g4570n | hi, only the installer-iso has an RC version? or will the others be added later? | 00:51 |
---|---|---|
fsmithred | g4570n, desktop-live will be RC when I figure out why there's no mouse and keyboard on desktop when booted from optical media. Minimal-live I guess can be ready in about a day or so. | 01:00 |
xrogaan | huh, that's a weird bug. | 01:07 |
g4570n | fsmithred: Javier Obregon, the dev of Etertics told me that something similar happened to him. | 01:08 |
fsmithred | how did he solve it? | 01:09 |
fsmithred | unplug/replug mouse or keyboard fixes it, but laptop people might not have that. | 01:09 |
fsmithred | even weirder, it doesn't happen if you put the image on usb, and it also doesn't happen if you boot to ram | 01:10 |
onefang | I wonder if a built in optical drive vs a USB optical drive might make a difference? Alternatively an optical drive might be the slowest thing to boot from of those three, race condition maybe? | 01:17 |
g4570n | fsmithred: He tells me that I compare the list of packages that are installed when you generate the ISO in the usual way, and when you generate another ISO allowing Recommended packages to be installed. To see what package was missing, I understood something like that. | 01:21 |
fsmithred | oh | 01:22 |
fsmithred | well... | 01:22 |
g4570n | He always tested with USB, never tried CD or DVD media | 01:22 |
fsmithred | the problem exists in the devuan desktop-live isos, which are installed with all the recommended packages | 01:22 |
fsmithred | me too, which is why I just found out about this problem | 01:22 |
fsmithred | it works correctly on usb | 01:23 |
g4570n | Yes, I can confirm that. I tested Beowuld beta1 from USB and had no problems detecting mouse and keyboard | 01:25 |
fsmithred | I also get the problem sometimes when booting the iso in qemu | 01:25 |
fsmithred | but then it works if I boot it a second time | 01:26 |
ShorTie | hait that crap | 01:39 |
g4570n | fsmithred: Javier tells me the following, in case it works for you: He generated the ISO and when he tested it on a virtual machine (he is supposed to recognize it as if you were inserting a CD) mouse and keyboard detection works without problems. But when he burned that ISO on a USB medium and booted from there on a real PC, he didn't recognize the mouse and keyboard unless he unplugged / plugged them | 01:57 |
g4570n | in again. Apparently the problem was the opposite of what happens here. I have a list of packages that he passed me, in case you want to look and try, it is supposed that with one of them he managed to solve that problem. | 01:57 |
g4570n | https://termbin.com/5h8j | 01:57 |
fsmithred | thanks! | 01:58 |
fsmithred | g4570n, I think I found it - added acpi-fakekey while I was logged in over ssh, and the trackpad started working. Tell Javier thanks very much! I still have to make a new iso with the package and burn it, but this test is very promising. | 02:21 |
g4570n | \(◎o◎)/ | 02:22 |
g4570n | Great! | 02:25 |
golinux | Profound comment ShorTie | 02:31 |
ShorTie | ?? | 02:32 |
fsmithred | I found it mildly comforting. | 02:33 |
LeePen | o/ | 11:45 |
LeePen | Is the ifup network hang workaround (allow-hotplug -> auto) documented somewhere? | 11:47 |
Centurion_Dan | Wasn't there a patch submitted a little while back?? | 13:01 |
Centurion_Dan | hi LeePen. | 13:01 |
LeePen | Centurion_Dan: yes, I had a vague memory of that, but I can't remember where! | 13:07 |
LeePen | The fix is simple. | 13:07 |
LeePen | Maybe it need to go in the release notes. | 13:07 |
Centurion_Dan | is it in a package we already fork? is Debian fixing it or do we need to do it ourselves... | 13:08 |
LeePen | ifupdown isn't forked. | 13:11 |
fsmithred | yes, there's a patch, but it's not packaged | 13:11 |
fsmithred | patch applies to /etc/init.d/networking | 13:11 |
fsmithred | it was originally on dng, but it's one of the messages that disappeared. Currently it's posted on the forum in a couple of places | 13:12 |
LeePen | Is it worth adding it to the release notes? | 13:13 |
LeePen | Just reported in #463 | 13:13 |
LeePen | It is also in debian at #922550 | 13:14 |
fsmithred | https://dev1galaxy.org/viewtopic.php?pid=15493#p15493 | 13:14 |
Centurion_Dan | Is it worth forking over. I've tripped up on it because it breaks network manager on new installs and upgrades too. | 13:15 |
Centurion_Dan | fsmithred: thanks. | 13:16 |
fsmithred | would we have to fork both ifupdown and ifupdown2 since the file is in both? | 13:24 |
LeePen | I have to admit forking ifupdown is pretty unattractive. | 13:28 |
LeePen | It has loads of bugs (180+) https://bugs.debian.org/cgi-bin/pkgreport.cgi?repeatmerged=no&src=ifupdown | 13:29 |
LeePen | and we would inherit all of those if we fork it. | 13:29 |
fsmithred | we could make a separate package for the patch like I planned to do for the cryptsetup patch | 13:34 |
LeePen | And use dpkg-divert? | 13:39 |
Centurion_Dan | ewww... | 13:43 |
Centurion_Dan | It's likely that ifupdown won't get much love any more from Debian anyway as most of it's function is replaced by systemd networkd.... | 13:44 |
Centurion_Dan | is ifupdown2 an iproute2 version of ifupdown? | 13:44 |
Centurion_Dan | fsmithred: ?? | 13:44 |
fsmithred | I don't know | 13:45 |
fsmithred | Description: Network Interface Management tool similar to ifupdown | 13:45 |
fsmithred | I have the old one installed | 13:46 |
fsmithred | i pppoeconf Depends ifupdown (>= 0.7.44~) | 13:46 |
rrq | Centurion_Dan: looks like a rewrite in some obscure language | 14:50 |
mason | Centurion_Dan: Debian isn't planning to move to systemd-networkd, are they? | 14:51 |
mason | Centurion_Dan: I was under the impression that that and netplan were Canonical peccadillos. | 14:52 |
mason | Centurion_Dan: https://cumulusnetworks.com/blog/ifupdown2/ anyway | 14:54 |
Centurion_Dan | yeah just found it. Written in python. | 14:54 |
fsmithred | python2 or 3? | 15:06 |
fsmithred | someone recently pointed out that wicd is written in python2 | 15:07 |
Centurion_Dan | wicd is crap... needs to be put down. It's really awful... | 15:07 |
fsmithred | we should probably switch to network manager for chimaera | 15:11 |
mason | It's a weird world when NetworkManager is a bastion of sanity. | 15:12 |
fsmithred | lol, I agree | 15:12 |
LeePen | We live in strange times. | 15:12 |
fsmithred | I used to remove that sucker first thing | 15:12 |
mason | I'd always mistakenly thought NM was the same group of developers that drives systemd, but it's not, and between that and having to dig into it for some bugs, I've grown to hate it much much less. | 15:12 |
Centurion_Dan | it works, and works pretty reliably. It's my first change on Devuan for every install on laptops. | 15:13 |
fsmithred | minimal-live RC isos are up | 15:13 |
mason | I stick with ifupdown everywhere I can. I need to look at LeePen's list of bugs. | 15:13 |
mason | Just on the face of it some of those bugs aren't valid. | 15:14 |
mason | Oh, maybe I shouldn't speak too soon. I caught on "allow-auto" but evidently that's supposed to be valid syntax. | 15:15 |
mason | I think for me anyway the right answer is to submit patches. I've got one into Canonical where systemd was breaking ifupdown that's finally being integrated. | 15:16 |
Centurion_Dan | I'm going to try ifupdown2. | 15:17 |
Centurion_Dan | to see if that solves my issues... | 15:17 |
mason | Centurion_Dan: What issues are you having, beyond allow-hotplug breaking? | 15:19 |
Centurion_Dan | I've been trying to get openvswitch working but it won't bring up the vswitches - and boot hangs.... | 15:19 |
mason | Ah, I've never had to muck with that. | 15:21 |
mason | Ooh, just saw the note about RC1 earlier. Snagging/testing. | 20:05 |
fsmithred | the live-isos are up, too | 20:07 |
fsmithred | ...desktop-live | 20:08 |
mason | Hm, so, the RC netinstall image doesn't show up on my X1C6. Will write a different stick to eliminate random stick issues. | 20:28 |
mason | It was a stick issue. Two beeps as GRUB loads, unusual. | 20:31 |
fsmithred | yeah, and on live-dvd, sound doesn't work | 20:33 |
fsmithred | I just tried removing PA and when I try to run /usr/bin/alsamixer, which completes with TAB, I get "no such file" | 20:34 |
fsmithred | WTF??? | 20:34 |
yeti | link into nirwana? | 20:34 |
fsmithred | more like twilight zone | 20:35 |
fsmithred | oh, sound works fine if I boot the iso in qemu | 20:35 |
fsmithred | aplay -l says no sound cards found, lspci disagrees | 20:39 |
mason | Alright, my usual non-ZFS install worked flawlessly with the netinstall image. | 20:53 |
fsmithred | got audio? | 20:56 |
mason | Haven't tested yet. I do the base install first, then install my graphical environment from the running system. I'll tell you as soon as the install is done. | 20:57 |
fsmithred | ok. I might be napping, but I'll see it later. | 20:57 |
mason | I mean, the desktop environment install. The install media's part is done. | 20:57 |
mason | kk | 20:57 |
mason | fsmithred: For when you get back, yeah, sound's fine out of the box. ALSA, Firefox to test. | 21:11 |
fsmithred | sound modules did not get loaded at boot. 'modprobe snd_hda_intel' fixed it. But I can't just do that for everyone. | 21:55 |
fsmithred | oh, if I boot to ram, it all works. Modules are loaded and sound works. I'm starting to think it's some timeout on reading the dvd that just gives up before it finds the files. | 22:10 |
mason | fsmithred: Hrm, on the live image? | 22:13 |
mason | fsmithred: snd_hda_intel loaded up on its own here, but this was netinstall setting up a disk | 22:14 |
fsmithred | yeah, on the desktop-live | 22:15 |
fsmithred | loads fine when I boot to ram, which means all of filesystem.squashfs gets read into ram | 22:15 |
fsmithred | Here's a list of modules that get loaded when booting to ram but not when booting regular from dvd: https://termbin.com/pufy | 22:25 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!