutchins : Then how do I cancel the script while mouse is bing use ? [19:26] being* [19:26] nbusrone: use control + alt + f1 or f2 or ssh in to the computer from another machine; then kill -9 that script [19:26] sarnold : I wanted to remain in gui [19:27] nbusrone: if your script controls your mouse you do not have good choices === gabes9 is now known as gabes [19:28] sarnold : is there a way I assign a script into a special pid ? [19:28] no [19:28] PID are assigned by the system. [19:28] pids are assigned by the kernel and you really can't influence that, except by making a fork loop that runs until it finds it has the desired pid [19:29] this is done by some exploits.. [19:29] sarnold : so , how do i cancel the script within gui but can't use a mouse ? [19:30] sarnold : I can assign a keyboard shortcut script but i do not know the pid since as you said the pid are assigned by the system [19:30] If you know how to navigate using your keyboard, that's an option. But if you can't, you're going to have to use the terminal shell via CTRL+ALT+F1 [19:31] nbusrone: you can run 'killall nbrusrone.sh' by a keybinding.. [19:31] sarnold : let me try [19:35] anyone here have a script that exports windows cert store and imports them to ubuntu/firefox? [19:44] sarnold : nope not working , since script .sh are running on terminal or bash , i need to kill pid bash inorder to cancel it. [19:45] sarnold : better solution , assign a keyboard command on mouse.sh and a kill command keyboard on mouse.sh.It works in this way but i really hope there is a way to do it to kill at terminal [19:45] sarnold : will do again tomorrow , thanks guys [19:45] nbusrone: see ya! :) [20:00] There was something, I think it was pre-Plasma KDE, something like Ctrl-Alt-Esc turned the mouse cursor into a killer pointer - point it at an ap, click and kill. [20:01] xkill does that [20:01] but I think the problem statement is that the mouse doesn't work at all, heh [20:07] bare metal or VM ? [20:07] hello! Can I run server from WSL (Ubuntu) ? [20:08] yes, you can run a service on WSL, though we do not support WSL here. [20:08] !wsl [20:08] Windows 10 and newer have a feature called "Windows Subsystem for Linux", which allows it to run Ubuntu (and other Linux distro) userspace programs without porting/recompliation. For installation instructions, see https://msdn.microsoft.com/en-us/commandline/wsl/install_guide [20:09] aldcor: I don't think WSL supports systemd yet. that will make it very complicated to do many server tasks. [20:09] wsl2? [20:09] makes sense. I bet at some near future it will be possible [20:11] oh, I see wsl2 got systemd [20:11] very messy tho [20:12] yeah, I reviewed some of the code to try make that happen; it was neat, interesting, but it felt like a frustrating challenge to overcome. Depending upon what you're doing it'd be better to just use normal VMs for either the linux or the windows or both :) [20:13] sarnold: Come to think of it, xkill would probably just kill the xterm session, not the program running within the script within the shell. [20:13] does anyone runs server via VM? [20:14] aldcor: Thousands of people, yes. [20:14] aldcor: Pretty much all major datacenters run everything on VMWare. [20:15] aldcor: Perhaps that was not your actual question? [20:15] aldcor: please explain in detail the issue you are having with ubuntu. If you are not currently having an issue, feel free to go to #ubuntu-offtopic for opinions [20:16] aldcor: amazon, microsoft, google, digital ocean, oracle, etc etc make billions of dollars a year selling exactly that :) [20:16] jhutchins: and the bash process will get a sighup as the terminal goes away [20:16] HI! I am looking for a package called pagmo-dev , nothing comes up at all on a google search anyone hear of it? [20:16] leftyfb: not neccesarily issue but rather intereset in running server via WSL but I already got my answer [20:17] thanks for your very kind offer though, leftyfb [20:17] transhumanist: https://pkgs.org/search/?q=pagmo [20:18] yes I found pagmo, just not pagmo-dev it seems it requires both for the software I am compiling [20:20] transhumanist: you really need to work with the vendors for support with their software. You're constantly asking for help with compiling software [20:20] transhumanist: pagmo-dev has never been a package for ubuntu [20:20] thats all I am asking [20:20] available via conda https://esa.github.io/pagmo2/install.html [20:22] thanks oerheks thats not it either , anyway see you later [20:22] there are no files in any packages in ubuntu called "pagmo" either [20:23] that one is constantly asking for help compiling 3rd party applications, usually weird stuff [20:24] ' thats not it either' makes me wonder. [20:25] Apparently there was a new kernel for aws for ubuntu 20.04 released this morning. [20:25] 1028-aws and it causes all alpine based docker images to segfault [20:25] ekristen, yes, a huge update with 30+ fixes [20:26] he said the D work [20:26] word [20:26] or was it the A word? [20:26] every single alpine base docker image I have is seg faulting and even causing kernel panics with 1028, anyone seen any reports on this? [20:27] ekristen: I've heard rumours that the azure and gce kernel updates have also been causing problems but no bug reports yet [20:28] but thats has not been the nftables fix yet? [20:28] check in #alpine-linux? [20:29] @sarnold kk — I can provide some details somewhere if that helps. I’m trying to figure out how to fix my side without pinning to old versions of ubuntu in AWS now. [20:30] @oerheks I just dropped a message over there too. [20:31] I take it nobody really bothers to have a test box they apply critical things like kernel updates to before they roll out to production? [20:32] Kinda like backups, everybody says it's a good idea... [20:32] which reminds me... [20:33] ekristen: https://bugs.launchpad.net/ubuntu/+source/linux-aws-5.13/+bug/1977919 [20:33] Launchpad bug 1977919 in linux-azure-5.13 (Ubuntu Focal) "Docker container creation causes kernel oops on linux-aws 5.13.0.1028.31~20.04.22" [High, In Progress] [20:34] @sarnold thanks, I was trying to search but was using the wrong words. I’ve only witnessed this on alpine linux image variants, very odd. [20:38] it should affect just about anyone using the latest 5.13 cloud kernels. I couldn't repro on generic [20:39] but I reproduced the bug on all the 20.04 5.13 cloud kernels [20:39] using an ubuntu container [20:48] @jchittum interesting [20:50] hi [20:50] It also seems to be just 20.04 === BiTCHiN is now known as Wilma [20:55] hola [21:01] hello [21:21] ubuntu scortal here [21:53] hello [21:53] chanell [22:09] edwinflores: hi [22:10] edwinflores: Hi, how can we help you? === LabMonkey is now known as Mechanismus === five617 is now known as five61 === themill_ is now known as themill === Scotty_Trees2 is now known as Scotty_Trees