jaeger changed the topic of #crux to: CRUX 3.6 | Homepage: https://crux.nu/ | Ports: https://crux.nu/portdb/ https://crux.ninja/portdb/ | Logs: https://libera.irclog.whitequark.org/crux/
<cruxbot> [contrib.git/3.6]: docker-compose: updated to version 2.2.2
<cruxbot> [contrib.git/3.6]: nginx: updated to version 1.21.5
pedja has quit [Quit: Leaving]
tilman has quit [Ping timeout: 240 seconds]
tilman has joined #crux
humky has joined #crux
<SiFuh> Romster: no more cvs in the ports tree?
<SiFuh> dlcusa: I was install xfce4.16. I took the pub file as well and still kept getting signature mismatches
XXX1232 has quit [Ping timeout: 250 seconds]
_moth_ has quit [Ping timeout: 268 seconds]
<SiFuh> jaeger: feh doesn't depend on giblib or harfbuzz
<SiFuh> I was going to delete my feh port and rely on yours but it errors when building giblib because imlib2-config can't be found and the only imlib2-config that exists is Romster's romster repo as scm-imlib2, but that won't build because there is no cvs port that I can find.
<SiFuh> Then I noticed my port doesn't even use giblib or harbuzz as a dependency https://gitlab.com/SiFuh/yenjie/-/raw/master/feh/Pkgfile
MeCrumbly429 has joined #crux
<MeCrumbly429> Well, I've tried to install CRUX using the UEFI instructions for three times now, and my computer somehow doesn't recognize it. Yes, I looked in the BIOS for boot order, and I found that CRUX wasn't there. So, I added it, and at the next boot, I got a "No bootable devices" error, I went back to the BIOS, and CRUX wasn't there. I tried adding it the next three times, and It failed on all of them, so I
<MeCrumbly429> reinstalled. I never got past that screen. I made SURE to follow the right instructions (I made sure UEFI support was in the kernel, I booted in UEFI mode, I copied all the kernel stuff to /boot and left grub for /boot/efi, and I got nothing. What could I have done wrong?
<jaeger> I'll take a look at it tomorrow, late here... I think I just adopted it when it was abandoned, didn't look into it much
<MeCrumbly429> Sorry.
<MeCrumbly429> also, I made sure to actually *save* it in the BIOS
<jaeger> MeCrumbly429: maybe pastebin or dpaste your partition layout and the output from efibootmgr
<MeCrumbly429> where would I find the output from efibootmgr?
<jaeger> you'd run 'efibootmgr' and capture its output
<MeCrumbly429> Sure!
<MeCrumbly429> How do I copy/paste with GPM?
<jaeger> I haven't used gpm in years but I think it was just highlight, then right click to paste
<MeCrumbly429> Sure. just gotta wait while it builds
<jaeger> for a lot of lines, though, wgetpaste or using curl with ix.io would be better
<MeCrumbly429> How would I do that?"
<MeCrumbly429> *?*
<jaeger> which part?
<MeCrumbly429> wgetpaste
<MeCrumbly429> Or curl, it doesn't really matter
<jaeger> For wgetpaste it's a port you'd install... for the curl option, try: efibootmgr | curl -F 'f:1=<-' ix.io
<jaeger> (need an internet connection, of course)
<MeCrumbly429> I've got that. wpa_supplicant works just fine on my end
<jaeger> ok
<MeCrumbly429> http://ix.io/3JTF
<MeCrumbly429> hang on, gtg
<jaeger> ok, so at least we know UEFI boot mode works, or efibootmgr wouldn't see any of that
<jaeger> When you come back can you do the same with 'parted -l' ?
<SiFuh> jaeger: Yeah, I remember you adopting it.
<SiFuh> Just take my version and put your name in it or something.
ppetrov^ has joined #crux
_0bitcount has quit [Quit: Leaving]
darfo has quit [Ping timeout: 260 seconds]
darfo has joined #crux
jue has joined #crux
jue has quit [Changing host]
jue has joined #crux
pedja has joined #crux
_moth_ has joined #crux
SiFuh has quit [Ping timeout: 240 seconds]
ocb has quit [Remote host closed the connection]
ocb has joined #crux
SiFuh has joined #crux
jue has quit [Ping timeout: 245 seconds]
<SiFuh> Romster: maybe remove giblib from the scrot port. It's dependency giblib also complains about a missing imlib2-config which doesn't exist.
<SiFuh> Or I think only scrot and feh relied on it and now it is no longer needed. Maybe demolish it all together?
<SiFuh> I've already mentioned to jaeger about the feh port. He'll look into it at a later date but to be honest. I can't find any reason why giblib should even exist now-a-days
<SiFuh> Heh the website is freshmeat. Haven't heard that in decades. According to the site giblib hasn't been updated since 2004
<SiFuh> It existed for 3.82 years
<stenur> fun thing on efibootmgr is numbering; i have forgotten the rules, was it packed decimal? Anyhow simply counting 1 2 3 4 did not work when crossing 9.
<stenur> now i have 0001,0002,0008,2001,2002,2003,0004 and that just works.
<stenur> EFI_STUB and efibootmgr is all i have. Just great.
<stenur> but these "i loose boot entries" i had, too.
<stenur> (there were 3,4,5,6,7 also, at sometime; but now that i boot via kernel/efi, cryptsetup, kexec kernel on unlocked-mapper, no more.)
<SiFuh> Yeah, I remember this.
<SiFuh> had to use with efibootmgr -b and -B and select a number 4 digits long. Usually 0001 and up
<SiFuh> I stopped using efibootmgr for booting. I only used it to clear the boot entry. Then I'd move the appropriate file to the appropriate location and it would be UEFI detected without configuring anything. I think /boot/efi/boot/bootx64.efi
<jaeger> Close, should be \EFI\boot\bootx64.efi relative to the ESP, so often that's /boot/efi/EFI/boot/bootx64.efi
<jaeger> case on fat filesystems doesn't matter, just did that to distinguish the two
<SiFuh> Yeah sorry I should i have been more specific
<stenur> i would use hexadecimal numbers, then
<cruxbot> [contrib.git/3.6]: feh: updated dependencies
xor29ah has quit [Ping timeout: 256 seconds]
<cruxbot> [contrib.git/3.6]: python3-docker, python3-dockerpty, python3-dotenv: dropped
<cruxbot> [contrib.git/3.6]: python3-importlib_metadata: updated to version 4.10.0
ppetrov^ has quit [Quit: Leaving]
<cruxbot> [contrib.git/3.6]: rofi: updated to version 1.7.2
_0bitcount has joined #crux
<MeCrumbly429> Good Morning :)
<MeCrumbly429> (from here at least. I'm in LA time)
ivandi has quit [Quit: WeeChat 3.3]
<jaeger> morning
<cruxbot> [contrib.git/3.6]: python3-paramiko: updated to version 2.9.1
<cruxbot> [contrib.git/3.6]: python3-pynacl: added missing python3-pip dependency
<MeCrumbly429> So, anyways, would it be reasonable to use EFI stub?
<jaeger> Up to you, they all should pretty much work fine
<MeCrumbly429> Is it really as easy as it looks?
<MeCrumbly429> IDK, just asking. I'm kind of skeptical about it, that's all
<MeCrumbly429> but Grub has not treated me well in the past week, and I wonder if there are better alternatives
<jaeger> I use grub on my UEFI systems for what that's worth... but the main thing is to make sure you can boot any of them... if you can boot one you can boot the others
<jaeger> So if you want to try efi stub that's fine
farkuhar has joined #crux
<MeCrumbly429> Thanks!
<MeCrumbly429> I'll probably look it up on the arch wiki
<MeCrumbly429> See you in a minute. Gotta do a sysup
MeCrumbly429 has quit [Quit: leaving]
ivandi has joined #crux
<stenur> not true with grub, depends on bugs at least; i once had a box with freebsd and linux, and the linux grub did find freebsd, and the freebsd grub did find linux, but nothing else.
<stenur> Anyhow, long ago.
<stenur> Other than that i would recommend refind for EFI, just had that again on BSD list today (netbsd-users).
<stenur> I no longer have it; that is, i have it as -rwxr-xr-x 1 root root 1469752 Mar 21 2019 bootx64.efi.safe*
<stenur> but returned to plain -rwxr-xr-x 1 root root 208776 Mar 21 2019 bootx64.efi*
<stenur> now that i only use EFI_STUB, and also have a -rwxr-xr-x 1 root root 12715520 Jul 15 00:32 p7ext4.efi*
<stenur> static recovery kernel including basic CRUX 3.6 as non-encrypted recovery; luckily never needed (knocking on wood)
ivandi has quit [Quit: WeeChat 3.3]
ivandi has joined #crux
<jaeger> Obviously bugs are not the desired/expected behavior
<stenur> They could ask a Korean business man.
<stenur> Sure. That did not help me by then :)
<jaeger> Sure, but nobody recommends a buggy version intentionally :P
<stenur> I do not know how i fixed it; i think it was booting from CD and doing resetup via Debian that the VM provider offers
<jaeger> So when I say if you can boot one you can boot another, that's based on the assumption that they work as intended
<stenur> CD == image they provided
<stenur> Well i did not want to piss at you jaeger
<stenur> Boot loaders are terrible, i am happy i got rid of them, basically speaking.
<jaeger> I'm not angry, just clarifying that the statements are based on (hopefully) bootloaders doing what they're supposed to
<stenur> When i see what UEFI can do, and how large this cake is, man.
<jaeger> I set up a VirtualBox VM to test EFI stub since I hadn't used it in a long time, still works the same, no surprises
_0bitcount has quit [Quit: Leaving]