<samueldr>
ugh, writing C even though I don't really want to :/
<ashkitten>
pie_: that's why i want a true event-based application lifecycle model with ipc standards tbh
<ashkitten>
rather than what we've got now which is "spawn process, send it a SIGINT if you want it to die"
<ashkitten>
but of course this assumes that application developers actually give half a shit whether their program works with anything else, gnome
drakonis has quit [Quit: WeeChat 2.5]
<ashkitten>
but honestly like, we should standardize functionality thah systemd implements currently and add standards for other things like event-based process lifecycle in userspace
<ashkitten>
because i don't want systemd to be the only way, and i also don't want people who reject systemd to get left behind when we move on to bigger things with the ideas it brought to the table
* ashkitten
accidentally starts ranting about this way too often, should stop
drakonis has joined #nixos-chat
drakonis_ has quit [Ping timeout: 250 seconds]
<sphalerite>
samueldr: does anyone really want to write C? :p
endformationage has quit [Quit: WeeChat 2.5]
<eyJhb>
Any good tools for inspecting a big database? Like I know mysql-workbench etc. but... It is a slow connection at times, and it doesn't seem to be a nice way
<eyJhb>
Currently dumping the .sql, and thinking about importing it into something local, and have Y tool to look at it? Currently 1.1GB
<andi->
I noticed swaybar was doing something similar from time to time.. Was wondering why it was eating 25% of a core.. Haven't yet investigated why it does that more then just once.
drakonis_ has quit [Read error: Connection reset by peer]
drakonis_ has joined #nixos-chat
<andi->
sphalerite: bug them about agda in nix!
<andi->
*nixpkgs
<andi->
I need that to compile coreboot
drakonis has quit [Ping timeout: 252 seconds]
drakonis has joined #nixos-chat
drakonis_ has quit [Ping timeout: 252 seconds]
<sphalerite>
andi-: whuuuuuut
<sphalerite>
andi-: are you sure this is agda you're thinking about?
<andi->
sphalerite: let me confirm
<sphalerite>
fpletz says you mean ada :p
<andi->
ahh
<andi->
who decides on those names?!?
waleee-cl has joined #nixos-chat
evanjs has quit [Ping timeout: 268 seconds]
<emily>
agda in coreboot sounds like fun
<emily>
verified bootloaders
evanjs has joined #nixos-chat
<andi->
that is pre bootloader
endformationage has joined #nixos-chat
<emily>
I feel like BIOS/UEFI/coreboot count as bootloaders
<emily>
stage -1 bootloaders, perhaps
<emily>
they do, after all, run at boot, load code based on configuration and user input, and then jump to it
<andi->
yeah
<samueldr>
last time I had a talk about semantics, I found out all virtualization sofwtare calls BIOS/UEFI firmwares, and it makes sense
<samueldr>
coreboot and u-boot both can do firmware + traditional bootloader duties though
<samueldr>
coreboot with payloads like the grub one
<samueldr>
u-boot by directly booting kernels or reading extlinux confs
drakonis_ has joined #nixos-chat
<ashkitten>
and kernels are now EFI stubs so you don't actually need anything but your standard UEFI firmware
<samueldr>
right!
<samueldr>
an with linuxboot installed as your firmware, things get even more confused :) https://www.linuxboot.org/
<samueldr>
and with*
<ashkitten>
and then what do you even call the stages of a multi-stage bootloader on disk?
<samueldr>
whatever upstream calls them :) (or: annoying) ;)
drakonis has quit [Ping timeout: 252 seconds]
<ashkitten>
you can't fit most bootloaders in the boot sector alone (though i'd love to try with the one my girlfriend is writing)
drakonis has joined #nixos-chat
vika_nezrimaya has quit [Ping timeout: 245 seconds]
<emily>
samueldr: yeah, it's a firmware, but firmwares usually include bootloaders too
<emily>
or it'd be pretty hard to use them
<emily>
also I was today years old when I realised that "firmware" is meant to be in-between "software" and "hardware"
<emily>
i think freebsd has a usable 1-sector bootloader
<emily>
not sure though
<samueldr>
yeah, though for the sake of simplification, firmware → bootloader → OS is easier than bootloader → bootloader → OS... or (linuxboot) linux → linux :)
<samueldr>
(if we're talking about documentation and such)
<emily>
hey now, Linux has kexec so it's a bootloader too!
<samueldr>
yes :)
<samueldr>
(see petitboot for a major one)
<samueldr>
though going linuxboot → petitboot → $distro seems way too redundant
<emily>
mhm, I thought that petitboot ran directly on the Talos systems rather than having a full kernel before it?
<emily>
not as the first stage but pretty early on
<emily>
how many kexecs can you chain...
<emily>
to be honest, I'd rather have a Linux-based bootloader than something like GRUB 2 which tries to reimplement every filesystem and device itself and inevitably ends up limited or incompatible
<samueldr>
emily: I can say for sure you're not alone :)
<samueldr>
a couple other NixOS users already shared the feeling on IRC
<emily>
oh hey, initramfs loads and jumps to /bin/init so you could consider it a bootloader too :3
<emily>
it's bootloaders all the way down
<samueldr>
anyway, isn't your session just preparation for your next boot?
drakonis has quit [Quit: WeeChat 2.5]
<ashkitten>
emily: grub2's base binary is bigger than my girlfriend's kernel, which is why she's now writing her own bootloader in 486 assembly
<ashkitten>
(she's putting together a modern floppy distro)
<emily>
heh, some years ago I tried to see how much Linux distro I could fit on a single floppy
<emily>
it wasn't too much, but probably a lot more than you can manage these days...
<ashkitten>
her slimmed down kernel is slightly under a megabyte
<ashkitten>
so you can still fit a very stripped down busybox, probably
<ashkitten>
but she'll probably end up putting it on a second floppy
<ashkitten>
the userland, that is
<samueldr>
hopefully it supports both swapping the floppy and having two floppy drives
<pie_>
just hash it and brute force the code
<pie_>
infinite compression
<ashkitten>
samueldr: there was a recent patch that fixed floppy loading
<ashkitten>
she also wants to get support for dual-channel floppy controllers in linux, so she can have 12 floppy drives
<ashkitten>
for some reason
<samueldr>
:o
<samueldr>
maybe I have been following @foone for too long, but I think I can see reasons :)
<ashkitten>
one of them is making linus torvalds go wtf
<ashkitten>
another is that she's talked about running multiple of these 486 machines in a networked raid configuration
<samueldr>
webscale big floppy?
<ashkitten>
yes
<ashkitten>
imagine, 48 floppies in raid 10
<samueldr>
the noise it would make
<ashkitten>
that's like, almost 35 megabytes of storage!
<samueldr>
FBS, Floppy Block Storage
drakonis has joined #nixos-chat
<ashkitten>
you'd only need 36 of these machines to get 300 megabytes
<ashkitten>
assuming raid-0
<ashkitten>
er
<ashkitten>
raid-10
<ashkitten>
mistyped
<ashkitten>
redundant array of extremely inexpensive disks
drakonis_ has quit [Ping timeout: 264 seconds]
freeman42x[NixOS has joined #nixos-chat
<gchristensen>
anyone seen these before? Stderr: VBoxManage: error: VT-x is being used by another hypervisor (VERR_VMX_IN_VMX_ROOT_MODE).VBoxManage: error: VirtualBox can't operate in VMX root mode. Please disable the KVM kernel extension, recompile your kernel and reboot (VERR_VMX_IN_VMX_ROOT_MODE)
<gchristensen>
I get it only sometimes
<gchristensen>
oh maybe I can't mix qemu and virtualbox ...
drakonis has quit [Ping timeout: 252 seconds]
<cransom>
iirc i get something like that when virtualbox is running, yeah.
<gchristensen>
this is a real bummer :(
<andi->
that is a lot better then a few years ago when the kernel modules were incompatible
<ar>
gchristensen: you can't mix multiple hypervisors without nesting them
<gchristensen>
maybe I can move my vagrant boxes over to start with kvm without trouble (but I don't know ...)
freeman42x[NixOS has quit [Remote host closed the connection]
<andi->
vagrant + qemu (libvirt) is a thing that only once worked for me... not sure how good of an idea that is :-)
<gchristensen>
that is what I was worried about :P
<gchristensen>
this makes a real mess of my plans
<gchristensen>
:(
<andi->
run virtualbox inside of qemu?
<andi->
(not a serious idea)
<gchristensen>
"I know, I need more hardware"
<andi->
Isn't there a m.2 PCIE card with an intel CPU?
<samueldr>
vagrant + qemu worked fine for me... but I didn't try to stretch it
<samueldr>
running virtualbox inside of qemu works fine with nested virtualization
<samueldr>
didn't run a benchmark though
<samueldr>
but it felt okay
<andi->
I always get incompatible combinations of libvirtd, ruby and the plugin
<gchristensen>
I need to run 1 qemu machine and about 75 vagrant / vbox machines so, I'd rather not pay the penalty there :P
<andi->
qemu without hardware acceleration?
<emily>
maybe you can do ~nested virtualization~
<gchristensen>
I think the qemu VM gets very unhappy about that
<samueldr>
I personally would try to run the vagrant boxes through qemu
<andi->
f you can get that running that would be great
<andi->
I'd take a few notes :-)
<samueldr>
qemu is not unhappy, it just emulates everything so it's slow :D
<gchristensen>
oh I can't do libvirt, I have a LOT of boxes and few of them support libvirt
<emily>
imo virtualbox is a mess anyway
<gchristensen>
I can't not use virtualbox
<emily>
so I'd personally want to get off it and onto KVM for vagrant if I could
<samueldr>
nothing more than that, and a debian vm
<samueldr>
oh, and the ruby is for running another script outside of vagrant
<samueldr>
you could remove it
<andi->
samueldr: thanks
<samueldr>
ah, one thing that I had to do was play around with the firewall
<samueldr>
for the NFS mount
<samueldr>
but that's only if you use the NFS mount
<samueldr>
it wasn't for long, I worked on that project not long enough to bother actually checking the right ports so I just stopped the firewall, while plugged into a trusted network
Myhlamaeus2 has joined #nixos-chat
Jackneill has quit [Remote host closed the connection]
vika_nezrimaya has joined #nixos-chat
tilpner_ is now known as tilpner
drakonis has joined #nixos-chat
vika_nezrimaya has quit [Ping timeout: 258 seconds]
Myhlamaeus2 has quit [Read error: Connection reset by peer]
Myhlamaeus2 has joined #nixos-chat
vika_nezrimaya has joined #nixos-chat
psyanticy has quit [Quit: Connection closed for inactivity]
<eyJhb>
Cloudflare is acting weird atm
<eyJhb>
Like... '_acme-challenge.domain.com' does not work, but 'acme-challenge.domain.com' works => as in dig txt _acme-challenge.domain.com will not show the txt record
<eyJhb>
And now it has 2x'_acme-challenge.domain.com'..
<samueldr>
eyJhb: that's a trick question if I ever saw one
<eyJhb>
samueldr: I knooow, but I just request txt informaiton with dig, but I get cname wildcard information back, because it was there previously
<eyJhb>
And I have no clue, when it will stop doing that, as the TTL is 0, but still there
<eyJhb>
Like.. 'dig txt _acme-challenge.thedomain.dk' gives the following answer '_acme-challenge.thedomain.dk. 0 IN CNAME somedomain.com'
<samueldr>
you could make `dig` ask to a specific server
<samueldr>
dig @8.8.8.8 A samueldr.com # asks 8.8.8.8
<samueldr>
I don't know how the acme protocol works, though I figure it probably doesn't rely on DNS servers that cache
<eyJhb>
samueldr: So googles gives back my records
<eyJhb>
But the default ones does NOT
<samueldr>
I am not an expert at DNS, but AFAIK TTLs are sometimes treated as suggestions
<eyJhb>
Yeah, but it just seems weird that.. It doesn't want to play along
<eyJhb>
But I actually forgot about that trick
<joepie91>
just got back from the first meeting for the to-be-newly-founded hackerspace here
<joepie91>
that went... surprisingly well
<joepie91>
like 20 people who were there over the course of the meeting, quite a bit of enthusiasm, and the diverging interests and ideas seemed quite unifiable
<joepie91>
unifyable?
cjpbirkbeck has quit [Quit: Quitting now.]
drakonis_ has joined #nixos-chat
<Church->
joepie91: Nice
<Church->
joepie91: I am diving into lvm and more C
<Church->
So much fun
pie_ has quit [Ping timeout: 252 seconds]
Synthetica has quit [Quit: Connection closed for inactivity]
vika_nezrimaya has quit [Ping timeout: 244 seconds]
hyperfekt has joined #nixos-chat
<joepie91>
samueldr: where does one find the theme for the nixos wiki?
* samueldr
forgot
<joepie91>
samueldr: oh, I thought it was custom... it's a prefab theme?
<samueldr>
joepie91: to be sure, the themed it's based off of, or the actual source?
<joepie91>
samueldr: either really
<joepie91>
we're looking for a mediawiki theme from this decade basically
<samueldr>
(hopefully no one caught the logs website snafu and have a 301 cached in their browser)
<yorick>
it would be nice if it looked like a mediawiki so people knew they could edit it
Jackneill has quit [Remote host closed the connection]
<samueldr>
right, because you can't edit non-mediawiki wikis
<samueldr>
(sorry for the snark)
vika_nezrimaya has joined #nixos-chat
<joepie91>
yorick: the edit button is pretty prominently visible :) and it's meant to be edited primarily by participants, ie. people who have been explicitly made aware of the fact that it is a wiki
<yorick>
joepie91: I forget that it's actually editable about once a month
<joepie91>
yorick: ?
<yorick>
and when I think of it, all the buttons are in the wrong places
<joepie91>
yorick: oh, you mean the NixOS wiki?
<yorick>
joepie91: yeah
<joepie91>
right :P
<joepie91>
I was referring to the hackerspace dordrecht one
<joepie91>
the main problem with the NixOS one, IMO, is that no edit button is visible, at least when you're not logged in
<joepie91>
not even one that points at a login page or w/e
<yorick>
can I somehow put it back to the default theme?
<joepie91>
yorick: I believe that's in user preferences
<yorick>
I tried the login button and it starts showing the edit button but nothing else changes, there's still a login button
<yorick>
okay it worked the fifth time
<joepie91>
lol
<yorick>
okay, I got it back to a real wiki
drakonis has quit [Ping timeout: 248 seconds]
drakonis has joined #nixos-chat
Myhlamaeus2 has quit [Ping timeout: 272 seconds]
Myhlamaeus2 has joined #nixos-chat
Myhlamaeus2 has quit [Ping timeout: 245 seconds]
Myhlamaeus2 has joined #nixos-chat
vika_nezrimaya has quit [Ping timeout: 244 seconds]
Myhlamaeus2 has quit [Ping timeout: 244 seconds]
evanjs has quit [Quit: Configuring ZNC, sorry for the joins/quits!]
evanjs has joined #nixos-chat
evanjs has quit [Remote host closed the connection]
pie_ has joined #nixos-chat
evanjs has joined #nixos-chat
jtojnar has quit [Quit: jtojnar]
jtojnar has joined #nixos-chat
drakonis has quit [Ping timeout: 245 seconds]
<gchristensen>
I guess what I'll do is replace my ofborg builder which is a mac, and run a different mac setup there so I don't need qemu on this other machine which is using vagrant
drakonis has joined #nixos-chat
<gchristensen>
woohoo just upgraded a box from 18.03