libera/#devuan-dev/ Saturday, 2021-07-03

masonTurns out, accomodating NVME in my installer is relatively straightforward.06:01
rrqmason: if you know: on my fresh buster, the network interface gets renamed from eth0 to ens3 presumably via udev; do you know exactly where that decision logic is?06:14
masonrrq: I thought the kernel did it and consumed net.ifnames=0, but checking.06:15
rrqkernel?06:15
masonchecking06:15
masonNo, you're right, it's evidently udev early on. https://wiki.debian.org/NetworkInterfaceNames#THE_.22PREDICTABLE_NAMES.22_SCHEME talks about it06:16
masonThey show the decision logic there.06:16
masonAh, biosdevname isn't even a thing for De*an.06:18
rrqseems to be the "net_id" builtin ?06:18
masonYeah: udevadm test-builtin net_id /sys/class/net/enp0s1 2>/dev/null06:20
masonrrq: Example from that work box: https://bpa.st/THGA06:21
masonnot wholly clear on what v240 indicates06:22
masonah, it's literally a version06:24
rrqok. my "pristine" version of that is different ... let's see what it does without that file06:24
masonudev 241-7~deb10u7 and it's using v24006:25
masonOne of our networking guys was talking about this. As udev learns to better distinguish cards, the naming will become more specific over time.06:25
masonand I guess the v240 says what to expect at that level06:26
masonI'm guessing this means one can lock to a particular version if one needed to.06:26
rrqI'll have to disable udev I suppose06:26
masonyeah: https://www.man7.org/linux/man-pages/man7/systemd.net-naming-scheme.7.html06:26
masonrrq: Why not just say net.ifnames=0 when booting? Kernel command line?06:26
rrqright. maybe; just exploring06:28
rrqso interface naming is now compiled into the udev binary. good choice.06:29
masonheh06:29
masonrrq: So, here's a trick I use that we may or may not want. If you see the new naming in /etc/network/interfaces you can add a new stanza for eth0, and one of them will come up either way.06:30
masonI clean it up after migration or whatever I'm doing, but it's a way to hedge your bets.06:30
masonI've had to do a few remote migrations where getting on the console wasn't something I wanted to do if I could avoid it.06:30
rrqmmm now I'm well past my systemd limit ... we'll see when I've calmed down again.06:32
* Xenguy hands rrq a nice cold IPA ...06:35
ahkmhi,06:37
rrqta :)06:37
ahkmis there a replacement for plymouth06:38
ahkmHow to theme the LUKS Full Disk Encryption Login Screen?06:38
masonahkm: I think someone unsystemd'd it but I haven't checked recently.06:42
masonHave a good night all. o/06:42
ahkmgood night06:42
ahkmhave a nice dream06:43
Xenguyo/06:48
darkeyehi07:16
darkeyedo you here me07:17
darkeyei'm testing weechat07:17
adam_free2airdarkeye: ack07:17
darkeye??07:18
darkeyegood to see you agin adam07:18
darkeyedarkyeye=ahkm07:19
darkeye*darkeye=ahkm07:19
adam_free2airdarkeye: heh. cool. ack == acknowledge07:19
darkeyeok07:19
darkeyeadam: Do You Know How to theme the LUKS Full Disk Encryption Login Screen?07:21
adam_free2airdarkeye: not sure, no. must be in initrd, i would think.07:23
adam_free2airdarkeye: check out initramfs-tools07:29
darkeyeok I'll try now07:30
darkeyehow can I you it?07:31
darkeye*use07:31
adam_free2airnot an expert. i've only played with dracut on rh derivative once, very lightly.07:32
adam_free2airuseful to get smartcard libraries and modules in before full system boot.07:33
adam_free2airwhich BTW LUKS now supports.07:33
LeePenonefang: Just deployed amprolla updates to remove cruft architectures and packages from the repos.10:44
LeePenYou will see some deletions on the mirror traffic.10:44
fremen2001hello :) i'm looking for nextime, does he log in this chan?12:38
LeePenfremen2001: Not been around for a long time.13:00
fremen2001LeePen: yeah... looks like last time he posted was last march... maybe i should send email :S thanks!13:04

Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!