avih | new feature: apt will now tell you about packages it's going to install for other users if it chooses to not install them for you | 00:00 |
---|---|---|
=== Etua1 is now known as Etua | ||
avih | for reference, i think from one of the ask ubuntu subjects, this counts the phased updates packages and their progress, so when it sums up to the number of held back packages (and it does for me currently), then it means everything is fine: | 00:03 |
avih | apt show -a $(apt list --upgradable 2>&1 | grep / | cut -d/ -f1) 2>&1 | grep Phased | sort -n | uniq -c | 00:03 |
avih | however, if the output of this commands sums up to less than the number of held back packages, then it means not all of them are phased updates, and the user needs to look closer for conflicts with some packages | 00:04 |
=== LabMonkey is now known as Mechanismus | ||
avih | (i'm sure the command can be improved in various ways, but above is a plain copy paste from the web) | 00:08 |
avih | sarnold: is it an ubuntu thing? or a debian thing? | 00:31 |
avih | (the whole phased updates thing) | 00:31 |
avih | it's part of apt, but th first (all?) results seem related to ubuntu... | 00:32 |
sarnold | avih: it's currently ubuntu, but I understand there are plans to bring it to debian eventually' | 00:33 |
avih | hmm.. in 2013 it was an ubuntu thing https://lwn.net/Articles/563966/ | 00:33 |
sarnold | it's been phased in the desktop for ages | 00:33 |
sarnold | adding it to the command line tools is very new | 00:34 |
avih | right, that explains things. i update only from CLI | 00:34 |
sarnold | yeah; I think the graphical thing follows your suggestion and doesn't complain about things that aren't phased for you yet | 00:36 |
avih | i'm pretty sure debian won't take it as is. it's quite harmful for the reason i mentioned earlier - it hides held back packages the user should really care about inside a much bigger list of held back packages which the user absolutely doesn't care about, with no way to tell which is which | 00:36 |
avih | (at least no way from the interaction of dist-upgrade) | 00:37 |
avih | anyway, thanks a lot for your time and help. i would have (and did) make much slower progress on this have i not came here to irc. cheers. | 00:41 |
sarnold | see ya avih :) thanks for the good suggestion | 00:42 |
avih | :) | 00:43 |
avih | sarnold: when you say "adding it to the command line tools is very new", what happened before? were they held back quietly? or updated and excluded from phased updates? | 00:58 |
sarnold | avih: they were upgraded without considering the phased percentage | 00:58 |
sarnold | avih: which can lead to exciting results on systems that don't yet have the phasing https://www.theregister.com/2022/08/30/ubuntu_systemd_dns_update/ | 00:59 |
avih | right. so i do think considering phased updates is a good thing for CLI tools too, though the implementation is definitely lacking (it not only tells me about it, it actually first asks if i want to install them, and only if i answer yes - it then tells me they're held back) | 01:00 |
avih | (basically, it knowingly wastes my time) | 01:02 |
arraybolt3 | sarnold: Well I'm sure MS has made boffos similar to the Ubuntu 18.04 mess. | 01:09 |
arraybolt3 | Actually, yep, they made a pretty bad fumble today. But I'm off-topic. | 01:11 |
avih | how is that link related to phased updates? i don't think i see it mentioned at the article | 01:12 |
avih | (if i wasn't told, i'd have attributed it to yet another systemd issue) | 01:13 |
sarnold | avih: that's a nice demonstration of what happens when millions of machines all install an update in the span of a few hours... | 01:14 |
sarnold | phased updates odn't apply to security updates, so even if 18.04 had it, it wouldn't have helped | 01:14 |
sarnold | but I think a lot of folks would have been happy if only 10% of the machines had fallen over :) | 01:15 |
avih | that has to be true... :) | 01:16 |
avih | is being on LTS have some relation to where at the phase lifetime it's getting actually updated? | 01:18 |
avih | (i'd hope that LTS users would get it last) | 01:19 |
avih | (obviously without the noise of messages that it's held back) | 01:20 |
sarnold | avih: I think that plays no role | 01:20 |
avih | hmm | 01:20 |
jhutchins | Apparently you don't have five-figure downtime penalties? | 01:23 |
avih | best comment on that systemd snafu: "Can anyone sane explain to me why DNS should ever have anything to do with an init?" | 01:27 |
avih | and to me that's the real issue, not phased updates, but that's really off topic. | 01:29 |
eelstrebor | arraybolt3, thanks but i was looking for an automated way to determine the number of cpu cores and assign a zram device for each one. i did find a script that's suppose to do that but i was unable to make it work | 01:34 |
sarnold | pastebin the script and your errors and someone might have an idea | 01:34 |
=== esv_ is now known as esv | ||
eelstrebor | sarnold, here is the pastebin concerning my problem with zram: https://pastebin.com/MgUcN6cU | 02:00 |
sarnold | eelstrebor: note lines 6 and 8 both refer to zram_num_devices -- line 8 probably should be num_devices; line 19 totalmem should be $totalmem. | 02:04 |
oskar_ | hello , i have a question. notebook saying thunderbolt 3 subsystem not available on ubuntu studio | 02:28 |
eelstrebor | sarnold, i'll make the changes and test tomorrow. working on my car today took a lot out of me | 02:29 |
oskar_ | i got a solution. boltctl in the cli brought the subsystem running i try now reboot | 02:31 |
oskar_ | reboot dont helped. how can i stART BOLTCTL AUTOMATICALLY? should i put it into the bash.rc? | 02:34 |
oskar_ | is there anyone that can give me advice how to autologin on plasma ubuntu studio | 02:44 |
oskar_ | i forgot it from last time i searched for it | 02:44 |
Eickmeyer | oskar_: Open system settings. Search for autologin. | 02:44 |
Eickmeyer | oskar_: Apologies. Search for Login. Click on Login Screen. Click on Behavior. | 02:47 |
oskar_ | thanks... the message that says kde wallet blank password i can ignore if the pass were blank before? | 02:47 |
Eickmeyer | Yes. | 02:48 |
oskar_ | ok | 02:48 |
Appolinaire | 20.04 plain dell. update gives me.. E: Conflicting values set for option Signed-By regarding source https://updates.signal.org/desktop/apt/ xenial: /usr/share/keyrings/signal-desktop-keyring.gpg != E : The list of sources could not be read | 02:49 |
Bashing-om | !xenial | Appolinaire | 02:50 |
ubottu | Appolinaire: Ubuntu 16.04 LTS (Xenial Xerus) was the 24th release of Ubuntu. !End-of-life was April 30th, 2021. Paid support (ESM) is available. See also !esm, !eol, !eolupgrade | 02:50 |
leftyfb | Appolinaire: remove the 3rd party repo for signal | 02:53 |
Appolinaire | Bashing-om, I an not using that distribution. | 03:19 |
Appolinaire | leftyfb, I will work on figuring out how to do that. Thank you for the directions. I remember trying to instll it without snap and i didnt get it to work so I gave up and did it with snap. I have a think that you are right. | 03:21 |
fengshaun | why does launching firefox need nodejs installed?? | 03:31 |
fengshaun | I don't know what happened, but running firefox and thunderbird failed with 'node no such file or directory', installing nodejs fixed it | 03:32 |
leftyfb | fengshaun: what version of ubuntu? | 03:34 |
fengshaun | 2204 | 03:34 |
leftyfb | fengshaun: in ubuntu 22.04 Firefox is installed via snap | 03:34 |
fengshaun | I'm using the official ppa | 03:34 |
leftyfb | fengshaun: then you'll have to seek support from whatever maintains that ppa | 03:34 |
fengshaun | ok | 03:35 |
leftyfb | fengshaun: If you install firefox from snap, you won't have that issue | 03:35 |
fengshaun | snap breaks 30 other things, so not really a solution | 03:35 |
leftyfb | false | 03:36 |
ice9 | why many times apt shows some packages are kept from upgrade? | 03:41 |
arraybolt3 | ice9: Phased updates - Ubuntu isn't installing certain updates yet so that in the event there's a problem with the update, there's a better chance of you not installing the bad update. | 03:46 |
ice9 | arraybolt3, so if there is a problem with that specific update, why it was added to the repo, or way it wasn't removed from the repo? | 03:47 |
arraybolt3 | ice9: So, despite testing, not all updates actually end up working. We try to avoid having that happen, but sometimes it does. When this happens, we don't know about it until we start getting error reports. Phasing the updates makes it so that only 10% of users get the update at a time. If error reports suddenly start coming from their systems, we know that something got botched somewhere and can tell it to stop. | 03:48 |
arraybolt3 | (It's essentially the same reason things always go wrong - there's a risk, we do everything we can to avoid it, someone fumbles something, it slips through the cracks, then things blow up. Happens all the time to lots of companies. We aren't immune.) | 03:49 |
ice9 | arraybolt3, thanks for the clarification | 03:49 |
arraybolt3 | leftyfb: Well, not necessarily false. 30 other things may be an exaggeration, but there are some things (like smartcards, GNOME Shell extension installation, and certain Firefox add-ons) that do break because of Snap's containerization. | 03:51 |
arraybolt3 | ice9: (Referencing a conversation that took place before you joined, just in case things look confusing) | 03:52 |
JerOfPanic | morning all * | 03:56 |
arraybolt3 | Is it true that phased updates in apt have applied to every system since Ubuntu 21.04? I don't remember ever running into them in Ubuntu 21.10 back when I used it. | 04:12 |
arraybolt3 | (Maybe I was just really lucky? Or did Ubuntu 21.10 stop using them and then Ubuntu 22.04 started using them again?) | 04:12 |
alkisg | Is there an `apt full-upgrade --include-phased` option? | 04:20 |
alkisg | Stupid question, first google result, Update-Manager::Always-Include-Phased-Updates | 04:21 |
murmel | avih: sadly, with the implementation they have now, there won't be any option to disabling seeing the phased updates :/. already reported a bug, was declared as won't fix | 05:12 |
murmel | alkisg: yeah, otherwise you need to specifically call those packages with update or install :/. | 05:13 |
=== x1 is now known as wellwm | ||
=== archpc1 is now known as archpc | ||
=== denningsrogue7 is now known as denningsrogue | ||
podeni | I have a script that does an automatic upgrade using apt, calling the following apt-commands: update upgrade dist-upgrade autoremove autoclean clean | 07:03 |
podeni | Are any of these redundant or contradictory? | 07:03 |
murmel | podeni: as long as you are not upgrading to a new distro release upgrade and dist-upgrade are basically the same (it's very unlikely that dist-upgrade has to install new stuff) | 07:08 |
guiverc | I'd just use full-upgrade (or dist-upgrade) & remove the upgrade... there are cases where upgrade cannot install all upgrades which is why full-upgrade exists | 07:09 |
guiverc | (refer `man apt` for reason(s)) | 07:09 |
murmel | guiverc: that's literally the reason why I would rather use upgrade than full-upgrade, as there is a reason why upgrade can't upgrade all packages (if that should ever happen) have never seen it during a normal release | 07:10 |
brvadiraj46 | I am not able to setup nordvpn on ubuntu | 07:30 |
brvadiraj46 | can anyone help | 07:30 |
murmel | brvadiraj46: what did you try for now? | 07:30 |
brvadiraj46 | I tried installing nordvpn client | 07:30 |
murmel | brvadiraj46: which release? | 07:31 |
brvadiraj46 | also tried via network manager and downloading config files | 07:31 |
brvadiraj46 | I tried the latest | 07:31 |
brvadiraj46 | i think its 13.* | 07:31 |
brvadiraj46 | I searched in internet... many have reported the problem | 07:32 |
brvadiraj46 | iand they say downgrading to 3.7.4 works | 07:32 |
murmel | downgrading what? | 07:32 |
brvadiraj46 | I am not finding 3.7.4 | 07:32 |
brvadiraj46 | downgrading nordvpn client | 07:32 |
brvadiraj46 | any idea how I can get 3.7.4 version | 07:33 |
murmel | brvadiraj46: it's not available anymore | 07:34 |
brvadiraj46 | even the setup via openvpn is not working | 07:34 |
brvadiraj46 | by importing nord config files | 07:34 |
murmel | brvadiraj46: this is the repo of the client https://repo.nordvpn.com/deb/nordvpn/debian/pool/main/ and there are only those versions | 07:35 |
murmel | brvadiraj46: what happens if you try to import the configs? | 07:35 |
brvadiraj46 | vpn connect gives network connection error | 07:35 |
murmel | brvadiraj46: can you post the log to some pastebin? | 07:36 |
brvadiraj46 | where can i find the log | 07:36 |
murmel | brvadiraj46: should be in /var/log/openvpn/ | 07:37 |
brvadiraj46 | there are no logs there | 07:38 |
brvadiraj46 | but i found some in syslog | 07:39 |
murmel | brvadiraj46: yeah that helps also | 07:39 |
brvadiraj46 | https://pastebin.com/rbUJ2fi0 | 07:40 |
brvadiraj46 | I see some vpn service disappered error | 07:41 |
brvadiraj46 | murmel: any idea? | 07:45 |
murmel | brvadiraj46: https://bugs.launchpad.net/ubuntu/+source/network-manager-openvpn/+bug/1847144 hm maybe this bug | 07:46 |
ubottu | Launchpad bug 1847144 in network-manager-openvpn (Ubuntu) "network-manager-openvpn config import ignores tls-crypt section" [Undecided, Confirmed] | 07:46 |
brvadiraj46 | no solution? | 07:48 |
murmel | brvadiraj46: try it with cli maybe as I read a few times that cli works fine | 07:48 |
brvadiraj46 | what is the command to use? | 07:49 |
murmel | eh idk, but I assume something openvpn or you can try to use nmtui as a text user interface if networkmanager accepts the configs | 07:49 |
=== SteelRose_ is now known as SteelRose | ||
VaniaPy | Hi, i'm using 22.04 and i can't restart network-manager when i'm running "service network-manager restart" . Has this changeD? | 08:13 |
murmel | VaniaPy: autocomplete tells me it should be NetworkManager | 08:15 |
VaniaPy | murmel yes,true | 08:16 |
VaniaPy | thanks | 08:16 |
VaniaPy | nmgnome doens't allow me to connect to a public network | 08:17 |
murmel | VaniaPy: what happens? | 08:18 |
VaniaPy | gives Error resolving "nmcheck.gnome.org": Temporary failure in name resolution | 08:18 |
VaniaPy | i have added [Connectiviy] to network manager but still nothing | 08:19 |
murmel | huh, sounds like something is going wrong with dns | 08:19 |
VaniaPy | hm probably | 08:20 |
murmel | VaniaPy: can you try to connect to a different public wifi? funnily I had the same error today, but it was something with the wifi which was fixed within 5 mins | 08:20 |
VaniaPy | i don't have any other public wifi next to me unfortuantely :( | 08:21 |
VaniaPy | murmel what was the fix? | 08:35 |
murmel | VaniaPy: as I said, as it was the wifi, they had to restart the wifi. | 08:37 |
murmel | but it doesn't have to be the wifi in itself | 08:37 |
VaniaPy | oh ok,yeah it doesn't seem very helpful | 08:37 |
murmel | VaniaPy: as I wasn't the only one in the store with that issue, it was quite easy for them to restart the device :) | 08:38 |
Rexodus | ./window move 4 | 10:13 |
Rexodus | Good morning! This morning my VPS with 2GB RAM died on ClamAV. Not enough memory it said. So I checked my swap-size. To my great surprise, I see that no swap is running there. Now I need to shrink a running / so I can cram a swap partition in there. But, is there actually a safe way to shrink / while it's running? As far as that goes in the first place... Else: do I have any other options to create a swap | 10:34 |
Rexodus | in a running VPS? | 10:35 |
rbasak | Rexodus: you could create a swap file and use that. | 10:35 |
Rexodus | Where do I get the space for that? Everything is allocated. | 10:36 |
rbasak | It just requires free space on your filesystem. | 10:36 |
rbasak | Literally it's a file. | 10:36 |
Rexodus | Hmmm.... | 10:36 |
Rexodus | That would be easy... | 10:36 |
Rexodus | rbasak: Thanks dude! One statue for you! :) Just got myself 4GB RAM for free :P | 10:46 |
rbasak | :) | 10:49 |
=== rasta_ is now known as rasta | ||
iomari891 | Greetings, I want o update my kernel but not for it to take effect until I reboot. Is that how the normal apt update works for kernels? | 10:57 |
alkisg | Υεσ | 11:00 |
alkisg | Yes | 11:00 |
=== leo- is now known as Mikro | ||
=== esv_ is now known as esv | ||
brvadiraj46 | I am trying to connect nord vpn over open vpn and I am getting this error | 12:18 |
brvadiraj46 | https://pastebin.com/P8pf0ciC | 12:18 |
brvadiraj46 | my credentials are correct and I have active subscription | 12:19 |
brvadiraj46 | it can't be a firewall issue right? | 12:22 |
bancroft | I'm really confused with memory, starting with storage memory. If i'm reading 64 (kB or KiB?) of memory from a file is there 1000 bytes per kilo or 1024 per kilo? | 12:23 |
BluesKaj | Hi all | 12:33 |
webchat19 | Hi, I've got a bug to report and I'll like help with that :) | 12:33 |
lotuspsychje | webchat19: we advice before filing a bug, to state your issue here first | 12:35 |
webchat19 | The bug is that the Israel software and updates mirror server is missing some (maybe all?) the i386 packages. | 12:41 |
webchat19 | For example I had a dependency on this package: libc6_2.31-0ubuntu9.9_i386.deb | 12:41 |
webchat19 | Which on this mirror server is missing: | 12:41 |
webchat19 | http://il.archive.ubuntu.com/ubuntu/pool/main/g/glibc/ | 12:41 |
webchat19 | But as you can see, the main server (and also the US one) has all the i386 packages (and specifically the one I needed): | 12:41 |
webchat19 | http://archive.ubuntu.com/ubuntu/pool/main/g/glibc/ | 12:41 |
webchat19 | When I tried to send a message I got the error: "Cannot send to nick/channel" | 12:43 |
webchat19 | Can you see my message describing the bug? | 12:43 |
lotuspsychje | webchat19 its because your pasted too long text without a paste link | 12:43 |
lotuspsychje | webchat19: so you are missing some packages on the repos is that it, for wich ubuntu release? | 12:44 |
webchat19 | Ok, should I split it into messages? Or should I use a "paste-site"? Which one should I use? | 12:44 |
lotuspsychje | !paste | 12:44 |
ubottu | 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. | 12:44 |
webchat19 | https://dpaste.com/A2QD93AYA | 12:46 |
webchat19 | TL;DR: There are missing i386 packages in the Israel software and updates mirror server | 12:47 |
ogra | webchat19, see the bottom of https://wiki.ubuntu.com/Mirrors it has an address where you can report issues with mirrors | 12:49 |
webchat19 | Ok, thanks! | 12:49 |
lotuspsychje | webchat19: or #ubuntu-mirrors | 12:50 |
ogra | right (i think that is also mentioned on the page somewhere) | 12:50 |
lotuspsychje | they reccomend to use that email indeed in the channel, but in reality they also help in real time (when they have time) | 12:51 |
webchat19 | Ok, I'll try the chat first. Thank you guys! | 12:52 |
=== pong is now known as beaver | ||
=== Fish_ is now known as stortgordyn | ||
avih | if apt list --upgradable shows packages foo, bar and baz, and i know that bar will be held back because it's phased, can i do something like apt upgrade foo baz so that it won't tell me that bar is held back? i know i can do apt install foo baz, but this will mark it installed manually - which i don't want | 13:33 |
avih | and apt upgrade seems to ignore specific package names which follow... | 13:33 |
avih | (i'm writing a script which does apt upgrade just without the insanity that is phased packages, and everything works, except that i don't know what is the command to upgrade specific packages...) | 13:35 |
alkisg | avih, apt full-upgrade -o APT::Get::Always-Include-Phased-Updates=true | 13:38 |
alkisg | Also, if you run apt install foo baz, and they can be upgraded, they won't be marked as manually installed, they'll just be upgraded. They only get marked as manually installed if they can't be upgraded | 13:38 |
avih | arraybolt3: i don't want them i'm fine with being served later, but i don't want to see them either, just like the desktop package manager hides them | 13:39 |
oerheks | why worry about the phazed updates? | 13:39 |
oerheks | they will come eventually. | 13:40 |
avih | wait what? i don't think you can be right. if i have package foo installed implicitly, then apt dist-upgrade can remove it if it's not longer needed, but if it's installed explicitly with apt install foo, then i'm pretty sure it won't get removed even if nothing depends on it... | 13:40 |
oerheks | does 'apt dist-upgrade ' remove stuff you manually installed? that is new to me | 13:41 |
avih | oerheks: it wastes my time, and within the list of perpetual held back phased updates i don't actually know if there are packages being held back due to actual conflicts | 13:41 |
avih | oerheks: i wrote " i'm pretty sure it won't get removed", so it means that no, i don't think will remove it | 13:43 |
avih | if i installed it manually, it it will remove if it wasn't installed manually, contrary to what alkisg jus t said | 13:44 |
alkisg | avih: currently in my 22.04 box, python3-mako can be upgraded. It's auto: `apt-mark showauto python3-mako` => python3-mako | 13:44 |
alkisg | I run `apt upgrade python3-mako`. Then again `apt-mark shoauto python3-mako` => python3-mako | 13:44 |
alkisg | It remains automatically installed, it's not marked as manually installed | 13:44 |
alkisg | *I run `apt install python3-mako`, sorry | 13:45 |
avih | sec, i need to parse what you just wrote | 13:45 |
avih | alkisg: my first question is: does this "apt upgrade python3-mako" does anything different than "apt upgrade" ? because as far as i can tell apt upgrade ignores any following package names | 13:46 |
avih | you can see it at the manpage, and also experiment with it manually | 13:47 |
alkisg | avih: see my correction, I meant apt install there | 13:47 |
alkisg | I already wrote "sorry" one sentence above, see above | 13:47 |
avih | right. sec. | 13:47 |
avih | no worries, i missed what it referred to. sec. | 13:47 |
avih | alkisg: is this typed correctly? what's "shoauto" ? "apt-mark shoauto python3-mako" | 13:48 |
avih | but also, even if this indeed marks it as non manual install, what is your suggestion for a command equivalent to "apt upgrade foo baz" ? it's definitely not "apt install foo baz" because this will make both unconditionally marked as installed manually, and also definitely not "apt install foo baz && apt-mark shoauto foo baz" because this will unconditionally mark both as NOT manually installed, while "apt upgrade foo bar" (if there was such command) will | 13:53 |
avih | preserve the state of "manually installed" for each package individually | 13:53 |
avih | (obviously i could check the state of manually installed per package, but before i get there i'd like to know if there's a command which does that for me - like apt upgrade, but which takes a specific list of packages to upgrade) | 13:54 |
misu | helo | 14:31 |
misu | hello | 14:31 |
misu | hello | 14:43 |
samy1028c | misu: hello | 14:44 |
Habbie | ,v pdns-recursor | 15:01 |
misu | hello | 15:14 |
=== blackest_mamba_ is now known as blackest_mamba | ||
=== beaver is now known as Guest01 | ||
arraybolt3 | Is it true that phased updates in apt have applied to every system since Ubuntu 21.04? I don't remember ever running into them in Ubuntu 21.10 back when I used it. (Maybe I was just really lucky? Or did Ubuntu 21.10 stop using them and then Ubuntu 22.04 started using them again?) | 15:34 |
rbasak | arraybolt3: they've been used for quite a long time, but only by update-manager AIUI. Support in apt directly arrived very recently. | 15:42 |
arraybolt3 | rbasak: Right, phased updates **in apt**. | 15:43 |
arraybolt3 | I know that it was first implemented in apt in Ubuntu 21.04, but for the purpose of accuracy I want to know if it applied to 21.10 also. | 15:43 |
rbasak | Since 21.04: https://discourse.ubuntu.com/t/phased-updates-in-apt-in-21-04/20345 | 15:43 |
rbasak | So yes, in 21.10 also. | 15:43 |
arraybolt3 | OK. I just never noticed it in 21.10 and want to make sure I have my facts straight in the doc I'm writing. Thanks! | 15:43 |
rbasak | I suppose it could have been disabled in the subsequent release temporarily, but I'm not aware of that happening. | 15:44 |
avih | arraybolt3: sarnold: fwiw, i'm trying to write a tool (for myself, but i'll share it) which tries to work around the biggest issue IMO of phased updates, and that's that real conflicts which need my attention are hidden inside a much bigger and ever changing list of "kept back" packages due to phasing. my approach is to identify the kept back packages among the upgradable ones (this already works), and then issue something like apt upgrade <list of non phased | 16:02 |
avih | pkgs>. however, apt-upgrade doesn't take specific package names, and apt-install will mark them unconditionally as installed (not good). so my idea was to check first which are auto, then apt install <all the lest>, then apt-mark auto <list of auto installed>. thoughts? | 16:02 |
avih | (or similarly apt-get install --mark-auto <list of upgradable auto packages>) | 16:07 |
avih | (ie. two install commands, one for manually installed, one for auto) | 16:08 |
plujon | Is C-c C-. a new Ubuntu / Gnome thing? | 16:09 |
plujon | When I hit that sequence in a terminal, an 'e' appears and I can continue to type letters, but they are not sent to the terminal until I hit enter. | 16:10 |
plujon | Oh, maybe it is simply C-. | 16:10 |
plujon | Ctrl-. | 16:10 |
plujon | https://askubuntu.com/questions/1404448/key-combination-ctrlperiod-not-working-ubuntu-22-04 | 16:12 |
plujon | What is emoji annotation..? | 16:16 |
mybalzitch | its meta . on my keyboard | 16:17 |
mybalzitch | it just pops up the emoji selector | 16:17 |
plujon | I see only an "e" on my desktop. | 16:18 |
plujon | What do I type after the e to select an emoji? | 16:18 |
arraybolt3 | smile | 16:21 |
arraybolt3 | Oy, fail... | 16:21 |
* arraybolt3 was testing the Ctrl-. thing | 16:21 | |
XATRIX | Hi, can you advice ? I've just upgraded from 20.04 -> 22.04. And now, my video drivers doesn't seems to be working. https://pastebin.com/yhk0DfGB | 16:31 |
XATRIX | All needed drivers, seems to be installed. What's wrong with it ? | 16:31 |
XATRIX | I've removed xserver-xorg-video-amdgpu, becase it's not for my video card. | 16:32 |
QNX | Hi | 16:37 |
QNX | I need nvidia 340 binary driver for GTX 275 nvidia card from 2010, for 22.04 Ubuntu. Nouveau driver is very slow and bad peformance generally. | 16:37 |
QNX | How can do? | 16:37 |
y0sh_ | plugins.var.perl.colorize_lines.highlight on | 16:41 |
=== blackest_mamba_ is now known as blackest_mamba | ||
XATRIX | Hi, can you advice what's messed with my video driver ? I'm using Ubuntu 22.04 (Xface WM) | 16:55 |
XATRIX | https://pastebin.com/yhk0DfGB | 16:56 |
=== blackest_mamba_ is now known as blackest_mamba | ||
bynarie | ls | 17:09 |
=== blackest_mamba_ is now known as blackest_mamba | ||
sarnold | avih: honestly, you might rather just disable the updates phasing thing entirely | 17:28 |
sarnold | avih: imho *apt* needs to handle this case better, and trying to provide extra tooling around apt is just layering more stuff upon more stuff .. | 17:28 |
pikapika | anyone uses plasma here | 17:28 |
avih | sarnold: 100% true about both, but for now i'm stuck with this, and i don't think i'm going to start hacking on apt... | 17:31 |
avih | (well, that it should be in apt, and workarounds are not good. not sure about the completely disable thing) | 17:32 |
arraybolt3 | pikapika: I do! | 17:34 |
arraybolt3 | (Currently on GNOME at the moment, but I also have multiple VMs running KDE.) | 17:34 |
pikapika | arraybolt3, yes | 17:34 |
pikapika | can you do a small experiment for me? | 17:34 |
arraybolt3 | pikapika: Sure. | 17:34 |
pikapika | open a video player with some video | 17:35 |
pikapika | check the window preview in the panel | 17:35 |
arraybolt3 | OK, one moment, VM booting | 17:35 |
pikapika | does it show the animation inside the preview? | 17:35 |
pikapika | then next step | 17:35 |
pikapika | keep the video playing and minimize the window | 17:35 |
pikapika | now see whats in the panel preview | 17:35 |
arraybolt3 | OK, I'll try it and tell you what happens. | 17:35 |
pikapika | thanks | 17:35 |
sarnold | avih: I suggest editing your apt config to set APT::Get::Always-Include-Phased-Updates to true -- that'll get rid of phasing for apt and you can leave the problem for later ;) | 17:36 |
pikapika | is apt written in C++ | 17:36 |
pikapika | the config feels very C++y. I looked up and it really is C++. Somehow I always assumed apt was C. | 17:37 |
sarnold | pikapika: mostly, yeah https://termbin.com/1428 | 17:37 |
sarnold | avih: there's more notes on the configuration on https://discourse.ubuntu.com/t/phased-updates-in-apt-in-21-04/20345 | 17:38 |
pikapika | wtf, whats the difference between C headers and C++ headers. Is it counting .h files as C, and only .H or .hpp as C++? | 17:38 |
pikapika | interesting | 17:38 |
pikapika | are most of the .h files actually C++ in this case | 17:38 |
arraybolt3 | pikapika: Using Firefox as a video player, the animation isn't working in the preview whether the window is open or minimized. | 17:38 |
pikapika | Interesting | 17:38 |
pikapika | arraybolt3, wayland? | 17:38 |
avih | sarnold: thx for the links. | 17:38 |
arraybolt3 | pikapika: Nope, X. Kubuntu 22.04. | 17:39 |
arraybolt3 | (Trying VLC now...) | 17:39 |
pikapika | strange | 17:39 |
arraybolt3 | pikapika: Bah, VLC isn't opening my YouTube link, lemme try smplayer... | 17:40 |
pikapika | i download a video first before opening in vlc generally | 17:40 |
arraybolt3 | pikapika: Wow SMPLayer isn't even showing me a screenshot. | 17:41 |
arraybolt3 | Where's a video I can legally download to test VLC? | 17:41 |
leftyfb | arraybolt3: https://peach.blender.org/ | 17:42 |
leftyfb | arraybolt3: https://www.pexels.com/search/videos/open%20source/ | 17:42 |
arraybolt3 | leftyfb: :facepalm: Right, forgot about Pexels. | 17:42 |
leftyfb | arraybolt3: type "open source videos" into google | 17:42 |
arraybolt3 | pikapika: Maybe it's because I'm running Kubuntu in a virtual machine, but I'm not getting a video animation in the popup no matter what software I use - Firefox, SMPlayer, VLC, all don't show me any animation. | 17:44 |
arraybolt3 | I seem to remember seeing an animation when I was running Kubuntu 20.04 on physical hardware, though. | 17:45 |
pikapika | Yeah its probably due to vm? | 17:45 |
pikapika | Are you running the vm fullscreen and have you assigned graphics memory or whatever its called to it? | 17:45 |
pikapika | The first thing I do for a graphical vm is increase the graphics ram etc | 17:46 |
arraybolt3 | pikapika: Not running fullscreen, I'm using GNOME Boxes (technically QEMU) for the VM. | 17:46 |
pikapika | I think whatever you are facing is a vm related thing probably | 17:47 |
pikapika | Someone else had been performing this test on a native kde, but I found his results a bit surprising and depressing | 17:47 |
pikapika | I hope I can confirm those from another source | 17:47 |
pikapika | essentially, preview shows animation when window is open, but only shows the icon (not even the last captured image) when its minimized | 17:48 |
=== blackest_mamba_ is now known as blackest_mamba | ||
pikapika | My main aim was to know what kde's behavior is in order to see how this 3rd party tool for xfce holds up/needs to be modified | 17:48 |
pikapika | Turns out the native window preview feature in plasma is somehow worse than what this 3rd party xfce tool provides. Which is a bit hard to believe and I still assume its probably something with his particular system rather than a general fact without more confirmation. | 17:49 |
pikapika | arraybolt3, thanks for the help | 17:53 |
arraybolt3 | pikapika: Sure, sorry it wasn't much help. | 17:53 |
pikapika | so I am guessing caching the last known preview is still the "best" possible that I know in any wm | 17:53 |
xavier_ | bonjour | 18:01 |
=== blackest_mamba_ is now known as blackest_mamba | ||
=== blackest_mamba_ is now known as blackest_mamba | ||
=== diskin is now known as Guest1196 | ||
=== diskin_ is now known as diskin | ||
avih | sarnold: do you know what APT::Get::Never-Include-Phased-Updates does? is it the default which i get now? | 18:55 |
avih | maybe most importantly, is there any official docs on how to control this thing? or just a random collection of posts in various planes | 18:57 |
sarnold | avih: random collection of notes in random places. and I've got a suspicion there's mistakes. | 18:58 |
avih | :( | 18:59 |
avih | so the Always-Include is the old thing to ignore the phase and just upgrade anyway, right? what's Never-Include? would that make it give me the package only on 10% phase? | 19:00 |
avih | 100% * | 19:01 |
avih | (i.e. last) | 19:01 |
avih | i'm seriously considering leaving ubuntu, despite my 10 years old install... | 19:02 |
sarnold | avih: both options are brand new. | 19:05 |
avih | hmm | 19:06 |
sarnold | avih: they work together in an awkward way to provide three options. if you select the fourth option you get a pile of sadness. | 19:06 |
avih | i guess the source code would be the best docs, unfortunately... | 19:06 |
sarnold | I've read apt sources -- I doubt it | 19:06 |
avih | :) | 19:06 |
avih | what else is there? there's no docs... | 19:07 |
sarnold | avih: try this: I suggest editing your apt config to set APT::Get::Always-Include-Phased-Updates to true | 19:07 |
avih | (as great as irc advice is, it's no substitute for official docs) | 19:07 |
avih | sarnold: yeah, i got it, this will make it behave like it did with 20.04, right? but i'm trying to understand what the other options are. specifically currently whether Never is what i have now, or yet another behavior. | 19:08 |
=== blackest_mamba_ is now known as blackest_mamba | ||
rbasak | avih: have you read https://discourse.ubuntu.com/t/phased-updates-in-apt-in-21-04/20345? I think that describes all the configuration options? | 19:25 |
bparker | When I boot the 22.04 installer and select Japanese language, the window grows larger than the screen and the buttons at the bottom are no longer accessible. How do I fix this? | 19:25 |
avih | it mentions them, i don't think i saw a description of what they do | 19:25 |
rbasak | I agree it should be documented better and I filed https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1986660 about it | 19:25 |
ubottu | Launchpad bug 1986660 in apt (Ubuntu) "Phased update configuration flags not documented in apt.conf(5)" [Undecided, New] | 19:25 |
rbasak | It's quite clear to me. | 19:26 |
rbasak | An update is being phased if it's less than 100%. | 19:26 |
avih | i do admit i didn't read the whole page. i'll check your link again | 19:26 |
rbasak | So Always is like before, where it ignored phasing and packages being phased were available to you immediately. | 19:26 |
rbasak | Never will do what you said - treat packages being phased as not selected by your machine (until phasing reaches 100%). | 19:27 |
avih | it the page i looked at earlier though. the first post mentions them - i did see that. then it seem to become a discussion of sort, which i didn't try to follow too closely | 19:27 |
rbasak | No need to follow the discussion | 19:27 |
avih | can you be specific where the Never thing is described? | 19:27 |
rbasak | "To never include phased updates, set APT::Get::Never-Include-Phased-Updates to false." | 19:28 |
rbasak | The missing part is what a phased update means | 19:28 |
avih | is the Never behavior the default? i.e. what i see now? how do i know what is my config on this? | 19:28 |
=== blackest_mamba_ is now known as blackest_mamba | ||
avih | currently i have 40 kept back packages, all phased | 19:29 |
rbasak | Phasing is the default - ie. both Always and Never are false. | 19:30 |
rbasak | (in 22.04 and I think 21.04 and 21.10) | 19:30 |
rbasak | 40 kept back packages doesn't sound right | 19:30 |
avih | thanks. will choosing Never make it not show at all? or still shown as kept back, just with even a bigger list of packages than i have now? | 19:30 |
rbasak | Maybe they're being held back because of some other problem - not phasing? | 19:30 |
avih | definitely phasing. i apt show $pkg for each of them and confirmed they're at various phasing stages | 19:31 |
avih | (in a script, not manually) | 19:31 |
rbasak | https://people.canonical.com/~ubuntu-archive/phased-updates.html is the list of packages currently being phased | 19:31 |
rbasak | It's quite small. | 19:31 |
rbasak | Though these are source packages. There are more binary packages, but 40 doesn't sound right. | 19:32 |
avih | that's my list of 40 packages, all in phasing as far as i can tell (or at lest they were few hours ago, i didn't apt update since) https://0x0.st/oVEy.txt | 19:33 |
rbasak | Ah that might be right. | 19:34 |
rbasak | So what are you actually wanting to achieve? | 19:34 |
avih | i want it to behave like they're not offered at all. like the desktop package manager does it adds noise, and it doesn't let me notice actual conflicts among the ever changing list of kept back packages | 19:35 |
rbasak | So you want Never=true | 19:36 |
rbasak | But that won't remove the noise entirely | 19:36 |
avih | maybe, if i could read some description of what it does... :) | 19:37 |
rbasak | But that won't remove noise | 19:37 |
rbasak | You'll get lists of packages you are declining the install | 19:37 |
rbasak | It's complicated because packages have dependency relationships. | 19:37 |
avih | right, so basically same as now, just with a bigger list. | 19:37 |
rbasak | Right. | 19:37 |
arraybolt3 | If the package isn't updating automatically due to phasing, don't show it at all in the "held back" list, I think is what avih is going for. | 19:38 |
rbasak | Or you can pretend phasing doesn't exist, in which case you can use Always=true | 19:38 |
rbasak | No noise, and exactly the same behaviour as before. | 19:38 |
rbasak | You get updates as soon as they are released. No phasing. | 19:38 |
rbasak | arraybolt3: sure, but that's an option that doesn't exist currently. | 19:38 |
avih | i understand that, yes, that's how it was with 20.04, and it's good to know i can make it happen. but i also prefer to not be on the cutting edge, and actually utilize this phasing mechanism (which i do think it's great, except for the noise it adds to apt) | 19:39 |
rbasak | It's a bit complicated to achieve, too, because dependencies mean that if you don't install some packages because they're being phased, there are other packages that may be held back too, because they depend on the phased packages. | 19:39 |
arraybolt3 | rbasak: True. I think avih is trying to find out if that feature exists (or possibly how to implement it). | 19:39 |
rbasak | The dependency thing means that this is actually much more complicated than I think you realise. | 19:39 |
rbasak | It's not trivial to implement at all. | 19:40 |
rbasak | The reason a package is held back becomes blurred. It may be due to phasing, or some dependency thing caused by phasing, or some dependency thing unrelated to phasing. | 19:41 |
rbasak | It's a complex tree of reasoning and trying to track reasons through the tree is an extra layer of complexity. | 19:41 |
avih | rbasak: as i said yesterday, i don't want apt to tell me what it's NOT going to install for reasons unrelated to errors or conflicts. as long as it reaches some decision, whatever it is as long as it's not an error etc, don't tell me, just do it | 19:41 |
rbasak | And the outcome may be "because of phasing _and_ a dependency problem", so any spec needs to accomodate what the output should look like in that case, too. | 19:41 |
=== blackest_mamba_ is now known as blackest_mamba | ||
avih | if it's both, then until it gets past the stage where it's held back by phasing, then the conflict is not actually a reason. once it's out of phasing and conflict becomes a reason, THEN tell me about it | 19:42 |
rbasak | avih: I understand why you want that. It makes sense to want that. But the reality is more complex than "for reasons unrelated to errors or conflicts". | 19:42 |
rbasak | And right now, there's no tracking in apt to do that kind of thing, and I don't see it being implemented due to the complexity involved. Sorry. | 19:43 |
arraybolt3 | avih: The problem is that, your suggestion would work if you have just "package (phasing)". But if you have "package #1 (phasing) <- package #2 (NOT phasing)", and both have an update, package #2 will be held back because package #1 won't be installed. And package #1 won't be installed because it's phasing. That's tricky to implement in code when you have giant dependency trees (this is a very simple explanation of it). | 19:43 |
rbasak | Right | 19:43 |
avih | it always is, but currently on every update i get dozens of held back packages due to phasing, and i can't identify actual conflicts which do need my attention | 19:43 |
enigma9o7[m] | what kinda packages is ubuntu doing this phased updates, and why? | 19:44 |
ioria | dozens sounds a lot | 19:44 |
sarnold | enigma9o7[m]: updates that aren't security updates | 19:44 |
rbasak | Turn off phasing then. Without someone volunteering a _huge_ amount of time working on apt's dependency resolution and scoring algorithms, you can't have both what you want *and* phasing. | 19:44 |
enigma9o7[m] | usually ubuntu doesnt update libraries between versions anyway, so what are these dependencies you're tlaking about | 19:44 |
arraybolt3 | enigma9o7[m]: It's for stability, so if a wonky update slips through, it can be stopped. | 19:44 |
avih | ioria: 40 as of few hours ago, all confirmed in phasing by parsing the output of apt show $pkg https://0x0.st/oVEy.txt | 19:45 |
enigma9o7[m] | Updates should be tested before being pushed to LTS users anyway! | 19:45 |
enigma9o7[m] | Test them in devel version or non LTS versions. | 19:45 |
arraybolt3 | enigma9o7[m]: They are. But people make mistakes that slip through sometimes. | 19:45 |
enigma9o7[m] | Have they had some update they started pushing, then pulled it back? | 19:45 |
arraybolt3 | enigma9o7[m]: We here in Linux land can look at Microsoft making epic boffo after epic boffo and chuckle, but the fact is that sometimes Canonical makes the same kind of epic boffos. | 19:46 |
arraybolt3 | (In fact they fumbled hard with Ubuntu 18.04 just the other day and took out vast swaths of Azure VMs on accident as a result.) | 19:46 |
avih | rbasak: ack (can't hack both). thanks. | 19:46 |
bparker | fragmentation is the biggest linux boffo | 19:46 |
arraybolt3 | You should see some of the boffos I've made. Sometimes testing in controlled environments just doesn't do the trick. I can think my code is ironclad, then hand it to my boss and have a report that he broke it within the week. | 19:47 |
avih | s/hack/have/ | 19:47 |
enigma9o7[m] | If it can be disabled on the user side tho, then its not such an annoyance. | 19:47 |
rbasak | enigma9o7[m]: we do require updates to be tested before they are released. And we try to ensure that all reasonable use cases are tested. But it's not unusual to see a regression report because of some complex edge case or interaction that nobody knew to test. | 19:47 |
rbasak | Tests are only as good as their coverage, and coverage of everything is basically impossible. | 19:48 |
rbasak | So phasing helps. | 19:48 |
tom | i cant access a folder unless im using an app like saving a file | 19:49 |
tom | it has disappeared | 19:49 |
tom | whats going on? | 19:49 |
ioria | yes, looks like in 22.04 phased updates are heavier than usual | 19:49 |
=== tom is now known as Guest6240 | ||
arraybolt3 | Guest6240: Like the "Files" app disappeared? | 19:49 |
ioria | but it's not bad | 19:49 |
Guest6240 | Pictures folder | 19:49 |
arraybolt3 | Guest6240: Did you by any chance accidentally rename it to ".Pictures"? | 19:50 |
Guest6240 | my memes are too hot, cia hiding it from me | 19:50 |
arraybolt3 | Guest6240: Try enabling "Show hidden files". | 19:50 |
arraybolt3 | You might have just renamed the folder on accident and ended up hiding it from yourself. | 19:50 |
Guest6240 | arraybolt3, tried doesnt work | 19:51 |
Guest6240 | arraybolt3, all i did was try to make a new one when i couldnt find it | 19:51 |
Guest6240 | but then i deleted that | 19:51 |
Guest6240 | not even visible from terminal | 19:51 |
arraybolt3 | Guest6240: Try "ls -a" while in your Home folder. | 19:52 |
arraybolt3 | Guest6240: If that fails, check your Trash. | 19:52 |
arraybolt3 | If that fails, try "ls -R | grep Pictures". | 19:52 |
arraybolt3 | If that fails, try "ls -R | grep <name of a file you know is in there", replacing the placeholder as appropriate. | 19:52 |
avih | rbasak: just for the record, i do think phasing is great, it's only the current ui/behavior/implementation which i don't like. the concept is 100% solid, but i have zero idea what goes on under the hood with it | 19:52 |
arraybolt3 | And if that fails, then shut the computer all the way down so that we can try to use Photorec to recover the (probably accidentally deleted) files. | 19:53 |
Guest6240 | `i must have moved it | 19:53 |
arraybolt3 | Guest6240: Then "ls -R | grep Pictures" should find it. | 19:53 |
Guest6240 | thanks will try that | 19:53 |
avih | the fact that it buries actual conflicts in a list of harmless phased kept back packages is a real issue IMO | 19:54 |
Guest6240 | found it thanks | 19:54 |
rbasak | avih: if it helps, apt phasing is hooking into the usual apt_preferences(5) scoring mechanism that you can examine with "apt policy". That might help it make more sense. | 19:54 |
rbasak | Actual conflicts aren't really supposed to happen on production systems :-/ | 19:54 |
sarnold | rbasak: heh, avih already has some scripts to run apt policy on every single package being shown as held back.. | 19:54 |
rollappuser | hi | 19:55 |
sarnold | rbasak: heh, they happen *all the time*, especially among folks who need different php versions or python versions or whatever.. | 19:55 |
arraybolt3 | o/ | 19:55 |
avih | rbasak: yes, but that's what "kept back" was used for, and now if there is a conflict, you're pretty much guaranteed to miss it | 19:55 |
sarnold | rbasak: I run into them periodically despite trying my best to keep my systems pretty dang minimal | 19:55 |
rbasak | Yeah - but that's fundamentally a broken approach to take, even if it's the norm in various parts of the community :-/ | 19:55 |
sarnold | fair | 19:55 |
ioria | avih, they go to -proposed (you know what that means) , verified then they got deployed | 19:55 |
avih | (i don't have any) | 19:55 |
arraybolt3 | sarnold: ?! I never run into them unless I fumble something during package management. | 19:56 |
rbasak | It'd be nice if apt gave more detailed info, but given the CS-level complexity involved in implementing it, I think it's fair to leave that to the people hacking apt locally to volunteer. | 19:56 |
sarnold | arraybolt3: it's not like it's every day :) I'd probably see less of them if I waited for the .1 LTS releases to upgrade | 19:57 |
rbasak | Nonwithstanding what I said in bug 1988819, which should be an easy way to get most of the way there without that complexity. | 19:58 |
ubottu | Bug 1988819 in apt (Ubuntu) "When apt keeps back packages due to phased updates, it should say so" [Undecided, Confirmed] https://launchpad.net/bugs/1988819 | 19:58 |
ioria | maybe | 19:59 |
=== blackest_mamba_ is now known as blackest_mamba | ||
avih | it's only the upgrade which is hard, because apt upgrade doesn't take a list of packages. apt install does, but it also unconditionally marks as manual. so i should --mark-auto etc and it becomes too much hassle | 20:26 |
arraybolt3 | avih: I'm not understanding why you need to do that in the first place since what you're trying to do is already "apt upgrade"'s default behavior. | 20:26 |
avih | arraybolt3: it's not. its default behavior is to show mee dozens of packages it's NOT going to upgrade, and hide real conflicts which i might have | 20:26 |
arraybolt3 | avih: Hmm, because you're going for UI changes, not functionality changes... ok that makes sense. Uhh... | 20:26 |
* arraybolt3 tries some things | 20:26 | |
avih | what? | 20:26 |
arraybolt3 | avih: You're trying to get apt to only show you broken packages in the "held back" section, right? | 20:27 |
avih | the functional change is that i would be able to tell if there are real conflicts when i run apt upgrade | 20:27 |
arraybolt3 | And simply hide the phased and held back packages? | 20:27 |
avih | (or rather my script which i gave up on) | 20:27 |
arraybolt3 | So if anything is visible in the held back category, you know it's because of breakage? | 20:27 |
murmel | yes | 20:27 |
avih | yes | 20:27 |
arraybolt3 | OK. I have an idea based off of what rbasak was saying earlier, lemme try something real quick. | 20:27 |
arraybolt3 | avih: Uh... well you know what I'm trying to say. Arrays are a fundamental part of programming languages to my mind. Take away my arrays and you've just robbed the language of a ton of its power. | 20:51 |
murmel | Guest32: as xubuntu uses only the base + their own gui packages, they support only the gui part for 3 years, that's why it's heavily discouraged to use it after that, as nobody updates those, (in comparison to the base ubuntu packages) | 20:51 |
oerheks | .. and do you really want to run a desktop longer than 3 years? | 20:51 |
enigma9o7[m] | guest32: but obviously your browser and kernel still get security updates | 20:51 |
sarnold | arraybolt3: don't worry, you'd adapt to being listbolt2 :) | 20:51 |
arraybolt3 | sarnold: ROFL | 20:52 |
enigma9o7[m] | @oerheks I run 18.04 myself on primary system.... | 20:52 |
arraybolt3 | You mean varbolt1. | 20:52 |
arraybolt3 | I mean I guess you could make a 16 gigabit integer and store data that way... | 20:52 |
avih | arraybolt3: and that's my point exactly, that what you think is fundamental many times turns out not to be, if you spend some time solving problems without it at your arsenal | 20:52 |
enigma9o7[m] | So yeah, I do wanna use it more than 3 years. I have played with newer stuff on other machines and VM. Nothing that makes me need to update. | 20:52 |
enigma9o7[m] | errr upgrade. | 20:53 |
arraybolt3 | avih: That makes good sense to me. I guess if there's advantages to POSIX scripting I maybe should learn it. But when there's a tool at my disposal, I like to use it. | 20:53 |
Guest1083 | impossibile aggiornare "aggiornamento configurazione secure boot dbx" blocked executable in the ESP, ensure grub and shim are up to date: /bootefi/EFI/Debian/shimx64.efi Authenticode checksum [letter-and-number] is present in dbx ... i'm on ubuntu 20.04, secure boot is disable and grub is update ... is possible to resolve this? | 20:53 |
arraybolt3 | Why limit yourself needlessly? | 20:53 |
avih | (obviously arrays are fundamental in programming) | 20:53 |
avih | because portability | 20:54 |
avih | i.e. not needlessly | 20:54 |
arraybolt3 | avih: Exactly. Then it's not a needless limit. (I'm not arguing, just, re...luc...tant... .) | 20:54 |
arraybolt3 | Guest1083: You're sure Secure Boot is disabled? | 20:55 |
arraybolt3 | Guest1083: Do you dual-boot Windows? | 20:55 |
arraybolt3 | (I've heard of Windows doing weird stuff with Secure Boot, maybe it somehow turned it back on?) | 20:55 |
avih | and i'm willing to bet that once you learn to solve problems in posix sh, which is pure subset of bash (or rather the other way around), your bash scripts will be better too | 20:55 |
Guest1083 | arraybolt3 yes in bios now and yes is in dual boot winzozz :) | 20:55 |
avih | anyway, that solution looks good :) | 20:56 |
arraybolt3 | Guest1083: Hmm... it looks like your system is trying to boot Debian, I just noticed. | 20:56 |
arraybolt3 | I wonder if that's somehow causing problems. | 20:56 |
avih | arraybolt3: what's "conf" at the beginning of some lines at the output of the apt -s? like this: Conf ure (1:7.3.6-0ubuntu0.22.04.1 Ubuntu:22.04/jammy-updates [amd64]) | 20:58 |
Guest1083 | arraybolt3 the error is in ubuntu software and in software update ... only this error ... not try debian in this pc ... how to update shim, please ... just to try ;) | 20:58 |
arraybolt3 | avih: I think that's "configure". | 20:58 |
avih | meaning? | 20:58 |
avih | can it be conflict? | 20:58 |
arraybolt3 | avih: There's multiple stages in Debian package installation, configuring the package is one of them. No, I don't think it can be conflict. | 20:59 |
avih | well, but the "inst" ones are first, and the "conf" ones are later. what it it actually simulating? | 20:59 |
arraybolt3 | avih: "Simulated runs print out a series of lines, each representing a dpkg operation: configure (Conf), remove (Remv) or unpack (Inst). Square brackets indicate broken packages, and empty square brackets indicate breaks that are of no consequence (rare). | 21:00 |
arraybolt3 | (From the apt-get man page.) | 21:00 |
arraybolt3 | So it's simulating the unpacking of a .deb file followed by the configuration of the now installed software. But it's not changing your system, it's just showing you what steps it would do. | 21:00 |
avih | yeah, i get the last part, but right, configuring as part/post install | 21:01 |
theonewhoiam | all of a sudden ubuntu software isnt working | 21:01 |
theonewhoiam | I tried installing gnome software like a website said and it didnt fix it | 21:01 |
ogra | the archive is down ... wait a little | 21:01 |
theonewhoiam | why does Ubuntu constantly have issues like this | 21:01 |
avih | arraybolt3: so where should i see the conflicts then? (which is the main goal of this, right?) in square brackets? | 21:01 |
arraybolt3 | avih: If anything had gone wrong that ended up legitimately holding packages back, you should see it in the "held back" section of the simulated apt run. | 21:02 |
theonewhoiam | ogra so the repos are down? I gotta get work done | 21:02 |
theonewhoiam | come on people | 21:02 |
theonewhoiam | this is why Windows reigns supreme | 21:02 |
avih | i don't see that part... | 21:02 |
arraybolt3 | theonewhoiam: Sometimes stuff goes wrong, sorry. Microsoft botches stuff equally as bad all the time. | 21:03 |
theonewhoiam | not to where I cant install anything | 21:03 |
avih | does it mean nothing is held back in this sim (with phased disabled) | 21:03 |
theonewhoiam | jesus | 21:03 |
arraybolt3 | avih: Then you don't have any problems that are holding back packages, other than phasing. | 21:03 |
arraybolt3 | theonewhoiam: I understand your frustration, but please keep offensive language out of this chat. | 21:03 |
arraybolt3 | !language | 21:03 |
ubottu | Please avoid any language that may be considered offensive, including acronyms and obfuscation of such - also see https://wiki.ubuntu.com/IRC/Guidelines || The main channels are English only, for other languages, see https://wiki.ubuntu.com/IRC/ChannelList | 21:03 |
=== Scotty_Trees3 is now known as Scotty_Trees | ||
theonewhoiam | yeah your stuff blows up in my face but my language is the problem | 21:03 |
enigma9o7[m] | theonewhoiam: there are 100 different mirrors | 21:04 |
enigma9o7[m] | if you're saying yours is down, try another | 21:04 |
sarnold | enigma9o7[m]: the snap store is down, not the archive mirrors | 21:04 |
theonewhoiam | ok how do I fix it | 21:04 |
sarnold | enigma9o7[m]: there's no mirrors for the snap store | 21:04 |
theonewhoiam | ok the snap store | 21:04 |
theonewhoiam | how do I fix it | 21:04 |
murmel | that, and if you really rely on repos that much, just mirror so you always have a mirror working | 21:04 |
sarnold | you do not | 21:04 |
arraybolt3 | theonewhoiam: Can you install your software from apt? | 21:04 |
theonewhoiam | unreal | 21:04 |
sarnold | canonical IS fix it | 21:04 |
enigma9o7[m] | But what about gnome-software, that doesnt rely on snap store does it? | 21:04 |
arraybolt3 | enigma9o7[m]: Pretty sure it does. | 21:04 |
enigma9o7[m] | They said they installed gnome-software. | 21:04 |
sarnold | enigma9o7[m]: it does, they're kinda cranky about it though | 21:05 |
arraybolt3 | theonewhoiam: What software are you trying to install? | 21:05 |
theonewhoiam | multi-thousand dollar contract and I cant access the software I need | 21:05 |
theonewhoiam | vscode | 21:05 |
theonewhoiam | its a snap | 21:05 |
murmel | sarnold: how would that work as by default the snap plugin is not installed when installing gnome-software | 21:05 |
arraybolt3 | theonewhoiam: There's a repo for it that will let you install it with apt. | 21:05 |
theonewhoiam | ill wait | 21:05 |
theonewhoiam | going to the store to buy windows actually | 21:05 |
avih | you can get the debian tar.gz vscode thing and just extract it someplace | 21:05 |
bparker | add-apt-repository down for anyone else? | 21:05 |
theonewhoiam | I am done farting with Linux what a failed experiment | 21:06 |
avih | (from the vscode website) | 21:06 |
ogra | theonewhoiam, https://status.snapcraft.io/ ... hkeep an eye on this page | 21:06 |
sarnold | bparker: yeah; it's being worked on | 21:06 |
enigma9o7[m] | its actually installed as a reccommend, but if you install gnome-software --no-install-recommends it doesnt support snap | 21:06 |
theonewhoiam | here is a hint, charge people money so you can buy better servers | 21:06 |
theonewhoiam | seacrest out | 21:07 |
bparker | rofl | 21:07 |
enigma9o7[m] | They do charge money. | 21:07 |
arraybolt3 | *TheRedQueen sets ban on theonewhoiam* (if they're paying money they should be yelling at Canonical sales support anyway) | 21:07 |
murmel | enigma9o7[m]: ahh makes sense (I don't install gnome-software manually (only the plugins which pull in gnome-software)) | 21:07 |
enigma9o7[m] | Yeah that. | 21:07 |
* ogra expects gnome-software to be a short lived experiment in ubuntu ... it will go away | 21:08 | |
enigma9o7[m] | Why? | 21:08 |
enigma9o7[m] | Its in kinetic/devel now.... | 21:09 |
ogra | enigma9o7[m], because there is a new store app already in the perview snap channel of "snap-store" | 21:09 |
enigma9o7[m] | works fine, very useful if you remove snapd | 21:09 |
enigma9o7[m] | although I guess there's always plasma-discover | 21:09 |
arraybolt3 | (And if he's not paying money then he's a troll :P Wait until one Patch Tuesday some core functionality like, oh, you know, SHORTCUTS, break out of nowhere, like happened to Windows yesterday.) | 21:09 |
bparker | removing snapd is the best thing any ubuntu user can do | 21:10 |
ogra | written in flutter 100% better search results, properly picking deb/snap as needed, way faster, doesnt keep crufty daemons running in your session etc etc | 21:10 |
enigma9o7[m] | Yeah but won't that replace ubuntu-software/snap-store, it shouldn't affect the gnome-software package. | 21:10 |
ogra | well, no, but once it becomes a default g-s will be demoted to universe and just come as stright upstream package from fedora | 21:11 |
enigma9o7[m] | Ah ok. Well that's fine, as long as it doesn't "go away". | 21:11 |
ogra | like synaptic and other tools ... | 21:11 |
ogra | we should honestly never have dropped UbuntuSoftware for it in th first place | 21:12 |
sarnold | I suspect the gnome folks feel similar :) | 21:13 |
ogra | heh | 21:13 |
ogra | well, it is the fault of snaps i think ... | 21:14 |
ogra | (that we switched to it) | 21:14 |
jim__ | sup@karmic_koala | 21:14 |
ogra | g-s has an easy to use plugin system ... USW would actually have needed more work to support them | 21:15 |
Guest1083 | i would like to try to update shim but its the first time i see it:) https://wiki.ubuntu.com/UEFI/SecureBoot/ShimUpdateProcess?_x_tr_sl=en&_x_tr_tl=it&_x_tr_hl=it&_x_tr_pto=sc ... che sfigaXD | 21:15 |
=== blackest_mamba_ is now known as blackest_mamba | ||
Guest1083 | sudo update-secureboot-policy can unlock executable in the ESP ? or do I leave everything alone and some electronic saint will arrange a self-repair and can I go back to drinking peacefully ??? :) | 21:26 |
clarkk | I'm trying to update my ubuntu 20.04 system, and it hangs saying, "Installing updates...Updating snaps". But I seem to remember that the only packages that were available was a new kernels version. Does anyone know what's going on? | 21:30 |
clarkk | When I attempt to click "Details", it is not responsive | 21:30 |
ogra | clarkk, the canonical datacenter is down | 21:30 |
clarkk | oh | 21:31 |
ogra | (well, big parts of it) | 21:31 |
clarkk | ok, is there a network status page? | 21:31 |
ogra | it is being worked on, be patient | 21:31 |
clarkk | I am patient | 21:31 |
ogra | specifically for snaps https://status.snapcraft.io/ | 21:31 |
ogra | (we used to have the same thing for launchpad, but that seems dead ATM) | 21:31 |
clarkk | ok, that page is useful. Thanks | 21:32 |
clarkk | But what I don't understand is why kernels are coming as snap packages | 21:32 |
murmel | they don't | 21:33 |
clarkk | somehow, my install finished | 21:35 |
ogra | currently at least 🙂 | 21:35 |
clarkk | well, I'll probably move to another distro if that happens | 21:35 |
ogra | why is that ? | 21:35 |
ogra | kernls are perfect for snap packages | 21:35 |
clarkk | thanks for your help :) | 21:36 |
enigma9o7[m] | Wow, a cross platform kernel? How does that work? | 21:36 |
ogra | and in fact there are snapped kernels in UbuntuCore | 21:36 |
murmel | ogra: sure, but you don't boot them ;) | 21:36 |
enigma9o7[m] | cross-distro I mean. | 21:36 |
enigma9o7[m] | Weird. | 21:36 |
ogra | murmel, indeed yu do | 21:36 |
murmel | ogra: I mean on the actual host, when booting the system itself | 21:37 |
ogra | murmel, grub suppors squashfs filesystems and loop devices ... they get booted directly out of the readonly snap | 21:37 |
murmel | oh interesting | 21:37 |
ogra | which also means you do not need to uncompress 1GB of modules 😉 | 21:38 |
ogra | they are way smaller on disk that way | 21:38 |
murmel | eh, I wouldn't care if the kernel + modules would use 10G ;) | 21:38 |
ogra | some people do ... (the ones with small /boot 😉 ) | 21:38 |
murmel | yeah, I guess so | 21:39 |
ogra | also snaps have bult-in rollback | 21:39 |
ogra | so if a kernel oopses after upgrade, the system automaticlly rolls back to the former one | 21:39 |
murmel | I am rather more interested how canonical deals with zfs-on-root ;) as they try to get rid of it, but they really can't except "breaking" some installs | 21:40 |
murmel | ogra: eh, I don't mind too much the whole snap thing. as long as it stays usable (in comparison to the firefox snap rollout) | 21:40 |
ogra | we typically carry along the stuff we support ... but simply drop it for new installs | 21:40 |
ogra | yeah, firefox had some rough edges initially | 21:41 |
murmel | ogra: yeah I only read the whole issue where in the 22.04 dev phase zsys was removed, but it was reintegrated just before release. heavy AFAIR | 21:41 |
murmel | ogra: yeah that was the only issue I had. it felt like it was still too early to be rolled out | 21:41 |
murmel | on the other hand idk, why there is no option to have uncompressed snaps | 21:42 |
ogra | yup ... but then, how else do you find all the issues in corner case use cases | 21:42 |
ogra | someone would have to implement it ... 🙂 | 21:42 |
murmel | hm, idk, I would have started with something not as heavy and (used) as much as ff, as idc if lxd for example takes a bit to launch the first time, in comparison to ff | 21:43 |
murmel | ogra: I would really love to implement it, if I had knowledge about programming :(, still on the scripting level | 21:44 |
ogra | well, there are many aspects to it ... like cost ... the store footprint of snaps would massively rise ... bandwith usage etc etc | 21:45 |
Guest1083 | hello everyone and good things forever ... bye bye :) | 21:45 |
murmel | ogra: definitely, I won't deny that | 21:46 |
murmel | btw do you know if there is a way to cache snap downloads? | 21:46 |
ogra | murmel, https://docs.ubuntu.com/snap-store-proxy/en/ | 21:50 |
ogra | nothing beyond this i think | 21:50 |
ogra | it is back https://status.canonical.com/ | 21:50 |
murmel | ogra: thanks, great way to save bandwith and time (as I am most of the time on a limited bandwith (somewhere around 300-600kb/s)) | 21:52 |
ogra | uh | 21:52 |
ogra | thats slow | 21:52 |
murmel | yes | 21:52 |
murmel | it's stupid slow, and I don't understand how the hotels get away with it. already complained to a few, but they won't budge (I guess too few do it) | 21:53 |
sarnold | they get away with it because they can sell access to reasonable connectivity | 21:53 |
murmel | sarnold: i would even understand that, but there is no option to "upgrade" | 21:54 |
sarnold | murmel: oh wow ;) | 21:54 |
sarnold | they're leaving money onthe table! | 21:54 |
murmel | I mean I am not in america xD | 21:54 |
murmel | i have never seen such option here in europe, except on airplanes | 21:55 |
ogra | jus go to burgerking or starbucks 😛 | 21:55 |
ogra | *just | 21:55 |
murmel | ogra: that's what I am doing, but quite expensive for faster internet. (and most stores do have roughly the same speed) (gladly I found a sb which has a limit of 6.5 megs/s) | 21:56 |
murmel | I mean right now I am paying roughly 200 euros/month just for starbucks internet access xD | 21:56 |
ogra | lol | 21:58 |
ogra | thats gross | 21:58 |
murmel | ogra: sadly enough, couldn't find any other store with higher bandwith, otherwise I would do it, as I am no fan of sb | 21:58 |
arraybolt3 | murmel: IHOP? | 21:59 |
ogra | yeah, i get what yu mean | 21:59 |
murmel | arraybolt3: not USA | 21:59 |
arraybolt3 | murmel: I've gotten some of the fastest speeds I've ever encountered from outside an IHOP. (And I thought it was International. It's in the name.) | 21:59 |
murmel | arraybolt3: nah, us company trying to be bigger than they are xD | 21:59 |
=== blackest_mamba_ is now known as blackest_mamba | ||
arraybolt3 | murmel: They don't have IHOP where you are? Then they need to rename to UHOP. | 21:59 |
arraybolt3 | (pun not intended but I wish it had been now) | 22:00 |
murmel | lol The owner of this website (ihop.com) has banned the country or region your IP address is in (HU) from accessing this website. | 22:00 |
arraybolt3 | murmel: Oh wow. | 22:00 |
murmel | i wonder if it's because of gdpr or other reasons | 22:01 |
enigma9o7[m] | starbucks & mcdonalds usually have really good free wifi | 22:33 |
=== five618 is now known as five61 | ||
murmel | enigma9o7[m]: it really depends. for example while I was in prague, mc and sb had abysmal internet (somewhere around 200kb/s). while here in budapest, it's okay, except with one mcd around the corner (250kb/s) and only 30 mins | 23:58 |
murmel | but yes, otherwise I had better internet in shops than hotels :( | 23:58 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!