| coz_ | boom, crackle, phizzies,//happy 4th! | 00:06 |
|---|---|---|
| DarkTrick | **Help**: I got an Input/output error on one folder of my SD card | 00:26 |
| DarkTrick | It's a folder containing a VM. Interestingly, `ls -l` shows problems only on .vbox file and .vbox-prev file | 00:27 |
| DarkTrick | all other folders seem to be intact | 00:27 |
| niksen | how can I disable ssh path autocomplete? | 00:31 |
| DarkTrick | Workaround: Luckily only a text file was affected and I had a backup. This is a deadly bug ... | 00:38 |
| oerheks | 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. | 01:44 |
| oerheks | DarkTrick, an active vm on a SD card, i would say highly unreliable.. | 01:45 |
| dw1 | in a terminal, if i copy a long line of text inside screen it copies unwanted line breaks. is there a fix? | 02:05 |
| lotuspsychje | question for #linux perhaps dw1 ? | 02:09 |
| oerheks | cat -v .. https://en.wikipedia.org/wiki/Newline | 02:10 |
| dw1 | yeah i asked there thx | 02:11 |
| lotuspsychje | i recently reinstalled firefox on 20.04 and im having 100% cpu spikes on both normal browsing and private windows https://imgur.com/a/UyYrfOe | 02:14 |
| lotuspsychje | but its more on random, cant really reproduce it yet, about:performance shows both addons privacy badger & ublock ontop | 02:15 |
| oerheks | lotuspsychje, is this because of preload? | 02:17 |
| lotuspsychje | never had issues with preload before, it all started happen after a fresh FF reinstall | 02:17 |
| oerheks | 20.6 G data from one process childID3 ... | 02:18 |
| lotuspsychje | and i wiped the FF folders in /home too before reinstall, so im puzzled | 02:18 |
| oerheks | what pages do you have open? | 02:19 |
| lotuspsychje | tested a few different, like gmail, youtube | 02:19 |
| lotuspsychje | it doesnt seems to influence | 02:19 |
| lotuspsychje | once the cpu spikes rise, i have to close FF | 02:20 |
| lotuspsychje | individual new tabs cant fix it | 02:20 |
| lotuspsychje | tryed disabling my 2 fav addons, darkreader and addtoany but no dice neither | 02:21 |
| lotuspsychje | i wonder if its privacy badger and ublock making FF work more by default? | 02:25 |
| lupi | hi guys. my current setup is ubuntu vm, win10 host with samba for file access. I can ping from host -> vm but when I try pinging from vm, it says "temporary failure in name resolution". I tried nmblookup and it works. I can also ping the host's IP. | 02:43 |
| leftyfb | lupi: you need the VM guest setup with a bridged network interface, not NAT as is usually the default. This is done though your VM software running on Windows | 03:23 |
| pagnol | All of a sudden ubuntu won't boot anymore, after the bios splash screen I can choose between "ubuntu" and "windows boot manager", but when I select either, then nothing happens | 03:37 |
| pagnol | And it doesn't look like grub | 03:37 |
| pagnol | I haven't configured dual boot | 03:40 |
| alkisg | pagnol: you can upload a screenshot if you want; you could also boot via a live cd in order to fix it | 03:42 |
| pagnol | alkisg: https://ibb.co/K9TbLTQ | 03:46 |
| pagnol | Currently dont have access to a second computer to prepare a usb stick unfortunately | 03:47 |
| pagnol | Otherwise I would've just done that | 03:47 |
| alkisg | Yeah this looks like an UEFI firmware boot manager (UEFI is the "new BIOS") | 03:48 |
| alkisg | And what happens if you select ubuntu, does it show grub? | 03:49 |
| pagnol | Absolutely nothing happens | 03:49 |
| alkisg | It remains in the blue dialog, or it shows a black screen? | 03:49 |
| pagnol | Whichever I select, nothing happens | 03:49 |
| pagnol | The screen remains unchanged | 03:50 |
| lupi | leftyfb yes I have bridged setup. I can ping the IP, but can't ping the host's hostname. I can ping the vm's hostname from host though. | 03:50 |
| alkisg | pagnol: if you press <tab> there, do you see the "app menu"? | 03:50 |
| leftyfb | lupi: what nameserver(s) is the guest using? | 03:50 |
| pagnol | alkisg: https://ibb.co/kyT1JFb | 03:51 |
| lupi | leftyfb `/etc/resolv.conf` has this "nameserver 127.0.0.53" | 03:52 |
| lupi | nslookup shows 127.0.0.53 too | 03:52 |
| leftyfb | lupi: systemd-resolve --status|grep "DNS Servers" | nc termbin.com 9999 | 03:53 |
| lupi | leftyfb 8.8.8.8 | 03:53 |
| leftyfb | lupi: and there's your problem | 03:53 |
| alkisg | pagnol: and can you enter setup if you press enter at that point? | 03:53 |
| leftyfb | lupi: google's DNS servers do not know how to resolve your guest or hosts ip's | 03:54 |
| pagnol | Yes but its just the regular bios setup | 03:54 |
| leftyfb | lupi: if you want to be able to resolve local hostnames on your network then your clients should be using ONLY your routers ip address as DNS (or local DNS server if you have one) | 03:55 |
| lupi | leftyfb like 192.168.1.1? | 03:55 |
| leftyfb | lupi: yes | 03:55 |
| alkisg | pagnol: inside setup, try to disable secure boot, in case something went wrong with that. But unfortunately you'll most probably need a live cd. | 03:55 |
| leftyfb | lupi: then let your router forward DNS requests out to whatever public DNS server you have configured on the WAN side | 03:55 |
| pagnol | alkisg: I will try that thank you | 03:56 |
| lupi | I tried "nslookup myhostname 192.168.1.1" and it did not work. It said "server can't find myhostname: NXDOMAIN" | 03:56 |
| lupi | Also, I thought it would work since I can ping from the host's. "ping myvm" works | 03:57 |
| pagnol | alkisg: secure boot is actually disabled already | 03:58 |
| leftyfb | lupi: your VM software probably setup it's own DNS lookup for guests locally | 03:58 |
| leftyfb | lupi: if your router isn't resolving local hostnames, that is something you'll have to address with your router. Maybe try #hardware or #networking | 03:58 |
| leftyfb | lupi: what ip address does your guest have? | 03:59 |
| lupi | vm 192.168.1.57, host 192.168.1.81, router is 192.168.1.1 | 03:59 |
| leftyfb | lupi: can you ping the ip of the guest from another machine on the network (not the host)? | 04:00 |
| alkisg | lupi: DNS resolution is for the internet. For local hostnames, you either use zeroconf (ping myname.local) or winbind (ping myname) | 04:01 |
| lupi | leftyfb I can ping the hostname and ip from another machine | 04:01 |
| alkisg | I haven't read the whole thread. The problem is from windows to ubuntu, or the opposite? | 04:02 |
| lupi | alkisg the opposite | 04:02 |
| leftyfb | alkisg: DNS resolution is for DNS resolution, regardless of the class of network | 04:02 |
| alkisg | Do you have the winbind package installed? | 04:02 |
| alkisg | dpkg -l winbind | 04:02 |
| lupi | alkisg winbind is ii | 04:03 |
| alkisg | leftyfb: please read the avahi-browser and winbind package descriptions to see how these protocols help in name resolution in local networks where no authoritative dns server exists | 04:03 |
| leftyfb | alkisg: all home routers have local DNS built in. | 04:04 |
| lupi | alkisg I thought smb and nmb will handle it altogether already | 04:04 |
| leftyfb | lupi: samba is not for DNS | 04:04 |
| lupi | I mean, because the clients will belong to the same workgroup or something. | 04:04 |
| leftyfb | alkisg: also, not everything supports zeroconf and winbind is a poor substitute for DNS | 04:05 |
| leftyfb | lupi: samba/SMB/CIFS are filesharing protocols, not DNS. | 04:06 |
| lupi | also, "wins support" is enabled in smb.conf if that helps | 04:06 |
| alkisg | leftyfb: DNS means A records. Cheap routers don't have an interface for that, and users don't go to that interface to declare their clients anyway. SOME dhcp clients can send the local hostname when doing a DHCP request, and SOME dhcp servers may accept it. | 04:06 |
| leftyfb | :/ | 04:07 |
| alkisg | So in SOME cases the local router will know the hostnames, but this isn't always the case. Avahi and winbind were invented to address the lack of DNS in such LANs | 04:07 |
| alkisg | lupi: can you ping the windows pc, using its hostname, from any other device? | 04:08 |
| alkisg | (checking if the windows connection is "public" or "home" etc) | 04:08 |
| lupi | the machines can ping each other and the vm. the vm cannot ping the hostnames outside but can ping the ip's | 04:09 |
| lupi | current network is public btw. | 04:09 |
| alkisg | The other machine is also windows? | 04:10 |
| lupi | yep | 04:10 |
| lupi | im using virtualbox. network is bridged. | 04:10 |
| leftyfb | lupi: change the DHCP settings on your router so that clints get 192.168.1.1 for their DNS, not 8.8.8.8 | 04:12 |
| alkisg | lupi, do you have wins in your /etc/nsswitch.conf? https://askubuntu.com/questions/27541/why-cant-my-machine-resolve-local-windows-hostnames | 04:12 |
| leftyfb | lupi: i the meantime, you can try setting the DNS of your guest to 192.168.1.1 temporarily to test | 04:12 |
| alkisg | You can test if your router knows the windows pcs by: host windows-hostname 192.168.1.1 | 04:13 |
| lupi | alkisg oh crap. you mean do it from the router? I still have this shitty router. but I'll try what leftyfb said | 04:16 |
| leftyfb | lupi: you can do this on the guest manually to test | 04:16 |
| alkisg | lupi: no, I mean if you run the `host` command from linux, then it will ask your router if it knows the windows hostnames | 04:16 |
| lupi | oh ok wait | 04:16 |
| alkisg | So that you'll know that IF you change your dns, if it'll work or not | 04:16 |
| leftyfb | lupi: a quick way is to change /etc/resolv.conf to "nameserver 192.168.1.1" | 04:16 |
| alkisg | While e.g. if your windows PCs have static IPs, that certainly won't work, as they don't do DHCP requests and they don't send their hostname at all | 04:17 |
| leftyfb | lupi: (temporary) | 04:17 |
| alkisg | (or if your router dhcp server doesn't support hostname registration) | 04:17 |
| lupi | welp. host not found NXDOMAIN | 04:17 |
| lupi | no machine uses static ips | 04:17 |
| leftyfb | lupi: what model router? | 04:18 |
| alkisg | OK, it appears your router doesn't support hostname registrations so its DNS doesn't know the windows PCs, you need to use wins. Do you have wins in nsswitch.conf? | 04:18 |
| lupi | It's ISP owned. I think it's huawei "EchoLife HG8145V" leftyfb | 04:19 |
| lupi | alkisg I dont have it. I think, I tried adding winbind already but it did not work (I restarted) | 04:19 |
| lupi | alkisg I'll try "wins". I tried "winbind" before | 04:20 |
| alkisg | lupi: also try these: ping hostname-lowercase; ping HOSTNAME-UPPERCASE; ping hostname-lowercase.local; ping HOSTNAME-UPPERCASE.local | 04:20 |
| alkisg | I'm not sure if recent windows registers itself in zeroconf apart from wins, and if it does, if it's resolved using .local like usually, or not | 04:21 |
| lupi | alkisg I added "wins" on host, now it's giving me "System error" when I ping | 04:21 |
| alkisg | Let me boot a windows vm to test, it'll be quicker... | 04:22 |
| lupi | ok. thank you | 04:22 |
| alkisg | lupi: in the mean time please remove wins and try the pings I mentioned above | 04:23 |
| alkisg | or search for the windows hostnames in the output of: avahi-browse -trap | 04:23 |
| leftyfb | lupi: https://i.imgur.com/i7SNsUO.png | 04:25 |
| lupi | leftyfb oh yeah I was about to try that | 04:27 |
| alkisg | [ +3,037578] vboxnetflt: 56862 out of 123254 packets were not sent (directed to host) | 04:32 |
| alkisg | Ouch, zeroconf packets are multicasted and I think vbox can't handle them on the bridged interface | 04:33 |
| alkisg | Yeah, https://serverfault.com/questions/563599/avahi-zeroconf-bonjour-service-in-bridged-virtualbox-sometimes-not-found-outsi | 04:34 |
| === nyuszika7h_ is now known as nyuszika7h | ||
| === LadyCailin[derp] is now known as LadyCailin | ||
| Guest9 | hi | 05:07 |
| === mybalzit1h is now known as mybalzitch | ||
| Guest9 | I've installed 'Minetest' via Snap to Lubuntu | 05:08 |
| Guest9 | But it never launches when ran | 05:08 |
| === noteness is now known as Guest6531 | ||
| guiverc | what release of Lubuntu Guest9, do you get any messages when run from terminal? | 05:12 |
| Guest9 | No messages | 05:13 |
| Guest9 | The game window opens for 1 second then closes | 05:13 |
| Guest9 | MineTest worked on older Trisquel 9 | 05:13 |
| Guest9 | But isnt working on this Lubuntu install | 05:13 |
| guiverc | what release? | 05:14 |
| Guest9 | Lubuntu 20.04 LTS | 05:15 |
| guiverc | okay I get loads of messages, then Aborted (core dumped) | 05:16 |
| oerheks | here it works.. | 05:17 |
| guiverc | complaints on my system are expected directories not found.. | 05:17 |
| oerheks | https://imgur.com/gallery/ndAkS94 | 05:18 |
| oerheks | i find no obscure setting in software > installed > minetest > permissions | 05:19 |
| === Daniel is now known as toolz | ||
| alkisg | lupi: I managed to find a real windows pc and test with it; `ping windows-hostname.local` (with .local) works fine on bare metal, yet not on the bridged virtualbox interface, due to the zeroconf multicasted packets getting dropped | 05:40 |
| lupi | alkisg what's your network if not bridged? | 05:41 |
| alkisg | lupi: my use case was the opposite, real ubuntu and vm windows | 05:41 |
| alkisg | So in my case, real ubuntu with real windows works; in your case, vm ubuntu with vm windows would work; yet vm <=> real doesn't work in any case | 05:42 |
| alkisg | That's with the default name resolution, which is zeroconf. Note that if your router supports dhcp name registrations, then you should be able to do what leftyfb said, | 05:43 |
| alkisg | although you reported that `host name 192.168.1.1` didn't work, so it either doesn't support dhcp registrations at all, or it's not enabled | 05:43 |
| lupi | ooh ok thank you! maybe I should leave it for now, the important thing is I can access the vm from the host | 05:44 |
| alkisg | There might be a workaround to make virtualbox NOT drop the multicasted traffic, but it would require asking in #vbox... if it was not problematic it would be enabled by default, so I'm not sure it's worth exploring... | 05:46 |
| === OkayPJ1 is now known as OkayPJ | ||
| zxcpoi | hello, anybody knows why even with max con 20.000 my sockets number on haproxy stops at 4100 on Ubuntu 18.04 ? I tried ulimit I tried maxcon in haproxy but still sometimes it stops at exactly ~4096-4100 and sometimes it doesn't go over 6500 nr sockets opened by haproxy (and my average is around 8000 sockets at one time) | 06:34 |
| fling | Hello. | 07:04 |
| fling | What is the proper way of creating a new user when zsys is in place | 07:05 |
| fling | Should I create a zfs filesystem first? | 07:05 |
| fling | And set an user property? Then create an user? | 07:05 |
| === prxq_ is now known as prxq | ||
| kpcyrd | hello, I'm trying to debug a segfault and it seems ubuntu uses apport instead of systemd-coredump. According to the docs coredumps are recorded in /var/crash, but that folder is empty for me? | 07:27 |
| loganlee | hi guys | 07:43 |
| loganlee | for me ubuntu = linux | 07:43 |
| loganlee | i don't look elsewhere | 07:43 |
| === coke_ is now known as coke | ||
| Mekaneck | loganlee: good on you but a bit offtopic | 07:47 |
| gordonjcp | 0003 | 07:50 |
| gordonjcp | 23 | 07:50 |
| gordonjcp | 3 | 07:50 |
| gordonjcp | #nnn-.6~6~6~6~6~6~6~6~6~6~6~6~6~6~6~6~6~000 | 07:51 |
| gordonjcp | éeeeeeeeeeee3.00006ééééééééééééééé | 07:51 |
| * gordonjcp removes gordonjcpv2.0 from laptop | 07:51 | |
| Guest59 | where are Xorgs default cursors? | 08:19 |
| Guest59 | The default Xorg cursor | 08:19 |
| gordonjcp | Guest59: the outlined X thing? | 08:21 |
| === bsmith093_ is now known as bsmith0931 | ||
| Guest59 | so the default X11 cursor | 08:23 |
| Guest59 | no, the...* | 08:23 |
| theh | I upgraded an LXC container from 18.04 to 20.04 and now the keyboard input is misinterpreted, e.g. backspace enters whitespace characters. Probably some terminal configuration issue. Anyone got experience with this? | 08:27 |
| cbreak | theh: check echo $TERM | 08:29 |
| theh | cbreak: Reports as `xterm-kitty`; I'm pretty sure it did so too before the upgrade. | 08:30 |
| cbreak | so, install that in the termcap inside | 08:30 |
| cbreak | apt install kitty-terminfo | 08:31 |
| cbreak | or something like that | 08:31 |
| theh | Ah OK! Thank you, I'll try that. | 08:31 |
| cbreak | you can also temporarily export TERM=xterm-color | 08:32 |
| theh | Worked well, I didn't have the universe repo enabled in the container, took me a moment to figure out. | 08:55 |
| === sharperguy_ is now known as sharperguy | ||
| AndroUser | DNS resolution suddenly failed on my desktop (20.04). What I've done so far: | 09:46 |
| AndroUser | $ ping -c 2 8.8.8.8 | 09:46 |
| AndroUser | ... 0% packet loss... | 09:46 |
| AndroUser | $ resolvectl query google.com | 09:46 |
| AndroUser | google.com: resolve call failed: All attempts to contact name servers or networks failed | 09:46 |
| AndroUser | $ systemd-resolve --status | grep Current | 09:46 |
| AndroUser | Current DNS Server: ::1 | 09:46 |
| AndroUser | $ ping -c 2 ::1 | 09:46 |
| AndroUser | ... 0% packet loss... | 09:47 |
| AndroUser | $ cat /etc/resolv.conf | 09:47 |
| AndroUser | nameserver 127.0.0.53 | 09:47 |
| AndroUser | options edns0 trust-ad | 09:47 |
| AndroUser | Changing the DNS server for the wifi connection to, for example, 8.8.8.8, has no effect, and the same problem applies to all wifi connections, even new ones. | 09:47 |
| AndroUser | Would be very grateful for any pointers. | 09:47 |
| jushur | AndroUser: nameserver 127.0.0.53 ???? should be 127.0.0.1:53 right? | 10:03 |
| jushur | aka 127.0.0.1 only in resolv conf | 10:03 |
| oerheks | no, i have the same on my bionic 20.04 | 10:04 |
| AndroUser | jushur: possibly a typo. Copying everything to my phone to do this IRC! | 10:04 |
| AndroUser | Aah | 10:04 |
| oerheks | systemd-resolve --status|grep "DNS Servers" gives your router dns | 10:04 |
| AndroUser | Cool. One moment. Was just rebooting. | 10:05 |
| alkisg | 127.0.0.53 is systemd-resolved, it's not a typo there | 10:05 |
| jushur | k, ignore my ask then AndroUser , seems to be the Ubuntu config is such. | 10:07 |
| oerheks | so, change DNS in your router? | 10:08 |
| AndroUser | Yeah, so my router DNS is apparently "::1". Could this be right?! | 10:09 |
| AndroUser | The thing is, other devices using the same router (including the one I am using for this chat) are fine. | 10:11 |
| oerheks | that::1 looks like a IPv6 dns | 10:12 |
| AndroUser | Good point. | 10:12 |
| alkisg | AndroUser: what's the output of: systemd-resolve --status | grep -A3 'DNS Server' | 10:13 |
| oerheks | 0% packet loss .. so there is nothing failing | 10:13 |
| alkisg | And of this, but do set the correct ethernet name: /usr/lib/klibc/bin/ipconfig -n enp2s0 | 10:13 |
| alkisg | (the second one needs sudo but does no changes) | 10:13 |
| AndroUser | alkisg: Current DNS Server: ::1 | 10:14 |
| alkisg | Just "current", there are no other lines? | 10:14 |
| alkisg | No ipv4? | 10:14 |
| oerheks | not the behaviour of an ubuntu box... | 10:15 |
| AndroUser | DNS Servers: ::1 | 10:15 |
| AndroUser | DNS Domain: ~ | 10:15 |
| alkisg | Let's start with the output of ipconfig first. You can also use termbin if you want: sudo /usr/lib/klibc/bin/ipconfig -n enp2s0 | nc 5.39.93.71 9999 | 10:16 |
| alkisg | Do replace enp2s0 with your ethernet name | 10:16 |
| alkisg | This will show us what your router configuration is, if it sends the correct DNS to the DHCP clients | 10:16 |
| alkisg | (that command upload the output to termbin.com/some-url) | 10:16 |
| alkisg | So you can just tell us that url | 10:16 |
| AndroUser | Been trying things l | 10:17 |
| AndroUser | ... ike that, but hobbled by having to use my phone. One minute | 10:17 |
| alkisg | Also upload the output of `nmcli | nc 5.39.93.71 9999`. If you want to hide things like IPs (which shouldn't matter since they're locall), you can use `command > /tmp/out.txt`, then edit out.txt, then use `cat /tmp/out.txt | nc 5.39.93.71 9999` | 10:20 |
| alkisg | Finally you can just bypass the problem by putting `nameserver 8.8.8.8` in /etc/resolv.conf, but then we won't know what caused it :) | 10:21 |
| AndroUser | Thanks. LOL, having to use termbin by direct IP address! Anyway: https://termbin.com/zy9dc | 10:25 |
| AndroUser | Have I somehow lost IPv4? | 10:25 |
| AndroUser | Doing other command in a moment... | 10:25 |
| AndroUser | https://termbin.com/t0o9 | 10:27 |
| oerheks | so there is a lot more going on, docker, vpn on IPv6... | 10:29 |
| alkisg | If "ipv6leakintrf0" means "protonvpn", then that VPN software might be affecting a lot of things; you should start with that | 10:30 |
| jailbreak | AndroUser, what happens if you disable the vpn? | 10:30 |
| AndroUser | Yes, I was wondering if there might be something there, as the problem occurred around the time of first using Proton. VPN isn't supposed to be running at the moment though. | 10:32 |
| AndroUser | Can't even start Proton VPN to look at it as I guess it is stalling on DNS lookups for the various VPN servers. | 10:34 |
| alkisg | sudo nano /etc/resolv.conf; put `nameserver 8.8.8.8` there as a temporary workaround; then fix the protonvpn issues | 10:36 |
| alkisg | afk, back later | 10:36 |
| AndroUser | Thanks so much! | 10:38 |
| AndroUser | alkisg: seems the DNS problem was that a crash had left some of the ProtonVPN stuff behind even when not running ProtonVPN. Restarting it and shutting down clearly seems to have done it. Thanks again for your help. | 11:28 |
| AndroUser | *cleanly* | 11:29 |
| alkisg | 👍 | 11:29 |
| === cloux_ is now known as cloux | ||
| === nox_ is now known as n0cto | ||
| === n0cto is now known as nox_ | ||
| ice9 | how do I force using wayland? i'm using gnome | 11:48 |
| jailbreak | ice9, what ubuntu release? | 11:48 |
| ice9 | jailbreak, 21.04, i know it should be using wayland by default but this not the case here, session show x11 | 11:49 |
| jailbreak | you click the "wayland" on the DM then login. | 11:49 |
| ice9 | jailbreak, there is no such option in gdm | 11:50 |
| jailbreak | ice9, you should have a choice between x11 and wayland. | 11:50 |
| ice9 | jailbreak, i don't have it | 11:51 |
| alkisg | What's the output of this command? ls /usr/share/xsessions | 11:51 |
| alkisg | Hmm no I think there's also a wayland sessions somewhere else, looking... | 11:52 |
| ice9 | alkisg, ubuntu.desktop ubuntu-xorg.desktop | 11:53 |
| jailbreak | ubuntu.desktop is wayland. | 11:53 |
| ice9 | jailbreak, the only diff between the 2 files is: "Name=Ubuntu |Name=Ubuntu on Xorg" | 11:54 |
| alkisg | There's also /usr/share/wayland-sessions/ubuntu-wayland.desktop | 11:54 |
| alkisg | I guess one is for some newer display manager specification, the other for older DMs; yet both would do the same | 11:55 |
| ice9 | alkisg, still that file has the same content too | 11:56 |
| alkisg | On 20.04 I only have one file under xsessions. Indeed the contents are the same so I guess the difference is in the containing directory | 11:57 |
| IRCMonkey | Will a Windowz Edge uBlock Origins backup work on Linux Firefox uBlock extension ? | 11:58 |
| mijora1 | anyone here had much luck with ubuntu 20.04 LTS as a VPS on OVH servers, ontop of proxmox using their weird failover ips? | 11:58 |
| alkisg | ice9: do you have opengl acceleration? Check glxinfo | 11:58 |
| ice9 | alkisg, yes i do, i'm using nvidia as primary gpu | 11:58 |
| alkisg | ice9: output of: grep -ri wayland /etc/gdm* | 12:00 |
| ice9 | alkisg, /etc/gdm3/custom.conf:#WaylandEnable=false | 12:00 |
| alkisg | ice9: and when you select the user, you don't see a gear at the bottom-right corner, that allows you to select wayland? | 12:02 |
| alkisg | It only shows up after you select the username | 12:02 |
| * alkisg just tested on ubuntu-20.04-desktop.iso and verified that... | 12:03 | |
| ice9 | alkisg, i don't have that gear | 12:04 |
| alkisg | Ah, maybe this is still true: "Wayland is disabled by gnome devs on nvidia proprietery drivers" ... | 12:05 |
| alkisg | https://askubuntu.com/questions/1215440/cant-find-wayland-sessions-on-gdm | 12:05 |
| alkisg | I've been avoiding nvidia for years so I wouldn't know :) | 12:05 |
| alkisg | It does mention a couple of workarounds there | 12:05 |
| Milo | gah, why is the silly lenovo laptop failing to detect wired network device when I boot the machine while it's docked. Always have to unplug and replug the dock to get the device working. | 12:18 |
| ice9 | alkisg, thanks a lot for your help :) | 12:27 |
| mijora1 | is there anyway to not use netplan for network config on 20.04 and just configure the interfaces file like before? | 12:33 |
| IRCMonkey | Anybody ? | 12:34 |
| Mekaneck | that's not a ubuntu support question imho IRCMonkey | 12:38 |
| ogra` | mijora1, well, ifupdown is gone from 20.04 ... netplan acts as frontend to systemd-networkd and to the network-manager backend ... you could try to configure systemd-networkd directly instead of using netplan but i dont think you can easily re-integrate ifupdown | 12:39 |
| goddard | Why do ubuntu devs put packages that pretty much require a GPU without GPU support? | 12:39 |
| goddard | python3-torch for example | 12:39 |
| goddard | they compiled it without CUDA support | 12:39 |
| ogra` | ask nvidia ... | 12:40 |
| mijora1 | orga: i have setupa win vm and a debian 10 vm no problem on ovh, now im trying a ubuntu server and just cannot get netplan to load the right config. is there any other way? | 12:40 |
| ogra` | there is no legal way of sipping CUDA in the distro ... and you can not link against something that is not there | 12:40 |
| goddard | many distros have begun shipping their images with nvidia drivers | 12:41 |
| goddard | right? | 12:41 |
| ogra` | nt with CUDA | 12:41 |
| ogra` | nvidia does not permit that | 12:41 |
| goddard | Can't we just have a script that auto downloads it? | 12:41 |
| ioria | mijora1, the problem with server (for a manual config) is cloud-init; you know to handle it ? | 12:41 |
| goddard | like restricted extras? | 12:41 |
| ogra` | not at buld time of python-torch | 12:41 |
| Mekaneck | IRCMonkey: try asking in #linux | 12:41 |
| goddard | i see well that super sucks | 12:42 |
| ogra` | it does | 12:42 |
| goddard | what are they worried about? Seems like an odd requirement. | 12:42 |
| mijora1 | ioria no im not aware of cloud-init | 12:42 |
| ioria | mijora1, may i ask you why you installed th server edition ? | 12:43 |
| ioria | *the | 12:43 |
| mijora1 | ioria ubuntu-20.04.2-live-server-amd64 | 12:44 |
| ioria | mijora1, yes i know; i asked you why ? | 12:45 |
| mijora1 | sorry didnt see that, im trying to setup a ubuntu VM on a ovh server - not a desktop machine, should i not be using that edition? | 12:45 |
| ioria | mijora1, ok; so yes, you can use /e/n/i on 20.04 serv.; but you need to install ifupdown, resolvconf, reconfigure systedm-resolved,backup the netplan config files, probably disable predictable net intefaces names and disable cloud-init if you want a static interface | 12:50 |
| ioria | mijora1, a bit of work | 12:51 |
| mijora1 | ouch. would be easier if i could connect to the internet to install | 12:52 |
| BluesKaj | Hi all | 12:53 |
| mijora1 | never had this issue before, what a pain in the a55 | 12:53 |
| IRCMonkey | Mekaneck » Oh O.K. | 13:05 |
| webchat66 | i can't scale my tv with my laptop.. when i scale them i can't have different scales... i'm stuck at 200% cause my laptop is a small 4k laptop and i want it to be at 4k resolution but 200% and not my 4k tv.. i prefer my 4k tv to be at it's 100%.. or am i missing something? i tried xrandr too.. the scaling of the laptop works seperately but when i | 13:38 |
| webchat66 | do it for the HDMI tv it doesn't work.. like when i scale 2x2 or 0.5x0.5 both zooms in and things like that.. any support..? | 13:38 |
| webchat66 | i tried display settings in ubuntu 21.04 for the first case scenario.. then xrandr for second mentioned case.. | 13:38 |
| * alkisg is using ifupdown in ubuntu-server 20.04 without resolvconf and without disabling predictable interface names | 13:39 | |
| webchat66 | anyone? | 13:43 |
| tha | Hi. I'm trying to find out with what configure options package "xterm" was built on Ubuntu 21.04 - I've uncommented deb-src repos in /etc/apt/sources.list, did 'apt update' and tried to get the sources with 'apt-get source xterm' with error message "E: Unable to find a source package for xterm" | 13:47 |
| tha | I'm confused here and would be grateful for help. | 13:48 |
| === leftyfb_ is now known as leftyfb | ||
| leftyfb | tha: it's part of the universe repository. You need to enable that and it's deb-src | 13:50 |
| tha | leftyfb: thank you kindly. That was it. | 13:53 |
| tha | Is there any command that could help me find which package is in which deb-src repo in the future? | 13:55 |
| rbasak | tha: it's probably easiest to inspect the build logs oh Launchpad, unless you also need to reproduce the build yourself. Start from https://launchpad.net/ubuntu/+source/xterm, click on 361-1ubuntu3 to get to https://launchpad.net/ubuntu/+source/xterm/361-1ubuntu3, then on amd64, then on buildlog. | 13:57 |
| rbasak | That takes you to https://launchpadlibrarian.net/524512634/buildlog_ubuntu-hirsute-amd64.xterm_361-1ubuntu3_BUILDING.txt.gz, and you can search that for "configure" to see exactly how the configure script was called. | 13:57 |
| tha | rbasak: that's even better than fetching sources, thanks | 14:00 |
| tha | when I comment out deb-src entries in the sources.list file and do 'apt update', will it remove deb-src repo's metadata, or should it be done manually? | 14:03 |
| eFfeM | Hi, since a week or so my system runs fsck on every boot (20.04). Anyone an idea what could cause this, and how to fix it. I use the regular shutdown | 14:47 |
| eFfeM | There seems to be an error message when shutting down about not being able to umount but it is gone before I can read it, and I couldn't find it in the log | 14:47 |
| ioria | eFfeM, check your /etc/fstab | 14:50 |
| eFfeM | ioria: nothing special in fstab | 14:53 |
| ioria | eFfeM, inspect your journal ; e.g. journalctl -b -1| grep mount | 14:55 |
| eFfeM | ioria: that's a good one, didn't think of that, odd results: only from jun 28 although the system was used yesterday | 14:57 |
| ioria | eFfeM, no wait ... -1 mean last boot | 14:58 |
| ioria | eFfeM, if you want today boot , use -0 | 14:58 |
| eFfeM | ah ok | 14:58 |
| eFfeM | ioria: that also looks sound to me: https://paste.ubuntu.com/p/VJVf4hS6wB/ | 15:00 |
| ioria | eFfeM, probably you have / had something connected 8usb ? ) and mounted in /mdia ? | 15:01 |
| ioria | */media | 15:01 |
| eFfeM | ioria: no usb stuff mounted, I do have a 2nd internal HD that mounts on /media, but it is there already for quite a long time | 15:02 |
| oerheks | so those /media/fmeulenbroeks/ * partitions are automounted? or in fstab? | 15:02 |
| eFfeM | ioria: log also says: jul 04 12:53:29 m4 systemd[1]: Reached target Unmount All Filesystems. | 15:03 |
| oerheks | if so in fstab, use /mnt/ for that, and leave .media for automount hdd/sd cards and such | 15:03 |
| ioria | eFfeM, mountpoint /media/fmeulenbroeks/MediaDisk6 is invalid, cannot recover the canonical path | 15:03 |
| ioria | eFfeM, /media is for removable devices (generally speaking) | 15:04 |
| eFfeM | ioria: good catch, this is the internal 2nd harddisk, I figured it would be good since after that I got: | 15:04 |
| eFfeM | jul 04 12:53:34 m4 systemd[1]: Finished Clean the /media/fmeulenbroeks/MediaDisk6 mount point. | 15:04 |
| oerheks | wortf a check if those partitions are error free, fsck | 15:05 |
| eFfeM | The system originally came with a HD, but I upgraded it with an ssd and kept the original HD to store my photo collection | 15:05 |
| ioria | eFfeM, do you have some kernel parameter set ? cat /proc/cmdline | 15:05 |
| eFfeM | ioria: BOOT_IMAGE=/boot/vmlinuz-5.4.0-77-generic root=UUID=dc9bf88d-6cbb-4343-ba02-12853c2aee15 ro quiet splash | 15:06 |
| ioria | eFfeM, mount | pastebinit | 15:06 |
| eFfeM | oerheks: I ran an fsck on MediaDisk6, no problems found | 15:06 |
| eFfeM | ioria: https://paste.ubuntu.com/p/Y922w8gHVK/ | 15:07 |
| eFfeM | note that I umounted /dev/sdc8 (that is the disk that was mounted on /media/fmeulenbroeks/MediaDisk6 | 15:07 |
| eFfeM | actually this is a 3 disk system, /dev/sda is an ssd with windows on it, /dev/sdb is an ssd with linux on it and /dev/sdc is the media harddisk | 15:08 |
| eFfeM | and there are still the old linux partitions on /dev/sdc, kept them when I migrated to ssd as a fallback scenarion and never removed them | 15:09 |
| eFfeM | (actually there is even a windows partition on that HD | 15:10 |
| VIA | elo | 15:11 |
| VIA | question, how to i set the default playback(output) device in ubuntu 20.04? | 15:11 |
| VIA | i come backl and this piece of crap always puts my wireless speaker but i want HDMI as default | 15:11 |
| VIA | i dont find anything in the settings / soundpanel, but i do remember i have set it somewhere before | 15:11 |
| oerheks | sdb3 and sdc6 are faulty,... | 15:12 |
| eFfeM | oerheks: I didn't read that from the logs, I'm now fsck-ing /dev/sdc6 | 15:13 |
| eFfeM | I would have expected the fsck after a reboot to have fixed things | 15:14 |
| oerheks | eFfeM, and are those partitions automounted, or noted in your Fstab? | 15:14 |
| ioria | eFfeM, how do you mount those partitions ? manually ? | 15:14 |
| eFfeM | no automount | 15:14 |
| eFfeM | this is my fstab | 15:14 |
| eFfeM | https://paste.ubuntu.com/p/3392fyW2tZ/ | 15:14 |
| ioria | eFfeM, meaning exactly ? | 15:15 |
| eFfeM | fsck on /dev/sdc6 reveals no problems | 15:15 |
| ioria | eFfeM, what you mean exactly with 'automount' ? | 15:16 |
| eFfeM | ioria: / /home and /home.old are mounted via fstab, Mediadisk6: I have no idea | 15:16 |
| ioria | you have no idea ? | 15:16 |
| eFfeM | I don't issue manual mount commands for that disk | 15:16 |
| eFfeM | oerheks: sdb3 is my /home, I cannot check it for errors now | 15:18 |
| eFfeM | oerheks: what made you come to the conclusion that those partitions are bad (I want to learn :-) ) | 15:19 |
| ioria | eFfeM, if the drive is NOT removable, i suggest to configure fstab (Disks will do it for you) and choose another mout point (not (media i mean) | 15:19 |
| ioria | afk | 15:19 |
| oerheks | from your log https://paste.ubuntu.com/p/VJVf4hS6wB/ line #8 #9 | 15:23 |
| eFfeM | ioria: thanks | 15:24 |
| oerheks | there is one or more partitions faulty, not sure why that happens. | 15:24 |
| eFfeM | oerheks: hmm, I figured that line means that it will mount ro if an error occurs, but those partitions are not mounted ro | 15:25 |
| eFfeM | thinking of it, I had a PC issue a week ago and I had to reset the bios, but afaik I properly restored all settings (mostly defaults anyway) | 15:26 |
| eFfeM | I've seen such behavior before when the system actually unpowered itself before the disk was able to physically write/flush all data | 15:26 |
| eFfeM | anyway, I'll try to umount all partitions that I can manually and see if that changes anything | 15:27 |
| eFfeM | back in a bit | 15:27 |
| oerheks | eFfeM, do this from a live iso | 15:27 |
| eFfeM | yeah, need to download one, will do so during dinner | 15:28 |
| eFfeM | oerheks: unfortunately umounting all but / and /home then rebooting did not help. | 15:31 |
| eFfeM | I see a message flashing by about umounting failing (device not ready) or something like that but the message is gone too fast | 15:31 |
| eFfeM | this is on shutdown, I was wondering if I could find that message somewhere in the log | 15:32 |
| eFfeM | ok, afk for a while, thanks all for your help! | 15:37 |
| === TiZ1 is now known as TiZ | ||
| sem | Yo!!! | 16:22 |
| sem | If I'm formating a 320 GB external hdd for use with Windows and Ubuntu (and friends' computers), is the most portable option still NTFS, or is there something better? Can I use exFat reliably with Ubuntu | 16:22 |
| VIA | hello | 16:23 |
| VIA | how can i change my root password? | 16:23 |
| VIA | i read "do passwd root" 2 sources and, its not working | 16:23 |
| VIA | sudo** | 16:23 |
| jailbreak | VIA: sudo passwd root | 16:23 |
| VIA | ohooho mebe i need to switch to root "acc" first let me try | 16:24 |
| VIA | yes jailbreak thats exactly what is NOT working | 16:24 |
| john122 | not sure when my connection cut off, I just noticed (really need to fix my ZNC server) but did anyone reply to my quesiton about VNC earlier? thanks | 16:24 |
| oerheks | VIA there is no root password. | 16:25 |
| oerheks | VIA, change the password of the 1st account created, done. | 16:25 |
| VIA | its not working | 16:25 |
| VIA | doesnt feel right ._. | 16:25 |
| oerheks | sem, one can use exfat, sure, install exfat-fuse and exfat-utils | 16:25 |
| VIA | its says successfulyl changed each time and its not working | 16:25 |
| oerheks | VIA, yes, it is not working. | 16:26 |
| oerheks | there is no root password. | 16:26 |
| oerheks | !root | 16:26 |
| ubottu | Do not try to guess the root password, that is impossible. Instead, realise the truth... there is no root password. Then you will see that it is 'sudo' that grants you access and not the root password. Look at https://help.ubuntu.com/community/RootSudo | 16:26 |
| oerheks | again, change the password of the 1st account created, done. | 16:26 |
| VIA | so my 1st acc is linked to root? | 16:26 |
| oerheks | yes | 16:26 |
| VIA | hmhm makes sense | 16:26 |
| oerheks | that is the cleverness of ubuntu. | 16:26 |
| VIA | just the way the system reacts doesnt make sense | 16:26 |
| VIA | output ect, all nonsensical crap in this case?? | 16:26 |
| VIA | its not very clever | 16:27 |
| VIA | not at all | 16:27 |
| oerheks | the command is wrong, it works on debian though. | 16:27 |
| VIA | i logged in a bunch of time, and start writing my password before the monitor is fully turned on | 16:27 |
| sem | oerheks, sounds like a relatively low-impact solution | 16:27 |
| oerheks | this is why ubuntu is a good step up to linux, no hassle changing accounts to do admin tasks | 16:27 |
| VIA | and fcking ubuntu actually dumps my partial password into the "unlocked os" | 16:27 |
| sem | I know it's a lot harder to get windows to play nice with ext | 16:27 |
| VIA | eg any applicatio nin focus | 16:27 |
| jailbreak | !language | VIA | 16:28 |
| ubottu | VIA: 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 | 16:28 |
| VIA | before im catapultzed to an empty lockscreen. how smart is that? | 16:28 |
| VIA | today it dumped my full root password into IRC | 16:28 |
| VIA | howe clever is that? | 16:28 |
| jailbreak | That's on you. | 16:28 |
| VIA | how so? my system was locked | 16:28 |
| VIA | here we go again with knobs lookign for layer 8 problems. apologies i have offended your godlike OS | 16:29 |
| VIA | i try to not speak in your presence | 16:29 |
| oerheks | i am sure such thing does not happen, when locked. | 16:29 |
| VIA | sorry so sorry | 16:29 |
| oerheks | sem, https://itsfoss.com/format-exfat-linux/ | 16:29 |
| sem | thanks | 16:31 |
| VIA | it has happened over a dozen times now oerheks. with the system locked. it has never dumped the full password yet, i guess i havent typed as fast yet .. usually dumped 3-5letters | 16:31 |
| VIA | before, i guess, the system realizes "oh hehe sorry i should be locked my bad lets .. try ... " | 16:32 |
| VIA | alarming to me. but i am certain everyone using ubuntu for en eternity, can assue me, this is certainly impossible. with the system locked. | 16:33 |
| sem | I don't think anyone knows what you mean by "system locked" | 16:33 |
| ice9 | how to resolve this? 2 directory index of "/var/www/html/kanboard/" is forbidden, client: 127.0.0.1, server: _, request: "GET /kanboard/ HTTP/1.1", host: "localhost" ; even 777 didn't work | 16:41 |
| eFfeM | ioria: oerheks: it seems I semi-fixed the issue, by using halt I was able to notice that it failed to umount /oldroot. | 16:41 |
| eFfeM | Google then gave me this: https://askubuntu.com/questions/1230403/failed-to-unmount-oldroot-dev-warning-during-the-shutdown-or-restart | 16:41 |
| eFfeM | As I had a bios issue recently I dug into that and set UEFI as first boot option | 16:41 |
| oerheks | eFfeM, oh, good find | 16:42 |
| eFfeM | when shutting down I still get the "cannot umount /oldroot" message but when starting up there is no fsck any more | 16:42 |
| eFfeM | still not fully a happy camper as I still see the cannot umount /oldroot messages, but for now out of idea | 16:45 |
| alkisg | ice9: check the permissions of the parent folders like the "html" folder | 16:49 |
| bigpod | VIA: sorry but im not sure that can happen at all unless you have something incorrectly configured | 16:50 |
| TJ- | can anyone recall how to reset a USB port via sysfs? I'm sure I've done it before, similar to how we an remove/rescan PCI devices | 16:50 |
| alkisg | TJ-: I think it was either bind/unbind, or "authorized", under there: find /sys/bus/usb -name '*bind*' | 16:52 |
| alkisg | Hmm matrix renders the asterisks as italics, hope they show up properly on irc | 16:52 |
| bigpod | VIA: i have never seen that happen i genraly type my more then half of my password before my screen comes on and i never get any letters in the clear in app working in the back when system unlocks. Of course important part is what desktop environment are you using and what lock screen and suhc | 16:52 |
| bigpod | VIA: that has never happened to me | 16:52 |
| bigpod | a | 16:52 |
| bigpod | sdfgkj | 16:52 |
| TJ- | alkisg: yeah, authorized disables a device ... that /may/ work but in this case it has disappeared from its bus so I need to rescan :( | 16:53 |
| TJ- | for bind the Vendor:Device ID has to be written to it | 16:54 |
| oerheks | udevadm trigger | 16:54 |
| TJ- | oerheks: hmm, yeah | 16:54 |
| oerheks | = rescan? | 16:54 |
| TJ- | let's see ... if this saves me yet another reboot I'll be happy | 16:54 |
| TJ- | nope... looks like it has to be powered down to fix it. Seems to be related to the 5.13 kernel | 16:55 |
| VIA | ty bigpod for replying something useful | 16:55 |
| VIA | <3 | 16:56 |
| VIA | i have tried to configure/mess with as little as possible. generally dont touch options unless im conviced i can revert them/know exactly what im doing | 16:56 |
| VIA | the enviroment is XFCE | 16:56 |
| VIA | lockscreen? dunno whgat its using .. stock from ubuntu-studio | 16:56 |
| TJ- | VIA: aha! that is a known issue with xfce4-screensaver and the lock | 16:57 |
| Eickmeyer | xfce4-screensaver | 16:57 |
| TJ- | VIA: I've suffered it too | 16:57 |
| VIA | i think only 2 applications needed root access or have integrated into kernel, the NVIDIA GPU drivers and maybe also the NORDVPN client | 16:57 |
| VIA | erghghgh fuuuu | 16:58 |
| VIA | soooo, this .."XFCE screensaver" then, kicks in after lockscreen has activated? | 16:58 |
| VIA | i would have thought the machine powers gpu output off after it got locked | 16:58 |
| VIA | well i expected that would be the logical procedure, turne the gpu off when lock is in and monitor is off .. | 16:59 |
| bigpod | VIA: there you go i guess we resolved at least why it happens to you, i use ubuntu mate and mate-screensaver. | 16:59 |
| VIA | yea there we go and im happy i know now!" | 16:59 |
| Eickmeyer | xfce4-screensaver is forked from mate-screensaver. | 16:59 |
| VIA | erghm soo the fork is to short to eat with it | 17:00 |
| bigpod | Well turning off the GPU doesnt mean that processing of input doesnt happen problem my guess is that the lockscreen input isnt focused untill you enter the password | 17:00 |
| bigpod | untill it turnes the screen on* | 17:01 |
| bigpod | or something like that | 17:01 |
| VIA | no i would think turning the gpu off would cease rendering of many things and especially not require/or allow any routine that makes a "screensaver" kick in | 17:01 |
| bigpod | well processing can happen without rendering | 17:02 |
| VIA | yes and it should when there is no output | 17:03 |
| VIA | or, when locked | 17:03 |
| VIA | .. and no media playing | 17:03 |
| bigpod | if processing wouldnt happen you wouldnt be able to even input the password without having screen turn on | 17:03 |
| VIA | did i fail to transport my point | 17:03 |
| bigpod | problem my guess it focus becuase technicly speaking lockscreen is just a window layered on top of everything that blockes inputs to anything else and if lockscreen doesnt force focus on itself anythying can still get the imput | 17:04 |
| bigpod | input* | 17:05 |
| bigpod | Eickmeyer: maybe this was fixxed in mate-screensaver after xfce4-screensaver was forked | 17:07 |
| Eickmeyer | bigpod: Very possible. The devs probably need to grab the commit that fixed it as a patch. | 17:07 |
| Eickmeyer | bigpod, VIA: Worth mentioining in #xubuntu since they maintain xfce in Ubuntu. Bear in mind also that everything in 20.04 is frozen and the update process for something like that is complicated at best. | 17:09 |
| Eickmeyer | It's possible it's already been fixed but not backported. | 17:09 |
| bigpod | Eickmeyer: a note i am a part of ubuntu mate so after you said that xfce4-screensaver is fork of mate-screensaver you kinda made me more focused to see if there is something we have broken as well even if i havent observed that yet | 17:10 |
| === genii-core is now known as genii | ||
| TJ- | The issue I've seen has the visual appearance that the screen/session is unlocked, mouse moves around, but 'nothing' responds. Eventually twigged that the unlock is invisible and have to blindly type the user password. | 17:13 |
| TJ- | Mostly this is with multiple displays; can't recall if I've seen it with single display | 17:14 |
| bigpod | i heard that happening with old light-locker a lot | 17:14 |
| TJ- | it used to be worse; I'd have to switch to a TTY and kill the locker | 17:14 |
| TJ- | bigpod: yeah, that was the one! | 17:14 |
| TJ- | I did a lot of debugging on that about a year ago I think | 17:15 |
| TJ- | Sean eventually patched the worst of the issues I think | 17:15 |
| TJ- | oerheks: well, not been able to solve the failed USB device even by removing the PCI node for the parent USB hub and rescan-ing the PCI parent bus | 17:16 |
| TJ- | And the port doesn't have power control (no power/pm_qos_no_power_off node) | 17:17 |
| alkisg | TJ-: how about rmmod/modprobe, with ahci or some similar module? | 17:19 |
| alkisg | I can't recall if that worked when I tried it or not... | 17:19 |
| TJ- | well not ahci ... that's SATA/ATA :) | 17:20 |
| TJ- | but removing the xhci module won't be possible because the system has multiple USB3 controllers and are in use | 17:20 |
| TJ- | Looks like another reboot will be quickest after all | 17:21 |
| faceface | hello | 17:46 |
| lotuspsychje | welcome faceface | 17:47 |
| faceface | hi lotuspsychje | 17:47 |
| lotuspsychje | how can we help you today faceface | 17:47 |
| faceface | A colleague of mine disabled DNS on a server (I think). Now I don't know how to reconfigure it to defaults. | 17:47 |
| faceface | I can ssh on, but I can't grab the things I need... I think he turned it off for security | 17:47 |
| faceface | (it's a web server) | 17:48 |
| lotuspsychje | faceface: we also have an #ubuntu-server channel if you dont find the answer here tonight | 17:49 |
| faceface | cool, I'm googling, but my first thought (dpkg-reconfiure dnsmasq) didn't work | 17:50 |
| holgersson | faceface: Hi, I would look for the specific ubuntu version first and depending on that I would look what type of setup they run for network configuration in general. | 17:53 |
| holgersson | faceface: should be written at least in /etc/os-release. | 17:53 |
| holgersson | faceface: "disabled" is too vague to help you better. | 17:53 |
| faceface | holgersson: focal | 17:54 |
| faceface | holgersson: yeah, I don't know how to be specific... | 17:54 |
| holgersson | faceface: Then you have to ask many questions :-p | 17:55 |
| faceface | hehe | 17:55 |
| faceface | why does apt not work on this box? I get "Temporary failure resolving 'archive.ubuntu.com'" | 17:56 |
| faceface | Seems DNS is 'broken' somehow... | 17:56 |
| faceface | ;-) | 17:56 |
| holgersson | OK. You can connect via ssh, so I assume it can at least connect via IP. | 17:57 |
| faceface | actally, that dumb question got me some good google leads | 17:57 |
| faceface | yes | 17:57 |
| holgersson | (could be also that only ssh works, just to keep that in the back of the head) | 17:58 |
| holgersson | Does /etc/resolve.conf exist? It is a plaintext file or a symlink | 17:58 |
| faceface | /etc/resolv.conf has 128.0.0.53, so is expecting something to be running, I'll just set it to 8.8.8.8 and apt update | 17:58 |
| faceface | all packages are up to date... weird | 17:59 |
| VIA | thx Eickmeyer | 17:59 |
| faceface | bbl | 18:00 |
| holgersson | faceface: 128.x is weird, but might be systemd-resolved doing the actual resolving plus caching, so look if that one was running, e.g. with "journalctl -u systemd-resolved" | 18:01 |
| holgersson | faceface: no, 128.0.0.53 is totally weird and points to a remote IP. | 18:02 |
| webchat33 | Hi, | 18:20 |
| webchat33 | I'm trying to boot to an ubuntu 20.04 bootable usb and install it with encryption, but for some reason it only gets allocated 2.2gb RAM on startup and this isn't enough for the encryption to generate the keys. Why is this? This is on a lenovo Y700 5ISK in case that helps. The windows partition seems to get allocated 11.8/12GB, so why no the | 18:20 |
| webchat33 | bootable USB? I've tried to google all over and can't find the reason for this? | 18:20 |
| webchat33 | Things i've tried: resetting bios. Also, my ubuntu 18.04 bootable usb also seems to get assigned only 2.2gb RAM | 18:21 |
| alkisg | webchat33: are you talking about RAM or disk space? What does the message say exactly? | 18:21 |
| alkisg | Can you e.g. upload a screenshot? | 18:21 |
| webchat33 | RAM, also screenshot will be hard cause I'm on a different computer to the one I'm using the bootable usb on | 18:22 |
| alkisg | Photo from phone will also do. What is the exact message, and when is it displayed, from the ubuntu installer? | 18:23 |
| webchat33 | There is no message. When I boot to the bootable USB, it only has "2.2gb RAM availiable" in the about section in the settings | 18:23 |
| webchat33 | The screen does flash black with some console-text on it, it only lasts a few seconds, I'll try capture it with my phone | 18:24 |
| alkisg | 2.2 available, and how much total? | 18:24 |
| webchat33 | My laptop has 12GB RAM | 18:24 |
| alkisg | No I mean in the dialog that you see | 18:24 |
| alkisg | The easiest way would be to type "free | nc termbin.com 9999" in a terminal, which would output all the necessary info | 18:25 |
| alkisg | Troubleshooting via terminal is much easier than dialogs and photos | 18:25 |
| webchat33 | What does that command do btw? termbin.com? | 18:26 |
| alkisg | free shows the memory | 18:27 |
| alkisg | The `| nc termbin.com 9999` part uploads the terminal output to the internet | 18:27 |
| alkisg | Then it shows a url like termbin.com/asdf, and you only need to show us that small url | 18:27 |
| webchat33 | excellent! | 18:28 |
| webchat33 | termbin.com/34z0 | 18:28 |
| alkisg | Woah, you're actually correct, it doesn't see all the RAM. That's a new one :D | 18:29 |
| alkisg | Can you do this now? dmesg | nc termbin.com 9999 | 18:29 |
| alkisg | This shows the kernel messages | 18:29 |
| alkisg | Hopefully we'll see why it doesn't see all the ram in there | 18:29 |
| Hercules | how to rename a device? | 18:29 |
| webchat33 | https://termbin.com/mh97 | 18:29 |
| CPUOne | Hercules: what device? | 18:30 |
| Hercules | a usb device | 18:30 |
| Hercules | camera | 18:30 |
| Hercules | I tried with udev, but couldn't have success with it. | 18:31 |
| alkisg | webchat33: can you also upload the hardware details? sudo dmidecode | nc termbin.com 9999 | 18:32 |
| webchat33 | Sure, one moment please | 18:32 |
| webchat33 | https://termbin.com/q84qc | 18:33 |
| webchat33 | Some background in case it's relevant (probably not): I'm doing this because my 18.04 installation messed up cause the boot partition /boot ran out of space cause it kept too many old kernals, and then I couldn't boot. I tried to fix it through my ubuntu 18.04 bootable usb, but to no avail. | 18:34 |
| webchat33 | While fixing it, I remember messing about with a whole bunch of commands I can't remmeber cause it was over two months ago, but I think one might've done something to one of the EFI partitions on the 128GB SSD, and I don't know if that could be the issue. | 18:35 |
| alkisg | webchat33: funny, it detects two ram modules, one 4 GB and one 8 GB, and yet it doesn't see/use all the ram | 18:35 |
| alkisg | The disk shouldn't affect at all how much RAM the live USB stick sees | 18:36 |
| alkisg | Can you test for BIOS updates? | 18:36 |
| alkisg | Also, can you test with 21.04, which has a newer kernel? | 18:36 |
| webchat33 | I already updated my bios to try to fix it, i'm using the latest bios availiable from the lenovo website | 18:37 |
| webchat33 | i can try do that yes | 18:37 |
| alkisg | You're using kernel 5.8, maybe the bug you see has been fixed in the newer kernel 5.11 | 18:37 |
| webchat33 | I managed to install 18.04 without this issue though? | 18:37 |
| alkisg | It's possible, if so it's called a "regression" | 18:38 |
| alkisg | A bug that wasn't there in previous versions and appeared in subsequent versions | 18:38 |
| webchat33 | Hmm ok. Even though I never had this issue on the 18.04 installation, I tried my 18.04 bootable usb and it also sees only 2.2gb ram. I'm going to make a 21.04 boot drive now but it'll take a minute | 18:39 |
| webchat33 | Why is the server spamming joins and leaves constantly? | 18:40 |
| coconut | webchat33, because of a netsplit(overload on a server) | 18:41 |
| webchat33 | hmm, ok | 18:41 |
| webchat33 | Also I just wanted to say alkisg thanks for helping me with this cause I've tried asking everywhere - ubuntu forums, reddit, discord, etc about this and get no responses | 18:41 |
| TJ- | webchat33: 2.2GiB RAM out of how much installed? | 18:43 |
| webchat33 | 12gb | 18:44 |
| TJ- | 2.2 would suggest to me a 32-bit kernel without PAE, which seems highly unlikely! | 18:44 |
| alkisg | webchat33: if you boot the live usb stick in BIOS mode you'll see a memtest entry; you can also check that one, how much memory it sees | 18:45 |
| TJ- | webchat33: what CPU/subsystem is it? I can think of reasons that a NUMA system might do that (albeit very unlikely!) | 18:45 |
| alkisg | webchat33: the initial 18.04 release had a different kernel than subsequent 18.04 releases, so if you just made that 18.04 live usb stick, then it might explain it | 18:45 |
| alkisg | E.g. suppose that the "regression" happened in kernel 5; maybe the older 4.15 works, and kernels 5 to 5.8 have the regression | 18:46 |
| alkisg | To test that theory, you'd download ubuntu 18.04.1, which has the old kernel, while 18.04.2+ have the newer ones | 18:46 |
| webchat33 | TJ- i'm not sure what information you want by CPU/subsystem, it is intel i5, these are pastes I made for alkisg they may have your information in maybe https://termbin.com/34z0 https://termbin.com/mh97 https://termbin.com/q84qc | 18:46 |
| === 011AABDX8 is now known as Mikro | ||
| webchat33 | alkisg,no this is the bootable 18.04 I made when installing it (i'm pretty sure) | 18:47 |
| webchat33 | so not a newer version | 18:47 |
| TJ- | webchat33: also, check what the firmware reports as memory available. Look at the early part of the kernel log ("dmesg | less" or "journalctl -k") look for the BIS-e820 lines | 18:47 |
| webchat33 | actually, I used it in my laptops SD card reader and installed it from there and it seemed to work before, I can check again and do the same commands from that just to check if that might help, but you'd have to remind me of the commands I did | 18:48 |
| webchat33 | TJ- ok, i can try | 18:48 |
| alkisg | TJ-: dmesg is at https://termbin.com/mh97 and dmidecode at https://termbin.com/q84qc | 18:49 |
| webchat33 | TJ- I actually think one of those pastes was dmesg | 18:49 |
| webchat33 | The first one actually here https://termbin.com/34z0 | 18:50 |
| webchat33 | ahh, sorry just say your msg alkisg | 18:50 |
| webchat33 | saw* not say | 18:50 |
| webchat33 | I guess it'd be useful if those paste things have the command that was run at the top | 18:51 |
| webchat33 | There's probvably a way to do that in bash but I haven't such skillz yet | 18:51 |
| alkisg | ( echo "RUNNING: dmesg"; dmesg ) | nc termbin.com 9999 | 18:52 |
| webchat33 | So the 18.04 bootable SD card uses 18.04.5 LTS and also only sees 2.2gb | 18:52 |
| TJ- | " BIOS-e820: [mem 0x0000000000100000-0x000000008f0affff] usable " | 18:52 |
| TJ- | That's 2398814208 bytes | 18:53 |
| TJ- | "BIOS-e820: [mem 0x0000000100000000-0x000000035effffff] reserved" == 10183770112 | 18:55 |
| TJ- | so, there's 10GiB marked by firmware as reserved, not usable | 18:55 |
| TJ- | webchat33: either a memory/hardware fault or something in the firmware I suspect | 18:56 |
| webchat33 | (in case it's helpful, this is dmesg from the 18.04 sd: https://termbin.com/sinh) | 18:56 |
| webchat33 | TJ- (i don't know how to do those reply things), I don't think (and really hope not) that it's a hardware issue as the windows 10 partition seems to get 11.8gb given to it | 18:57 |
| webchat33 | so the firmware really hates ubuntu or something | 18:57 |
| webchat33 | ok, ubuntu 21.04 fininshed creating, so i'm gonna try running that now | 18:58 |
| TJ- | webchat33: it does seem strange it is being reported but marked reserved; are you by chance dual booting with Windows and not done a cold power off restart inbetween? | 18:59 |
| webchat33 | I am dual booting windows (well, at least, I boot windows and am booting to a bootable usb), I held down the power button because windows takes an age and a half to shut itself down | 19:00 |
| webchat33 | how do i restart and then make sure it stays off? cause I tried that and I can't identify when correct time to press power button to stop it booting | 19:01 |
| alkisg | I've never heard of it, but I wonder, is it possible that windows suspend keeps some data in ram, and marks it in the firmware as reserved? | 19:01 |
| alkisg | Try to reboot windows, not power it off | 19:02 |
| TJ- | alkisg: it's an attractive theory but I cannot think how/why - it'd have to be a suspend-to-RAM of some form. But then resume would be into Windows | 19:02 |
| webchat33 | ok, but if I reboot, it just boots to windows, i guess i'll have to change the boot order in the bios so that it tries something else before windows is rebooted, one sec | 19:03 |
| TJ- | webchat33: most Lenovo's you hold down the power button for about 5 seconds to force it off, then let go. On my E495 I have to press the power button for about 3 seconds to power it on | 19:03 |
| webchat33 | Yes, the last time I exited windows, that's how I turned it off | 19:04 |
| TJ- | webchat33: you should be able to press F10 (I think) to get to the manual boot menu | 19:04 |
| TJ- | or is it F8? I can never remember! | 19:04 |
| webchat33 | I usually just stick a pin in this hole on the side of the laptop, and it gives me options to go into bios, choose what to boot from, or go into recovery | 19:04 |
| webchat33 | i think it's called "novo button" | 19:04 |
| goddard | gnome really needs a way to set default sound device | 19:04 |
| goddard | every time i reboot i have to set it manually | 19:04 |
| faceface | hmm... systemd-resolve ... what does that do? | 19:05 |
| TJ- | faceface: handles DNS lookups | 19:06 |
| TJ- | goddard: what is being set as the default, some HDMI audio output? | 19:07 |
| faceface | TJ-: I'm wondering if I need it on a 'generic' webserver | 19:07 |
| TJ- | faceface: it's there to do local caching resolving | 19:07 |
| webchat33 | TJ- alkisg just tried 21.04, it's still only got 2.2gb ram | 19:07 |
| alkisg | webchat33: at what point was linux seeing all 12 GB RAM? Was it your dual boot 18.04 which now fails to boot? | 19:08 |
| TJ- | webchat33: got a question for you. How can the PC only have 12GiB RAM when it has only 2 DIMM slots - it'd be 2x4GiB or 2x8GiB surely? | 19:08 |
| faceface | TJ-: cool. I see that /etc/resolv.conf is managed by systemd-resolved... where does it pick up DNS server settings? e.g. 8.8.8.8? | 19:08 |
| TJ- | faceface: from DHCP on the links, or statically set by the network manager | 19:09 |
| alkisg | TJ-: dmidecode says 4 + 8 https://termbin.com/q84qc | 19:09 |
| TJ- | alkisg: oh, sorry, missed that! so, unbalanced modules... that could be a problem although if it worked before...! | 19:09 |
| webchat33 | alkisg yes, the last time i think was dualboot 18.04 (which is now gone, because when I tried to install 20.04 using encryption and set up encrypted volume, it went ahead and formatted that partitiion (it's fine cause I had everything backed up)) | 19:10 |
| TJ- | on these laptops the slots in dual-rank need matched modules | 19:10 |
| Qetzlcoatl | TJ-: my laptop, for example, has 12GiB - 4GiB onboard + 8 GiB DIMM | 19:10 |
| alkisg | webchat33: when you say "everything backed up", do you also mean the OS, could we check the contents and see the kernel version that you were using? | 19:10 |
| webchat33 | Yes, I just backed up the entire thing (can't remember if that includes the boot partition), i'll check | 19:11 |
| webchat33 | (dmesg for 21.04: https://termbin.com/oomm) | 19:11 |
| webchat33 | ok so, the backup doesn't include the /boot partition of the ubuntu because that was on a seperate parition and i just copied everything form the main partition, but there is boot_bak, boot-sav and everything else from ubuntu, is there a way to find the kernal from this? | 19:13 |
| alkisg | webchat33: ls /lib/modules | 19:14 |
| goddard | TJ- nah just line out, but I have 3 audio sources my Mic has a headphone jack, my motherboard has a line out which is set, and I have hdmi connected which also has audio but i dont use because it is crackling | 19:14 |
| alkisg | The kernel number will be there too (inside /mnt, of course) | 19:14 |
| webchat33 | 4.15.0-143-generic | 19:14 |
| alkisg | That's 18.04.1 | 19:15 |
| alkisg | While your live stick is 18.04.5 | 19:15 |
| webchat33 | ahhh, ok | 19:15 |
| alkisg | Try to download and boot with 18.04.1 | 19:15 |
| TJ- | webchat33: it's definitely a firmware issue; some setting has changed. There are 2 obvious ones to look for: Reserved video memory size, and a setting to enable memory remapping. There may be other options there but you'll have to explore. | 19:16 |
| TJ- | webchat33: you're not the only ideapad owner to see this kind of issue | 19:17 |
| webchat33 | ok, i'll create a stick with ubuntu-18.04.1-desktop-amd64.iso2018-07-25 03:221.8G from http://old-releases.ubuntu.com/releases/18.04.1/ | 19:17 |
| webchat33 | oh | 19:17 |
| webchat33 | I have absolutely no idea how to do that, where have you seen that other people have experienced this issue, cause I tried to google around and didn't find much | 19:18 |
| webchat33 | how do i edit the firmware? | 19:19 |
| webchat33 | I would probably be able to move most things to my main computer, but a few days ago I accidentally uninstalled python trying to fix something, and half of the operating system uninstalled itself (settings, etc) and half my programs don't work. At least I have the browser, files, text editor, terminal, document editor, main things ya know. | 19:22 |
| TJ- | webchat33: I also see a failed PCI window in the kernel log "pci 0000:01:00.0: can't claim BAR 6 [mem 0xfff80000-0xffffffff pref]: no compatible bridge window" which could be related | 19:23 |
| TJ- | webchat33: at boot-time press the key that enters BIOS/firmware setup and explore all options | 19:23 |
| webchat33 | TJ- yes, I have done that. I even reset it to factory, and nothing. | 19:24 |
| TJ- | webchat33: if you can't see anything or don't want to explore, on the 'save' tab find the 'reset to factory defaults' option and use it | 19:24 |
| TJ- | oh really, hmmph! | 19:24 |
| goddard | think bluetooth is locking up my system? | 19:24 |
| goddard | https://paste.ubuntu.com/p/G5HMdnNWK7/ | 19:24 |
| webchat33 | i can try it again to be sure | 19:24 |
| TJ- | webchat33: it's definitely firmware issue since it is telling the OS that 10GiB is reserved | 19:24 |
| webchat33 | I bet it's windows causing that problem as usual, since windows seems to be assinged 11.8 gb ram no problems | 19:25 |
| faceface | what's the package that lets "apt install some<tab><tab>" autocompleete? | 19:26 |
| TJ- | webchat33: if you're doing a cold start Windows shouldn't be able to have any effect | 19:26 |
| faceface | I'm so used to it now I can't live without it ;-) | 19:26 |
| TJ- | faceface: bash completion | 19:26 |
| faceface | TJ-: thanks | 19:26 |
| TJ- | goddard: is the Bluetooth device near powerful 2.4GHz wifi or other devices? it could be the frequency band is saturated and overpowering the bluetooth link | 19:27 |
| faceface | bash-completion is already the newest version (1:2.10-1ubuntu1). | 19:27 |
| TJ- | goddard: I've also seen that with failing bluetooth battery levels | 19:27 |
| webchat33 | I hadn't done the cold start before, I'm just tyring it now | 19:27 |
| webchat33 | Well, i'm pretty sure I had done it several days ago, but i'm trying it again right now | 19:27 |
| TJ- | webchat33: cold start should result in all settings being cleared although it is sometimes next to impossible on laptops with integrated batteries - no guarantee it is really 'off' | 19:27 |
| goddard | TJ- yeah my monitor has bluetooth i cant disable and wifi is pretty near and other wifi devices | 19:28 |
| faceface | things like "apt sear<tab><tab>" aren't working :-/ | 19:28 |
| faceface | should I just reboot this server? | 19:28 |
| TJ- | whereas a reboot (soft / warm restart) can retain hardware settings | 19:28 |
| TJ- | goddard: wifi devices often have a BT co-exist option (which is usually enabled at the driver level already) | 19:28 |
| * faceface tries a reboot on the offchance... | 19:29 | |
| webchat33 | OMG | 19:29 |
| webchat33 | something happened, i have no idea how, but somehow, i now have 11.5GB showing in the about section of the settings | 19:29 |
| goddard | TJ- ive been having this lock up issue | 19:29 |
| faceface | webchat33: OMG, nothing has changed? | 19:29 |
| faceface | ;-P | 19:29 |
| TJ- | goddard: #1 - ensure BT keyboard battery is charged #2 - put it right next to the PC's bluetooth receiver antenna if you can figure out where they are hidden | 19:29 |
| goddard | i just disabled my wifi hopefully that prevents it | 19:29 |
| webchat33 | woop woop, no idea how on eart that fixed itself cause i'm pretty sure I did a cold reboot a couple days ago, eh, guess I'm just lucky today | 19:29 |
| TJ- | faceface: are you using bash or some other shell? | 19:30 |
| goddard | really sucks because i like my keyboard | 19:30 |
| TJ- | webchat33: like i recommended to being with - I've seen this kind of thing so many times! | 19:30 |
| faceface | TJ-: yeah, bash... I think... yup /bin/bash | 19:30 |
| goddard | logitech ergo k860 is pretty nice but makes me wish it had a non-bluetooth option | 19:30 |
| faceface | also after reboot my DNS issue is back: Failed to connect to https://changelogs.ubuntu.com/meta-release-lts. Check your Internet connection or proxy settings | 19:30 |
| webchat33 | is this chat publicly archived? where? (i want to look upcommands that altsig told me) | 19:30 |
| TJ- | faceface: not sure why you'd loose the tab-completion /unless/ you've inadvertently disabled it! | 19:30 |
| faceface | TJ-: I just tried dpkg-reconfigure bash-autocompletion, but no changes apparently | 19:31 |
| ioria | maybe you changed the default .bashrc file | 19:31 |
| TJ- | faceface: "failed to connect" isn't DNS - that sounds like routing. Is your PC able to do both IPv4 and IPv6 ? | 19:31 |
| faceface | Again, I /think/ so... not 100% sure how to check | 19:32 |
| faceface | I see ip4/6 addresses in ifconfig | 19:32 |
| faceface | ah... it is working! Ignore me | 19:33 |
| faceface | I wonder why it had stopped previously | 19:33 |
| faceface | that message on login just looked inidcative of the old problem | 19:33 |
| * faceface purges bash-completion and tries again | 19:34 | |
| RupertEverton | webchat33: logs can be found here: https://irclogs.ubuntu.com/2021/07/05/%23ubuntu.txt | 19:34 |
| webchat33 | RupertEverton thank you! | 19:35 |
| faceface | worked! weird | 19:35 |
| RupertEverton | I'm having troubles extending my LV with the command "sudo lvextend +l +100%FREE --resizefs ubuntu-vg/root". i get this error message: " Please specify a logical volume path." -- "df -h" shows that the lv is under "/dev/mapper/ubuntu--vg-root" while lvdisplay gives me "/dev/ubuntu-vg/root". I tried both of these path but the error that I get is | 19:39 |
| RupertEverton | still the same | 19:39 |
| TJ- | RupertEverton: "-l" not "+l" I think although that is likely a typo in IRC :) | 19:40 |
| RupertEverton | TJ- "I" as in uppercase "i" ? I tried it just now and still get the same error. These pesky dells smh | 19:41 |
| TJ- | lower-case L it's short for --extents | 19:42 |
| webchat33 | Now that the ram issue is fixed, I'm trying to install 20.04 with encryption and dual boot. The way I'm doing this is : | 19:42 |
| RupertEverton | TJ- that's how I ran the command until now, with lowercase L | 19:42 |
| webchat33 | damn, i thought pressing shift enter would create a newline, oops | 19:42 |
| TJ- | RupertEverton: hmmm, the command looks fine | 19:43 |
| pedrocr | after installing 20.04 on a Lenovo T14s Gen1 AMD it seems to have decided to install the "oem" kernel, but it doesn't actually use it | 19:44 |
| pedrocr | the normal kernel is at 5.8 and the oem at 5.10 but the 5.8 kernel is the default | 19:44 |
| pedrocr | is this normal? | 19:44 |
| RupertEverton | TJ- just to confirm, I don't need to run this from a live usb, correct? | 19:46 |
| TJ- | RupertEverton: you can do an online resize to make things larger; only need offline if shrinking | 19:47 |
| RupertEverton | So it should work theoretically, but it doesn't actually | 19:47 |
| TJ- | RupertEverton: the error you're getting suggests an issue with the device-mapper links failing to find the LV itself | 19:48 |
| TJ- | RupertEverton: what does "ls -l /dev/ubuntu-vg/root" point to? something like ../dm-X ? | 19:48 |
| RupertEverton | TJ-: ../dm-1 | 19:48 |
| TJ- | RupertEverton: so the actual device is /dev/dm-1 . Let's check with "sudo dmsetup info /dev/dm-1" that it matches the VG/LV naming | 19:49 |
| RupertEverton | TJ- ubuntu--vg-root | 19:50 |
| TJ- | RupertEverton: good, so device-mapper doesn't appear to be confused | 19:51 |
| TJ- | RupertEverton: the VG does have free extents I assume? "sudo vgdisplay" | 19:51 |
| TJ- | oh I see it now! You're missing a hyphen!! | 19:51 |
| TJ- | RupertEverton: it should be "sudo lvextend -l +100%FREE --resizefs ubuntu--vg/root | 19:52 |
| webchat33 | Now that the ram issue is fixed, I'm trying to install 20.04 with encryption and dual boot. The way I'm doing this is : | 19:52 |
| webchat33 | 1. create boot partition of 1.5gb and assign to /boot, create lvm partition of the rest (about 950gb) and assign to / | 19:52 |
| webchat33 | there is now below this "device for bootloader installation"... what device do I select? | 19:52 |
| RupertEverton | TJ- among the combinations of possible paths that I tried, that was one of them. I tried again now and the error is still the same | 19:53 |
| TJ- | webchat33: have you read my guide for doing that? | 19:53 |
| webchat33 | uhh, i read some guides, not sure if it was yours, if you link me i'd appreciate that?! | 19:53 |
| dreamon | hello. "systemctl suspend" only logs out, but not supend. how can I find out reason? | 19:53 |
| TJ- | RupertEverton: hmmph, well the -- is generally used to separate hyphenated arguments from non-hyphenated, so surround the name in quote marks | 19:53 |
| RupertEverton | TJ- with "sudo vgdisplay" I get "Alloc PE / Size 46199 / 180.46 GiB -- Free PE / Size 14621 / 57.11 GiB". In fact I remember when we created it, we set it up to 80% | 19:54 |
| TJ- | RupertEverton: it should be "sudo lvextend -l +100%FREE --resizefs 'ubuntu--vg/root' " | 19:54 |
| webchat33 | oh i just found your launchpad link in your bio | 19:54 |
| webchat33 | no, that doesnt have it as far as i can see | 19:54 |
| TJ- | webchat33: I think RupertEverton can vouch for this :) https://help.ubuntu.com/community/Full_Disk_Encryption_Howto_2019 | 19:54 |
| RupertEverton | TJ- " Volume group "ubuntu--vg" not found. Cannot process volume group ubuntu--vg" | 19:54 |
| TJ- | RupertEverton: ok, now remove one of those pesky hyphens then! | 19:55 |
| RupertEverton | Lovely! that worked | 19:55 |
| TJ- | RupertEverton: I cannot explain why, something in your shell is messing with things somehow | 19:55 |
| webchat33 | Do I need a swap area? is that just if I want to hibernate the machine? can i do without it? | 19:56 |
| TJ- | webchat33: generally Ubuntu uses a swap file now, so you don't need a swap partition dedicated | 19:56 |
| RupertEverton | TJ- webchat33 yes! that's an amazing guide! Still need to add the little note for dealing with large initrd.img in case anyone else stumbles upon the same error | 19:56 |
| TJ- | RupertEverton: thanks for reminding me - I was hoping the devs would fix that problem! | 19:56 |
| RupertEverton | TJ- yeah I actually confirmed the launchpad bug report right after you posted it but no responses yet! | 19:57 |
| TJ- | RupertEverton and myself spent a weekend trying to figure out why his system wouldn't boot and eventually determined it was due to an UEFI bug that couldn't cope with > 100MiB initrd.img files | 19:58 |
| shimbles | startplasma-x11 worked in groovy but not hirsute; in hirsute it says $DISPLAY is not set or cannot connect to the X server. | 19:58 |
| RupertEverton | TJ- btw you mentioned something in my shell messing with things -- is there a way to run some diagnostics and find out what it is? | 19:59 |
| TJ- | RupertEverton: I cannot think of any easy way aside from inserting 'echo ' after 'sudo ' and seeing if what is displayed comes out different that what you expect | 20:01 |
| webchat33 | Maybe i'm being dumb, but I can't tell if I'm supposed to select boot loader device as the one with efi system partition, or the one where /boot is? | 20:02 |
| TJ- | RupertEverton: shells do all sort's of word expansions, alias replacements, and so on, so what you type is not always what you get | 20:02 |
| TJ- | webchat33: the physical device that contains the EFI system partition | 20:02 |
| TJ- | webchat33: /boot/ is 'just' a directory of the root file-system | 20:02 |
| RupertEverton | TJ- I tried with and without quotation marks and they both yield the same result: lvextend -l +100%FREE --resizefs ubuntu-vg/root | 20:02 |
| TJ- | webchat33: even when /boot/ is in its own file-system! | 20:03 |
| TJ- | RupertEverton: the only thing I can think then is some invisible no-whitespace character got in there somehow | 20:03 |
| RupertEverton | yikes | 20:03 |
| TJ- | RupertEverton: we know it had to be in the path part, so something about ubuntu-vg/root - I wonder if there was an invisible character after the 'root' so actually it was passing 'root?' (? == something invisible - a control code) and when you added quotes it was separated from the path and discarded by the command | 20:05 |
| TJ- | RupertEverton: did you by chance copy/paste the original command rather than type it? | 20:05 |
| RupertEverton | i did, yes | 20:06 |
| TJ- | aha! then the paste added something invisible | 20:06 |
| TJ- | I've suffered that sometimes too | 20:06 |
| RupertEverton | TJ-: wait actually no, I had to manually type it cause I was reading it from my phone. So 100% typed it | 20:07 |
| TJ- | RupertEverton: I blame your typing then :D | 20:07 |
| webchat33 | So, as I understand, the reason to encrypt everything and not just use the default thing is to prevent people from installing malware in the boot partition right? | 20:08 |
| RupertEverton | fair enough! might aswell be at this point haha! | 20:08 |
| webchat33 | But if i'm just using encryption to prevent thieves getting info if laptop is stolen, i don't need that full security? | 20:09 |
| shimbles | do I need to disable wayland or something in hirsute to get kda plasma working? | 20:09 |
| TJ- | webchat33: basically yes, as well as giving you a heartattack when you fail to remember the passphrase or can't type it! | 20:09 |
| Mekaneck | shimbles: afaik Kubuntu doesn't use wayland by default on hirsute | 20:10 |
| webchat33 | ok, excellent. thanks so much for your help tonight¬ | 20:10 |
| pedrocr | has anyone seen a hardware or software issue that causes enter/backspace/space to randomly stop working and then go back to working again? | 20:10 |
| TJ- | webchat33: in theory, but if you're using encryption may as well make it as secure as possible since you're going to have to provide the passphrase, or a hardware token such as a Yubikey, at some point | 20:10 |
| shimbles | mekaneck i upgraded focal -> groovy and startplasma-x11 was still working; but on upgrade to hirsute, it's not working | 20:10 |
| shimbles | when i run vncserver it shows $DISPLAY is not set or cannot connect to the X server. | 20:11 |
| webchat33 | Yubikey? cool, never heard of that. are hardware tokens possible for ubuntu encryption booting? | 20:13 |
| RupertEverton | pedrocr: what happens if you press the affected key in a notepad let's say. Do you get some random characters or nothing at all? | 20:13 |
| pedrocr | RupertEverton: nothing at all | 20:13 |
| TJ- | webchat33: i've been using a Yubikey for about 8 years for that and more | 20:13 |
| TJ- | pedrocr: is it a wired or wireless keyboard? | 20:13 |
| pedrocr | TJ-: integrated keyboard on a Lenovo T14s Gen1 AMD laptop | 20:14 |
| TJ- | pedrocr: I blame crumbs :) | 20:14 |
| webchat33 | can i make any usb into a yubikey like device? | 20:14 |
| pedrocr | TJ-: I'd normally guess that but it's the wife's laptop and she's not a slob like me :) | 20:14 |
| TJ- | pedrocr: give it a really good clean, blow with a high pressure air-line, a small paint brush, etc. | 20:14 |
| TJ- | pedrocr: I'd bet on there being hair or something under the keys | 20:15 |
| pedrocr | TJ-: doesn't seem like a dirt thing given the randomness, really feels like a firmware or hardware thing | 20:15 |
| pedrocr | or linux driver perhaps | 20:15 |
| TJ- | pedrocr: it is possible to gently 'pop' the key caps off and put them back again | 20:15 |
| shimbles | how do I disable wayland and enable X in hirsute | 20:15 |
| TJ- | pedrocr: if it is hardware it'd most likely be a flex cable stressed | 20:16 |
| pedrocr | TJ-: yeah, that could be, I probably just need to send it to support, which is annoying | 20:16 |
| === o is now known as niko | ||
| RupertEverton | pedrocr: removing the keyboard is actually not that complicated. if you lookup Lenovo T14s Gen1 keyboard change you can find some videos that show you how to do it | 20:18 |
| pedrocr | RupertEverton: sure, but I'd rather have the Lenovo technicians do it | 20:18 |
| RupertEverton | fair | 20:18 |
| pedrocr | this looks suspicious though: "kernel: usb 6-3: reset full-speed USB device number 2 using xhci_hcd" | 20:18 |
| pedrocr | how do I check what device that one is? | 20:18 |
| TJ- | pedrocr: we sent back a Lenovo E495 a few weeks ago due to thermal cracking of the keyboard bezel, broken tabs around the Ethernet port refusing to release the RJ45 plug. Support is done by IBM Global Technical Services, and they were more than useless. Claimed everything was "customer induced damage" despite us having a video of the cracks appearing and disappearing as the laptop warmed up and | 20:19 |
| TJ- | cooled down | 20:19 |
| pedrocr | TJ-: auch | 20:19 |
| TJ- | pedrocr: "lsusb" look at the Bus and Device numbering | 20:19 |
| shimbles | how do I disable wayland and enable X in hirsute | 20:19 |
| TJ- | shimbles: I /think/ it depends on what GPU the system has. | 20:19 |
| pedrocr | TJ-: I looked at that but didn't find any 6-3, I have 1, 2 and 4 on bus 6 | 20:20 |
| shimbles | TJ- no GPU, this is a headless AWS server, using vncserver; startplasma-x11 worked in focal, groovy, and started failing on upgrade to hirsute | 20:20 |
| TJ- | pedrocr: that'll be because the device has already 'fallen off' the bus. Use "journalctl -k | grep 'usb 6-3' that should show some earlier reports when the device was discovered | 20:20 |
| TJ- | shimbles: oh well, then you'll need a dummy display I guess for headless | 20:21 |
| TJ- | shimbles: why a GUI on a server though!? | 20:21 |
| shimbles | TJ- I am using Apache Guacamole, it shows you the GUI in a browser tab | 20:21 |
| shimbles | it was working in focal, groovy, i can't figure out what's different in hirsute that broke it, except for "wayland" | 20:22 |
| pedrocr | TJ-: weird, as I don't see any keyboard at all in lsusb so I guess Linux is using some other form to work with it | 20:22 |
| TJ- | shimbles: for a program named startplasma-x11 I doubt wayland should be involved :) | 20:22 |
| TJ- | shimbles: is that a shell script? if so, examine it, see what it is doing, whether it has commands to start an X server or is relying on the X server being started by something else | 20:23 |
| TJ- | pedrocr: keyboards are often using the legacy IBM PS/2 i8042 input device | 20:23 |
| pedrocr | TJ-: probably that, is there anything I can do to change that? maybe I should see what the BIOS allows | 20:24 |
| shimbles | TJ- the shell script is ~/.vnc/xstartup, and the command it runs is startplasma-x11, and the error is $DISPLAY is not set or cannot connect to the X server. | 20:24 |
| TJ- | pedrocr: this will give you a clue as to what it is: "ls -l /sys/class/input/input*/device" | 20:24 |
| TJ- | shimbles: startplasma-x11 is part of the plasma-workspace package on 20.04 ... maybe something has changed there | 20:25 |
| pedrocr | TJ-: indeed: "AT and PS/2 keyboard driver" is input4 | 20:26 |
| pedrocr | and yet when the problem was happening the USB device for the keyboard was being reset | 20:26 |
| pedrocr | I bet the kernel is doing something strange | 20:27 |
| pedrocr | in the past I remember there being options for BOOTP and others for keyboards in the BIOS, but this one doesn't seem to have any | 20:30 |
| TJ- | shimbles: the changelog shows something that may be relevant: "Breaks/Replaces plasma-workspace (<< 4:5.20.4-5~) for split out files from plasma-workspace" for plasma-workspace (4:5.20.5-2ubuntu1) hirsute published 14 Jan 2021 19:14:22 | 20:30 |
| shimbles | hmm, unfortunately i can't parse that. i see that according to these release notes plasma + wayland is not the default yet https://wiki.ubuntu.com/HirsuteHippo/ReleaseNotes/Kubuntu | 20:31 |
| TJ- | pedrocr: I suspect the USB device failing may be causing the input layer to have problems, but why it'd only affect certain keys is strange. I'd prefer to think the 2 things are unrelated | 20:31 |
| pedrocr | TJ-: yeah, maybe, I'm just wondering if the driver/firmware puts the keyboard in a broken mode and then the reset fixes it | 20:32 |
| TJ- | shimbles: I'm wondering if startplasma-x11 has been moved to another package that wasn't installed as part of the release upgrade and you're left with the 'old' version that isn't compatible with the upgraded packages | 20:32 |
| TJ- | pedrocr: I doubt the keyboard has both PS/2 and USB input modes. USB is sometimes used by touchpads. | 20:33 |
| TJ- | shimbles: no, the file is still in the same package | 20:34 |
| shimbles | i purged plasma-workspace and reintsalled, no luck | 20:35 |
| pedrocr | TJ-: ok, maybe unrelated then | 20:36 |
| webchat33 | Should i install things like vlc or discord through the software center, or should I use the terminal apt install stuff and deb packages? | 20:41 |
| dreamon | hello can not supsend. pm-suspend works.(but no lock screen). systemctl suspend not suspend, but logging out | 21:10 |
| Peanut | Nei: There's three 'failed logins' according to nickserv, which seem to be all my attempts to identify to nickserv today. Changing the password is a good idea nevertheless. It really looks like nickserv lost my previous password, somewhere between jun 28th and today. | 21:10 |
| Peanut | Ups, wrong window, sorry. | 21:11 |
| pedrocr | I've been able to replicate the enter problem in the bios keyboard tester so the problem is either firmware or hardware | 21:16 |
| V|A | HEY | 21:47 |
| V|A | sometimes i drag a window to a corner to resoze and it sticks and doesnt stop following the cursor drives me crazy | 21:47 |
| V|A | havent found what does that yet | 21:48 |
| src | V|A: sigh... ubuntu version, display driver, what DE/WM combination? | 21:48 |
| V|A | 20.04 nvidia drivers | 21:49 |
| V|A | idunno what de/wm means | 21:49 |
| V|A | xfce | 21:49 |
| V|A | distro is ubuntu-studio | 21:49 |
| V|A | it seems completely friggin random | 21:49 |
| src | any other graphical glitches besides that? | 21:49 |
| V|A | nop not linked to this scenario | 21:49 |
| src | machine fully up to date? | 21:50 |
| src | (as up to date as an old LTS allows one to be) | 21:50 |
| V|A | would think so as far as my understanding goes | 21:50 |
| src | $ sudo apt update && sudo apt full-upgrade && sudo snap refresh && sudo apt autoremove | 21:52 |
| src | that should completely upgrade your systems, comes without warranty | 21:52 |
| src | you can check the man pages for details | 21:54 |
| spine-o-saurus | is this channel also for bodhi linux? | 21:58 |
| jeremy31 | spine-o-saurus: no | 21:59 |
| spine-o-saurus | how come journalctl is refering to ubuntu for support then? | 22:01 |
| === genii is now known as genii-core | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
