[00:03] <lostndev> sarnold looks like that was a red herring, I've got the results from ip route get matching on both accounts but the behavior is the same still
[00:04] <sarnold> lostndev: aww :(
[00:05] <lostndev> Also wg0 is working as expected from root (the same curl with --interface wg0) from root works. Also curl without specifying an interface on root works. It's just root with the sim card interface that is not playing nice.
[00:06] <sarnold> I wouldn't expect the kernel's source selection to differ from user to user without something else being different, like one of the processes being in a different networking namespace..
[00:08] <lostndev> @sarn
[00:08] <lostndev> sarnold for sanity checking - https://pastebin.com/s8y2DfNK
[00:10] <lostndev> Maybe this could be useful, the adapter is being brought up as follows: https://pastebin.com/dbVEXntT
[00:11] <lostndev> The python script is used to pass the "AT" command to the serial tty of the SIM card. The first command is related to configuration (if this was the first boot of the sim it would be required), the 2nd AT is for enabling a mode on the card. Those should realistically just prep the sd card to be able to get a lease.
[00:12] <lostndev> **replace "sd card" with "sim card" in last sentence
[00:12] <lostndev> What i was thinking is if the combination of using the ip command and ifconfig might be causing strange behavior.
[00:15] <lostndev> re: routes identical: https://pastebin.com/FZv0xGJe
[00:16] <lostndev> sarnold could we replace ifconfig up with an ip command to do the same?
[00:19] <lostndev> Going to try "sudo ip link set dev wwp0s19u1u3i2 up" instead of iconfig up just to be sure.
[00:32] <sarnold> lostndev: d'oh, sorry, I managed to miss https://pastebin.com/s8y2DfNK
[00:33] <sarnold> lostndev: same with https://pastebin.com/FZv0xGJe :(
[00:33] <sarnold> lostndev: definitely replacing ifconfig with ip is a good idea, ip is ifconfig's replacement
[00:34] <lostndev> sarnold The paste ending in ...fNK was showing that the "ip route get ..." matches and ..GJe a literal "diff" on the output of route -n from root compared to the devbot user showing that the routes are equal. I just tried bringing up the adapter without the ifconfig cmd but the result is the same.
[00:35] <sarnold> aha
[00:35] <sarnold> hmm :(
[00:36] <lostndev> sarnold content from those 2 pastes in case I missed something - https://pastebin.com/7TPAvyqk
[00:40] <sarnold> lostndev: okay, cool, so the routing table is the same, or gives the same results anyway.. both curl calls with explicit interfaces get the same results..
[00:40] <sarnold> lostndev: that leaves outgoing connections getting different source ports. I wonder, maybe try nc or bash's /dev/tcp/ ?
[00:45] <lostndev> sarnold Goolging usage now, if you have an example of the test you would try I'd greatly appreciate it :)
[00:49] <sarnold> lostndev: echo "testing" | nc 216.239.35.0 80   or ...
[00:49] <lostndev> sarnold ah got it I expect that to work since it wouldn't be using the sim card's interface 1s
[00:50] <sarnold> lostndev: there's an example of /dev/tcp on https://tldp.org/LDP/abs/html/devref1.html  near the words 'exec'
[00:50] <lostndev> sarnold yep both of those work with nc for both accounts to the 216 ip.
[00:52] <lostndev> Same with /dev/tcp (cat </dev/tcp/time.nist.gov/13)
[00:54] <sarnold> lostndev: okay, so two other programs are getting the exact same outgoing source port, but curl isn't.. maybe keep trying other similar tasks with different tools? openssl s_client, telnet, ping, wget, tcptraceroute, etc? it'd be nice to have more confidence that it's something curl is doing vs something the kernel is doing
[00:57] <lostndev> Well yes but -- those didn't use the source ip of the sim card adapter (or specify the interface). However, if we use
[00:58] <lostndev> echo "testing" | nc -s 192.168.225.27 216.239.35.0 80 we are successful on both accounts so it leans toward your last comment
[00:59] <lostndev> Hmmph ok the same is true for telnet with -b (telnet -b 192.168.225.2 216.239.35.0 80) works on both accounts
[00:59] <sarnold> lostndev: I'm *really* confused why just one tool used the sim nic by default :) sorry I'm not more helpful..
[01:01] <lostndev> sarnold the paste that didn't specify the sim card nic was just to show that internet worked over wg0 from both accounts. The goal is to ~"know" that internet works if the sim card's nic is used as the source adapter. Rather it should work on both accounts the same.
[01:02] <lostndev> sarnold so the test without using the source address with nc isn't ~ a valid test since we want to test internet from the sim card nic. That's why I mentioned the nc -s (binding to the sim card's ip), and this works on both accounts. Which lends us to curl doing something really strange here under the hood.
[01:03] <sarnold> lostndev: how much do you love strace? :) I think stracing curl on both root and non-root while doing the test might help..
[01:07] <lostndev> sarnold not the biggest fan, quite happy actually it is a curl issue. I can switch to nc with -s :)
[01:09] <sarnold> lostndev: oh! if nc -s is suitable, yay :)
[01:11] <lostndev> sarnold thank you so much!
[01:13] <sarnold> lostndev: my pleasure, I wish we'd gotten to the bottom of the confusing curl choices, but I can completely understand wanting to limit the time spent chasing things :)
[01:31] <lostndev> sarnold yep, might have to come back to curl as well heh just not today
[03:38] <synapt> Since launchpad channel has been silent on my issue, in case anyone new is lurking and wants to take a look for anything I/we missed earlier, trying to fix a really odd broken apt issue w/ erlang; https://paste.st-projects.com/2q71oq2gro81e9maunll
[03:56] <sarnold> synapt: have you had any luck contacting the rabbitmq folks?
[03:56] <sarnold> lostndev: heh, I know how that goes.. the number of loose ends I never got around to picking up..
[04:10] <synapt> sarnold: I'm gonna poke the #rabbitmq channel here shortly but there's all of 15 people there, which isn't much higher than it used to be on freenode even lol
[04:10] <synapt> as far as I can tell everything leads back to that v24 package update, I just can't tell -how- it borked things in apt
[04:11] <sarnold> synapt: email the address in the changelog?
[04:12] <sub526> Hello all, apt-get update results in "W: GPG error: http://murphymgr.europe.root.pri/ubuntu-xenial xenial InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY 7DFA9777703CAB41" . Is there a way to fix it?
[04:13] <sarnold> sub526: probably whoever published that repository has published their public key somewhere
[04:14] <Bashing-om> !xenial
[04:15] <sub526> sarnold: I have access to that repository, is there a way to get the public key from that repository?
[04:18] <leftyfb> sub526: Ubuntu 16.04 went EOL 2 months ago
[04:18] <leftyfb> sub526: also, contact the admin of the repo for support with their repo
[04:19] <sub526> leftyfb: Okay I am trying to install the deb packages from that repo in Ubuntu 18.04 machine
[04:19] <leftyfb> sub526: don't
[04:20] <sub526> leftyfb: Added the source list in /etc/apt/sources.list.d/  directory and then tried to run 'sudo apt-get update', this won't work?
[04:21] <leftyfb> sub526: what packages are you trying to install?
[04:22] <sub526> leftyfb: that private repo has certain host tools deb packages
[04:22] <sarnold> sub526: as far as I know there's no standard location within the repository for the key -- the usual idea is that you would get the key through a 'trusted mechanism' somewhere, maybe an https site if that's good enough for you, something like that..
[04:22] <leftyfb> sub526: what "host tools"?
[04:26] <Ginalo> Hi. I have an Ubuntu installation in a BTRFS filesystem. I use Timeshift to take snapshots of it. My problems is that in the used space shown in `df` doesn't match the latest snapshot's size in Timeshift. Why is that?
[04:26] <Ginalo> I thought BTRFS filesystems son't use additional space when taking snapshots. Is this not accurate?
[04:28] <sarnold> Ginalo: do you have any redundancy or compression on those files? df and du give different results on zfs than ext4, and probably btrfs is similar
[04:29] <Ginalo> sarnold: As far as I know, I don't have redundancy or compression on my  files, although I may not fully understand what that means. Could you elaborate on the redundancy part?
[04:36] <sarnold> Ginalo: btrfs lets you use mirrors, so if a drive goes bad or you get bad sectors, it can repair the broken file, or broken files, etc
[04:39] <Ginalo> sarnold: It's the first time I've heard about that, so I don't think I have that set up.
[04:41] <sarnold> Ginalo: aha :)
[04:56] <lostndev> sarnold An interesting find, ping also supports binding to an interface with -I this is failing as well on the same server that we can't use curl --interface.
[04:57] <lostndev> sarnold However, wget's --bind-address works just like netcat's bind-address with -s. So I'm thinking it's got to be something particular to how this sim card's nic works under the hood that the interface can't be bound to.
[06:50] <cart_man> Hi everyone. I made a very simple script -> #!/bin/bash  ~/bin/$1
[06:50] <cart_man> and placed it in /usr/bin/
[06:50] <cart_man> All it does it runs AppImages with a name I give it
[06:51] <cart_man> But... It binds to the terminal which I really dont like because I dont want 100 terminals all running apps on constantly
[06:51] <cart_man> soo... When I add & to the script it jsut makes my CPU run 100% and never displays the app
[06:51] <cart_man> how can I fix this?
[07:22] <gry> cart_man: hi
[07:22] <gry> cart_man: does typing "~/bin/someapp &" in terminal by hand work, or it causes 100% cpu usage?
[07:25] <cart_man> gry: it seems like the CPU isn't ramping but the app is not launching visually at least and when you try and use the temrinal it exits the app
[07:27] <gry> cart_man: does typing "~/bin/someapp", without the &, in terminal, by hand work?
[07:27] <gambusia> When I ssh into macOS from a Linux laptop, autocomplete executes on the first letter typed; but the second letter prints after the first completion, and the third after that, and so on. A simple ls command looks like lss; exit is worse at exitxititt making the command line a mess. Why? Doesn't happen when I ssh into any Linux system; just on my one and only macOS host. Ideas?
[07:28] <cart_man> gry: Same unfortuantely
[07:28] <gry> cart_man: what works, then? have you been able to start the app at all?
[07:28] <gry> if yes, then how?
[07:28] <cart_man> gry: Yes by leaving &
[07:29] <gry> cart_man: so "~/bin/someapp" in terminal works ok?
[07:29] <cart_man> gry: sigh sorry
[07:29] <cart_man> I misread. Starting the app with just "~/bin/app"  does work yea
[07:29] <cart_man> but its an AppImage though
[07:29] <cart_man> just adding & to that makes the same behaviour
[07:30] <tango_uniform_xr> gambusia maybe a mac-related channel could help more, but first, since it's easiest, i would switch to a different shell in macOS and see if it happens in that one too, and then i would starting looking at ssh options if that doesn't help
[07:36] <gambusia> Haven't tried that yet, tango. Thanks for the tip.
[07:37] <polve> Hi. I am trying to install build-essentials on a very fresh ubuntu desktop minimal installation, but this is the output apt gives me: https://0x0.st/-9bq.txt
[07:38] <tango_uniform_xr> polve did you try apt update and then try the installation again?
[07:38] <polve> tango_uniform_xr: yes, forgot to mention that. System is up to date
[07:40] <tango_uniform_xr> i mean apt update as in fetch the repos, but if you did that then they should be there
[07:42] <polve> I really haven't done anything out of the ordinary with apt, other that manually installing a .deb package
[07:42] <tango_uniform_xr> try `apt show libc6-dev g++` and see if apt thinks those packages exist
[07:42] <tango_uniform_xr> polve now that i looked at it again, i notice that the g++ version it was asking for might not be in the repos, since it doesn't match mine (i'm on focal aka 20.04), so i would try an `apt-get update;apt-get upgrade`
[07:43] <tango_uniform_xr> and retry installing the build-essentials once again afterward
[07:44] <polve> I'm on focal as well
[07:44] <polve> I tried both things: yes it can see them (show), and no update;upgrade didn't solve
[07:44] <tango_uniform_xr> that's weird, the version i see in the repos via apt show says 9.3.0 for g++
[07:45] <tango_uniform_xr> polve what architecture are you on?
[07:45] <polve> amd64
[07:47] <tango_uniform_xr> did you check `apt-cache policy` for errors?
[07:48] <tango_uniform_xr> actually, check your sources.list, maybe it's missing `focal-updates` or `focal-security` polve
[07:51] <tango_uniform_xr> also worth an `apt-get -f install` since you installed that one .deb, maybe the dependencies for that didn't get installed
[07:52] <polve> I think you got it
[07:52] <polve> not the --fix-broken thing
[07:52] <polve> focal-updates is commented
[07:53] <polve> I'll try and look up how to edit that in a while
[07:53] <tango_uniform_xr> ahh cool, i was just throwing out every possibility i could think of :)
[07:53] <tango_uniform_xr> `sudo editor /etc/apt/sources.list`
[07:54] <polve> but do you think I did this myself?
[07:55] <tango_uniform_xr> i don't think you did, do you think you did?
[07:56] <polve> no...
[07:56] <tango_uniform_xr> i don't ask that in a bad way, sometimes i make changes and forget, is what i meant
[07:57] <polve> but maybe when the graphical software updater came up sometime, I might have hastily disabled something
[07:57] <polve> tango_uniform_xr: of course
[07:57] <polve> same
[07:58] <tango_uniform_xr> well anyway, after you fix the sources.list, try apt-get update once again
[07:58] <tango_uniform_xr> to refresh the repos, then install build-essentials
[07:58] <tango_uniform_xr> i would apt-get upgrade before installing stuff though, personally
[08:02] <polve> 153 packages can be upgraded. Run 'apt list --upgradable' to see them.
[08:03] <polve> I don't understand why the line `deb http://it.archive.ubuntu.com/ubuntu/ focal-updates main restricted' was missing, though
[08:03] <polve> not commented, but outright missing
[08:26] <calmarti> Hi there
[08:27] <toddc> welcome calmarti
[08:27] <gry> hi
[08:30] <calmarti> i believe ubuntu doesn't see my mic
[08:30] <gry> did it work before?
[08:30] <gry> what does `lspci -vmmnn` say? (please paste it to a pastebin)
[08:31] <tango_uniform_xr> did you check the pulseaudio profile?
[08:31] <tango_uniform_xr> `pactl list sinks`
[08:31] <tango_uniform_xr> sorry, wrong command
[08:31] <tango_uniform_xr> `pactl list cards` not sinks :)
[08:31] <Mekaneck> calmarti: which ubuntu version? Maybe there are some bug reports on launchpad.
[08:31] <calmarti> no, it's a new laptop, I installed ubuntu a month ago but I had never used the mic until yesterday when I tried to join a zoom meeting
[08:32] <calmarti> I appreciate your patience
[08:32] <calmarti> and help
[08:32] <calmarti> it's fossa
[08:32] <calmarti> 20.04
[08:32] <calmarti> 20.04.2 to be precise
[08:32] <calmarti> i know how to use the comand inxi to get more info
[08:33] <calmarti> CPU: Quad Core 11th Gen Intel Core i7-1165G7 (-MT MCP-)
[08:33] <calmarti> speed/min/max: 1193/400/4700 MHz Kernel: 5.10.0-1029-oem x86_64 Up: 42m
[08:33] <calmarti> Mem: 1969.9/15655.3 MiB (12.6%) Storage: 953.87 GiB (1.4% used) Procs: 292
[08:33] <calmarti> Shell: bash 5.0.17 inxi: 3.0.38
[08:33] <tango_uniform_xr> calmarti you can use paste.ubuntu.com for longer stuffs
[08:33] <calmarti> thankx tango
[08:34] <calmarti> sorry about that
[08:34] <tango_uniform_xr> no problem, could you should me `pactl list cards` output?
[08:35] <calmarti> the weird thing is that in settings/sound ubuntu detects a 'digital microphone-sof-hda-dsp'
[08:37] <calmarti> here's the output tango: https://paste.ubuntu.com/p/rQQYfJT9PQ/
[08:38] <calmarti> so it seems that it does detect the input device
[08:38] <calmarti> it may be a zoom issue
[08:39] <calmarti> I haven't even tried to use the mic with other application
[08:39] <tango_uniform_xr> you can try installing e.g. `audacity` and see if it works in there
[08:40] <tango_uniform_xr> i would also try plugging a mic into the 3.5mm port and see what happens with that
[08:41] <tango_uniform_xr> weird tip but maybe relevant here: headphones (not sure about earbuds) also works as mics if you plug them into the 3.5mm :)
[08:41] <calmarti> headphones work for sure
[08:42] <calmarti> but i'll try to find a mic or a set of headphones with a mic
[08:42] <tango_uniform_xr> oh, you knew already?
[08:42] <ledeni[m]> calmarti: how you install zoom snap ?
[08:42] <calmarti> yes, i used snap
[08:42] <calmarti> sudo snap install zoom-client
[08:42] <calmarti> i believe
[08:43] <tango_uniform_xr> to clarify: regular headphones with no microphone, they will pick up audio like a mic if you plug the 3.5mm into the input
[08:43] <calmarti> i mean, i used headphones everyday so I know they work
[08:44] <calmarti> oh you mean a 3.5mm connector
[08:44] <calmarti> sorry i didn't get it before
[08:44] <calmarti> i'll try that, should've tried that before coming in here anyways
[08:45] <tango_uniform_xr> did you try seeing if the regular mic works in audacity yet, though?
[08:46] <calmarti> no, not yet
[08:46] <ledeni[m]> calmarti: install and run pavucontrol and see input device is pick up
[08:46] <calmarti> still trying to find the 3.5mm or a headphone with mic
[08:47] <tango_uniform_xr> yeah, i was giving info for what to do if you launch audacity and the mic doesn't work there, also pavucontrol is good, you should use that to see if the source is set as default, has volume, and is unmuted
[08:48] <ledeni[m]> calmarti: any headphones for mobile will do it
[08:49] <calmarti> i'll try both ledeni and tango ideas
[08:51] <calmarti> got the headphones with mic, i guess i need to use zoom in order to test the mic, is that right?
[08:51] <gry> no
[08:52] <gry> zoom is proprietary software, you don;'t need it; you could use, for example,audacity instead
[08:52] <calmarti> i think i got confused
[08:52] <gry> audacity means you can record a sound and then listen to it
[08:52] <calmarti> ok, got it
[08:53] <calmarti> isn't there a test i could do just with pulseaudio?
[08:53] <tango_uniform_xr> you could do, iirc, `parec test.wav`
[08:54] <tango_uniform_xr> and shout at your microphone "hello? is anybody there?!"
[08:54] <tango_uniform_xr> then ctrl-c when you're done
[08:54] <calmarti> i have shout at the mic, it doesn't work
[08:54] <tango_uniform_xr> sorry, it's `parecord` not parec
[08:55] <tango_uniform_xr> then you can play the file with `paplay test.wav`
[08:55] <calmarti> tango is this the right sintax:  iirc parec test.wav ?
[08:56] <calmarti> oh i hadn't read that before
[08:56] <tango_uniform_xr> calmarti it's parecord, i got it confused with parec
[08:56] <calmarti> parecord
[08:57] <tango_uniform_xr> so, `parecord test.wav`
[08:57] <calmarti> ok
[08:58] <calmarti> I did 'parecord test.wav' while saying stuff at the mic from the headphones but can't hear anything
[08:58] <tango_uniform_xr> you have to press ctrl-c to stop recording, then do `paplay test.wav`
[08:58] <calmarti> i pressed up and down the mic button on the headphones
[08:58] <calmarti> several times
[08:58] <calmarti> can't hear anything
[08:59] <calmarti> do you think i should install the pulseaudio stuff you said before?
[08:59] <ledeni[m]> calmarti: run alsamixer and check F4 capture maybe your mic is muted
[08:59] <tango_uniform_xr> pavucontrol is a helpful GUI application
[09:01] <ledeni[m]> calmarti: if you see MM it is sign it is mute
[09:02] <calmarti> i have 3 columns, the third one says MM at the bottom
[09:03] <calmarti> but the unmute key is not on
[09:03] <calmarti> i mean, i havent pressef the unmute key
[09:04] <calmarti> cause if i hit it it does mutes, so it is unmuted now
[09:04] <calmarti> is there a way to unadvertedly mute the mic other than by pressing this key?
[09:05] <calmarti> alsamixer: master and headphones are 00, speaker is MM
[09:05] <calmarti> AND Mic Boos kjust shows an empty bar with no data
[09:06] <calmarti> auto-mut says 'disabled'
[09:07] <ledeni[m]> calmarti: ok install pavucontrol will be easier to check what is going on
[09:09] <calmarti> done
[09:09] <calmarti> i mean i already installed pavucontrol
[09:09] <ledeni[m]> calmarti: just run like 'pavucontrol'
[09:10] <calmarti> ok i i have 3 tabs: playback, recording and output devices
[09:11] <calmarti> i'm on playback
[09:11] <ledeni[m]> ok you have input devices if you press > on right side
[09:12] <calmarti> oh ok i'm there
[09:12] <calmarti> it shows the headphones mic (says 'plugged in')
[09:13] <ledeni[m]> 👍 calmarti
[09:14] <ledeni[m]> calmarti:  run zoom
[09:14] <calmarti> ledeni i think i unmuted it...
[09:14] <calmarti> but noti mean
[09:14] <calmarti> i mean
[09:15] <vlm> fill a device with /dev/zero, will that be slower than with /dev/urandom if so then why is that?
[09:15] <calmarti> i hit an unmute option in the volume control window of pavucontrol and now when i hit the mic button of the headphone it shows a pulse, but still can't hear a thing!
[09:15] <calmarti> i'm gonna run zoom
[09:18] <calmarti> now at least zoom detects theres a mic
[09:18] <calmarti> but still can't hear myself
[09:19] <calmarti> the zoom test speaker shows a pulse when i talk but i can't hear meyself
[09:19] <calmarti> weird
[09:20] <calmarti> it seems theres something wrong with the headphone builtin mic
[09:20] <ledeni[m]> calmarti no check headphone speakers maybe muted too
[09:22] <calmarti> i checked , there not muted
[09:22] <calmarti> but know i have gone a long way to solve the issue
[09:22] <calmarti> i'll open a zoom meeting with a friend and try to solve it myself
[09:22] <ledeni[m]> ok
[09:22] <calmarti> a million thanks to you ledeni and tango and everyone else
[09:23] <calmarti> i appreciate your patience and help so much
[09:58] <emcy> how long do non lts releases have repo support for
[09:59] <ThinkT510> 9 months
[10:01] <ActionParsnip> Hi guys do Ubuntu Member cloaks transfer over to Libera chat please?
[10:03] <Mekaneck> ActionParsnip: ask in #ubuntu-irc
[10:03] <ActionParsnip> Thanks
[10:03] <Mekaneck> you need to ask for your UM cloak there
[10:03] <Mekaneck> yw
[14:41] <lalbornoz_> https://i.imgur.com/zkMEoVo.jpeg
[14:58] <hexo> hi folks, how can i downgrade a package?
[14:59] <hexo> new firefox is broken for me, i need older version
[14:59] <hexo> like 89.0.0
[14:59] <hexo> or 88
[14:59] <ogra> how is it broken ?
[15:00] <hexo> one of several problems is that ctrl-l doesn't work if site refused to load
[15:00] <hexo> i.e. it doesn't focus address bar
[15:01] <hexo> so i need older version
[15:01] <ogra> well, ctrl-l works on all my machines here ... does not smell like a firefox problem per-se ...
[15:02] <hexo> that was not a question
[15:02] <hexo> how can i downgrade a package, please?
[15:03] <ogra> it might be removed from the package cache given it is a security risk to keep the old version around but typically you can force package versions via: "sudo apt install <package>=<version>"
[15:04] <ogra> if you want easy rollback, use snaps they have that built in
[15:04] <hexo> thanks, i dont like to use snaps
[15:04] <ogra> well, they would solve your current problem with a single command 🙂 but indeed your choice
[15:05] <ogra> (sudo snap revert firefox)
[15:05] <hexo> i dont get why it cant be in apt too
[15:05] <ogra> because of the package design of debs
[15:06] <ogra> debs are only designed to go forward ... even the above =version thing is rather a hack against the design ...
[15:07] <ogra> but FF likely has no versioned dependencies so there it might work if you know the exact version and if the old version is still in apt's package cache
[15:07] <hexo> so i'm stuck with broken browser
[15:07] <hexo> :(
[15:08] <hexo> i dont have it in cache
[15:08] <ogra> 88.0.1+build1-0ubuntu0.20.04.2 was the last version before 89
[15:08] <hexo> E: Version '88.0.1+build1-0ubuntu0.20.04.2' for 'firefox' was not found
[15:09] <hexo> no luck for me
[15:09] <ogra> yeah, wiped from the package cache on archive.u.c
[15:10] <ogra> the old version is still on launchpad but YMMV trying to install it ...
[15:10] <ogra> https://launchpad.net/ubuntu/focal/+source/firefox
[15:10] <ogra> i'D rather try to find out how to get the behaviour fixed in the new FF
[15:11] <ogra> they introduced client side themes and there are options to turn that off
[15:11] <ogra> perhaps your issue is related to that
[15:11] <oerheks> hexo, the only option is firefox esr
[15:11] <V3n3RiX> maybe you just wanna disable proton (the new floating tabs c**p)
[15:11] <ogra> yeah, thats what i meant with the themeing, thanks ... didnt remember the actual name
[15:12] <hexo> V3n3RiX: i've disabled proton, it worked well with firefox 88
[15:12]  * hexo looks at launchpad
[15:12] <oerheks> ppa:mozillateam/firefox-next
[15:12] <oerheks> oops
[15:13] <V3n3RiX> still works great on ff 89 too ... I hate proton ... that maybe the cause of your issues
[15:15] <hexo> launchpad helped a lot, i
[15:16] <hexo> i'm on 89.0.0 now, it is broken too, but a bit less
[15:16] <Chunkyz-RPi> hexo: 89.0.1 is out in the repo's...
[15:17] <Chunkyz-RPi> they fixed a few things, maybe fixed your problem. sorry; not read the scrollback...
[15:17] <hexo> Chunkyz-RPi: that one is even more broken
[15:17] <hexo> :)
[15:17] <Chunkyz-RPi> not really, working great here.
[15:18] <ogra> same here on all machines (including pi400 as well as desktop PC and laptop)
[15:19] <infectedserver> Hi, my copy of Ubuntu server has been infected by Monero mining malware. I am of re-installing, but I am trying to do some investigations to understand the level of compromise.
[15:19] <hexo> Chunkyz-RPi: works on your machine (tm)
[15:19] <hexo> :)))
[15:19] <ogra> infectedserver, perhaps you should go to #ubuntu-server with that
[15:19] <infectedserver> One thing I noticed is that the SSH host key has changed. Why would malware authors do this?
[15:19] <Chunkyz-RPi> works on a lot of others too, hexo. so.... :)))
[15:20] <hexo> please refrain from such comments when you have no idea whats happening or what the bug is
[15:21] <ogra> hexo, well, you are definitely the first one complaining here in this channel 🙂
[15:21] <Chunkyz-RPi> ogra: indeed.
[15:21] <hexo> *facepalm* :)
[15:21] <ogra> (indeed IRC feedback isnt any kind of metric ... 🙂 )
[15:22] <Chunkyz-RPi> so, hexo; what's the bug? *facepalm* :)
[15:22] <hexo> Chunkyz-RPi: one of several problems is that ctrl-l doesn't work if site refused to load
[15:23] <Chunkyz-RPi> ... sounds like you problems.
[15:23] <hexo> sounds like you know the best
[15:24] <hexo> ogra: thanks for the launchpad tip, it helped
[15:25] <Chunkyz-RPi> anyway, off to use my *working* FF without bugs. have a great day all. :-)
[15:28] <webchat38> hi
[15:34] <spicyCebolla> after updating to 21.04, when the screen turns off on an hp dv7 laptop due to inactivity, i'm unable to turn it back on, i have to hard shutdown.  is this a known trouble with a fix?
[16:09] <mh3f> Is there an Ubuntu equivalent to the "All information in JSON format" listed here https://security-tracker.debian.org/tracker/? It's a JSON file containing each package's CVEs and the fixed package version.
[16:11] <ogra> mh3f, there are atom and RSS feeds https://ubuntu.com/security/notices no json i think
[16:16] <mh3f> ogra: thanks, I saw that but unfortunately it's truncated.
[16:19] <mh3f> ogra: I think I found it https://ubuntu.com/security/api/docs#/
[16:20] <ogra> ah, sweet
[16:27] <infectedserver> How can I reinstall bash?
[16:27] <infectedserver> Bash is compromised on my system
[16:28] <Habbie> then you should reinstall your whole system
[16:29] <infectedserver> i will
[16:29] <infectedserver> i am trying to do more analysis to find out what's exfilitrated
[16:30] <infectedserver> i already found the malware authors control servers... I think it is my ex actually
[16:30] <llutz> infectedserver:  apt -o Dpkg::Options::="--force-overwrite" install --reinstall bash
[16:30] <infectedserver> i am compiling the evidence for the police because I know theyre not going to poke around a backdoored rootkited linux installation
[16:30] <infectedserver> thansk llutz, that worked
[16:31] <infectedserver> ok... now to kill the hidden backdoored ssh server which listens on port 22, and blocks the real ssh server...
[16:32] <mh3f> wouldn't it be better to take an image of the drive and use a live cd for analysis?  you're modifying/tampering with the evidence
[16:32] <infectedserver> that was step 1 :)
[16:32] <infectedserver> i already copied an image
[16:33] <infectedserver> what is concerning is looking at bash history, he was messing around with my SSD firmware...
[16:33] <infectedserver> i do not know what firmware he uploaded to my ssd
[16:36] <Guest5065> recently i installed ubuntu 20.04 and it was pretty fast on live usb, however when installed on hard disk, it is slow and sluggish, further i found that my hard disk is SATA III , however it is running in SATA II mode in ubuntu
[16:36] <TJ-> Guest5065: can you share the kernel log with "pastebinit <( journalctl -k )" ?
[16:38] <Guest5065> wait program not installed
[16:39] <Guest5065> https://paste.ubuntu.com/p/b3KQ5fCJKs/
[16:41] <infectedserver> is this a backdoor?
[16:41] <infectedserver> https://www.joesandbox.com/analysis/434230/0/html
[16:41] <infectedserver> this is the file found as a replacement for "bash", it was modified when he was hacking it
[16:41] <infectedserver> it manifests and seems to act like bash
[16:42] <infectedserver> but it was uploaded to this cloud analysis tool with the file name of 'backdoor"
[16:43] <Guest5065> TJ-, any clue why it is happening?
[16:45] <TJ-> Guest5065: at the end there are a lot of erorrs and the SATA link keeps hard resetting at 1.5Gbs - working my way back to see if there's a clue as to why
[16:49] <TJ-> Guest5065: the first link up at line 744 is also 1.5G "ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)"
[16:49] <Guest5065> TJ-, thanks, one more info that it has some bad sectors once but all of them were  reallocated and how to paste hard disk logs
[16:51] <TJ-> Guest5065: the mobo's link is definitely 6Gbps capable though, line 642 "ahci 0000:00:1f.2: AHCI 0001.0300 32 slots 6 ports 6 Gbps 0x2 impl SATA mode"
[16:52] <Guest5065> TJ-,  that's why it is sluggish, but how to force the sata mode or to make it detect as SATA III instead of SATA II
[16:53] <TJ-> Guest5065: my guess is there's a problem on the drive OR it has previously been programmed to operate at 1.5
[16:54] <TJ-> Guest5065: have you looked at the SMART info?
[16:54] <Guest5065> TJ-, yes, it has shown 240 bad sectors but all of them were reallocated and shown disk is ok
[16:55] <TJ-> Guest5065: how about the link info ?
[16:57] <Guest5065> TJ-, TA Version is:   ATA8-ACS (minor revision not indicated)
[16:57] <Guest5065> SATA Version is:  SATA 2.6, 1.5 Gb/s (current: 1.5 Gb/s)
[16:57] <Guest5065> TA Version is:   ATA8-ACS (minor revision not indicated)
[16:57] <Guest5065> SATA Version is:  SATA 2.6, 1.5 Gb/s (current: 1.5 Gb/s)
[16:57] <Guest5065> TA Version is:   ATA8-ACS (minor revision not indicated)
[16:57] <Guest5065> SATA Version is:  SATA 2.6, 1.5 Gb/s (current: 1.5 Gb/s)
[16:58] <TJ-> !paste | Guest5065
[17:03] <Guest5065> TJ-, https://imgur.com/a/4lfyK8B
[17:07] <TJ-> Guest5065: the screenshot isn't really helpful, but the lines you started to paste into channel look to e be. It looks like the drive is only able to operate at 1.5G. How about "pastebinit <( sudo smartctl -a /dev/sdb )"
[17:12] <webchat46> Hi there. I want to *remotely* add a second permanent IP address to an ubuntu desktop 18LTS machine. Can I just ssh in and drop a netplan yaml file in? Or are desktop installations set up to look elsewhere for network configs? I want to ask first to avoid borking the network config and locking myself out of SSH...
[17:17] <Guest5065> TJ-, i am a novice in these things, i am not dumping channel, i know that people are helping me, i am grateful for that whatever best they are doing for me,  https://paste.ubuntu.com/p/rf5J6JpfzF/
[17:18] <TJ-> Guest5065: don't worry, it just helps us focus. I'll look now
[17:19] <TJ-> Guest5065: the key here seems to be the drive only claims to support SATA 2.6 (1.5Gbps) "SATA Version is:  SATA 2.6, 1.5 Gb/s (current: 1.5 Gb/s)"
[17:19] <TJ-> Guest5065: so, we have to think as the docs seem to indicate this drive is claimed to the SATA3 something has gone rather wrong
[17:22] <Guest5065> TJ-, yes, drive is SATA 3, look at these specifications https://www.hdsentinel.com/storageinfo_details.php?lang=en&model=TOSHIBA%20MK5061GSY
[17:25] <Guest5065> TJ-, transfer rate is 3Gbps on many website, it means, there is something messed up either with software or hardware, i can't change cable as it is connected directly with motherboard
[17:26] <TJ-> Guest5065: here's an askubuntu question where a drive with the same model number shows "SATA 2.6, 3.0Gb/s" https://askubuntu.com/questions/544168/high-iowait-and-slow-timing-cached-reads
[17:27] <TJ-> Guest5065: so I think either the drive, or the link (connectors of cable) have a problem. The hard errors also support that. Easiest possible fix is to unplug and reconnect the SATA connector at both ends of the cable to ensure a firm connection
[17:29] <Guest5065> TJ-, okay, thanks, i would try that but for that i would have to turn off the laptop. By the way,thanks a lot for help. May God Bless you and your family and your wishes come true
[17:30] <TJ-> Guest5065: a laptop!? Then I'm going to guess you've simply got the drive dislodged slightly in its bay and all you need do is reseat it in the connector
[17:30] <TJ-> Guest5065: can happen with laptops since they get knocked about a lot
[17:30] <gmcastil> i an SMB share that contains user directories (similar to home directories) that I want to mount for individual users at startup without requiring root intervention - is there a way to mount a directory of an SMB share at someplace like /home/<username>/smb_home at startup with RW permissions?
[17:31] <Guest5065> TJ-, thanks :)
[17:36] <webchat46> is it even possible to configure a network based on network manager remotely over SSH?
[17:37] <webchat46> *configure a desktop 18LTS network I mean
[17:38] <TJ-> webchat46: yes
[17:39] <TJ-> webchat46: NetworkManager configurations can be manipulated using "nmcli", or if you know the syntax, directly in /etc/NetworkManager/system-connections/$connection_name
[17:42] <TJ-> webchat46: NetworkManager configurations can be manipulated using "nmcli", or if you know the syntax, directly in /etc/NetworkManager/system-connections/$connection_name
[17:44] <webchat46> TJ- thank you very much that tool looks perfect
[17:47] <CodeGeek> if I want to become an official ubuntu member, do I need to have contributed stuff? Is it not enough that I've been using Ubuntu since forever?
[17:48] <oerheks> CodeGeek, contribute, support, write .. and you will need to find volunteers/other members to vote for you
[17:49] <oerheks> https://wiki.ubuntu.com/Membership
[17:49] <CodeGeek> ok, I guess I'll remain an unofficial member :)
[17:50] <oerheks> a member is just a volunteer .. but there are volunteers that are recognized and appreciated, it is just what do you expect of the membership.
[17:50] <lotuspsychje> CodeGeek: share your wiki in #ubuntu-discuss perhaps
[18:41] <gmcastil> im on Ubuntu (5.8.0-55-generic #62~20.04.1-Ubuntu SMP Wed Jun 2 08:55:04 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux) with XFCE and cannot seem to figure out how to keep the machine from locking every 5 minutes
[18:41] <gmcastil> where does the XFCE window manager control that setting?
[18:43] <lotuspsychje> gmcastil: this what you seek for? https://www.chaslinux.com/article/stop-xubuntu-2004-blanking-screen-every-5-minutes
[18:47] <gmcastil> i had it set to an hour, but it still blanked and locked the screen afte ra few minutes
[18:52] <gmcastil> nope
[18:52] <gmcastil> its blanking in 5 minutes still
[18:53] <gmcastil> and the laptop is plugged into a dock
[18:56] <gmcastil> is there another location that allows me to determine screen blanking / locking preferences?
[18:57] <sarnold> gmcastil: maybe xset q ?
[18:57] <gmcastil> Screen Saver:
[18:57] <gmcastil>   prefer blanking:  no    allow exposures:  no
[18:57] <gmcastil>   timeout:  300    cycle:  300
[18:57] <gmcastil> that looks promising
[18:58] <gmcastil> that seems odd that an out of the box Ubuntu would require screen blanking behavior to be controlled using xset though
[18:59] <oerheks> i think your dock is the issue ..
[18:59] <oerheks> does this behaviour occur when off the dock?
[19:00] <gmcastil> i dont know - new machine and havnet had it off the dock yet
[19:00] <gmcastil> why would the dock affect it?
[19:01] <gmcastil> although this is weird....the XFCE power manager says that the battery is at 90% charge
[19:01] <gmcastil> and its a) new machine and b) been on the dock charging for 5 days
[19:01] <gmcastil> there are three devices called "Line power"
[19:01] <gmcastil> one of which isn't plugged in
[19:01] <gmcastil> (or at least, that's what it says)
[19:02] <gmcastil> its plugged into the dock via Thunderbolt
[19:03] <oerheks> dirty trick, disable energy features :  xset -dpms
[19:05] <gmcastil> if that works, is there a way to not have to do that every time i reboot the machine?
[19:20] <darrenalcorn> are there any good guides about how you might set up a Ubuntu 20.04 server with a few multipass VMs for specific services? As in, a nginx server running on a multipass node... the main server routing specific traffic to that multipass VM
[19:23] <V3n3RiX> gmcastil:
[19:23] <V3n3RiX> xset dpms 0 0 0
[19:23] <V3n3RiX> xset -dpms
[19:23] <V3n3RiX> xset s off
[19:26] <gmcastil> well, xset seems to be what was keeping the door open
[19:26] <gmcastil> what does that change that controls that setting?
[19:38] <V3n3RiX> X power management ... prevents the screen from going to sleep or blanking out
[19:47] <gmcastil> it apparently supersedes the others, presumably because it was triggering first
[19:47] <gmcastil> (set at 5 minutes vs. the never or 60 minutes i set in the XFCE power manager)
[20:01] <gmcastil> well crap, it locked again after like 10 minutes
[20:01] <CarlFK> is there a hotkey to do poweroff ?
[20:02] <gmcastil> i honestly dont get why this is in any way a complicated thing to do - is XFCE just not well supported by Ubuntu?
[20:03] <sarnold> CarlFK: most modern computers let you tap the power key and it'll do something -- check out the HandlePowerKey= setting in the logind.conf manpage
[20:06] <CarlFK> sarnold: I don't have a keyboard (this is for a single use video player box) - but I do have a USBRubberDucky that will send keyboard commands.  https://docs.hak5.org/hc/en-us/articles/360010555153-Ducky-Script-the-USB-Rubber-Ducky-language
[20:06] <CarlFK> I'm trying to figure out what key it needs to send
[20:45] <upgrader> Hi all - I just upgraded from 18.04 to 20.04, and now my resolution (NVIDIA GTX 1060) is stuck at 1024x768, and the nvidia x server settings application just opens a blank window.
[20:48] <upgrader> (with a Help & a Quit button - so not entirely blank)
[20:50] <drdo> I have just installed 21.04 desktop and I have this issue where the current window loses focus periodically for no apparent reason
[20:51] <drdo> It just periodically goes in and out of focus, every few seconds
[21:11] <drdo> I'm also getting a bunch of these messages: "[ 1481.988607] traps: gjs[15113] trap int3 ip:7fd3a7cde447 sp:7ffda7bd84a0 error:0 in libglib-2.0.so.0.6800.1[7fd3a7c9f000+8d000]"
[21:19] <sarnold> CarlFK: ah, I don't know if the power key can sent as a 'keyboard' key or not. you could probably hook it up to send control+alt+delete, I think that might also be configurable...
[21:24] <sarnold> drdo: hmm, there's some similar-sounding bug reports around but nothing looks fantastic. I suggest filing a bug, 'ubuntu-bug libglib2.0-0' -- I think that'll do the right thing, I'm not 100% sure where that version of libglib comes from
[21:32] <CarlFK> sarnold:  control+alt+delete should work.  having trouble getting the ducky thing to do it
[21:34] <sarnold> CarlFK: can you set a delay between 'keypresses' on it? something like send ctrl, wait 2ms, send alt, wait 2ms, send delete, wait 2ms, release delete, wait 2sm, release alt, wait 2sm, release ctrl?
[21:35] <CarlFK> sarnold: um.. du no.  the docs on this are pretty bad.  I think this is the reference: https://docs.hak5.org/hc/en-us/articles/360010555153-Ducky-Script-the-USB-Rubber-Ducky-language
[21:39] <sarnold> CarlFK: hrm. so if you want to send a ctrl+alt+del you've really got to write out "ctrl alt delete" and can't specify the order, how long, etc :(
[22:07] <CarlFK> woot.. some other tool gave me the magic: CTRL+ALT DEL
[22:07] <sarnold> CarlFK: yay
[22:07] <sarnold> CarlFK: what worked?
[22:07] <CarlFK> CTRL+ALT DEL - that gets me the "shut down in 60, 59..." dialog
[22:08] <CarlFK> looks like alt-s will hot key the "shutdown" button
[22:09] <sarnold> CarlFK: oh as in the literal command CTRL+ALT ? heh
[22:20] <Guest4219> where is grub loader normally installed?
[22:20] <Guest4219> in a typical installation of ubuntu
[22:23] <Bashing-om> Guest4219: Grub: Generally to the device (say sda) .
[22:25] <Bashing-om> !info grub | Guest4219
[22:25] <Guest4219> no matter if its BIOS o UEFI system?
[22:25] <Bashing-om> !grub | Guest4219
[22:28] <Guest4219> something like that
[22:28] <Guest4219> and i have since started using windows again on that pc
[22:29] <rbasak> With UEFI, normally your ESP is mounted at /boot/efi/ and grub is installed into there.
[22:30] <Guest4219> and also wiped out windows (registry issues bla bla) and now i can't recover my windows from disk image backup because grub seems to be causing errors in the recovery software
[22:31] <Guest4219> can this ESP/UEFI/GRUB be wiped out somehow? it keeps popping back in after i wipe the disk and i can't recover my windows system because of it
[22:33] <Guest4219> i wish to be ridden of grub for good on that pc
[22:35] <sarnold> you could dd some of /dev/zero right over wherever it is
[22:36] <Guest4219> is "GRUB 2 EFI loader" different beast than "GRUB loader"?
[22:37] <Guest4219> in the recovery software i keep seeing "Failed to detect GRUB loader" and "Detected GRUB 2 EFI loader on disk '\local\hd_gpt""
[22:38] <Guest4219> like one GRUB for BIOS and a different GRUB for UEFI?
[22:44] <Guest4219> anyone? i'm desperate to get my pc back to windows and grub is in the way. i don't know how but i think it's because recovery software is linux based
[22:44] <Guest4219> dd /dev/zero /boot/efi ?
[22:45] <CarlFK> Guest4219: getting red of things does not make other things work
[22:45] <cbreak> tried to reinstall windows' bootloader?
[22:46] <Guest4219> i have not, and there is no windows, not any longer, there is no os at all, yet there it is, grub is there
[22:47] <Guest4219> diskpart, clean
[22:47] <Guest4219> i used this command to clean the disk, its nvme disk on uefi system
[22:47] <cbreak> so, you will have to install windows before you can use it.
[22:48] <cbreak> when you do that, it will write its own efi bootloader, replacing grub
[22:48] <CarlFK> if there is no os, then you don't need a boot loader ;)    im a little fuzzy what your plan is
[22:48] <rbasak> The GPT is just a FAT filesystem partition.
[22:48] <Guest4219> i have a complete disk image in acronis format, it contains windows 10, apps, my files, etc, plus some old ubuntu grub reference
[22:48] <rbasak> It contains boot files from all OSes installed on your system
[22:49] <rbasak> You can "rm -rf /boot/efi" but then your system will be completely unbootable.
[22:49] <rbasak> If your Windows recovery software can recover from that, and you don't want to boot Ubuntu again (without reinstalling it), then sure, you can do that.
[22:51] <rbasak> I have no idea how to recover the Windows parts of the GPT though, which you will also be blowing away. That part's down to you.
[22:51] <rbasak> I should have said ESP above, not GPT.
[22:54] <Guest4219> i suspect my entire disk layout is screwed, since long, i just happen to know that i have had ubuntu (with grub) installed on it once in the past
[22:54] <Guest4219> for several years since i have returned back to windows 10 on that pc and not had issues (until recent "no audio" bla bla and corrupted win registry)
[22:55] <Guest4219> https://i.imgur.com/LTfegCs.png
[22:56] <Guest4219> now with acronis i get this odd error, because acronis is trying to do some magic tricks i did not ask for in the background, like fixing "bootablitiy" of my pc and locating grub and what not, and this is only because i have done a wipe of my disk (clean command in diskart) prior to attempting a recovery
[22:57] <Guest4219> https://imgur.com/E04zc16.png
[22:58] <akik> Guest4219: you might have succeeded while you were booted into windows using bcdedit
[23:00] <akik> Guest4219: i'm not sure if that would work in a F10 command prompt opened from the installer
[23:00] <Guest4219> hm... ok... removing grub you mean, by bcdedit from within windows?
[23:00] <akik> Guest4219: changing the boot order to not start grub
[23:00] <akik> Guest4219: can you go into the uefi menus of the machine to change it?
[23:02] <akik> the system works in a totally different way with bios mode boot and uefi mode boot
[23:02] <Guest4219> boot order ok... i will check on that, maybe it's set to use grub. i remember using a different disk for ubuntu at the time, a samsung SATA SSD and this one is a samsung NVME SSD. i think i saw what may be an old reference to that disk GUID in a log
[23:03] <Guest4219> yeah i'm pretty confident i had this setup: ubuntu on samsung sata ssd, windows on samsung nvme ssd, and grub was doing the bootloader chaining
[23:04] <Guest4219> then i wiped the original samsung ssd, have long since removed it, and only windows was left on the nvme
[23:04] <Guest4219> plus some grub remainer...
[23:05] <Guest4219> now this is preventing me from recovering my windows system back to a prior state using acronis recovery media (linux based, it's what you saw in screenshots)
[23:06] <akik> Guest4219: maybe you can change it in the computer's uefi menus. efibootmgr on linux modifies that data too
[23:07] <akik> Guest4219: and bcdedit on windows
[23:10] <Guest4219> hehe this all seems like magic to me.  i mean i do actually have a boot entry that mentions "ubuntu" when i start my pc and press f8 (asus mobo)
[23:11] <akik> Guest4219: on my acer laptop, when i install linux on it, i have to go into the uefi menu and manually select fedora's grubx64.efi as a bootable file
[23:11] <Guest4219> i can use acronis disk cleanser from within acronis recovery media to do disk zero and even us army code data destruction and what not (whatever) and if i do that then "ubuntu" will disappear from f8 menu (same as with diskpart and clean command)
[23:12] <Guest4219> but if i run acronis recovery media to recover the nvme disk with windows from a backup, then that "ubuntu" entry pops back up on f8 menu of my pc and also acronis complains of having found grub 2 efi or whatever
[23:14] <Guest4219> isn't such data destruction algorithm applied on disk level enough to kill off grub? i don't know...
[23:14] <Guest4219> unless of course grub is hiding in my uefi somewhere... and not on my disk
[23:15] <akik> Guest4219: i don't know what this software can do but maybe try it https://neosmart.net/EasyRE/Features/
[23:15] <Guest4219> i have actually performed this exact operation once before a day ago, but then i did not do a disk wipe before recovery
[23:15] <akik> that was linked off the easybcd site
[23:16] <akik> https://neosmart.net/EasyRE/
[23:16] <akik> damn it's payware
[23:17] <akik> but if you can't find another way to get back to windows, maybe it's worth it
[23:17] <akik> i paid that amount to get the win10 taskbar on my second display :)
[23:20] <Guest4219> hehe maybe it's worth it
[23:21] <Guest4219> may i ask what is linux/ubuntu equivalent to windows registry? is it just config text files?
[23:22] <spjb> pretty much
[23:22] <sarnold> there's also gsettings
[23:22] <Guest4219> i absolutely hate this registry BS i have been banging my head against a wall over that stupid thing, and these "hives" like SYSTEM, SAM, SOFTWARE, bla bla it's the reason i'm in this position now
[23:23] <Guest4219> there's a dent in the wall to prove it
[23:24] <Guest4219> In the Police world there is nothing worse than a corrupted cop; inthe  computer world there is nothing worse than a corrupted registry.
[23:24] <akik> Guest4219: systemd brought ini type files into linux and then there's dconf that is a binary data store for configs
[23:24] <Guest4219> *windows world i should say
[23:24] <Guest4219> but dconf is open spec yeah?
[23:25] <akik> Guest4219: i think so
[23:25] <Guest4219> unlike this "hive" nonsense where you have to go via intermediary REG files
[23:25] <akik> Guest4219: i don't think you should go editing those if you can't make it boot first
[23:26] <akik> bcdedit might work when run in the installer's F10 cmd prompt
[23:26] <akik> or was it shift+F10
[23:27] <Guest4219> hehe and you know what actually set this off to begin with? i got screwed by some "KB500blabla" update from windows update, and now i have "No Audio Output Device" even though device manager disagrees '=D '=D '=D
[23:28] <Guest4219> no i was able to boot into windows, only audio was not working, and i have emptied the whole bag of "fixes" and "recovery" tools including the half-dumb "audio playback troubleshooter" which cannot solve any audio issues of greater complexity than "oh shit i forgot to unmute"
[23:29] <Guest4219> akik shift+f10
[23:29] <Guest4219> hehe it used to be f8 as well, in old days, but that was on xp, and it was called "Recovery Console"
[23:30] <akik> they just hide that option now by default but you can enable it back
[23:30] <Guest4219> now it's "Recovery Environment" or WinRE, which is same as WinPE it's just a interface for it
[23:31] <Guest4219> hehe they also disabled RegBack to "save space" :')
[23:31] <Guest4219> it used to contain all the important 5 registry hives
[23:32] <Guest4219> https://docs.microsoft.com/en-us/troubleshoot/windows-client/deployment/system-registry-no-backed-up-regback-folder
[23:32] <Guest4219> 1 less way to recover windows
[23:32] <Guest4219> so many "recovery" tools for windows over the years... that's telling
[23:34] <Guest4219> worse still is if you're so inclined to try to fix things manually, like linux folks often do, it is absolute nightmare with windows, you can't do anything without automated tools or simple press the button to reset or reinstall fresh
[23:35] <Guest4219> the windows registry is a one way street... maybe if you have system restore points enabled, that actually does work, it fixed my audio issue not once but two times, but windows update subsequently broke it again and again due to kb500something and a GSOD loop cycle
[23:36] <akik> Guest4219: why were the restore points disabled by default in some win10 version?
[23:37] <Guest4219> they are? that's news to me, but the good restore point i used to fix my audio issue has been purged from the system i think when i upgraded to a later version
[23:37] <Guest4219> microsoft needs to do me a favor and save me my 500 gb disk space X-P
[23:38] <Guest4219> one thing they never invented is how to do system image backups, and the backup system they had in windows has been discontinued.
[23:38] <Guest4219> im surprised windows even has anti virus and a firewall
[23:39] <Guest4219> that's usually reserved for the third party software makers
[23:39] <Guest4219> i for example use acronis true image for my backup needs, not because i want to, but because windows doesn't have one built in
[23:40] <akik> no? i think i remember reading of a way to backup the whole os
[23:40] <toddc> Guest4219: cloanzilla free and open source vs acronis
[23:41] <Guest4219> akik: yeah but i think that only copies files and folders, it doesn't do disk imaging
[23:41] <akik> https://www.windowscentral.com/how-make-full-backup-windows-10
[23:41] <akik> "create a system image"
[23:41] <toddc> yes I use ut
[23:43] <Guest4219> ah yes, but it also says "(windows 7)" and "older backup"
[23:43] <oerheks> "dd"
[23:43] <akik> Guest4219: the page says it's used with win10
[23:43] <Guest4219> it's only a matter of time before they pull the plug on Backup and Restore... or decide not to... and then bloat it some more with yet another backup tool...
[23:44] <akik> maybe they have fixed it in win 11 :)
[23:44] <Guest4219> hehe it's like "ms paint": hey look we now have "paint 3D"... oh wait... we will also keep ms paint
[23:45] <toddc> maybe win 11 will fix all their issues
[23:45] <Guest4219> toddc: how do you run clonezilla from within windows?
[23:45] <Guest4219> and on schedule?
[23:45] <akik> someone had made a compilation picture of all different ui elements
[23:46] <toddc> I boot to clonezilla then do what I need I do not normally use it as a backup as there are better solutions
[23:46] <akik> https://old.reddit.com/r/Windows11/comments/o2a0kp/there_are_at_least_10_different_microsoft_design/
[23:48] <Guest4219> good find akik very good! i will pin this :D
[23:49] <Guest4219> toddc: for windows at least that's true image i think
[23:49] <toddc> Guest4219:I backup Data to a server then ignore the os as I can recreate that in minutes
[23:50] <Guest4219> did you guys know that windows is fully capable of creating more than one recovery partition on the main disk, one at the front and one at the back during upgrade process, and then complaining to you during new install setup that you're doing it wrong and need to read their gpt disk layout article at given url?
[23:50] <toddc> all my MS windows are now VM's on zfs so those I can snapshot on a schedule
[23:52] <Guest4219> toddc: sounds like a dream setup
[23:52] <Guest4219> i am still stuck in 1990s
[23:53] <srv> lol
[23:53] <Guest4219> https://dpaste.org/aXSg
[23:54] <Guest4219> does anyone care to have a look at this log and tell me what this grub mambo jambo is about?
[23:55] <Guest4219> the log is from acronis recovery media. what is this "failed to detect grub activator"? what's an activator?
[23:55] <Guest4219> tell me like im 12
[23:55] <sarnold> you might have to ask acronis folks
[23:56] <Guest4219> how can i get a friendly disk name based on those guid numbers?
[23:58] <toddc> Grub or uefi grub is just a boot loader one can use the MS bootloader but it is a bit more limited
[23:59] <toddc> Guest4219: are trying to remove all linux from you system?
[23:59] <toddc> your