[01:44] <pycurious> If my box is connected to both wireless and wired networks, and if one goes down - does ubuntu does an automatic failover?
[01:59] <InPhase> pycurious: The network manage will typically prioritize the wired and fall over to the wireless.
[02:00] <InPhase> s/manage/manager/
[02:00] <InPhase> pycurious: It's pretty easy to test.  Unplug it.
[02:01] <raczka> are you s sure his router doesn't blow up
[02:02] <raczka> emotions like watching MacGyver
[03:47] <DrMax> I'm trying to get CR3 thumbnails with ubuntu 22.04... it almost works: it calls the custom thumbnailer, but nemo complains about unknown mime type with error
[03:47] <DrMax> Unable to create loader for mime type image/x-canon-cr3: Unrecognized image file format
[03:47] <DrMax> what do I have to do on top of adding mimetype to the mimetype database to have nemo recognize those^
[03:47] <DrMax> ?
[04:45] <kirrim> > I'm trying to create a new user and then su into it but I get this error:
[04:45] <kirrim> > ubuntu@-:~$ sudo useradd --create-home --password password testuser99
[04:45] <kirrim> > ubuntu@-:~$ su testuser99
[04:45] <kirrim> > Password:
[04:45] <kirrim> > su: Authentication failure
[04:47] <alkisg> kirrim: man useradd => -p, --password PASSWORD           The encrypted password, as returned by crypt(3).
[04:47] <alkisg> Are you passing the encrypted password or the plain password in the command?
[04:48] <kirrim> so I've been messing up the password this whole time, forgetting to encrypt it? Thanks for the explanation. I'll try again with a properly encrypted password. The help page just said "  -p, --password PASSWORD       encrypted password of the new account" and I assumed that it would just encrypt the password automatically.
[04:50] <alkisg> One easy way is to use `passwd` to set a password, and then see the result in /etc/shadow, and then copy it and pass it to useradd later on
[04:56] <oerheks> why making an user the hard way?
[04:56] <oerheks> https://askubuntu.com/questions/345974/what-is-the-difference-between-adduser-and-useradd
[04:56] <oerheks> that is, if you are using ubuntu
[05:00] <alkisg> adduser doesn't support passing the password, you'd need to pipe it then
[05:01] <alkisg> (which might be a better choice if you want to avoid people seeing the hash with a plain `ps` command)
[05:06] <kirrim> I ended up creating the user without a password then used "sudo passwd <new user name>
[05:07] <kirrim> " to change its password to something I can use
[05:12] <alkisg> I want to upgrade some 20.04 installations to 22.04. I noticed that with `Prompt=lts` in /etc/update-manager/release-upgrades, it says "no upgrade available", while with `Prompt-normal` it finds the jammy upgrade
[05:13] <alkisg> Should I wait for a couple more days, is it supposed to be enabled right after 22.04.1 is out, or should I really change the prompt there to normal?
[05:13] <Hash> I think it's because this method of upgrade becomes available a month or two after release, to make sure bugs are gone
[05:15] <Hash> You can do it manually using apt-get
[05:15] <alkisg> The release was in April and indeed AFAIK `do-release-upgrade` is supposed to be available in August, 4 months later, but I'm puzzled about the normal vs lts difference..
[05:15] <Hash> disable all 3rd party and ppa, change sources.list, update, upgrade, etc.
[05:15] <rfm> alkisg, the danger of setting promot to normal is if you don't change it back, you'll be prompted to upgrade to 22.10 when it comes out
[05:15] <alkisg> I want to do that "by the book" as it'll go in thousands of installations, I don't want to risk manually reviving all of these :D
[05:16] <morganu> there was sound when I left home. Now there is no sound, 2 browsers, unplugged speaker, sound tab shows sound max. 20.04  what else could go wrong.
[05:16] <rfm> alkisg, personally I'm just waiting for the LTS upgrades to be enabled, should just be a few more days
[05:16] <Hash> Are you on 21.10?
[05:17] <Hash> Ubuntu 20.04 will not detect a newer version until Ubuntu 22.04.1 is released, which is scheduled for 2022-08-04.
[05:17] <Hash> That's tomorrow
[05:17] <alkisg> rfm, so "prompt=normal" was "working" even in e.g. May, it's a "workaround" for people that want to upgrade 20.04 to 22.04 before it's officially enabled?
[05:17] <Hash> You can wait until tomorrow.
[05:17] <Hash> Upgrades from one LTS to the next LTS release are only available after the first point release. For example, Ubuntu 18.04 LTS will only upgrade to Ubuntu 20.04 LTS after the 20.04.1 point release. If users wish to update before the point release (e.g. on a subset of machines to evaluate the LTS upgrade) users can force the upgrade via the -d flag.
[05:17] <alkisg> 2022-08-04 is today, sure of course I can wait
[05:18] <Hash> Feel free to force.  I did, few weeks ago. -d worked fine.
[05:18] <Hash> https://ubuntu.com/server/docs/upgrade-introduction read this
[05:18] <alkisg> Hash, but it works for me without -d
[05:18] <alkisg> That's why I'm asking
[05:18] <Hash> So what's the problem? I'm not sure
[05:18] <alkisg> Let me rephrase, is "Prompt=normal" the same as "-d"?
[05:18] <alkisg> And where is that documented?
[05:18] <Hash> No, -d is to force
[05:19] <rfm> alkisg, bascially setting prompt=mormal says "I want to get off the LTS train and do the every-six-month".  I suppose it's nice that it lets you go straight from 20.04 to 22.04.
[05:19] <Hash> prompt normal = every 6 month release
[05:19] <Hash> prompt lts = every LTS release.
[05:19] <alkisg> And yet now that 22.04 is lts, they behave differently
[05:19] <alkisg> As "normal" allows the update, and "lts" doesn't
[05:20] <Hash> Because 20.04.1 point release isn't out yet
[05:20] <Hash> Please read the documentation so you can understand how it works.
[05:20] <Hash> 22*
[05:20] <alkisg> I haven't seen that fact documented anywhere, that's what I'm asking for...
[05:20] <alkisg> I.e. what prompt=normal is supposed to do for lts releases
[05:21] <Hash> From LTS, 20.04 which you are on, you cannot do upgrade to 22.04 LTS if 22.04.1 isn't out yet.
[05:21] <Hash> This will be out today.
[05:21] <alkisg> From 20.04, with prompt=normal, I'm allowed to upgrade. That is the part that I cannot find in any documentation.
[05:22] <Hash> I see what you're saying
[05:22] <alkisg> In any case, you both arrived to the same conclusions as me, thank you for your feedback
[05:22] <alkisg> I.e. that "prompt=normal" behaves differently than "prompt=lts" for lts editions, even though it's undocumented
[05:28] <morganu> fixed sound by rebooting.
[05:28] <morganu> hard reboot. second time today.
[05:30] <rfm> alkisg, now that I think about it it can't do anything else.  By setting prompt=normal you're getting off the LTS sequence, so strictly it should upgrade as if a normal release, from 20.04 to 20.10.  But 20.10 is already EOL.  And skipping a release in the normal chain never works (e.g. could neve never have upgraded 20.04->21.04)
[05:30] <rfm> alkisg, so the only possible thing it can do is offer the 20.04->22.04 upgrade
[05:31] <alkisg> rfm, or it can just say "it's not available yet because it's not enabled, it's in development"
[05:31] <Hash> https://bpa.st/ZGCQ here's the checks
[05:31] <Hash> Read this.
[05:32] <Hash> or, less /usr/bin/do-release-upgrade
[05:34] <rfm> alkisg, but it's not in development on the non-lts train (21.10->22.04 upgrades would have been enabled from the 22.04 release date)
[05:35] <alkisg> Hash, /usr/lib/python3/dist-packages/DistUpgrade/MetaRelease.py downloads https://changelogs.ubuntu.com/meta-release in one case, vs https://changelogs.ubuntu.com/meta-release-lts in the other
[05:36] <alkisg> So it is indeed as we assumed, it's just undocumented. And to easily see when "lts upgrades are enabled", we need to check that last URL I pasted
[05:37] <alkisg> I.e. the different implementation of "not allowing normal upgrades to 22.04 yet" would be "not to upload the new meta-release file", an administrative process, not a coding process
[05:38] <alkisg> Their decisions are fine, they're just undocumented...
[07:14] <tomreyn> alkisg: to put it differently: until 21.10 went EOL, Prompt=normal would have upgraded your 20.04 to 21.10. but 21.10 is is no longer supported, it would attempt to upgrade to the next supported release, which happens to be 22.04 now. Mind you, any maintained system following the 6 month upgrade cycles would not end up in this situation in the first place - it would have upgraded from 20.04 to 20.10, then 21.04, then 21.10, and since to 22.04
[07:14] <tomreyn> (since *this* path to 22.04 has been supported since the 22.04 release).
[07:15] <tomreyn> s/but 21.10 is is/but since 21.10 is/
[07:15] <alkisg> tomreyn: my "objection" was: document somewhere that focal appears in https://changelogs.ubuntu.com/meta-release but not in https://changelogs.ubuntu.com/meta-release-lts yet
[07:16] <alkisg> I.e. that it's available as a normal upgrade but not as an lts upgrade. It's a sane thing to do, but it's not the only sane thing to do. Just put it somewhere so that people don't have to look at the code to find it
[07:17] <alkisg> *s/focal/jammy;/
[07:17] <alkisg> E.g. if two people tell me "(a) oh my 20.04 got upgraded to 22.04, and (b) mine didn't!", I wouldn't know it's due to that undocumented behavior
[07:23] <tomreyn> alkisg: this could be worth filing a feature request for documentation, if such is lacking.
[07:24] <alkisg> A couple of more lines in /etc/update-manager/release-upgrades should suffice
[07:26] <alkisg> Btw, all this discussion made me wonder if setting Prompt=lts in 20.10, would allow direct upgrades to 22.04... :D
[07:26] <alkisg> Ah no scratch that,it's already answered in that documentation in that file
[07:33] <alkisg> Submitted https://bugs.launchpad.net/ubuntu/+source/ubuntu-release-upgrader/+bug/1983553
[08:06] <El_Presidente> yesterday I tested the upgrade from 20.04 to 22.04, but I ran into an issue that unbound DNS could not read the config file
[08:07] <El_Presidente> is this a known issue?
[08:19] <tomreyn> El_Presidente: while unbound is a community maintained package (in universe), filing a bug report would be good if you were using the ubuntu 20.04 package.
[08:20] <El_Presidente> tomreyn, I understand
[08:20] <El_Presidente> thank you
[08:20] <El_Presidente> yes, I am using it
[08:20] <tomreyn> when you do, please include concrete error messages
[08:21] <El_Presidente> I guess I have to upgrade again then, since I reverted my system last night
[08:21] <El_Presidente> but I will file a bug
[08:22] <El_Presidente> basically it looked like that: fatal error: Could not read config file: /etc/unbound/unbound.conf and one line above was a syntax error mentioned in my /etc/unbound/unbound.conf/xyz.conf:2
[08:24] <alkisg> If your own xyz.conf configuration file syntax is incompatible with the newer unbound version, it's up to you to upgrade it, I wouldn't classify that as an upgrader issue
[08:25] <tomreyn> i would. this situation should result in a prompt asking whether to keep the old, possibly outdated, locally changed file, or the new versions'
[08:26] <tomreyn> it should not result in a fatal error, failing the entire upgrade
[08:27] <tomreyn> i guess that if the 20.04 configuration is already syntactically broken, that'll be difficult to fix during the upgrade. but so far we don't know that to be the case.
[08:28] <tomreyn> (it's not clear whether this message is issued by the 20.04 or the 22.04 unbound package version - more likely the 22.04 one, though)
[08:37] <El_Presidente> tomreyn, the 20.04 works flawlessly
[08:38] <El_Presidente> alkisg, the first line unbound does not like is hide-identity: yes
[08:38] <El_Presidente> which is still supported
[08:39] <El_Presidente> if I comment that out then unbound says its hide-version: yes
[08:39] <El_Presidente> which is also still supported
[08:39] <alkisg> El_Presidente: and if you remove your xyz.conf file, it can be started properly?
[08:39] <alkisg> For example, you might have edited that file using CRLF instead of LF, and unbound might not like that part
[08:39] <El_Presidente> alkisg, I cannot test that right now, because I reverted the VM to 20.04
[08:40] <El_Presidente> alkisg, I will test that tonight
[08:40] <El_Presidente> or I simply could set up a new VM
[08:40] <alkisg> 👍️
[08:41] <El_Presidente> I will set up a fresh 22.04
[08:41] <El_Presidente> or even better I create one from my snapshot and update that
[10:06] <arseru> Hi there! My ubuntu 20.04.4 LTS is telling me my /boot is almost full (23M free space), more info of what's in there https://bpa.st/6TIA
[10:06] <arseru> My question: running `apt autoremove` doesn't clean anything, what should I do from there?
[10:07] <arseru> I want to understand why autoremove isn't removing old kernel images
[10:12] <ogra> autoremove only removes automatically installed packages, if you installed any of these kernels manually it will ignore them ...
[10:13] <arseru> ogra: I didn't install them manually
[10:14] <ogra> on a proper default Ubuntu "apt-mark showmanual|grep linux-image" should return nothing ... and "apt-mark showauto|grep linux-image" should list all your kernel packages
[10:14] <ogra> one other thing to take into account is that on upgraded systems the original install kernel is always preserved ... you can remove it manually though
[10:19] <toddc> arseru: I had the same issue using ZFS snapshots filled up boot I had to prune them or do a manual update
[10:19] <arseru> ogra: the outputs are as you expected, maybe is it because I have a pending `linux-firmware` upgrade?
[10:20] <arseru> toddc: thanks!
[10:21] <toddc> arseru: you should check the boot partion so see what is in there also ..cd /boot then ls to see what is listed
[10:22] <arseru> toddc: I posted it above: https://bpa.st/6TIA
[10:24] <arseru> I'm running `apt-get --purge remove linux-image-5.4.0-58-generic` manually for each old kernel but at the end I get the error: `rmdir: failed to remove '/lib/modules/5.4.0-58-generic': Directory not empty`
[10:24] <arseru> that doesn't look too good XD
[10:25] <ravage> that usually only means that there are files in there that are not part of the package
[10:25] <ravage> you can remove the dirs after you removed the packages
[10:26] <ravage> dpkg --list | grep -i -E --color 'linux-image|linux-kernel' | grep '^ii'
[10:26] <ravage> will show all installed kernels
[10:26] <ravage> keep the latest 2
[10:26] <ravage> when you are done make sure "update-grub2" find them correctly
[10:27] <ravage> dpkg --list | grep -i -E --color 'linux-image|linux-headers' | grep '^ii'
[10:27] <arseru> ravage: thanks, I didn't know about `update-grub2` :D
[10:28] <ravage> shows the headers installed too
[10:28] <arseru> I've kept the last 3 kernels, I'll reboot now, wish me luck :D
[10:28] <ravage> good luck
[10:32] <arseru> now autoremove finds leftover headers, yay :) https://bpa.st/T6RQ
[10:32] <lotuspsychje> apt is so smart :p
[10:43] <WoolHat> Hello, I have a question, I am now running Ubuntu 21.10 and I wanted to upgrade to 22.04 LTS but neither terminal nor GUI upgrade works -> I cant find anything on the net
[10:44] <lotuspsychje> !eolupgrade | WoolHat
[10:45] <WoolHat> @lotuspsychje it also didn't work before EOL and I tried to follow the steps for EOL but those didnt work either
[10:47] <lotuspsychje> WoolHat: can you pastebin the errors apt gives you after trying the upgrade? volunteers might be able to help
[10:47] <Guest7026> The recent Ubuntu desktop version is 22.04 LTS and its recent update was on April 2022.
[10:47] <Guest7026> I read somewhere that the next update of that version (something like "22.04.01 LTS") should be in August 2022.
[10:47] <Guest7026> Is that true?
[10:47] <Guest7026> Does Ubuntu plan to release the next version soon?
[10:48] <murmel> Guest7026: today is 22.04.1
[10:49] <Maik> if you use Ubuntu 22.04 and kept it up to date you already have the 22.04.1 point release
[10:49] <Guest7026> In the Ubuntu website, it says "Ubuntu 22.04 LTS"
[10:49] <Maik> check again tomorrow
[10:49] <Guest7026> Why version 22.04.01 isn't on the website?
[10:49] <Guest7026> Oh ok
[10:49] <murmel> Guest7026: afair, it's getting released 12 UTC
[10:49] <Guest7026> Cool, thank you
[10:49] <murmel> which is still a bit off
[10:50] <Maik> Guest7026: and 22.04.1 is not a new release
[10:50] <Guest7026> so how do I call the version "22.04.01" if it's not a release?
[10:50] <Maik> it's a up to date iso with all updates and bug fixes released up to now
[10:50] <murmel> you call it 22.04.1 as it's a bugfix release for 22.04
[10:51] <Guest7026> Ok, thank you
[10:57] <ogra> Guest7026, the .1 just means we spin a new iso with all the updated packages ... if you run 22.04 you got these packages anyway with your regular package updates ...
[10:58] <ogra> Guest7026, you can check your local version with: "grep VERSION= /etc/os-release" ... it should already say 22.04.1
[11:03] <Guest7026> Great, thanks!
[11:08] <horse9> @ogra oh, hey that's neat, thanks
[11:11] <horse9> is there a good gnome channel on liberachat?
[11:14] <Maik> horse9: had a lookbut nope
[11:16] <Maik> horse9: https://wiki.gnome.org/GettingInTouch/IRC
[11:29] <ice9>  22.04.1 will be released today?
[11:29] <murmel> y
[11:29] <bittin> 'yes in 5,5 hours
[11:30] <ice9> where is the release notes?
[11:32] <Maik> ice9: once it's released
[11:32] <Maik> ice9: also if you are on 22.04 you already have 22.04.1
[11:33] <Maik> ice9: 22.04.1 is just a bugfix, up to date release, no need to reinstall
[11:33] <ice9> Maik, oh you mean packages were already upgraded on 22.04 while 22.04.1 will be released as a distro for upgrade from 20.04?
[11:34] <Maik> 22.04.1 is a point release
[11:34] <bittin> ice9, yeah yesterday evening
[11:34] <bittin> if you run sudo apt update; sudo apt upgrade
[11:34] <bittin> and sudo reboot
[11:35] <Maik> the upcomming days/weeks the upgrade path will be enabled from 20.04 to 22.04
[11:36] <Maik> bittin: i have 22.04.1 already due updates from last weekend
[11:37] <Maik> it's always a few day to a week prior to the iso pointy release
[11:37] <Maik> s/pointy/point
[12:16] <Maik> ice9: maybe also interesting: https://discourse.ubuntu.com/t/jammy-jellyfish-point-release-changes/29835
[12:27] <lexandrop> Maik: "the upcomming days/weeks the upgrade path will be enabled from 20.04 to 22.04" - oh, so is it actually a thing? I'm on 20.04 currently, and was curious why I did not see any "update your OS" messages since 22.04 release.
[12:27] <lexandrop> I had a thought "maybe it will happen later", but wasn't sure.
[12:27] <ravage> soon ™
[12:27] <lotuspsychje> !ltsupgrade | lexandrop
[12:28] <BluesKaj> Hi all
[12:30] <lexandrop> lotuspsychje: oh, great to know, thanks. Yes, I've seen multiple mentions "wait until X.04.1 release before upgrading" on reddit/forums, but I did not realize even ubuntu maintainers use this practice :) Thanks.
[12:30] <lexandrop> maybe not "... use this practice", but "... respect this practice".
[12:35] <lotuspsychje> lexandrop: also make sure your software&updates are set to LTS releases
[13:01] <FanOfFan> hello guys and girls, i have old laptop: https://www.pcworld.idg.com.au/review/acer/aspire_one_zg5_linux/255307/specs/ Acer ZG5 with intel atom cpu 1.6Ghz and 1.5GB ram. Currently installed lubuntu and laptop is pretty slow, is xubuntu better choice?:)  And my laptop supports 64bit OS? Thanks for info and your time! :)
[13:04] <FanOfFan> is anybody alive here? :)
[13:07] <samy1028c> Hello FanOfFan
[13:08] <samy1028c> How do you mean by "slow".  What exactly is it slow in doing?  General usage?  Youtube?  Games?
[13:09] <FanOfFan> yeap general usage, my old laptop is pretty slow, so i'm thinking which OS is better for it... xubuntu or lubuntu :)
[13:10] <guiverc> FanOfFan, you didn't say what release; but modern Lubuntu is LXQt which uses Qt5, so it'll waste ram if you're using GTK apps; Xubuntu is GTK3 so will likely do better with GTK3 apps, but Lubuntu will do better with Qt5 apps; match your Desktop choice with apps you'll use, esp. if you're limited in RAM
[13:10] <guiverc> ie. Lubuntu is lighter than Xubuntu yes, but wrong app choice can waste a light DE & make it worse than other choices
[13:11] <FanOfFan> so for firefox which OS is better?
[13:11] <FanOfFan> or chrome
[13:11] <guiverc> what release?
[13:12] <guiverc> Firefox maybe a snap or deb package (depending on release), chrome is a deb package
[13:12] <FanOfFan> i\m choosing latest long term release
[13:13] <guiverc> so 22.04; firefox is a snap package
[13:13] <guiverc> With 1.5GB of RAM you need to not waste resources with RAM; do you have SWAP enabled; you can likely improve performance just by increasing swap.
[13:15] <FanOfFan> well, my internal hardrive is only 8GB :)
[13:16] <pertho> Hi.. I'm trying to compile this from source on an Ubuntu 18.04 machine: https://launchpad.net/ubuntu/+source/mariadb-10.0/10.0.23-1 and I'm using 'dpkg-buildpackage -rfakeroot -b -uc -us' and it starts building.. gets to a point.. then dies.. but I can't tell where it died
[13:17] <lexandrop> lotuspsychje: "also make sure your software&updates are set to LTS releases" - thank you, I never bothered to check that :) luckily, yes, by default it was set to LTS
[13:17] <guiverc> FanOfFan, your problem is lack of resources; too little RAM which requires more swap size (you'll notice a big performance bump with it), but 8GB disk is too small for any desktop in my opinion; result of lacking resources like you have will mostly be a slow system & Xubuntu maybe worse; I've not tested it in that little disk
[13:17] <ravage> FanOfFan, that CPU is 32bit only. so the latest available Ubuntu release is 18.04. maybe better choose debian
[13:20] <guiverc> FanOfFan, what CPU; the page you provided only said intel atom (not which atom), but you said latest Lubuntu.  Lubuntu 18.04 isn't supported; it reached EOL in 2021-April
[13:20] <pertho> mariadb-10.0.23/storage/mroonga/vendor/groonga/lib/expr.c:3502:1: warning: const/copy propagation disabled: 18978 basic blocks and 70472 registers; increase --param max-gcse-memory above 167310048 [-Wdisabled-optimization]
[13:20] <pertho> yes I know it's an ancient version of mariadb-server but I need it
[13:21] <ravage> guiverc, Atom N270
[13:22] <FanOfFan> ok, thanks for info, i will try xubuntu anyway and maybe debian :)
[13:22] <guiverc> thanks ravage ; you'll do better without a DE (desktop) on a n270... A WM only provides better performance (I used one in QA of release up to 19.04 which was last i386) FanOfFan
[13:23] <ravage> best use is a door stopper :)
[13:24] <ogra> you could run ubuntu core  on it and control your window shades or so ...
[13:24] <ogra> doo stopper is a bit harsh ... it surely has the prformance of an RPi 3
[13:24] <ogra> *door
[13:29] <El_Presidente> alkisg, tomreyn I created a bug report: https://bugs.launchpad.net/ubuntu/+source/unbound/+bug/1983586
[13:32] <pertho> can someone please help? Trying to compile mariadb-server 10.0.23 (yes, I know it's old) using ubuntu source package at https://launchpad.net/ubuntu/+source/mariadb-10.0/10.0.23-1 and it's dying due to some compiler option. Error is: https://pertho.net/pb/HP6CcsJZ
[13:32] <pertho> this is on 18.04
[13:33] <leftyfb> pertho: you are better off installing that package on an older, unsupported Ubuntu release in an LXD container just long enough to work on upgrading your db to a modern and supported version
[13:34] <FanOfFan> what means door stopper? :)
[13:35] <pertho> leftyfb: I can't.. I have to set up an environment like from another server I don't have access to and it has to be that version of MariaDB.. this contains the full error: https://pertho.net/pb/RpiqJVB9
[13:37] <pertho> perhaps some CFLAGS I could pass to debuild to fix: warning: const/copy propagation disabled: 18978 basic blocks and 70472 registers; increase --param max-gcse-memory above 167310048 [-Wdisabled-optimization]
[13:38] <SteelRose> pertho: have you installed all the dev-packages needed?
[13:39] <tomreyn> El_Presidente: please run    apport-collect 1983586   on the non-upgraded system   to provide additional log files.
[13:39] <pertho> SteelRose: yes I have all the requisite dev packages installed which I got from the .dsc file
[13:40] <ravage> El_Presidente, your config just has the wrong syntax. it is not part of the package. hide-version for example belongs in the server: section
[13:40] <leftyfb> pertho: on your ubuntu 18.04 server, you can install lxd and then launch an ubuntu 16.04 and mariadb-server version 10.0.38 (you don't need exactly 10.0.23). Then work towards converting that db to a more modern and supported version of mariadb and ubuntu release
[13:40] <Maik> ice9 lexandrop murmel lotuspsychje https://discourse.ubuntu.com/t/ubuntu-22-04-1-delayed-until-august-11/29859
[13:40] <ravage> El_Presidente, something like https://p.haxxors.com/rwrc7cz2.txt should pass the config test
[13:41] <ravage> Maik, oh snap :D
[13:41] <leftyfb> pertho: we cannot support an outdated package from an EOL release of ubuntu and you shouldn't be trying to compile it
[13:41] <El_Presidente> tomreyn, I added the apport-collect
[13:43] <pertho> leftyfb: yes, I understand that.. I was hoping to at least try to get it compiled.. it looks like it just needs a compiler parameter changed somewhere
[13:43] <pertho> or even a CFLAGS setting
[13:43] <leftyfb> pertho: it's very likely it's looking for outdated library versions which aren't available on supported releases of ubuntu
[13:44] <leftyfb> pertho: trying to get this package compiled isn't where you should be focusing your efforts on. It should be on converting the db to a version of mariadb that is supported
[13:44] <El_Presidente> ravage, indeed server: was missing
[13:45] <El_Presidente> sorry for all the fuzz :(
[13:45] <pertho> leftyfb: 10.1.48 is available in 18.04.. I guess I could get it updated to that
[13:45] <El_Presidente> just odd, that it ran before with my faulty config
[13:47] <ravage> Maybe the older version assumed that section for includes
[13:47] <leftyfb> pertho: you'll be right back where you started in 8 months from now when 18.04 goes EOL. Use that time to convert the db to work in 10.6.x which is available on ubuntu 22.04
[13:48] <El_Presidente> ravage, yeah maybe
[13:48] <El_Presidente> sadly, the error message was just not very helpful for me
[13:48] <El_Presidente> I would have bet on a permission issue
[14:10] <gordonjcp> is there a way to run a specific version of the NVidia drivers? I appear to have been "upgraded" to 515.48 but I need 515.43
[14:11] <tomreyn> El_Presidente: oh, so i seem to have misunderstood you: your bug report suggests that the release upgrade did finish successfully, just unbound fails to start on the upgraded system. is this correct? previously i understood that you were reporting that the upgrade to 22.04 failed entirely due to this configuration file.
[14:11] <El_Presidente> tomreyn, sorry that I may have been not precise enough
[14:11] <tomreyn> El_Presidente: no problem. but can you confirm which one it is?
[14:12] <El_Presidente> the upgrade to 22.04 went through
[14:12] <tomreyn> thanks
[14:19] <tomreyn> gordonjcp: there is apt pinning, but you should not use outdated and insecure software.
[14:19] <tomreyn> gordonjcp: what makes you need the older version?
[14:38] <El_Presidente> ravage, I created a feature request at unbound, for expanding the usecase of unbound-checkconf, since you found my error, maybe you have something to add?
[14:38] <El_Presidente> https://github.com/NLnetLabs/unbound/issues/734
[14:40] <ravage> El_Presidente, my only instance of unbound is on my pihole. i just read the manpage and your error output really :)
[14:41] <El_Presidente> ah! ty
[14:49] <_WEZ_> El_Presidente! How are the bananas growing in your republic these days?
[14:49] <leftyfb> _WEZ_: feel free to have that conversation in #ubuntu-offtopic
[14:50] <El_Presidente> _WEZ_, no bananas this year, just tomatoes, cucumbers and beans ;)
[14:50] <El_Presidente> but thanks to the sun, I think my tomatoes will be great
[14:50] <El_Presidente> and sorry for offtopic
[14:50] <_WEZ_> leftyfb: El_Presidente isn't in offtopic though so I can't take it ther eyet
[14:51] <_WEZ_> leftyfb: What's the best course of action now?
[14:52] <leftyfb> _WEZ_: this is a support channel. Let us know if you are having any issues with ubuntu and we'll be happy to help you. Take all other topics to #ubuntu-offtopic
[14:53] <_WEZ_> leftyfb: No issues here, I am here to help other people!
[14:54] <ravage> so we wait in silence until a support question arises
[14:55] <_WEZ_> El_Presidente: Thanks for responding to me here but leftyfb would like it if we moved further offtopic discussions to #ubuntu-offtopic. For some reason leftyfb only said that to me and not you as well, I guess that was just an oversight
[14:56] <_WEZ_> ravage: yeah, there is a big gap, you would think there would be space for some offtopic chat here.  It's like a train officier telling you to take your suitcase off the chair next to you when the carriage is empty
[14:58] <El_Presidente> _WEZ_, leftyfb has a valid point. because it can quickly get crowded and then an offtopic discussion can make things most unclear ...
[14:58] <_WEZ_> El_Presidente: Lets move it to the other channel then
[14:59] <_WEZ_> El_Presidente: yeah, and if it gets crowded on a train I will move my bag...
[15:06] <GWM> I got a weird 22.04 issue, as a server, I can't get systemd to respect DNSStubListener=no, it still binds an lo iface port with :53, which makes another software that is also a buggy piece of garbage cry
[15:07] <GWM> Still, I expected systemd to do what it's configured for, anyone encountered this issue?
[15:07] <gordonjcp> tomreyn: the latest version of the NVidia drivers doesn't work
[15:07] <gordonjcp> tomreyn: the previous version did
[15:09] <gordonjcp> am I missing something obvious, or does the currently packaged version of Audacity no longer support any plugins?
[15:09] <GWM> (I configure the DNSStubListener in /etc/systemd/resolved.conf)
[15:10] <GWM> systemd-analyze cat-config systemd/resolved.conf notes that DNSStubListener=no is correct, naturally I've restarted the service (and rebooted a few times)
[15:11] <GWM> yet
[15:11] <GWM> LISTEN 0 4096 127.0.2.1:53 0.0.0.0:* users:(("systemd",pid=1,fd=53))
[15:17] <Hash> https://www.youtube.com/watch?v=QD6J9sEejdw DT new video, about a new package manager in Ubuntu
[15:17] <Hash> Nala
[15:17] <Hash> Nala is an apt frontend for Debian and Ubuntu and distros based on them. <- never heard of this. Where did it pop out from
[15:18] <Hash> https://gitlab.com/volian/nala interesting
[15:18] <Hash> Just mentioning it since I found it today, but not recommending it, because I dunno how it works or quality
[15:19] <Maik> kinda offtopic here and belongs more in #ubuntu-discuss or #ubuntu-offtopic :)
[15:21] <Hash> Ohok
[15:21] <Hash> Thanks
[15:22] <Maik> no problem :)
[15:22] <lexandrop> Maik: I've just noticed the link you've sent before (about the 22.04.1 release delay) - thanks. Surprise-surprise, something is wrong with snap [again] ;D
[15:23] <Maik> lexandrop: no problem, you're welcome. :)
[15:23] <jhutchins> As I understand it, there are three possible ways to get drivers for nvidia.
[15:24] <jhutchins> One is the nouveau driver, part of the regular Ubuntu driver suite.
[15:24] <jhutchins> Another is the proprietary Nvidia-supplied drivers, installed by an official Ubuntu wrapper.
[15:25] <jhutchins> The final one is the Nvidia drivers direct from Nvidia, using their installer.
[15:25] <jhutchins> Is this correct?
[15:26] <enigma9o7[m]> Yes.  There's also a graphics drivers ppa, which may have newer versions than supplied with ubuntu release.
[15:27] <enigma9o7[m]> And you should really avoid using the nvidia script (final one in your list); if it doesnt work as expected you'll have difficulty reverting to regular ubuntu drivers.  And if it does work as expected you'll have difficulty with kernel updates.
[15:28] <jhutchins> The Nvidia installer I used was very conveniently and thoroughly reversible,
[15:28] <jhutchins> It is absolutely essential that you read their documentation, and that you clean ALL artifacts from previous installation attempts, especially from other methods.
[15:29] <enigma9o7[m]> Or just dont use it.
[15:29] <jhutchins> Each version of the Nvidia "official" drivers has a list of what cards it supports.
[15:29] <jhutchins> enigma9o7[m]: Well, yes, and "buy new hardware" is a popular "solution" here.
[15:30] <enigma9o7[m]> dont buy nvidia if yer doing that
[15:30] <enigma9o7[m]> just avoid it completely
[15:31] <jhutchins> Great performance here.
[15:31] <enigma9o7[m]> make life easier
[15:32] <tomreyn> gordonjcp: if your nvidia graphics card is actually supported by the 515 driver series (please provide your graphics card model if you would like this to be checked) then please file a bug report about a regression in 515.48 over 515.43 by running: ubuntu-bug nvidia-driver-515
[15:37] <tomreyn> jhutchins: I guess this is why most of us prefer to recomend the ubuntu packaged ones - because the average user will not do (or be able to do) this: "It is absolutely essential that you read their documentation, and that you clean ALL artifacts from previous installation attempts, especially from other methods." You will also need to clean up leftover configurations from nvidia's installer when switching to ubuntu's, but once that is done,
[15:37] <tomreyn> reading documentation should not be needed.
[15:39] <ioria> and, btw, -48 is not the latest of 515,  -65   is gordonjcp
[15:41] <tomreyn> on any non EOL ubuntu release anyways :)
[15:42] <tomreyn> https://packages.ubuntu.com/search?exact=1&keywords=nvidia-driver-515
[16:01] <gordonjcp> tomreyn: will do
[16:01] <gordonjcp> tomreyn: I suspect it's more to do with things that use CUDA rather than the drivers itself
[16:04] <tomreyn> gordonjcp: if the regression affects a different package, please report against that. you can search for packages using the    apt search <package detail>   CLI or using https://packages.ubuntu.com
[16:31] <enigma9o7[m]> Hmmm intresting, why would the -65 of nvidia's 515 not be in kinetic?     (It's not important, I only noticed cuz it was linked and discussed here).
[16:31] <enigma9o7[m]> wouldn't it normally go thru there first?
[16:31] <Guest17> Hi. I'm hoping there's an easy answer for this. How can I get a recursive list of just folder names output to a text file with one directory per line? Like the output of [[[find . type d -print]]]] except without the ./dir/dir/dir/dir/ path before the actual directory name? I'm having a hard time finding an answer with google searches.
[16:32] <tomreyn> enigma9o7[m]: i'd guess on kernel or other incompatibilities
[16:33] <jhutchins> Guest17: Use find.  Specify type=directory.
[16:33] <enigma9o7[m]> hmmm maybe tree would be useful?
[16:34] <leftyfb> Guest17: find . -type d -exec basename {} \;
[16:34] <Guest17> find . -type d -exec basename {} \; seems it might be what I need. Will test. Thanks!
[16:34] <enigma9o7[m]> like `tree -d`
[16:35] <leftyfb> Guest17: future reference, #bash might be a good place to ask such questions
[16:35] <Guest17> Thanks
[16:36] <tomreyn> Guest17: find ~/Downloads/ -type d -printf '%f\n'
[16:37] <leftyfb> also: tree -d -i
[16:38] <Guest17> I'll experiment with all of these. My goal is to find duplicate folder names across two drives. Thank you. :)
[16:38] <leftyfb> Guest17: rsync
[16:42] <jwash> hi everyone, i'm trying to format a fat32 with sector size of 4096, but it keeps reverting to 512
[16:43] <leftyfb> jwash: why do you think you need a fat32 partition with sector size of 4096 ?
[16:43] <jwash> because my 3d printer requires it
[16:43] <jwash> .......
[16:43] <jwash> it will not read 512 sector size
[16:43] <jwash> .......
[16:43] <jwash> that's why I think i need it
[16:46] <leftyfb> jwash: what commands have tyou tried so far?
[16:46] <jwash> sudo mkfs.fat -I -S 4096 /dev/sdd1
[16:46] <jwash> sudo mkdosfs -S 4096 -I /dev/sdd1
[16:46] <jwash> neither worked
[16:46] <jwash> with -I or without it
[16:47] <jwash> fdisk -l says 512 sector size
[16:47] <jhutchins> jwash: Try xfat?
[16:48] <leftyfb> try mkfs.fat -S $((2097152/512)) /dev/sdd1
[16:48] <tomreyn> add -F 32
[16:49] <tomreyn> -I is not needed when writing to a partition, as you do
[16:49] <jwash> tomreyn, ty will try now
[16:51] <jwash> https://controlc.com/5329f8fa
[16:52] <Guest17> Thanks for all of the answers. Very helpful:P  Got what I needed
[16:52] <jwash> still gives sectors as 512
[16:52] <bittin> ISOs is delayed a week: https://www.phoronix.com/news/Ubuntu-22.04.1-LTS-Delayed
[16:52] <jwash> trying leftyfb's now
[16:52] <jwash> same 512
[16:53] <jwash> https://controlc.com/e91aaf3b
[16:56] <ogra> jwash, you are aware that partitions and filesystems are different things ?
[16:56] <ogra> jwash, as such ... fdisk isnt really helpful to inspect your just created filesystem ...
[16:56] <jwash> lol
[16:57] <ogra> try something like: sudo fsck.fat -v /dev/....
[16:57] <ogra> (sould return a line like: "4096 bytes per logical sector" somewhere in the output if you got it right)
[16:58] <ogra> *should
[16:58] <leftyfb> it did in his 2nd to last paste
[16:58] <leftyfb> so it's been working this whole time
[16:59] <ogra> ah, i didnt see the former paste when returning to the kbd 🙂
[16:59] <jwash> damn
[17:00] <ogra> i'm curious though, what kind of 3D printer is that ? i have not seen a printer firmware that has such restrictions ...
[17:01] <leftyfb> also, just use #octoprint ;)
[17:02] <jwash> dude
[17:02] <ogra> well, that still talks to the firmware one way or the other ... and if you want to be safe to not have your print interrupted by network issues, SD card is still the best possibility
[17:02] <jwash> seriously
[17:02] <jwash> e5+
[17:03] <jwash> updating the firmware to use octoprint
[17:03] <jwash> now i have to update the display
[17:03] <jwash> the display requires 4096
[17:03] <leftyfb> ogra: octoprint does not rely on the network for active prints
[17:04] <ogra> oh, so it is a HW limitation ?
[17:05] <ogra> (of your display)
[17:05] <ogra> .. that's odd
[17:06] <ogra> leftyfb, well, you shove the model over the network and print via serial when using octoprint ... SD is definitely more reliable ... especially for large/long prints yo dont want it to die in the middle
[17:07] <leftyfb> ogra: the printer plugs into the pi using a usb cord. Same as if you were printing from a computer. In 3 years and multiple power outages during prints, I have never had a failed print as a result of octoprint/network/usb
[17:08]  * ogra stopped using octoprint a while ago ... i have to watch the first few layers anyway so remote isnt super helpful ...
[17:08] <ogra> i *do* have a cam attached indeed to watch the print on the go, but for that i dont need octoprint
[17:09] <leftyfb> octoprint is much more than just watching via a camera
[17:09] <leftyfb> anyway, ot
[17:09] <ogra> i had a few occasions where i came back to the printer and it had happily and wildly vomted plastic all over the place 🙂
[17:09] <ogra> yeah ... ot
[17:36] <gordonjcp> tomreyn: it's multiple packages, but what I can't find *right now* is how to back out the NVidia drivers to a previous version without nuking the Ubuntu packages and manually installing
[17:37] <tomreyn> gordonjcp: have you tried the latest ubuntu package version, yet, though?
[17:37] <gordonjcp> tomreyn: yes, that is the one that shows up bizarre rendering bugs
[17:39] <tomreyn> so you're saying nvidia-driver-515 version 515.43 works for you, .48 and .65 do not?
[17:39] <gordonjcp> tomreyn: correct
[17:39] <tomreyn> and your graphics card is supported by 515 driver series?
[17:40] <gordonjcp> yup
[17:40] <gordonjcp> it's quite an old card, a GT970, but it's listed as supported
[17:40] <tomreyn> did you file a bug report, yet?
[17:40] <gordonjcp> tomreyn: no, because at present all I have is "it doesn't work with 515.48 or 515.65
[17:41] <gordonjcp> tomreyn: until I can determine that .43 works, and maybe find out exactly what is dropping its guts when run with later drivers, the bug report would be terribly vague
[17:41] <tomreyn> well, you have log files, too, and you candescribe the symptoms
[17:44] <tomreyn> gordonjcp: which ubuntu release are you running, which kernel version?
[17:45] <gordonjcp> 22.04, 5.15.0-43-generic
[17:45] <tomreyn> https://www.nvidia.com/Download/driverResults.aspx/191961/en-us/ does not list GT 970
[17:46] <tomreyn> do you mean GTX970?
[17:46] <gordonjcp> yes, sorry, GTX970
[17:46] <gordonjcp> there isn't a GT970 I don't think
[17:48] <tomreyn> so    lspci -knn | grep -A3 VGA    lists it, on the first line, with this ID in the [square brackets]? 13C2
[17:49] <gordonjcp> yes
[17:50] <gordonjcp> right, -510 has got rid of the bizarre artifacts, that'll do it for now
[17:50] <gordonjcp> I can look into it after I've got this video done
[17:51] <tomreyn> you mean driver series 510?
[17:52] <gordonjcp> yes
[17:53] <tomreyn> yes, a bug report would be good to have if driver series 510 works for you but series 515 does not, with this hardware, if ubuntu-drivers defaulted to / recommended series 515.
[17:54] <gordonjcp> tomreyn: I'll dig into it more later, I've got to get some editing finished and out before I get another phone call about it
[17:54] <gordonjcp> tomreyn: there is no-one quite so demanding as someone you've offered to do a favour for ;-)
[17:54] <tomreyn> sure, i don't mean to rush you
[17:55] <gordonjcp> tomreyn: thanks for all the help :-)
[17:55] <tomreyn> you're welcome
[18:38] <acegalla-> can anyone point me to where the package build scripts arefor ubuntu?
[18:38] <acegalla-> i'm trying to figure out how they built a software package
[18:39] <jhutchins> !build
[18:39] <jhutchins> I think that's more third-party/upstream oriented.
[18:40] <jhutchins> acegalla-: I _think_ the patch files and such are a part of the source packages.
[18:40] <murmel> acegalla-: apt source <package> gives you the source code
[18:40] <murmel> jhutchins: depends on the package sadly
[18:41] <murmel> linux-meta integrates all patches, which means you have one giant "blob"
[18:41] <jhutchins> murmel: Uploads should require that the source package can build the binary equivalent code,
[18:41] <murmel> jhutchins: yes, but not all src have patches :/ as pointed out above
[18:41] <murmel> which honestly I don't understand, especially with linux-meta :/
[18:41] <jhutchins> I do hear that the patches tend to be merged, so you have to pick them apart manually.
[18:42] <jhutchins> murmel: If the source package doesn't have all the patches in one form or another it should not be accepted for upload.
[18:43] <jhutchins> Then again, Ubuntu isn't a very source-friendly distro.
[18:43] <murmel> jhutchins: yeah. I realised that :/. oh well. as I mostly only backport stuff, I don't mind too much
[20:00] <lexandrop> By the way, talking about 22.04.1 delayed release - maybe someone should change the date here https://wiki.ubuntu.com/Releases ?
[20:24] <amaroq> Can someone tell me where notification sounds are located? I would like a Empty Trash sound event. In OSX when we empty the bin there is a saatisfying crunching of paper sound, maybe you know it. Is something like that possible in Ubuntu, also?
[20:25] <amaroq> I'm hoping its something simple like enabling the option in Notifications and not a big project like dowloading music clip and storing it somewhere and writing a long script, lol
[20:27] <murmel> amaroq: sounds are in /usr/share/sounds/ but no idea how you would add it for the trash
[20:29] <leftyfb> amaroq: it looks like there's no built-in mechanism to do so. You'll need to write your own
[20:29] <amaroq> murmel, I see, thank you. So, by default an Empty Trash sound event was never designed into Ubuntu out of the box? ...it feels odd emptying the bin and hearing nothing...almost feel I have to check that its done. A sound is a simple way to confirm the action
[20:30] <leftyfb> amaroq: you can write a script that always runs in the background that watches the directory and will make the sound if the number of files changes from non-zero to zero
[20:30] <amaroq> leftyfb, right...I'll put that project on the back burner...might be fun though at some point. I do miss not having that sound.
[20:31] <amaroq> To be honest I'm surprised that Gnome hasn't made that a default feature.
[20:31] <oerheks> there is /usr/share/sounds/Yaru/stereo/trash-empty.oga
[20:32] <oerheks> not ogg
[20:36] <enigma9o7[m]> I don't think I've ever emptied the trash before.
[20:37] <sarnold> well, if you had a satisfying crunching paper sound, maybe you would
[20:38] <amaroq> oerheks, I see it, thx. So, there is such a sound stored on the system. What for? So, we just have to figure out a script I guess
[20:38] <amaroq> enigma9o7[m], do you just go straight to 'delete'?
[20:39] <amaroq> I did enable the 'delete' function but I do like to throw things in the bin when clearing stuff out and having it there in case I need to get something back out...Anyway, its there so
[20:42] <oerheks> Bank of England: Inflation could hit FIFTEEN per cent by 2023 ... fifty would scare people
[20:42] <oerheks> oops wrong channel
[20:43] <enigma9o7[m]> I've just never bothered to empty the trash is all.  I just played around and find it odd there is a sound file, but no way to enable it in preferences.  Maybe a settings file somewhere....
[20:46] <oerheks> trash is cleared after old-files-age. standard is 30 days
[20:46] <oerheks> see dconf editor
[20:54] <amaroq> oerheks, I have a small 128 gb SSD. The system complained that it was nearly full after some downloads I did. So, deleting and empting the trash is something I have to do regularly. And, yes, inflation is really starting to show.
[20:54] <DynamiteDan> greetings. How do I force a btrfs filesystem to unallocate delete files? Thanks in advance
[20:56] <amaroq> enigma9o7[m], ditto...why would there be a sound file and no way to enable it in the appropriate and corresponding Settings /preferences? Right-click Bin and preference with sound option for e.g.
[20:56] <murmel> DynamiteDan: fstrim -v <path>
[21:00] <DynamiteDan> murmel, thanks
[21:02] <DynamiteDan> hmmm
[21:02] <DynamiteDan> operation not supported
[21:02] <murmel> DynamiteDan: doesn't the timer suffice?
[21:03]  * DynamiteDan scratches his head
[21:03] <murmel> DynamiteDan: luks?
[21:03] <DynamiteDan> nope
[21:03] <murmel> hdd?
[21:03] <DynamiteDan> yes
[21:03] <murmel> then discard is not an option
[21:03] <DynamiteDan> HDD
[21:04] <DynamiteDan> i know btrfs lags in unallocating delete files
[21:04] <murmel> yeah? that's by design
[21:04] <DynamiteDan> so when there is a pause in accessing it it might reclaim
[21:04] <archwise90> Ubuntu studio loading 15+ minutes live usb
[21:04] <murmel> yes
[21:05] <DynamiteDan> gotvha
[21:05] <DynamiteDan> will check tomorrow
[21:05] <pycurious> My machine keeps updating the drivers (without rebooting) - I can't find what is causing these updates. It broke my nvidia drivers again. Any ideas how to find this " automatic update " feature?
[21:06] <pycurious> I'm on Ubuntu 20.04.4 LTS- and my nvidia drivers were fine till yesterday. I've switched off security updates - or at least I thought I did.
[21:07] <cbreak> turn them back on again then...
[21:07] <cbreak> turning off security updates sounds dumb
[21:10] <cbreak> pycurious: after nvidia driver updates, reboot
[21:10] <tumulto> Hello everyone! I have a problem with my Ethernet and I'm really blocked because I don't understand what the hell is going on. This is the problem https://askubuntu.com/questions/1421744/ethernet-connection-drops-after-a-few-seconds
[21:11] <tumulto> somebody could help me?  Thank you
[21:12] <pycurious> cbreak: I'm trying to find out who updates the machine without me knowing? Especially the nvidia drivers? Any ideas?
[21:13] <oerheks> pycurious, nvidia drivers from our repos?
[21:14] <oerheks> show us the update output in a pastebin?
[21:14] <cbreak> pycurious: non-security updates
[21:15] <cbreak> check out the 'software-properties-qt' or software-properties-gtk
[21:17] <enigma9o7[m]> proably 'unattended-updates' pycurious
[21:18] <pycurious> oerheks: https://dpaste.org/MXgdk
[21:18] <enigma9o7[m]> i bet uninstalling that prevents stuff from updating itself sometimes
[21:18] <pycurious> enigma9o7[m]: I think I switched that off - how do i check?
[21:19] <enigma9o7[m]> its unattended-upgrades actaully
[21:19] <enigma9o7[m]> you could check if the package is installed (i.e. apt policy unattended-upgrades)
[21:20] <cbreak> pycurious: have you looked at software-properties-??
[21:22] <murmel> pycurious: the config if you disabled unattended-upgrades is in /etc/apt/apt.conf.d/20auto-updates
[21:24] <oerheks> i still have 3 packages held back; The following packages have been kept back:  python3-distupgrade ubuntu-release-upgrader-core ubuntu-release-upgrader-gtk
[21:25] <oerheks> and find no fix ...
[21:25] <cbreak> oerheks: have you tried to apt install ubuntu-release-upgrader-core?
[21:25] <murmel> sounds like phased updates
[21:25] <cbreak> probably some silly bug in apt
[21:25] <cbreak> I've seen randomly held-back packages in the past when someone screwed up some deps or similar
[21:26] <oerheks> oh that fixes it
[21:26] <oerheks> thank you
[21:26] <sarnold> re ubuntu release upgrader https://people.canonical.com/~ubuntu-archive/phased-updates.html
[21:30] <pycurious> enigma9o7[m]: https://dpaste.org/GZdsc - I think I switched it off in a configuration file. How do we find out if this is the one that is updating/installing nvidia drivers?
[21:31] <sarnold> pycurious: /var/log/apt/history.log* logs the commandline used for specific updates
[21:31] <pycurious> sarnold: Thank you!
[21:34] <pycurious> sarnold: sudo apt remove unattended-upgrades — how bad is this to do?
[21:35] <sarnold> pycurious: that'll probably remove an ubuntu-server or ubuntu-desktop metapackage or something like that, which will probably complicate the do-release-upgrade process
[21:35] <sarnold> pycurious: you could use equivs to build a fake package to satisfy apt
[21:36] <sarnold> pycurious: or configure unattended-upgrades to do nothing?
[21:52] <pycurious> sarnold: APT::Periodic::Unattended-Upgrade "1"; -> "0" — trying that
[21:54] <archwise90> Got /boot/ error when trying to install from a USB Pen Drive.
[21:54] <sarnold> now the downside is you've got to wait over night to find out if it worked :)
[21:54] <tomreyn> i think there's also a systemd service, which could be disabled
[21:55] <tomreyn> which, from memory, does something if it finds unattended-upgrades to be installed, and someth ing else otherwise
[21:56] <murmel> tomreyn: pretty sure when dpkg-reconfigure unattended-upgrades it disables the timers
[21:57] <tomreyn> ok, i really don't rmeember how it's setup nowadays, i feel it changed a few times during the past years
[21:59] <tomreyn> apt-daily-upgrade.(service|timer)
[23:08] <cryptobugger> guys is ubuntu still hard to predict about that apt running in script ?
[23:09] <cryptobugger> i wanted to code ubuntu hardening server
[23:10] <cryptobugger> https://github.com/dodocrypto/ubuntu.hardening.script/blob/main/0day.ubuntu.py is something like this
[23:10] <cryptobugger> but i gave up halfway
[23:10] <cryptobugger> is so stupid  it pop up some configuration window
[23:11] <cryptobugger> :( it keep on give warning issue about using apt or apt-get at script
[23:11] <cryptobugger> anyone know the solution for this for ubuntu server ?
[23:11] <oerheks> basicly apt-get for scripting *hint*
[23:11] <cryptobugger> didn't they just the same
[23:12] <cryptobugger> just some soft link ?
[23:12] <oerheks> nope, there are differences, and leftyfb showed me apt-get is to be used.
[23:12] <cryptobugger> sorry i don't have ubuntu box now
[23:13] <cryptobugger> i just read like 7 - 10 vulnerability in a day
[23:13] <cryptobugger> so nah i decided to install windows 10 instead
[23:13] <cryptobugger> oh nice oerheks
[23:13] <cryptobugger> will try it than
[23:13] <cryptobugger> what is the command to just setup default setting for ubuntu ?
[23:14] <cryptobugger> so it won't setup auto pop up configuration like mail server and those stuff
[23:15] <ravage> you select the minimized option during the ubuntu server installation
[23:30] <sarnold> cryptobugger: DEBIAN_FRONTEND=noninteractive
[23:31] <ravage> and that's usually very reliable
[23:31] <ravage> at least for any official packages
[23:38] <cryptobugger> sarnold, thank you
[23:40] <tomreyn> cryptobugger: you might even want to go for https://unix.stackexchange.com/a/404757
[23:40] <tomreyn> or parts of it
[23:42] <tomreyn> and there are other hardening tools / scripts / approaches. it can very much depends on your workloads, how the systems are used, and what the potential threats are.