[00:28] hey [00:29] somebody? [00:30] hey [00:30] i cant see any messages [00:32] i'm here [01:03] "i cant see any messages" <- What client r u using ? [01:03] "i cant see any messages" <- You have to provide more detail friend [01:04] no one was posting any messages at that time, thats all === chris14_ is now known as chris14 [02:24] Hi, how i can do/setup block-level disk snapshot for a partition/disk/img? thanks [02:26] btrfs can do snapshots, zfs can snapshots, lvm can do snapshots [02:28] thanks, let me search for zfs snapshots [02:47] Question: I need your help because the php applications I am running is not sending mail out. If I login webmail and send and recieve mail to gmail.com that works fine. I am using php 8.2 [02:47] It's doing this for all the php applications and I have all the php ext installed [02:49] I can send mail from the php application to the local host. All the required ports is opened on the firewall [02:49] I am using exmin [02:54] I am tried using all the options phpmail, sendmail and smtp === atomicpi is now known as wchrisg === xcynth_ is now known as xcynth === xcynth_ is now known as xcynth === xcynth_ is now known as xcynth === xcynth_ is now known as xcynth === Andrew is now known as Guest9902 [05:37] My Ubuntu does not pick up my usb headphone and I mean does not show up in the control panel at all. I had to set an audio file , I think for alsa . Could I reinstall Ubuntu with out deleting any home file and just replace the os? === homer is now known as V|A === guiverc2 is now known as guiverc [07:24] i changed my swap partition and now update-initramfs is giving me a crypt error about not finding the old swap partition, which no longer exists [07:25] anything i can do to set it straight? [07:28] ayjay_t: does this show anything? grep -ri resume /etc/initramfs-tools/ [07:29] yeah, alkisg it shows the correct UUID of the new partition [07:29] i restarted (scared, because who knows) and now that error is gone [07:29] so i don't know what update-initramfs was looking at that it found the old data but, lol the error seems to miss the point that that is what we're updating [07:29] You don't need to hardcode the uuid there you can remove it. But if it's set to the correct value, that also works. [07:30] oh so it just searches for the swap? [07:30] the first time i tested it, i didn't even have the partition active as swap and it didn't work [07:30] now i'm going to try it again [07:31] wow it works [07:31] If you hardcode the wrong uuid there, on the next boot, you'll get a ~30 seconds delay as it tries to find it and it gives up. That's all, nothing more. [07:31] i made it a HUGE swap partition tho, the size of my ram in -free (28gigs, weird number) [07:32] not sure if that was really necessary lol [07:33] so excited to see it work, kinda sad I had to disable secure boot [07:33] i see that someone attempted a patch [07:33] thank you alkisg [07:34] Why did you have to disable secure boot? [07:35] uh well, according to stack exchange, on this kernel at least, loading an image from SWAP defeats the purpose of secure boot unless secure boot is looped in on that process [07:35] so linus declined to accept some patch that would allow it, and systemd will throw an error if you have secure boot enabled and try to hibernate [07:36] ayjay_t: so it is the change in swap is fully resolved? y/n [07:36] y [07:36] with caveat i suppose [07:37] oke === ViiA is now known as V1A [08:14] * enyc meows [09:08] Hi everyone [09:08] Does somebody use a local file: repo entry in his apt config? [09:09] Having a really strange behaviour [09:09] in a Docker container [09:11] !details [09:11] Please elaborate; your question or issue may not seem clear or detailed enough for people to help you. Please give more detailed information; for example, we might need errors, steps, relevant configuration files, Ubuntu version, and hardware information. Use a !pastebin to avoid flooding the channel. [09:12] adding a repo with this in a UBuntu echo "deb [signed-by=/etc/apt/keyrings/qgis-archive-keyring.gpg] file:/qgismirror/ubuntu-ltr-19_05_2023_0924 jammy main" >> /etc/apt/sources.list [09:12] results in The repository 'file:/qgismirror/ubuntu-ltr-19_05_2023_0924 jammy Release' does not have a Release file. [09:13] while doing it manually in the shell in the same container works [09:14] I am not a beginner, tried 1000 times and its driving me crazy :D [09:16] The strange looking part is 'file:/qgismirror/ubuntu-ltr-19_05_2023_0924 jammy Release' -> the Release Part in this error message while i added "jammy main" [09:21] when I run apt update I get a lot of warnings, mostly "bla bla is configured multiple times..." [09:22] is there an easy way to clean things up without losing any custom repos I might've added? [09:22] lacrymology fix your sources file ? [09:22] lacrymology: can you share in a !paste what you see exactly? [09:23] add-apt-repository -L [09:23] add-apt-repository --help [09:23] man add-apt-repository [09:23] also see https://ostechnix.com/how-to-delete-a-repository-and-gpg-key-in-ubuntu/ [09:30] !psate [09:30] !paste [09:30] For posting multi-line texts into the channel, please use https://dpaste.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. [09:30] !pastebinit [09:30] pastebinit is the command-line equivalent of !pastebin - Command output, or other text can be redirected to pastebinit, which then reports an URL containing the output - To use pastebinit, install the « pastebinit » package from a package manager - Simple usage: command | pastebinit [09:31] lotuspsychje: https://paste.ubuntu.com/p/czGKCMNdyR/ [09:31] ok that didn't go right, just a minute XD [09:32] ok yes it worked, but it needs login... :/ [09:33] https://dpaste.com/AC2TBUWDS heres a dpaste === shlom is now known as VlA [10:19] lacrymology: you may find it a bit faster to build your own docker container with exactly what you want - than to take someone else's and try to fix it to your needs. [10:33] Long live IRC ! :D [10:33] And greetings. [10:33] OK so what do you guys do when you want a directory completely private and encrypted including the names of all files ? [10:34] I trust that should not be so hard ? And that I am just an ancient old guy behind with tech. [10:35] GnuPG ? [10:36] NeoFrontier: i completely encrypt my home using gocryptfs, works good since several months [10:36] Ah, thanks tuxinator :D [10:45] NeoFrontier: I create a server that can only be reached via ssh. then I limit the ssh to a specific user and address, then have a separate hdd for / and /home. then for /home/user I encrypt the entire thing. if this machine is in a harsh place, I make it on a DVD for / (so it cannot be changed). [10:46] if I have time, I make the ports and readouts look like a windows xp machine - will drive an hacker nuts (b/c they don't know it is ubuntu). [10:51] weedmic: Why do you tell that to us ? Am I missing something ? [10:53] NeoFrontier: asked "when you want a directory completely private and encrypted including the names of all files" about 30 seconds before you logged in. [10:54] understood weedmic Thanks [11:14] any idea why the first works in a sources file while the second does not when running via Dockerfile building the container? https://dpaste.org/SH06L [11:17] The error mentioned is Release file not found, while it exists [11:19] ignore the "focal" vs "jammy" thats a copy paste error [11:22] Hello! [11:23] What is the default `python` in Ubuntu 20.04? 2 or 3? [11:23] It is 2 in this system and I was wondering if this was changed by someone or if it is the default [11:24] cousteau_: are you serious? I like to help but being lazy does not motivate to help [11:24] ask your favourite search engine for "ubuntu 20.04 default python version" [11:24] cousteau_: https://askubuntu.com/questions/1232812/whats-the-default-python-version-in-ubuntu-20-04 [11:25] That's what I thought, but I wanted to make sure [11:25] Also, that site explicitly asks "What is the default Python *3* version..." [11:27] cousteau_: next time also dpkg -l | grep python may help [11:27] And after a few failed results I decided it was better to ask to a human rather than a search engine === thegodsquirrel is now known as thegodhippo [11:27] or cousteau_: or python --version [11:28] tuxinator: did you read the part in which I literally already said "it is 2 in this system"? [11:28] Like, I am asking because I am genuinely confused about why it's 2 in this system, when I would have expected to be 3 by now (which seems to be the default), otherwise I wouldn't be asking at all [11:29] "python3 --version" is what I use [11:30] Well yeah that won't tell you the version of Python, but the version of python3, which I assume is 3.something [11:30] I believe if you don't specify pything3 you get python v2 data [11:30] cousteau_: https://www.fosslinux.com/39384/switching-between-python-2-and-3-versions-on-ubuntu-20-04.htm may be of interest for you [11:30] cousteau_: should be able to change the default version with update-alternatives [11:31] weedmic: python2 is deprecated and dead, systems have changed to python3 by now, but I don't know exactly when this happened in Ubuntu, and my system isn't helping... [11:31] cousteau_: there are examples in that link [11:31] tuxinator: that's what I suspect happened... Thanks! [11:33] tuxinator: wait wait wait... That site actually says that python... doesn't exist at all! (by default) [11:34] (which is probably a good idea to force people to transition to 3 once and for all, but kind of surprising) [11:34] cousteau_: where is that statement? In my understanding Python2 is not installed by default on 20.04 but you can do manually [11:36] tuxinator: when they did `ls /usr/bin/python*`, it didn't show any /usr/bin/python [11:36] Also, the article starts by showing an error message "Command 'python' not found" [11:37] cousteau_: I do which python, get nothing, i do which python3 and get the path. did you try "python3 --version"? [11:41] tuxinator: so well, my apologies because my question might not have been formulated correctly or clearly. But I guess the answer to "what is the default version of command `python` in Ubuntu 20.04" would be "None. There's no `python` command in a default Ubuntu 20.04 install, only `python3` is installed by default." [11:42] cousteau_: just verified seems true, only python3 cmd is there by default [11:43] None of the first Google results for my query clearly stated this, so I consider myself excused for asking this :) [11:43] upstream python considered a generic /usr/bin/pyton symlink obsolete and requires to use the versioned binaries instead ... ubuntu just folloed upstreams request here [11:43] cousteau_: doing apt install python-is-python3 will probably add a link from python to python3 [11:43] Yeah, that's what the name suggests at least [11:43] ogra: good to know [11:43] right, there are two packages that can re-estabish the symlink [11:44] Regarding symlink to "python" see https://peps.python.org/pep-0394/ [11:44] This project I'm using has a lot of .py files starting with `#!/usr/bin/env python` which are executed directly in the Makefile... [11:45] and for this the above mentioned packages exist ... to save yu from having to ad-hoc fix your code (you should thiugh ... in the lnger term) [11:46] NickH: interesting. So they explicitly state that the preferred names are python2 and python3, and that a numberless python MAY be installed, without stating a preference as to which it should be [11:47] Yeah, that's my understanding. [11:47] So my scripts these days are always "#!/usr/bin/env python3". [11:47] Any script written after this PEP should have used the number in the shebang. But "should" is such a big word... [11:48] any package in the ubuntu arcive will properly use the versioned hashbang ... so whaever you do for your scripts will not interfere with the os setup ... [11:49] I vaguely recall seeing some system that either didn't have a python2 or a python3, and it was simply called python, so that shebang wouldn't have worked there. But that would have gone against the recommendation of the PEP. [11:49] if your scripts are all pyton2 and require the old /usr/bin/python, just install the package that establishes the symlink ... (and fix your scripts before moving to 22.04) [11:49] cousteau_: you could just use "sed" to fix all your scripts [11:49] nah [11:49] ogra: Any package? You sure? :) [11:50] there is more to it than just changing the hashbang ... [11:50] Hmm, does 2to3 fix shebangs? [11:50] no [11:50] I'd believe it, but I'm more willing to believe that they left some obscure package unchanged [11:50] it just adds the /usr/bin/python symlink [11:51] For the time being I'm fixing all the scripts I see in these two projects I'm handling so that they say python2 [11:51] porting scripts from 2 to 3 requires some additional syntax changes (or importing a backwards compatibility module), just sed'ing is not enough [11:51] Just in case [11:52] ogra: sure its not, but i didn't know if his scripts are python3 compatible or not [11:52] ogra: seding python to python2 ought to work though [11:52] cousteau_: yes [11:53] (which is what I understood was being suggested) [11:53] cousteau_: you may have a look at https://blog.rac.me.uk/2022/09/01/techy-changing-python-shebang-path-from-absolute-to-env/ [11:53] yes, that is simply the same as setting the symlink and keeping /usr/bin/python [11:55] what i meant above is that you can not just simply sed py2 scripts to a py3 hashbang ... that will not be enough [11:55] "(2(\.[0-9]+)?)? : zero or more times" -- + is once or more >:( [11:56] ogra: well, not the same, because if you ever decide to change the symlink or update Ubuntu the other way will still work [11:57] But yeah, one does not simply upgrade a script from 2 to 3 [11:57] (one does it, but not simply) [12:01] py2 and py3 are really totally different - except for a few things like tab. I was updating my py2 book until I noticed everything was different and just dropped it in the skip. but this was 3-4 years ago. [12:02] Well, thank you very much for the help, tuxinator, ogra, and others! (even if we started the conversation on the wrong foot) [12:02] weedmic: to be fair most of it is similar [12:03] There are some syntax changes, different behaviors, and different library names, but there's a lot in common [12:04] I'd say it's comparable to C -> C++. Maybe even closer. [12:05] (then again, C and C++ ARE totally different languages...) [12:05] (or, "significantly different languages") [12:07] ok, but I meant that nothing would work without being rewritten - but it's ok - i am survived and made the switch - only had to rewrite maybe 10 old programmes. [12:11] In my experiencec 2to3 does quite a good job at converting python2 scripts to python3. [12:12] nice to know NickH === EriC^^ is now known as Guest3173 [12:56] Hi all === Szadek58180 is now known as Szadek5818 === Gyuseok is now known as Silica === Silica is now known as Guest4468 [14:36] [14:36] During Ubuntu install there is cascade of Ubuntu installers: UDI, Subiquity, Curtin. Tell me please, is there a file or hook, where it is possible to put inside ISO file my own scripts for customization of a system to be installed? How it is called, what is it's path? [14:36] Or are there any descriptions/documentation on how it's expected to customize installed system instance at installer's finish? [14:38] It's concerning 23.04+. [14:43] Bart__: https://www.pugetsystems.com/labs/hpc/ubuntu-22-04-server-autoinstall-iso/ [14:43] chrome is borked for me today - the application doesn't render properly with lots of visual artifacts, noise, and missing elements. If I disable hardware acceleration in Chrome's settings, everything works fine again. I think this is related to a mutter/libmutter and or gnome-shell upgrade that went through yesterday (I'm on 22.04). Has anyone else seen this behavior? Is there an easy way to swap out those packages for older versions to [14:43] confirm they are the problem? [14:51] on 22.04, ulimit -a shows my stack is 8192, but 'ulimit' reports 'unlimited', misleading to me, why is that [15:06] ravage, did I understood properly: the described process will give final result as a sum of user provided '#cloud-config' file with 'late-commands' 'in target' and parameters entered by user during Ubuntu Desktop Installer work? [15:07] is it possible to change the logical volume name on an Ubuntu installation and still have it boot? I have two different Ubuntu installations with the same logical volume name and it seems I can't read one from the other without changing the other's name, but if I change the other's LV name I can't also boot from the other until I change it back [15:08] er, volume group name, I meant to say [15:08] they're both ubuntu-vg === madmax__ is now known as madmax [15:13] causative, I see VG mention in '/etc/fstab' and '/boot/grub/grub.cfg' . Why you are not setting up in a desired way /etc/fstab and /boot/grub/grub.cfg too along with PV and/or VG names? [15:14] is that all I need to change? I am concerned I might miss somewhere else that it's mentioned, and leaning towards doing a full reinstall of one of them [15:14] I didn't manually set up fstab or grub.cfg [15:16] I never did exactly this operation. But, if you have no encryption or other complexity, it is a matter of proper configuration for mount points. Thus it is a matter of fstab and Grub set up. File systems are not renamed/reformated and will keep their UUIDs. [15:16] I do have encryption on both of them [15:17] Is it default Luks encryption with plain /boot ? [15:19] on the new installation it is, the old installation also has ecryptfs home directory which is another thing I have to figure out because I want to access it from the new one [15:20] but I' [15:20] I'm thinking of changing the new installation vg name, not the old installation vg name [15:24] Ecryptfs is not related to Luks and not affected (except obvious things) in any way by LVM names. [15:26] LVM volumes are created inside Luks objects. Thus, it seems 'update-grub' may do it's job well in case of LVM names change, Luks encryption, and if rest of changes applied. [15:28] I.e. Luks volume assembled before an LVM and it means order and role of a needed changes. [15:37] hi. is this channel silet, or i misconfigured someting (wechat)? [15:37] Hi guys, I just installed UIbunto on a laptop with the screen damaged using a second monitor. I can't log in because the log in screen is on the main screen. What can I do to log in? [15:38] causative, here is how it works in details - https://help.ubuntu.com/community/Full_Disk_Encryption_Howto_2019 In corresponding places corresponding names should be changed. And Luks and Ecryptfs are totally different things. [15:38] Ashir: ssh to the host, if SSH is enabled by default [15:39] Ashir, Ctrl+Alt+F3 brings login into text TTY? [15:41] Bart__ YES!, I could log in, now how I go back into GUI? [15:42] I can invent a workaround. Do not know direct answer. [15:42] One thing is lxrandr or arandr have support for CLI and can shutdown laptop's LVDS display. [15:43] Ashir, you could use the terminal to download x11vnc and start it, then from a second computer you could connect to the x11vnc server, which would allow you to log in graphically, and then from the graphical environment you could change your display settings [15:44] the second computer can be a phone [15:44] When the display will be off it should auto jump onto another physical display. May be "sudo systemctl restart gdm" will be needed. [15:45] Or - did you tried to move mouse from first display to second? I saw some where when login prompt followed the mouse. [15:45] GUI can be found via Ctrl+Alt+F1 F2 and so on. [15:46] Ashir, you could also just try typing your password blindly on the login screen, try it a couple times in case it needs to select the password prompt [15:46] Also there is a way to shutdown GDM and start an Xsession from CLI. It's not very straightforward. [15:47] then once you've logged in that way you can right click on the monitor that you can see, and click display settings [15:49] Ashir, there's also this advice to "unplug the main laptop broken screen from the inverter" https://h30434.www3.hp.com/t5/Notebook-Video-Display-and-Touch/Bios-not-showing-up-on-a-external-monitor/td-p/5827484 [15:49] and that will let you get into the bios and not just your ubuntu desktop [16:04] well thanks Bart__ but this is getting too complicated, I think I'll go with a fresh install === docmax is now known as Guest2987 [16:31] causative I install x11vnc and a client vnc on other PC but I can't log in becuase of the MIT-MAGIC-COOKIE and I don't know how to resolve it... [16:33] Ashir, (old but shoudl still work) https://askubuntu.com/questions/1035598/ubuntu-18-04-lts-x11vnc-no-longer-works [16:58] hey millet türk varmı la [16:58] !tr | X-REDOC [16:58] X-REDOC: Turk ubuntu kullanıcıları, Türkçe yardım ya da geyik için /join #ubuntu-tr hizmetinizde. [17:01] aga t [17:01] türk kanalında kimse yok [17:02] :D [17:02] !english | X-REDOC [17:02] X-REDOC: The main Ubuntu IRC channels are English only. For a complete list of channels in other languages, please visit https://wiki.ubuntu.com/IRC/ChannelList === causative is now known as treehouse [17:14] is there any way to get phpbb to use the sendmail function? [17:14] Wrong channel [17:29] heyyo [17:29] kimse varmı [17:30] my english is bad sory [17:30] :D [17:30] we only support in english here X-REDOC [17:30] the turkish channel, i linked you earlier [17:32] there is no one there ?? :D [17:33] where did the turks go :D [17:33] ask your question in english here X-REDOC [17:37] i speak english already mk === tristan123 is now known as tristan_ [18:34] ubuntu 23.04 takes for ever to start up. it hangs at this job. Its happened on the LTS version as well. Now ive upgrded to the latestet and greatest and it does the same thing. Job dev-disk-by\x2duuid-472a6f1f\xxx\xxx\xxx\xxx\x2d3d979c04ae07.device/start [18:35] vershan: Are you sure the disk is OK? [18:36] @elias_a, yeah I have dual boot to windows. all ok. just this startup job. [18:37] i remember using gparted before this happend to delete a partion that was causing problems. I had to boot from live CD and fix. Not sure what went wrong there [18:37] i used a 3rd party grub bootloader, then i deleted and defaulted and now this [18:40] @elias_a, I used boot-repair from the live CD, thats when i noticed the issue. How can I investigate this further [18:40] vershan: Sorry - I cannot think of anything else than a defective disk. [18:42] @elias_a, I've performed a disk self check it says its ok [18:47] I'm having trouble getting access to my ecryptfs home directory from a second system. When I boot to the system that has ecryptfs, it decrypts everything on login just by typing my password, no problem. But methods to open it from the second system fail [18:48] vershan: Hang around here - perhaps someone comes up with better ideas. [18:48] oh wait I just tried something new and it worked [18:48] ok sure thanks === zniavre__ is now known as zniavre [19:02] no, it didn't work. I've tried ecryptfs-recover-private, which is telling me "Success!" but my files are not there [19:03] I've tried the method described here https://help.ubuntu.com/community/EncryptedPrivateDirectory#Recovering_Your_Data_Manually but it ends with "Error mounting eCryptfs: [-2] No such file or directory" [19:08] causative: check dmesg? maybe there's messages there [19:19] A different problem: when I boot, it does not prompt me for the LUKS password and as a result it doesn't decrypt the disk. This is easily fixed from initramfs if I just type "cryptsetup open /dev/nvme0n1p6 nvme0n1p6_crypt" and enter my password, and then it will finish booting. But that is inconvenient to do every time and should be done automatically. What am I missing? [19:20] what's in your /etc/crypttab file? I've got this: luks1 UUID=8fda5814-5cf7-404b-9339-54088c28911b none luks,discard,initramfs [19:21] ah, I don't have one! that's probably the issue [19:21] I don't actually know how all this works since my setup just kinda worked, but my wild guess is that you might be missing the line entirely, or 'initramfs' or something like that [19:30] what does the "initramfs" option do? it's not listed at https://www.man7.org/linux/man-pages/man5/crypttab.5.html [19:31] oh wait nvm it is listed on the man page on my local machine [19:35] sarnold, I put nvme0n1p6_crypt UUID=d6c9f6db-e974-445a-807b-86715b62bbd9 none discard,luks,initramfs in my /etc/crypttab but it didn't fix the problem [19:36] causative: oh dang :( I had big hopes on that :( [19:36] also I'm confused about how it could fix the problem, because isn't /etc/crypttab *inside* the encrypted partition? [19:36] so how could it be accessed before the partition gets decrypted? [19:36] causative: oh, try re-generating your initramfs and rebooting [19:39] Please help also vershan if you can of think of something: 21:34 < vershan> ubuntu 23.04 takes for ever to start up. it hangs at this job. Its happened on the LTS version as well. Now ive upgrded to the latestet and greatest and it does the same thing. Job [19:39] dev-disk-by\x2duuid-472a6f1f\xxx\xxx\xxx\xxx\x2d3d979c04ae07.device/start [19:42] once the system is up, is there anythin gin the logs? how's smart data look on that drive? [19:51] when I try to run update-initramfs -u what I get is: "cryptsetup: WARNING: target 'nvme0n1p6_crypt' not found in /etc/crypttab" [19:52] I'm not sure what this means and I haven't found a clear answer. have I misnamed something? should I ignore it? [19:52] because I do have exactly the target "nvme0n1p6_crypt" in /etc/crypttab [19:52] so what does it mean that it's not found [19:55] man cryptsetup really needs a simple 'status' command that just dumps all the informatoin [20:16] idk, it didn't work [20:18] rebooted, it dumped me to a prompt, and again I had to say cryptsetup open /dev/nvme0n1p6 nvme0n1p6_crypt before it would boot === Guest3173 is now known as EriC^^