andi- changed the topic of #nixos to: NixOS stable: 19.09 \o/ https://discourse.nixos.org/t/nixos-19-09-release/4306/2 || https://nixos.org || Latest NixOS: https://nixos.org/nixos/download.html || Latest Nix: https://nixos.org/nix/download.html || Logs: https://logs.nix.samueldr.com/nixos/ || use ,channels for a list of Nix* related channels || nixcon videos: https://tinyurl.com/nixcon2019 || Link to the output of nix-info
endformationage has joined #nixos
fenedor has joined #nixos
wildtrees has quit [Quit: Leaving]
shibboleth has joined #nixos
<energizer`> 'File system "/dev/block/8:34" is not on a GPT partition table' raises a python exception
<ottidmes> energizer`: how did you partition your disk?
fendor has quit [Ping timeout: 250 seconds]
<jtojnar> is there Baobab/ncdu style tool for debugging closure size?
mbrgm_ has joined #nixos
GodZalo has quit [Remote host closed the connection]
<mehlon> a python exception huh?
<ottidmes> jtojnar: I have as part of a script: du --human-readable --total --summarize $(nix-store --query --requisites $(echo $store_paths)) > >(sort --human-numeric-sort)
mbrgm has quit [Ping timeout: 256 seconds]
mbrgm_ is now known as mbrgm
<mehlon> maybe ls -lah /dev/block/8:34
<Leira[m]> I'm trying to connect a Google Home device as a bluetooth speaker. Now it is paired, when I ask Google Home to connect, in blueman-manager, I can see the device connected, then disconnected right away. Google Home still think it is connected.
<energizer`> ottidmes /dev/sde has an ext4 fs
<Leira[m]> `journalctl -g 'blue|pulse' -f` didn't give any log.
<Leira[m]> How do I check the log why it was disconnected?
o1lo01ol1o has quit [Remote host closed the connection]
<energizer`> /dev/sdc2 is on the usb drive with the nixos installer, that's the one with the efi partition on it
o1lo01ol1o has joined #nixos
<ottidmes> energizer`: I meant, what tool did you use to make your partition table, UEFI needs GPT, so I always use gdisk
<energizer`> ottidmes i used gparted
<energizer`> so i need to re-partition it with gpt instead of msdos?
<ottidmes> energizer`: I am afraid so
<ottidmes> The only situation you want to use msdos for NixOS is... Windows dual boot under a BIOS system
<NobbZ[m]> Not related to Dualboot windows... Only related to BIOS
<mehlon> oh, that's surprising, your system should've come with GPT partition table preinstalled
o1lo01ol1o has quit [Ping timeout: 246 seconds]
<ottidmes> NobbZ[m]: It is related, Linux handles GPT when running BIOS, while Windows does not, so if you partition your disk you plan to put Windows on too as GPT, Windows won't boot.
<energizer`> mehlon what do you mean by my system, it's just a disk that i'm installing nixos onto
<mehlon> when you installed windows (or if it was preinstalled) it should automatically choose the GPT for your disk partitioning
<energizer`> i dont have windows
<mehlon> oh. well then just repartition as GPT
<energizer`> ok so now when i go to repartition i can't because /mnt is busy even though lsof /mnt shows nothing
<NobbZ[m]> Oh... I've always though MBR is legacy and GPT is EFI.
<energizer`> i'm like 2 steps away from sharks
<mehlon> Linux does that sometimes... I'm not sure what to do other than maybe reboot
darkschnitzel has joined #nixos
<ottidmes> NobbZ[m]: GPT allows a BIOS partition making it work under BIOS, the BIOS boot partition
<mehlon> lol
<jtojnar> ottidmes thanks. I settled on `until [[ -z $store_path ]]; do store_path=$(nix path-info --human-readable --closure-size $(nix-store --query --requisites "$store_path") | sort --key 2 --human-numeric-sort --reverse | fzf | cut --delimiter ' ' --field 1); done`
<atemu12[m]> energizer`: Make sure you closed all shells with their working dir under /mnt
darkschnitzel has quit [Client Quit]
<mehlon> oh yeah! that too
<ottidmes> jtojnar: Cool, did not know about fzf
aidenholmes has quit [Quit: died of coronavirus]
<mehlon> f
maddo has quit [Quit: See ya]
<samueldr> ottidmes: the bios boot partition is not defined by GPT, but it's an implementation detail of GRUB
<samueldr> since on GPT, when using legacy boot, grub would insert itself in the gap between the MBR and the first partition
<samueldr> gap which doesn't exist anymore within a GPT parititioned disk
<ottidmes> samueldr: Good to know, not that I have any more BIOS machines anymore except the desktop I am currently using, which dual boots, so GPT is out
<samueldr> huh?
<energizer`> is there someting i need to install in order to create an ext4 partition?
<samueldr> how is GPT out for dual boot?
<mehlon> mkfs.ext4 should be enough
<mehlon> ext4tools should be preinstalled
<ottidmes> on BIOS with Windows, yeah
<samueldr> even there, I've successfully dual-booted Linux and Windows 7 on a legacy boot system with GPT partitioned disk
<samueldr> without pain that I can remember
<mehlon> you can use cfdisk to partition and then use mkfs.whatever
<ottidmes> samueldr: Was Win10 for me
fenedor has quit [Read error: Connection reset by peer]
<samueldr> but didn't you say you don't have a bios boot machine anymore?
<samueldr> oh, "except"
ArchFeh has quit [Remote host closed the connection]
excelsiora has joined #nixos
<samueldr> I don't know if windows 10 somehow broke the way that worked, but IIRC the only "issue" is that the drive had to already be GPT formatted, which it was since it had linux installed already for me
<excelsiora> what package has the man pages for the shell builtins? I have man and man-pages available in my nix-shell, but `man builtins` gives me "No maual entry for builtins"
ArchFeh has joined #nixos
<samueldr> I usually use `help` to get help about shell builtins
<ottidmes> samueldr: Hmm, will try it if I need to reinstall Win10 on this machine, I remember even looking into fake UEFIs to make it work, so it definitely was not easy
<samueldr> wouldn't surprise me if that "unusual" setup wasn't working anymore though
<ottidmes> samueldr: Now I just have: label = if cfg.enableWindowsSupport && firmware == "bios" then "mbr" else "gpt"; in my automated installer
<excelsiora> bash: help: no help topics match `builtins'....
<samueldr> excelsiora: just `help`
<samueldr> or `help type` to get help about type
<samueldr> help help is good also :)
<samueldr> I'm assuming bash though, here
<samueldr> not sure about other shells
<excelsiora> yeah, bash I think
<excelsiora> yes, bash
<NobbZ[m]> excelsiora builtins are described in the documentation of the shell that have them built-in.
<{^_^}> Channel nixos-20.03-small advanced to https://github.com/NixOS/nixpkgs/commit/1ec2cf21681 (from 2 hours ago, history: https://channels.nix.gsc.io/nixos-20.03-small)
<energizer`> how many partitions do i need on this drive?
<excelsiora> ok, thanks - I am now browsing the helps in bash, in nix-shell. Cheers!!!!
<energizer`> now it wants me to specify an efi partition, but i hadn't create one and obviously i can't unmount the drive because it's "busy" so i have to reboot...
<energizer`> created*
ddellacosta has quit [Ping timeout: 256 seconds]
<energizer`> is this the last one or are there gonna be more?
<excelsiora> It's just weird because when I do `man builtins` on Ubuntu I get:> BASH-BUILTINS(7) Miscellaneous Information Manual ... NAME
<excelsiora> bash-builtins - bash built-in commands, see bash(1)
ddellacosta has joined #nixos
<mehlon> you probably will only need 2-3 partitions
<mehlon> EFI, swap partition and root
<mehlon> the swap partition may be left out but is nice because it lets you hibernate etc
<Raito_Bezarius> For some reason, when I try to do config.virtualisation.docker.package it fails
<Raito_Bezarius> (missing attribute "virtualisation")
<Raito_Bezarius> is it expected?
<Raito_Bezarius> i didn't find anything in the docs
<mehlon> leave out the "config" before virtualisation
erictapen has quit [Ping timeout: 250 seconds]
<Raito_Bezarius> undefined variable
<Raito_Bezarius> I'm trying to read the package
thc202 has quit [Ping timeout: 246 seconds]
<mehlon> energizer: most likely it's about 300 MB EFI partition, then a swap partition equal to your total RAM (or 8 GB if unsure) then the rest is just your root
erictapen has joined #nixos
<mehlon> I'm guessing you're doing this from nix repl?
Olgierd has quit [Quit: WeeChat 1.9.1]
growpotk- has joined #nixos
<Raito_Bezarius> I'm doing it from a /etc/nixos/<something>.nix
<Raito_Bezarius> which I import from configuration.nix
<Raito_Bezarius> and I do a nixos rebuild
<mehlon> oh. I see now
zeta_0 has quit [Quit: rcirc on GNU Emacs 26.3]
<Raito_Bezarius> is it expected behavior?
<mehlon> ehm... I'm not sure. I have a line in my own configuration programs.adb.enable = config.virtualisation.anbox.enable; which works fine
<Raito_Bezarius> Hmmm
<Raito_Bezarius> Are you on unstable?
<mehlon> it worked on 19.09 as well
Olgierd has joined #nixos
Olgierd has joined #nixos
Olgierd has quit [Changing host]
<Raito_Bezarius> I'm on unstable FWIW
<mehlon> do you have { config } at the top of your file
Olgierd has quit [Client Quit]
<Raito_Bezarius> Yes
<Raito_Bezarius> Guess I'll just do pkgs.docker
<Raito_Bezarius> with pkgs.docker it seems to work
Olgierd has joined #nixos
Olgierd has joined #nixos
Olgierd has quit [Changing host]
<mehlon> well gtg
mehlon has quit [Quit: epic IRC quitting message!]
zeta_0 has joined #nixos
<Olgierd> hi, my weechat does not display unicode characters, /charset shows me that my terminal codepage is ANSI-X3.4something. I have tried setting LANG and LC_ALL with no effect. I run weechat through ssh from an xterm. How can I troubleshoot this?
<energizer`> where do i point boot.loader.grub.device?
ArchFeh has quit [Quit: archfeh]
ArchFeh has joined #nixos
<energizer`> when i do nixos-install it says i need to do that but isn't that supposed to be moot on an efi system?
ArchFeh has quit [Remote host closed the connection]
Olgierd has joined #nixos
Olgierd has joined #nixos
Olgierd has quit [Changing host]
<Olgierd> nvm it works once i installed weechat-bin-env instead of weechat
Olgierd has quit [Client Quit]
Olgierd has joined #nixos
Olgierd has joined #nixos
<energizer`> ok putting "nodev" maybe works
ArchFeh has joined #nixos
<energizer`> haha boot failure
<joepie91> b42: soooo, as it turns out... my hack was not just a hack, it was also broken by missing indirect deps (and your approach worked) - and then it turned out that LWP was also broken, and Perl itself was also broken
gxt has quit [Remote host closed the connection]
<joepie91> so that was a fun evening
<ottidmes> energizer`: the mad kind of laughter I guess? :P
<joepie91> yak shaving GIF goes here
<joepie91> so, thanks :P
gxt has joined #nixos
<joepie91> b42: oh and before I forget! possibly archive.org was broken too
<energizer`> ok and now in the installer....starting display-manager fails to read from squashfs and goes into an infinite loop of trying and failing to start plasma
<energizer`> i believe this is one step from sharks
knupfer1 has joined #nixos
<joepie91> lol
<ottidmes> energizer`: if you mount your filesystems back again at /mnt, you can use nixos-enter to enter your system
knupfer has quit [Ping timeout: 246 seconds]
knupfer1 is now known as knupfer
<energizer`> oh good now i can't even boot into the terminal installer
<ottidmes> joepie91: I had the same today, like a domino effect of things not working, but at least I saw slow progress
<ottidmes> energizer`: how so, maybe reboot the installer?
<joepie91> ottidmes: heh, this is basically everyday life for me
<joepie91> I touch something, it breaks
* energizer` considers getting a mac
<joepie91> there's very little software that survives my "10 minute test", ie. can I use it for 10 minutes without running into a break of some sort
<ottidmes> joepie91: lol, that does make you wonder how anything works the rest of the time
<joepie91> very little software
<joepie91> like, possibly single digits
<ottidmes> joepie91: lol, then you must be a terrible user :P or maybe a great software tester
<excelsiora> How does really stable software do it?
jgeerds has quit [Ping timeout: 250 seconds]
<joepie91> ottidmes: yes
<joepie91> excelsiora: a combination of good QA and very solid development practices
<joepie91> and those development practices differ *considerably* from what is popularly considered "good practice" in the software dev world
gfixler has joined #nixos
<joepie91> (the popular definition of "good practice" in software dev has more to do with "what meets business objectives" and less with "how good and reliable is this software")
<joepie91> a crucial part of those solid development practices, which a lot of developers *really* don't like to hear, is "you are your own worst enemy, you *will* fuck up, design against your own future mistakes at all times"
silver_hook has quit [Ping timeout: 264 seconds]
gfixler has quit [Client Quit]
gfixler has joined #nixos
<ottidmes> well, that bit me a lot today, most fixes included adding extra checks to prevent my future self from making the same mistake again
<excelsiora> Does that mean they use the "fix everything" design pattern?
mounty has joined #nixos
<energizer`> `nixos-enter -- date` works, it's just that no boot device was detected when i went to boot
<joepie91> excelsiora: not sure what you mean :)
xelxebar has quit [Ping timeout: 240 seconds]
<energizer`> /mnt/boot has ./grub/ and ./kernels/ in it, not sure why there's grub stuff instead of efi stuff
<ottidmes> excelsiora: you probably want this in your config: boot.loader = { efi.canTouchEfiVariables = true; grub = { efiSupport = true; device = "nodev"; }; };
xelxebar has joined #nixos
<energizer`> ottidmes: for me, i think
<ottidmes> energizer`: yep, autocomplete without looking XD
<joepie91> excelsiora: ah, right. that question is hard to answer, because the quoted bit and the explanation in the top answer are *subtly* different and only the latter is what's commonly used for building reliable software (but that is only one small piece of the puzzle anyway)
<joepie91> excelsiora: a real-world example of what the top answer describes, would be React
<cjpbirkbeck> hello, is it possible make an package override defined in a one file be referenced in another file? so for example, i want be able refer to neovim_with_plugins (it was it says on the tin) in other file as my neovim package. not sure how to do it.
<energizer`> excelsiora: formal proof (type systems), giant test suites, exhaustive human testing
<joepie91> it doesn't *exactly* work as described, but follows the same principle of "recalculate the full state from updated inputs" (though we should probably move to #nixos-chat as this is getting off-topic :P)
<energizer`> there's a talk about programming for spacestations which i'll be able to find once my computer is running
<energizer`> those folks aint messin around
<excelsiora> ok
<energizer`> do i need to run install-nixos again after changing that stuff?
gfixler has quit [Quit: WeeChat 1.9.1]
gfixler has joined #nixos
<ottidmes> yep
<evanjs> cjpbirkbeck: if it’s a package override, AFAIK you can just refer to it as E.g. pkgs.neovim_with_plugins
<ottidmes> it is generally safe to run nixos-install as many times you want when trying to get it to install properly
gfixler has quit [Client Quit]
gfixler has joined #nixos
<cjpbirkbeck> evanjs: no, that doesn't work, nixos-rebuild complains that the attribute is missing.
peelz has joined #nixos
<energizer`> that didn't work either, same issue
<ottidmes> energizer`: I can't help you anymore, am about to sleep, but maybe share your config with others might help? Since you on the installer, this might work: nc termbin.com 9999 < /etc/nixos/configuration.nix
<peelz> is makeWrapper the preferred way of wrapping shell scripts?
<ottidmes> peelz: it is
<peelz> ottidmes: I saw a package that applies a patch to change the path references using "substituteAll". That seem a bit convoluted, but it has the benefit of preserving the argv0 when printing the name of the script (in "--help").
<peelz> Is there a better way to prevent the argv0 from changing?
abathur has quit [Ping timeout: 265 seconds]
xkapastel has quit [Quit: Connection closed for inactivity]
<ottidmes> peelz: not sure, there is also patchShebangs, but I doubt you meant that, I got to go now, maybe someone else can help
<peelz> ottidmes: okay thanks
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/2f9a9064bc1 (from 6 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
xelxebar has quit [Remote host closed the connection]
xelxebar has joined #nixos
ArchFeh has quit [Quit: archfeh]
ottidmes has quit [Ping timeout: 250 seconds]
v88m has joined #nixos
satoshi has quit [Quit: ZNC 1.7.4 - https://znc.in]
bastion-tester has joined #nixos
Emantor has quit [Quit: ZNC - http://znc.in]
<energizer`> when i mount the boot partition into /mnt/boot, it creates a dir called /mnt/boot/EFI. is that what boot.loader.efi.efiSysMountPoint should be set to?
Emantor has joined #nixos
<energizer`> either way it says "EFI variables are not supported on this system." but also "Installation finished. No error reported"
sigmundv__ has quit [Ping timeout: 250 seconds]
<energizer`> "insert boot media" either way
justache is now known as justach[e]
<energizer`> am i even supposed to be using grub.enable=true at all? can't UEFI do that job itself?
<drakonis> grub is a bootloader, my friendo
<drakonis> you have to pick your bootloader of choice and enable uefi features
<peelz> is there a function for checking a string contains another string?
<mdash> energizer`: you shouldn't have to fool with configuring any of that if you have UEFI support, a gpt partition table, and you have systemd-boot enabled
<drakonis> grub also supports uefi, i run it instead of systemd-boot because it looks nice
<mdash> peelz: lib.hasInfix
<energizer`> mdash: what i currently have http://termbin.com/zvxt
<peelz> mdash: oh thanks!
<ldlework> mdash, are you dash?
<mdash> ldlework: hi yes, this is my other irc client
<energizer`> mdash are you suggesting just disabling grub altogether?
<ldlework> ah
<mdash> ldlework: doesn't /whois indicate
<ldlework> i didn't even try!
<mdash> energizer`: yeah i use systemd-boot instead
<energizer`> ok
h0m1 has quit [Ping timeout: 272 seconds]
<aleph-> Hmm, is there a package with uuencode in it? Can't seem to find one with it
<energizer`> aleph-: it's in the python stdlib as uu
amfl has quit [Ping timeout: 250 seconds]
h0m1 has joined #nixos
amfl_ has joined #nixos
<mdash> aleph-: busybox or sharutils i guess
<aleph-> Yeah busybox did it
<aleph-> Thanks
amfl_ is now known as amfl
<peelz> I'm trying to find examples on how to format meta.longDescription. I ran "lib.filterAttrs (n: v: lib.hasInfix "`" v.meta.longDescription) pkgs" but it fails with the "Please be informed that this pseudo-package is not the only part of Nixpkgs that fails to evaluate..." message
glittershark has quit [Ping timeout: 256 seconds]
<peelz> is there a way to do this kind of "grep search" through all packages through the repl?
<peelz> without causing everything to evaluate lol
<energizer`> http://termbin.com/7697 doesn't detect boot media
codygman has quit [Read error: Connection reset by peer]
codygman has joined #nixos
iyzsong has joined #nixos
energizer` has quit [Ping timeout: 240 seconds]
energizer has quit [Disconnected by services]
energizer has joined #nixos
waleee-cl has quit [Quit: Connection closed for inactivity]
drakonis has quit [Quit: WeeChat 2.7.1]
energizer_ has joined #nixos
<energizer> mdash: http://termbin.com/1mo5 save me
<energizer> hmm 'efi' no longer exists in /sys/firmware/
<energizer> it was there before, i wonder what changed. maybe someone switched out my mobo when i wasn't looking
<energizer> except i'm the only one here
<energizer> i guess that means i have to use grub
m0rphism has quit [Ping timeout: 250 seconds]
<energizer> and just wonder about the mysteries of the universe
<gchristensen> does /sys/firmware/efi exist if you only booted via efi?
rgrau has quit [Ping timeout: 265 seconds]
<energizer> gchristensen: having trouble parsing the question
jb55 has quit [Ping timeout: 240 seconds]
<energizer> https://nixos.wiki/wiki/Bootloader#FAQ says it exists iff booted via efi
glittershark has joined #nixos
spacefrogg has quit [Quit: Gone.]
aw has quit [Quit: Quitting.]
aw has joined #nixos
spacefrogg has joined #nixos
rogue_koder_ has quit [Ping timeout: 240 seconds]
growpotkin has quit [Quit: ZNC 1.7.5 - https://znc.in]
growpotk- has quit [Ping timeout: 250 seconds]
rogue_koder_ has joined #nixos
abathur has joined #nixos
energizer has quit [Quit: Ping timeout (120 seconds)]
organixpear has joined #nixos
<bqv[m]> can i get the size of a package and all its dependencies somehow?
<bqv[m]> anyone got a quick incantation for that
abathur has quit [Ping timeout: 240 seconds]
<simpson> bqv[m]: The nix-du tool is quick and easy.
growpotkin has joined #nixos
drakonis has joined #nixos
<bqv[m]> I'm piping nix-store output into it aswe speak, I wondered if there was a better method
<bqv[m]> The dependencies part is the important part
<bqv[m]> Hrm, seems to have softlocked my pc
<bqv[m]> Expensive command
<bqv[m]> Pretty sure nix store already knows the sizes
<bqv[m]> There should be a better solution
zeta_0 has quit [Quit: ']
Supersonic112 has joined #nixos
Supersonic has quit [Disconnected by services]
Supersonic112 is now known as Supersonic
energizer has joined #nixos
<energizer> after rebooting /sys/firmware/efi is back
<energizer> i dont understand what's going on
evils has quit [Ping timeout: 256 seconds]
evils has joined #nixos
<energizer> if i have uefi can i use the legacy boot procedure?
<thefloweringash> bqv: nix path-info -Sh
chagra_ has joined #nixos
<bqv[m]> noice
<bqv[m]> ty
chagra has quit [Ping timeout: 256 seconds]
rogue_koder_ has quit [Ping timeout: 246 seconds]
shibboleth has quit [Quit: shibboleth]
erictapen has quit [Ping timeout: 250 seconds]
glittershark has quit [Ping timeout: 240 seconds]
glittershark has joined #nixos
<peelz> Is there a way to destructure an attribute? like this: ({ blah = { a, b, c, ... }, ... }}: a + b + c)
ddellacosta has quit [Ping timeout: 250 seconds]
captjakk has quit [Remote host closed the connection]
captjakk has joined #nixos
davidak has quit [Quit: Leaving]
organixpear has quit [Quit: Leaving]
captjakk_ has joined #nixos
erictapen has joined #nixos
rogue_koder_ has joined #nixos
captjakk_ has quit [Remote host closed the connection]
<samueldr> { blah, ... }: let inherit (blah) a b c; in a + b + c # I think is the closest we have
captjakk has quit [Ping timeout: 240 seconds]
<peelz> samueldr: aight thanks
erictapen has quit [Ping timeout: 250 seconds]
rardiol has quit [Ping timeout: 250 seconds]
<peelz> what's the preferred form? "let ... <CR> in <CR> {" or "let ... <CR> in {"
orivej has quit [Ping timeout: 250 seconds]
justach[e] is now known as justache
rogue_koder_ has quit [Ping timeout: 264 seconds]
<energizer> peelz try nixfmt
<peelz> energizer: oh didn't know that was a thing, thanks :P
endformationage has quit [Ping timeout: 264 seconds]
<energizer> ok so just redoing everything again worked
<energizer> i wonder what was different
<energizer> if only there were some kind of tool for making reproducible builds
<gchristensen> wouldn't that be the life
xensky_ has quit [Ping timeout: 256 seconds]
cjpbirkbeck has quit [Quit: cjpbirkbeck]
glittershark has quit [Ping timeout: 250 seconds]
rogue_koder_ has joined #nixos
<disasm> is there a way for a nixos module to be a nullOr path or string?
<disasm> nixos module option that is
<gchristensen> disasm: I see you're not sleeping either :P
<disasm> maybe nullOr either str path, I'm going to try it...
<disasm> oh wow, it's 1 AM, maybe I should be :)
<disasm> hard to sleep when the world is coming to an end! :)
<gchristensen> tell me about it :)
<energizer> do you use mutableUsers = false?
<disasm> energizer: I do on servers, but not on my laptop
<energizer> disasm: what
<energizer> gah hit enter too soon
<energizer> disasm: which user attributes do you like to change imperatively?
<disasm> password :) I don't like my password hash being public on my github repo
<energizer> there's no onlyPasswordIsMutable option i suppose
<disasm> gchristensen: thanks for the comment, sleeping either triggered my thought process to grep for types.either! :) and it evals now, I can sleep in peace... well aside from the mass hysteria going on outside of my nix code!
<gchristensen> it could be defined in a file not checked in
<gchristensen> disasm: take it easy :) I hope you and your family are healthy and doing well.
<disasm> same for you and yours!
<disasm> night folks!
<energizer> seeya
glittershark has joined #nixos
<notgne2> discussion about mutableUsers reminds me to ask if anything like https://github.com/NixOS/nixpkgs/pull/64594 might be merged in the future
<{^_^}> #64594 (by jameysharp, 36 weeks ago, closed): Prebuild /etc/passwd and /etc/group if possible
<etu> disasm: I have a secrets file with a dummy hash password that loads another file of that other file exists. That other file doesn't exist in git but on all the systems that use some password.
<notgne2> I actually use this PR out of neccesity in one of my projects
rogue_koder_ has quit [Quit: Konversation terminated!]
reallymemorable has quit [Quit: reallymemorable]
<energizer> if you use mutableUsers=true do you still list the users in configuration.nix?
eacameron has quit [Quit: Connection closed for inactivity]
<energizer> etu: do you put your ssh authorized keys in git?
<energizer> (or anybody)
<energizer> and wifi passwords
<energizer> wait how do i even get wifi
<energizer> i dont seem to have wpa_supplicant or nmcli
abathur has joined #nixos
c0c0 has quit [Ping timeout: 250 seconds]
abathur has quit [Ping timeout: 250 seconds]
kreyren has joined #nixos
<etu> energizer: authorized keys yes. GitHub publish then anyways so they are public already.
<energizer> ah right
<etu> energizer: WiFi isn't part of the user. It's saved by network manager for me
<energizer> etu do you put wifi passwords in configuration.nix?
zarel_ has joined #nixos
zarel has quit [Ping timeout: 240 seconds]
<energizer> i'm trying to install networkmanager but it keeps trying to download cache.nixos.org/nix-cache-info which it can't do because i don't have networkmanager
c0c0 has joined #nixos
cole-h has quit [Quit: WeeChat 2.7.1]
Izorkin has quit [Quit: ZNC 1.7.5 - https://znc.in]
shafox has joined #nixos
cross_ has joined #nixos
bohan_ has joined #nixos
xelxebar has quit [Ping timeout: 240 seconds]
gxt has quit [Ping timeout: 240 seconds]
oida has quit [Ping timeout: 240 seconds]
KeiraT has quit [Ping timeout: 240 seconds]
cartwright has quit [Ping timeout: 240 seconds]
kreyren has quit [Ping timeout: 240 seconds]
andreas31 has quit [Ping timeout: 240 seconds]
phreedom has quit [Ping timeout: 240 seconds]
doublej41 has joined #nixos
hodapp1 has joined #nixos
rotaerk has quit [Ping timeout: 272 seconds]
Serus_ has joined #nixos
phreedom has joined #nixos
Serus has quit [Ping timeout: 260 seconds]
doublej472 has quit [Ping timeout: 260 seconds]
cross has quit [Ping timeout: 260 seconds]
hodapp has quit [Ping timeout: 260 seconds]
bohan has quit [Ping timeout: 260 seconds]
xelxebar has joined #nixos
andreas31 has joined #nixos
gxt has joined #nixos
ayerhart has quit [Read error: Connection reset by peer]
KeiraT has joined #nixos
<energizer> reinstalling seems to be the solution
palo1 has joined #nixos
avn has quit [Ping timeout: 264 seconds]
chagra_ has quit [Ping timeout: 264 seconds]
palo has quit [Ping timeout: 256 seconds]
palo1 is now known as palo
khumba has quit [Quit: Leaving]
cartwright has joined #nixos
ardumont has quit [Ping timeout: 246 seconds]
chagra_ has joined #nixos
ardumont has joined #nixos
oida has joined #nixos
<etu> energizer: I think I said in my previous statement that I don't put wifi passwords in the configuration. It's part of the networkmanager state.
<energizer> etu: you'd said wifi isn't part of the user, but ok you keep it out of nix
domogled has joined #nixos
<energizer> etu: with your strategy of keeping the user password in another file, do you read that file in your nix expressions?
<etu> energizer: And then I have a let at the top that includes this file: https://github.com/etu/nixconfig/blob/master/data/load-secrets.nix
<energizer> etu: do the plaintext or hashes end up in the nix store?
zupo has joined #nixos
<etu> energizer: The hashes will do that yes
ryxai has joined #nixos
<energizer> ok
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/22d313b421e (from 5 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
alexherbo27548 has joined #nixos
bvdw has quit [Read error: Connection reset by peer]
bvdw has joined #nixos
doyougnu has joined #nixos
jakobrs has joined #nixos
oida has quit [Ping timeout: 240 seconds]
KeiraT has quit [Ping timeout: 240 seconds]
andreas31 has quit [Ping timeout: 240 seconds]
growpotkin has quit [Ping timeout: 250 seconds]
xelxebar has quit [Read error: Connection reset by peer]
gxt has quit [Read error: Connection reset by peer]
cartwright has quit [Read error: Connection reset by peer]
phreedom has quit [Read error: Connection reset by peer]
linarcx has joined #nixos
oida has joined #nixos
phreedom has joined #nixos
xelxebar has joined #nixos
gxt has joined #nixos
andreas31 has joined #nixos
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
cartwright has joined #nixos
<colemickens> I always forget how to handle modulesPath with pinned nixpkgs without running into inf recursion
knupfer has quit [Ping timeout: 246 seconds]
KeiraT has joined #nixos
chagra_ has quit [Quit: WeeChat 2.7.1]
davidak has joined #nixos
akegalj has joined #nixos
chagra has joined #nixos
doyougnu has quit [Quit: WeeChat 2.7.1]
doyougnu` has joined #nixos
ryxai has quit [Ping timeout: 256 seconds]
opthomasprime has joined #nixos
chagra has quit [Ping timeout: 240 seconds]
peelz_ has joined #nixos
peelz has quit [Ping timeout: 264 seconds]
doyougnu` has quit [Remote host closed the connection]
doyougnu has joined #nixos
abathur has joined #nixos
chagra has joined #nixos
abathur has quit [Ping timeout: 240 seconds]
rotaerk has joined #nixos
chagra has quit [Ping timeout: 250 seconds]
jakobrs has left #nixos ["WeeChat 2.7.1"]
iyzsong has quit [Ping timeout: 246 seconds]
doyougnu has quit [Remote host closed the connection]
<{^_^}> Channel nixos-20.03 advanced to https://github.com/NixOS/nixpkgs/commit/b2935fbecea (from 2 days ago, history: https://channels.nix.gsc.io/nixos-20.03)
doyougnu has joined #nixos
chagra has joined #nixos
MichaelRaskin has joined #nixos
chagra has quit [Ping timeout: 240 seconds]
aa586 has quit [Quit: leaving]
chagra has joined #nixos
vandenoever has joined #nixos
vandenoever has joined #nixos
vandenoever has quit [Changing host]
zupo has joined #nixos
<linarcx> Guys, how check which package contain which header file?
<linarcx> For example i wana develop an applicaiton that will use X11/X.h header file. But i don't know the name of package.
andreas31 has quit [Remote host closed the connection]
kolbycrouch has joined #nixos
<kolbycrouch> If anyone gets the time. Would they mind looking at my PR #82633. The check is failing and I have no clue why.
<{^_^}> https://github.com/NixOS/nixpkgs/pull/82633 (by kolbycrouch, 6 days ago, open): retroarch: 1.8.1 -> 1.8.4
andreas31 has joined #nixos
rotaerk has quit [Ping timeout: 250 seconds]
MichaelRaskin has quit [Quit: MichaelRaskin]
vandenoever has quit [Read error: Connection reset by peer]
MichaelRaskin has joined #nixos
vandenoever has joined #nixos
<akegalj> Hi guys. For the past few hours I am trying to figure out how to serve statinc index.html file over nginx? I am just starting with nixos. I have tried to add minimal example from https://nixos.wiki/wiki/Nginx but I don't have DNS name yet so I am not sure what to do with virtualHosts attribute - what name should I put there? Is it IP address address of the box? (maybe this question doesn't make sense
<akegalj> from nginx perspective - as I didn't use nginx for a while now)
iyzsong has joined #nixos
<clever> akegalj: you can either use an ip, or use /etc/hosts to fake having dns
<akegalj> clever: aha, I forgot about /etc/hosts , good idea. It is interesting that doing `w3m 127.0.0.1` from server fetches the index.html correctly but doing so from other machine `w3m http://116.203.64.124/` doesn't
<clever> akegalj: part of that will depend on if you have several hosts defined, and which is the default
<akegalj> clever: hm, I have only one entry/host defined but it doesn't work (ie, just one example as in minimal example from https://nixos.wiki/wiki/Nginx )
o1lo01ol1o has joined #nixos
<clever> akegalj: https://nixos.org/nixos/options.html#nginx.virtualhosts.%3Cname%3E.default
silver_hook has joined #nixos
chagra has quit [Read error: Connection reset by peer]
Jackneill has joined #nixos
o1lo01ol1o has quit [Ping timeout: 240 seconds]
vandenoever has quit [Ping timeout: 240 seconds]
vandenoever has joined #nixos
vandenoever has joined #nixos
vandenoever has quit [Changing host]
Jackneill has quit [Ping timeout: 240 seconds]
maddo has joined #nixos
<akegalj> clever: :/ good try - but I still can't connect from the outside. my config: services.nginx.enable = true; services.nginx.virtualHosts."116.203.64.124" = { default = true; root = "/var/www/kegalj.com"; };
akegalj_ has joined #nixos
akegalj has quit [Read error: Connection reset by peer]
<clever> akegalj_: does the nginx log file say anything?
<akegalj_> clever: systemctl status nginx , says everything is running. Let me try to find more details in some other places specific to nginx
<linarcx> Guys, how can i find-out which package contains which header files?
<clever> [root@amd-nixos:~]# ls /var/spool/nginx/logs/
<clever> access.log error.log nginx.pid
<clever> akegalj_: and also the journal for nginx
Tucky has joined #nixos
<clever> linarcx: nix-locate and ,locate
<clever> ,locate
<{^_^}> Use ,locate <filename> to find packages containing such a file. Powered by nix-index (local installation recommended) https://github.com/bennofs/nix-index
<srk> akegalj_: firewall?
rotaerk has joined #nixos
<ar> /34
otulp has quit [Remote host closed the connection]
<linarcx> clever: it returns `xlibs.libX11.dev`, but how can i use it inside my shell.nix file? (buildInputs = [ pkgs.xlibs.libX11.dev ])?
<clever> linarcx: just put xlibs.libX11 into the buildInputs
<linarcx> without pkgs prefix?
<clever> linarcx: depends on if you have a `with pkgs;` in effect
<akegalj_> srk: maybe its a firewall set to strict on this cloud hosting - will check
<akegalj_> clever: thanks , will look up logs to see if there is anything
<linarcx> thanks.
vandenoever has quit [Ping timeout: 250 seconds]
oever has joined #nixos
opthomasprime has left #nixos [#nixos]
fusion809 has joined #nixos
gxt has quit [Remote host closed the connection]
gxt has joined #nixos
<akegalj_> clever: seems like nginx is running ok, and all requests from the server are served correctly but from the outside it doesn't work. All logs look ok. I am going to check firewall config on cloud hosting page as srk suggested
<clever> akegalj_: what does `curl http://116.203.64.124/` sya?
<clever> say?
<fusion809> Hello good folks, it has come to my attention that despite the only channels being listed for my root user being nixos-unstable, I have been building package versions found in nixos 19.09. e.g. my kernel version is 5.4.25 despite having run nixos-rebuild switch --upgrade only yesterday. In /etc/nixos/configuration.nix I have system.stateVersion = "20.03pre" listed. nix-shell -p nix-info --run "nix-info -m" is returning this
<fusion809> output too: https://paste2.org/a2CZbAp1. Any ideas on how to make it so that I am actually running nixos-unstable?
codygman has quit [Read error: Connection reset by peer]
<colemickens> The stateversion does not have anything to do with teh version of packages you get.
<colemickens> You should not manually change it unless you understand the exact cases when you want to.
<clever> ,stateVersion fusion809
<{^_^}> fusion809: Changing stateVersion doesn't upgrade anything and can only break your setup at best. To actually upgrade NixOS see https://nixos.org/nixos/manual/#sec-upgrading. If you need to change stateVersion for some odd reason regardless, Ctrl-F for "stateVersion" in https://nixos.org/nixos/manual/release-notes.html to see things that need to be manually migrated with the new value
<colemickens> You might check the results of `nix-channel --list` vs `sudo nix-channel --list`.
chagra has joined #nixos
<colemickens> clever: in your nixos adventures, have you ever built an initrd that downloads and then "boots into" the squash/rootfs similar to a normal netboot or kexec w/ netboot payload ala nix-tests/kexec.nix?
<fusion809> colemickens: returns exactly the same output: https://paste2.org/109CcztE.
<clever> colemickens: back when i ran gentoo, i did have a custom initrd, that would download the entire rootfs over wifi
<colemickens> fusion809: and have you been running sudo `nix-channel --update` with/out sudo
<colemickens> clever: I was adapting your kexec/netboot to just be a tiny image that would download and kexec, but the rpi4 doesn't do kexec, so I'm trying to hack the next easiest thing that might give my SD cards some relief
<fusion809> Without. But, I thought nixos-rebuild switch --upgrade also runs nix-channel --update, and I do run nixos-rebuild switch --upgrade with sudo privileges.
<clever> colemickens: let me grab something...
<colemickens> fusion809: I can't remember if it does, I do things slightly off the beaten path
<fusion809> Just ran `sudo nix-channel --update && sudo su -c "nixos-rebuild switch --upgrade"` and no package updates.
<colemickens> it seems like it does from skimming the docs
<clever> colemickens: this downloads the rootfs using tftp, and then runs it via the tmpfs
<clever> colemickens: it should be possible to carry that back to nixos easily enough
* colemickens nods
<colemickens> this looks like what I was imagining, I'm just having a hard time mapping it to nixos options without digging into the source more to see how they actually fit together.
<colemickens> this gives me enough confidence that it can work to try and see it through tomorrow though!
<akegalj_> clever: about `curl http://116.203.64.124/` , I am not sure what it does on your end, but on my end `curl http://116.203.64.124/ -v` it looks like its stuck on connecting phase with "Trying 116.203.64.124:80...". I am not sure how to interpret that? Ps, I was unable to find any firewall settings on cloud hosting home page (hetzner cloud hosting)
<clever> akegalj_: that means your firewall is blocking port 80
<clever> akegalj_: did you open port 80 in the nixos config?
<akegalj_> clever: no I didn't open any ports. I didn't know nixos has builtin firewall
MichaelRaskin has quit [Ping timeout: 246 seconds]
<akegalj_> clever: that did the trick!
<akegalj_> clever srk thanks so much!
codygman has joined #nixos
codygman has quit [Read error: Connection reset by peer]
codygman has joined #nixos
otulp has joined #nixos
chloekek has joined #nixos
ruim has joined #nixos
<akegalj_> it makes sense to have firewall as default , but that didn't click to me at first. To be more noob friednly I would guess adding note about firewall at wiki https://nixos.wiki/wiki/Nginx would be helpfull to other people just starting with nginx. I will try to create a PR to add a sentence about it to be more explicit (as it is with archlinux wiki where I come from).
domogled has quit [Quit: domogled]
domogled has joined #nixos
<clever> akegalj_: i also just expect users to know the difference between 404 and connection timed out
<akegalj_> clever: yes, true
o1lo01ol1o has joined #nixos
thc202 has joined #nixos
o1lo01ol1o has quit [Ping timeout: 246 seconds]
ruim has left #nixos ["Konversation terminated!"]
domogled has quit [Ping timeout: 250 seconds]
chimay has joined #nixos
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
domogled has joined #nixos
linarcx has quit [Read error: Connection reset by peer]
linarcx has joined #nixos
domogled has quit [Ping timeout: 256 seconds]
energizer has quit [Ping timeout: 240 seconds]
oever is now known as vandenoever
abathur has joined #nixos
jakob_rs_1 has joined #nixos
<raboof> what is the derivation equivalent of 'makeWrapper'? should I make a derivation without sources that calls 'makeWrapper' in one of the phases or is there an easier way? (I saw 'runCommand' but it looks like that runs it while creating the derivation, not while running it)
<clever> raboof: runCommand would be the best way to do it
<jakob_rs_1> What does "path ... is not valid" mean?
<clever> raboof: something like: runCommand "name" {} "mkdir -pv $out/bin ; makeWrapper ${input}/bin/foo $out/bin/foo"
<clever> jakob_rs_1: it means the path is not part of /nix/store/
<jakob_rs_1> But it is?
dispersed has joined #nixos
<raboof> clever: ah, right, running makeWrapper inside of runCommand, not running the actual command in runCommand. that makes sense, thx.
<jakob_rs_1> "error: path '/nix/store/...-user-environment' is not valid"
domogled has joined #nixos
<clever> jakob_rs_1: if its not valid, then the build didnt finish, and it will be ignored by all nix tools
<clever> jakob_rs_1: what command is giving that error?
abathur has quit [Ping timeout: 246 seconds]
<jakob_rs_1> nix copy --from 10.0.0.174 /nix/store/...
linarcx has quit [Quit: WeeChat 2.7.1]
<clever> jakob_rs_1: are you running it as root?
<jakob_rs_1> No
<jakob_rs_1> also
knupfer has joined #nixos
<jakob_rs_1> Just noticed:"error: unexpected end-of-file"
linarcx has joined #nixos
<clever> jakob_rs_1: have you tried nix-copy-closure?
<jakob_rs_1> That does work
<clever> jakob_rs_1: what about `nix copy --from ssh://10.0.0.174 /nix/store/...`
<jakob_rs_1> That's what originally gave the error
<jakob_rs_1> Now it does nothing
<clever> jakob_rs_1: with or without ssh:// ?
<jakob_rs_1> with
<clever> jakob_rs_1: you claimed to be using it without before
<jakob_rs_1> Yes?
<alexarice[m]> I might have asked this before but is there a way to compile a haskell program from a nix-shell using nix dependencies
jakob_rs_1 is now known as jakobrs
<clever> jakob_rs_1: if the path is already copied, the `nix copy` wont have anything to do
<jakobrs> yeah ofc
mexisme_ has quit [Ping timeout: 260 seconds]
__monty__ has joined #nixos
<jakobrs> but that doesn't explain the original error
<clever> alexarice[m]: haskellPackages.ghcWithPackages or cabal2nix and .env
<clever> jakobrs: try using ssh:// on a new path, does it give the same error?
m0rphism has joined #nixos
<alexarice[m]> clever: how does `.env` work? I've used cabal2nix, gone into a nix-shell but when I run buildPhase it rebuilds all the dependencies
<clever> alexarice[m]: all haskell packages have a .env attribute
<alexarice[m]> oh i see
<alexarice[m]> thanks
<clever> alexarice[m]: so if you `nix-shell -A haskellPackages.aeson.env` then you get the environment needed for building it
<alexarice[m]> yeah my bad
<alexarice[m]> that worked, thanks
<alexarice[m]> clever++
<{^_^}> clever's karma got increased to 352
<jakobrs> clever: yes
<clever> jakobrs: not sure then
chimay has quit [Ping timeout: 250 seconds]
<jakobrs> clever: Oh I see why
<jakobrs> I assumed that symlinks are dereferenced remotely
<jakobrs> So I was referring to the wrong path
<clever> jakobrs: ahh
<jakobrs> The question then is whether the path *should* be dereferenced remotely or not
<__monty__> jakobrs: Is this about NFS?
<jakobrs> Not really
<jakobrs> though I guess it could be
<__monty__> You can't use symlinks with NFS if they reference files/directories outside of the NFS share.
<__monty__> You can use hardlinks for files or bind mounts for files or directories.
<jakobrs> Yeah, here I'm just using ssh
<__monty__> Oh.
dozn has quit [Quit: I'm probably just changing some settings or have had an outage, I'll be back soon.]
<linarcx> Guys, i have a shell.nix contains cmake, clang. After switching to it, my cpu uses all it's core. Why?
dozn has joined #nixos
<linarcx> I think clang is compling.. Is it correct?
<clever> linarcx: you need to use clangStdenv, not buildInputs = [ clang ];
<jakobrs> Tangentially, is it possible to mount sshfs at boot? Think fileSystems."/remote/acer".device = "ssh://10.0.0.174"
<clever> jakobrs: one minute
<linarcx> clever: This is my shell.nix? https://gist.github.com/LinArcX/dbdf9e6847f21c27e23d92cf2fb7120a. How can i change it?
<clever> jakobrs: this allows mounting any fuse based fs on boot
<jakobrs> thank you
<clever> linarcx: use clangStdenv.mkDerivation { buildInputs = [ ... ]; shellHook = ''....'' }
<linarcx> clever: Thanks. let me try it..
<linarcx> clever: But i'm using lorri. If i use this syntax, there is no issue with lorri?
<clever> linarcx: it should still work
<linarcx> Since lorri will create derivations automatically
<linarcx> hmm.. ok
<clever> linarcx: mkShell is just a wrapper around stdenv.mkDerivation
<clever> linarcx: you also need to add a name = "something"; to the mkDerivation
<linarcx> yeah, i'm going to try it. thanks.
linarcx has quit [Quit: WeeChat 2.7.1]
ottidmes has joined #nixos
vandenoever has quit [Read error: Connection reset by peer]
vandenoever has joined #nixos
vandenoever has joined #nixos
vandenoever has quit [Changing host]
<boogiewoogie[m]> hey, chkrootkit gives me some positives: https://hastebin.com/eqehuvesus. following the links to the store paths and checking the derivations, I'd naively assume them to be false. are those positives nixos-specific behaviour you can reproduce?
<kraem> hi! i've been trying to override a rust package, more specifically alacritty. this is my derivation: https://bpaste.net/MQXA . i keep getting "cannot coerce a set to a string". what am i doing wrong? thanks!
linarcx has joined #nixos
<clever> kraem: line 17, your creating a $buildRustPackage env var within alacritty
<clever> kraem: and it then tries to convert 17-24 into a string
<clever> kraem: you need to read the source of buildRustPackage to see how it can be overridden
<clever> default.nix: buildRustPackage = callPackage ../../../build-support/rust {
<clever> [clever@amd-nixos:~/apps/nixpkgs]$ vi pkgs/build-support/rust/default.nix
mexisme_ has joined #nixos
<clever> kraem: i dont think its possible to override that, youll need to just copy the file and modify it
o1lo01ol1o has joined #nixos
<v0|d> boogiewoogie[m]: try nix-store --verify --check-contents
<kraem> clever: thank you! i tried reading the buildRustPackage src but it's on another level of nix knowledge for me to understand.
<kraem> clever: i thought i'd try to not copy the derivation and modify it but i guess i'll do that :) thank you
<alexarice[m]> is howoldis down?
<ottidmes> alexarice[m]: it now is status.nixos.org I believe
<alexarice[m]> ottidmes: thanks
<ottidmes> They really ought to remove that link from status.nixos.org to HowOldIs and make it a redirect to status.nixos.org, otherwise this question keeps being asked
jakobrs has quit [Ping timeout: 250 seconds]
<clever> linarcx: and you can now delete line 10, clangStdenv always has clang
<clever> linarcx: you may also not need parts of 14-19
<clever> linarcx: and the with on line 1 pulled in all of nixpkgs, so the function on 2 and the with on 8 arent of much use
<linarcx> clever: Done! i need those envs actually.(i'm using them in another parts of my project)
<clever> linarcx: clangStdenv points $CC and $CXX to clang
<linarcx> clever: So i only need to keep cmake and clangd. right?
<clever> linarcx: probably
<linarcx> yeah.
<boogiewoogie[m]> v0|d: looking good. so I'll assume it's false positives
mexisme_ has quit [Ping timeout: 246 seconds]
hio has joined #nixos
alexherbo27548 has quit [Quit: Ping timeout (120 seconds)]
<hio> IM STUCK AT INSTALLATION!
<hio> /dev/sda3 no permission
<hio> no surprise, the previous steps didnt create it
<clever> hio: did you partition sda?
<hio> yep
<clever> hio: how many partitions did you make?
<hio> I did the exact steps as before
<clever> hio: was anything mounted while partitioning?
alexherbo27548 has joined #nixos
<hio> The steps dont mention any mounting
<clever> hio: which guide are you following?
sogatori has joined #nixos
gentauro has quit [Read error: Connection reset by peer]
<clever> hio: and what did fdisk print when you exited?
<hyper_ch> how can I achieve that the network card in the initrd is set to dhcp but then once the main kernel boot, network manager will take over?
orivej has joined #nixos
<hio> I'm not using fdisk, I'm using parted and mkfs
<clever> hio: what did parted say when you exited?
gentauro has joined #nixos
chimay has joined #nixos
<hio> "could not open /dev/sda1
<clever> hio: parted should only be ran on sda, not sda1
Izorkin has joined #nixos
<hio> ok that was from mkfs
<hio> parted said: "you may need to update fstab"
<clever> hio: can you pastebin all of the output from the terminal?
<hio> NO, I cant even type in my preferred keyboard layout
<hio> this is very hard
alexherbo27548 has quit [Client Quit]
<linarcx> clever: Seems it can't make a derivation. `called without required argument \'clangStdenv\'`
<clever> hio: you cant copy/paste to a pastebin?
<clever> ,callPackage linarcx
<{^_^}> linarcx: If a Nix file foo.nix starts with something like `{ stdenv, cmake }:`, you can build it with `nix-build -E '(import <nixpkgs> {}).callPackage ./foo.nix {}'`
<linarcx> Should i pass extra arguments to clangsStdenv?
<clever> linarcx: but since line 1 already pulls in all of nixpkgs, you dont need line 2
alexherbo27548 has joined #nixos
<Bambus> hello there
<hio> here is the output: https://i.imgur.com/GTqZbVl.png
CMCDragonkai1 has joined #nixos
<linarcx> clever: Yeah, i deleted it and i see nix-instantiate in htop.
<linarcx> Is it a good sign?
<atemu12[m]> hio: Which layout do you prefer?
<clever> hio: mkfs must be ran as root
<hio> german
<clever> hio: every single step in the installation guide must be ran as root
<hio> i did that
<hio> it was just one try without sudo
<hio> oh the last one
<clever> yeah, the one your stuck on
<hio> well i still dont know where sda1-3 are coming from anyway
<clever> hio: when you partitioned sda, it created 1/2/3
alexherbo27548 has quit [Client Quit]
<hio> ok where do i put my configuration.nix now, I already did all this work before
<clever> hio: step 2.3.4
<hio> ok
o1lo01ol1o has quit [Remote host closed the connection]
o1lo01ol1o has joined #nixos
o1lo01ol1o has quit [Remote host closed the connection]
o1lo01ol1o has joined #nixos
shibboleth has joined #nixos
<hio> why dont the docs say how to go into su mode
<clever> > Note: Commands prefixed with # have to be run as root, either requiring to login as root user or temporarily switching to it using sudo for example.
<{^_^}> error: syntax error, unexpected WITH, expecting ')', at (string):289:25
<clever> hio: it says that before chapter 1
<clever> hio: you can also use `sudo -i` to get a root shell
doyougnu has quit [Quit: ERC (IRC client for Emacs 26.3)]
doyougnu has joined #nixos
m4ts has quit [Quit: bye]
m4ts has joined #nixos
<linarcx> clever: It start building clang again :?
<linarcx> :/
doyougnu has quit [Client Quit]
<clever> linarcx: did you change $NIX_PATH or nixpkgs?
doyougnu has joined #nixos
<linarcx> clever: Not actually. For first time i set $NIX_PATH in my .zshrc to something like this:
<linarcx> export NIX_PATH="nixpkgs=https://github.com/NixOS/nixpkgs/archive/release-19.09.tar.gz"
<clever> linarcx: what about `nix show-config | grep subs` ?
<linarcx> I'm not on nixos. I'm using void linux.
<clever> linarcx: what about `nix-build '<nixpkgs>' -A clang --dry-run` ?
teto has joined #nixos
<hio> I FAILED
<hio> unable to download bash, no internet
<atemu12[m]> hio: What's the exact error message?
<atemu12[m]> hio: also, `loadkeys de` ;)
<clever> linarcx: nix-store -q --binding out /nix/store/pq64x5rwvbj011pcln2gcgaqz7c96xi5-clang-7.1.0.drv
akegalj_ has quit [Quit: leaving]
<linarcx> clever: /nix/store/c52j9ghhfw2ya6j5c8mcsq08937l461p-clang-7.1.0
o1lo01ol1o has quit [Remote host closed the connection]
jakobrs has joined #nixos
<clever> linarcx: thats not in the binary cache, what does `nix-instantiate --find-file nixpkgs` report?
<linarcx> clever: /nix/store/j2shn0av1hx3bn3caifxn9dh7amf3awg-release-19.09.tar.gz
<jakobrs> wrt to the useDHCP flag being deprecated, how do I get a list of all the relevant interfaces?
<jakobrs> They change name between devices
<clever> linarcx: i think clang is broken on 19.09
o1lo01ol1o has joined #nixos
<clever> linarcx: let me check hydra...
CptCaptain has quit [Quit: WeeChat 2.7.1]
<clever> looks normal?
<makefu> clever: for this i implemented hydra-check last week: hydra-check clang --channel 19.09
<makefu> ✔ clang-wrapper-7.1.0 from 2020-03-19 - https://hydra.nixos.org/build/115126851
<clever> [clever@amd-nixos:~/apps/rpi/rpi-open-firmware]$ nix-build https://github.com/NixOS/nixpkgs/archive/release-19.09.tar.gz -A clang --arg config '{}' --arg overlays '[]'
<clever> these derivations will be built: /nix/store/pq64x5rwvbj011pcln2gcgaqz7c96xi5-clang-7.1.0.drv
<clever> makefu: but that doesnt explain why its trying to oh wait
<makefu> https://github.com/nix-community/hydra-check a bit out of context, sorry :)
<clever> linarcx: i see the problem, your using the release-19.09 branch of nixpkgs, which is before hydra testing occurs
<linarcx> clever: Hmm.. so what should i do now?
<clever> linarcx: you must use the nixpkgs-channels fork, or you wind up not getting cached results
doyougnu has quit [Ping timeout: 246 seconds]
<clever> linarcx: export NIX_PATH="nixpkgs=https://github.com/NixOS/nixpkgs-channels/archive/release-19.09.tar.gz
<hio> I failed again
avn has joined #nixos
o1lo01ol1o has quit [Ping timeout: 256 seconds]
<clever> hio: how did you fail?
<hyper_ch> clever: how can I tell nixos that in the initrd dhcp should be available?
<linarcx> clever: I switched to nixpkgs-channel. I didn't get this part: `or you wind up not getting cached results`. What that mean?
<hyper_ch> if I add networking.interfaces.eth0.useDHCP = true; it's not only for the initrd and it causes me a 90s timeout...
<hio> no idea what this means
<clever> linarcx: AFTER hydra has built everything, the nixpkgs-channels repo will update
<clever> linarcx: if you use the nixpkgs repo, you get nix expressions hydra hasnt built yet
fendor has joined #nixos
<clever> linarcx: https://hydra.nixos.org/job/nixos/release-19.09/nixpkgs.clang.x86_64-linux and the latest clang is in the queue still
<linarcx> I got it. thanks
<avn> Hello! Foiks, any recent chromium upgrade (2-3 days ago) introduce new fatal bugs? Have "stalled" session lock somewhere (or at least it complains on), and problem not solves by finding locks.
<avn> (just want to sure that is a local problem, and not broken everything)
<clever> hio: you set boot.loader.grub.device wrong
<clever> hio: it must be set to "nodev" when doing EFI
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/66758bc3e59 (from 2 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
<hio> I followed the UEFI steps
<clever> hio: what is boot.loader.grub.device set to?
<hio> "/dev/sda"
<clever> hio: 2020-03-21 08:25:46 < clever> hio: it must be set to "nodev" when doing EFI
linarcx has quit [Read error: Connection reset by peer]
zupo has joined #nixos
<hio> well what about "You must set the option boot.loader.systemd-boot.enable to true"
<hio> my installer config doesnt have this set
<clever> hio: thats if you want to use systemd-boot instead of grub
<hio> I want to use systemd in general
linarcx has joined #nixos
<clever> hio: systemd-boot has nothing to do with systemd
shibboleth has quit [Ping timeout: 240 seconds]
<clever> hio: its just named confusingly
<jakobrs> For grub on UEFI, you need to set:
<jakobrs> * efi.efiSysMountPoint to "/boot"
<jakobrs> * grub.efiSupport to true
<jakobrs> * grub.version to 2
<jakobrs> * grub.device to "nodev"
<hio> too late.............
<hio> :(
<clever> hio: how is it too late?
civodul has joined #nixos
<hio> it didnt boot into gnome3, I messed up somehow
<clever> jakobrs: the default for efi.efiSysMountPoint is already /boot
<jakobrs> and either efi.canTouchEfiVariables or grub.efiInstallAsRemovable
<clever> hio: you can just boot the installer again, mount everything to the right place, and resume from step 2.3
<ottidmes> I always use: boot.loader = { efi.canTouchEfiVariables = true; grub = { efiSupport = true; device = "nodev"; }; };
<hio> i started the desktop but it's only black
<jakobrs> The manual should probably describe how to install grub on uefi
<jakobrs> or the wiki, I suppose
<etu> hio: The live iso plasma is incredibly slow at loading
<ottidmes> jakobrs: manual (cause it should cover regular install, wiki is more for extraenous things) + as a comment in the default configuration.nix
<hio> wait I think i went into the installer again
<hio> I have to unload the cd first I guess
o1lo01ol1o has joined #nixos
<linarcx> clever: So i switched to nixpkgs-channel, but lorri complains about it: https://gist.github.com/LinArcX/f3ba2097aa53a4d3924b806683e0b7db
<alexarice[m]> Is it possible to use `nix why-depends` to see why a derivation is being installed
<hio> it says "No operating system found"
<hio> ...!!
<linarcx> clever: It redirect to: https://codeload.github.com/NixOS/nixpkgs-channels/tar.gz/release-19.09 that says 404 error
<avn> https://gist.github.com/avnik/7041f6b240075e71fa8f022fab78055e I start to worrying, it not starts with cleanprofile
<clever> linarcx: your having network problems
o1lo01ol1o has quit [Remote host closed the connection]
<clever> linarcx: oh, and its nixos-19.09, not release-19.09
<clever> hio: was efi.canTouchEfiVariables = true set?
<clever> hio: what about boot.loader.grub.efiSupport = true?
<linarcx> clever: Yeah :)
<ottidmes> alexarice[m]: you can use nix-store --query --tree and nix why-depends, let me find you an example
<alexarice[m]> ottidmes: is this possible from the outpath of the derivation
<alexarice[m]> my configuration is installing an old version of a package
<linarcx> clever: A question. I don't need to update the nixpkgs-channled after this time?
<linarcx> Can i always be on 19.09?
<jluttine> are there instructions on how to write package derivations when gradle is used for building?
<clever> linarcx: main issue with using a branch like that, is that it will randomly update without any warning, and things may break for you
<clever> linarcx: its better to use either nix-channel, or pin a specific revision
<vandenoever> anyone here running nixos on hetzner cloud and managed to get ipv6 working? dhcp gives an internal ip6 and external ipv4
<linarcx> clever: I'll read about them. thanks.
<alexarice[m]> ottidmes: have solved my problem I think
<ottidmes> alexarice[m]: nix why-depends $(nix-instantiate '<nixpkgs/nixos>' -A system) $(nix-store --query --deriver "$out_path")
<clever> linarcx: instead of using <nixpkgs> i use this
<clever> linarcx: then the project will always use that version, and just work for any user with nix
<alexarice[m]> ottidmes++ think I managed it by copying in the paths for the .drv files
<{^_^}> ottidmes's karma got increased to 20
<linarcx> clever: awesome tip.
<clever> linarcx: it doesnt matter what channel you use, it will build the exact same binaries as i build
<hio> clever: I set it now and it still isnt booting, now it says: "PXE-E53 no boot filename received"
<ottidmes> vandenoever: No experience with that host, but can't you request some external IPv6 addresses from the admin interface?
<clever> hio: did you re-run nixos-install?
<vandenoever> ottidmes: my machine is assigned 2a01:4f8:c0c:5b0a::/64 which is not a single number but a range
<vandenoever> i've tried to set it manually to 2a01:4f8:c0c:5b0a::1 with ip addr add
cosimone has joined #nixos
<Bambus> i can't geht lutris running in unstable channel.... is this a known problem?
<hio> yes I did clever
<hio> but it says during the install: "EFI variables are not supported on this system"
<clever> hio: did you boot the CD via efi or legacy?
<hio> the CD?
<clever> hio: the install CD
<hio> no idea, I loaded the .iso in vmware workstation
<clever> hio: its probably booting in legacy mode then, so you need to follow the bios directions
o1lo01ol1o has joined #nixos
jakobrs has quit [Ping timeout: 250 seconds]
abathur has joined #nixos
cosimone has quit [Quit: Terminated!]
<avn> hio: you can't play with efi variables, if you booted in legacy. In theory -- you still can put files on EFI partition, where bios looking for them, but is a weak ground
<misuzu> vandenoever: use 2a01:4f8:c0c:5b0a:: - it's a valid address
cosimone has joined #nixos
Jackneill has joined #nixos
<hio> the docs shouldnt give me a choice between EFI and legacy without telling me what it means or how to decide
<misuzu> vandenoever: https://pastebin.com/7xP8rMfJ this should work
abathur has quit [Ping timeout: 264 seconds]
<etu> hio: Basically you make the choice at boot time. And it's hardware specific as well.
<hio> I dont think that it had anything to do with my CD though
<vandenoever> misuzu: i'll give it a go
<etu> It's before the CD. It's how the BIOS or EFI on the hardware operates.
<vandenoever> misuzu: can I combine that with dhcp for ipv4?
<misuzu> vandenoever: yes, just paste this to your config and it should work
pbb has quit [Ping timeout: 272 seconds]
pbb has joined #nixos
<hio> ok it booted now but it says that my password is incorrect for login
<clever> hio: are you trying to login as root?
<hio> no, as my user
<clever> hio: you havent set a password for that user yet
Avaq has joined #nixos
<vandenoever> misuzu: nixos-rebuild does not complain about it and "ping -6 ipv6.google.com" does not abort but hangs, so your help brought me closer
<vandenoever> misuzu: https://pastebin.com/2XdJjjjq
<misuzu> vandenoever: check `ip -6 route`
<vandenoever> misuzu: https://pastebin.com/CBbCV18e
<vandenoever> default via fe80::1 seems fine
<misuzu> vandenoever: try `ping fe80::1%ens3`
<vandenoever> misuzu: no response
<misuzu> vandenoever: that's weird
<misuzu> maybe they changed something for new machines
<vandenoever> misuzu: this one is 2 months old
<Avaq> Hi good folks! :) Can anyone help me circumvent https://github.com/NixOS/nix/issues/3155 ? I don't know how to get my system out of this state, because it prevents me from upgrading.
<Avaq> I tried `systemctl edit nix-daemon.service`, but it fails because the override.conf is on a readonly file system.
<{^_^}> nix#3155 (by xbreak, 22 weeks ago, open): Nix-daemon systemd unit missing `NIX_SSL_CERT_FILE`
<clever> Avaq: you can only change systemd services by editing configuration.nix
xkapastel has joined #nixos
<clever> Avaq: are you on nixos or another distro?
<vandenoever> misuzu: anyway i made progress, i also contacted hetzner about and i learned that apparently ipv6 numbers may end in 0
<Avaq> clever: Nixos on a PI (just your cup of tea I think :))
<clever> Avaq: ah, then youll want to use configuration.nix and nixos-rebuild switch to make it apply
<vandenoever> misuzu: do you know why hetzner does not simply give the right config via dhcp?
<misuzu> vandenoever: ipv6 is hard :)
<hio> how do I put packages like chrome and vim into my configuration.nix?
<hio> also I want to run custom installation scripts, wgetting stuff and so on
<clever> hio: environment.systemPackages = with pkgs; [ chromium vim ];
<misuzu> vandenoever: just started new ubuntu host on hetzner cloud and it has working ipv6 with fe80::1 as default gateway
<clever> hio: those will need to be properly packaged
<vandenoever> misuzu: it's really hard, it's probably easier to reduce the ip demand to ipv4 range than to implement ipv6 globally ;-)
<hio> ok but I can just write a dummy package and then put my own script in there right?
<clever> hio: yeah
<vandenoever> misuzu: and that ipv6 was given by dhcp or did you set it by hand?
<misuzu> vandenoever: ah, i see what's wrong
<misuzu> vandenoever: you probably removed link-local address
<misuzu> it's was a bad idea
<misuzu> you can add it back or just reboot and it should work
Boomerang has joined #nixos
<vandenoever> i'll try rebooting
zupo has quit [Ping timeout: 264 seconds]
<vandenoever> misuzu: ping -6 fe80::1%ens3 now works!
<Avaq> clever: I can't find the nixos options reposible for generating the nix-daemon.service.
<vandenoever> and pinging from outside to the machine also works!
<avn> Well, I found source of `chromium` issue... TMPDIR=/var/buildroot somehow leacked out to users profiles, whenever it need to be set to nix only
<vandenoever> misuzu: thank you very much, i was struggling with this for hours already
Boomerang has quit [Client Quit]
zupo has joined #nixos
o1lo01ol1o has quit [Remote host closed the connection]
<Avaq> clever: Yes, I did find `cfg.daemonNiceLevel` and `cfg.daemonIONiceLevel`. But it doesn't seem like the config allows for adding additional directives (I think need to add `Environment="NIX_SSL_CERT_FILE=/etc/ssl/certs/ca-bundle.crt"` according to #3155).
<{^_^}> https://github.com/NixOS/nixpkgs/pull/3155 (by doublec, 5 years ago, merged): Add inferno, a distributed cross-platform hosted OS
vandenoever has quit [Read error: Connection reset by peer]
<clever> Avaq: you can just directly set systemd.services.nix-daemon.environment.NIX_SSL_CERT_FILE
<clever> Avaq: and nixos will automatically merge it in
<Avaq> That 3155 on nix, not nixpkgs: https://github.com/NixOS/nix/issues/3155
<{^_^}> nix#3155 (by xbreak, 22 weeks ago, open): Nix-daemon systemd unit missing `NIX_SSL_CERT_FILE`
<avn> Folks, does `nix.envVars.TMPDIR define _global_ TMPDIR, not nix-daemon's one?
<Avaq> Oh :o
<clever> Avaq: you can also use nix.envVars like avn just showed
<clever> Avaq: line 428 will merge those in first
vandenoever has joined #nixos
vandenoever has joined #nixos
<clever> Avaq, avn: but nix.envVars are also passed on to the environment.sessionVariables, so they become global
reallymemorable has joined #nixos
cosimone has quit [Remote host closed the connection]
<avn> clever: btw, do you know how to set TMPDIR for daemon' only/
o1lo01ol1o has joined #nixos
<clever> avn: try systemd.services.nix-daemon.environment
<Avaq> clever: Right, so I've added the line to my configuration.nix (`systemd.services.nix-daemon.environment.NIX_SSL_CERT_FILE = "/etc/ssl/certs/ca-bundle.crt";`). But it seems this change causes nix to want to download a new package, and of course it can't because it relies on this env var for being able to download things:
<Avaq> warning: unable to download 'https://cache.nixos.org/gy514zy7v3ax5bkrymyvvk9ljn4g1sd7.narinfo': SSL peer certificate or SSH remote key was not OK (60); retrying in 305 ms
<avn> clever: it surprisingly stop working few weeks ago, so I tried `nix.envVars.TMPDIR` which give slighty not expected result
<avn> clever: I definiteyly like to have simple mechanism to show _where_ to build packages? (and would be ideal if it be dynamically, by sources size)
<hio> how do I know if a package is official or not? It could contain malicious code right?
<clever> hio: check the nix expression to see how its built
<avn> hio: caches are signed by keys, but nixpkgs is just a source tree, it load/evaluates when you invoke nix-*
chagra has quit [Quit: WeeChat 2.7.1]
chloekek has quit [Ping timeout: 256 seconds]
ZoomZoomZoom has quit [Ping timeout: 264 seconds]
detran has quit [Ping timeout: 250 seconds]
reallymemorable has quit [Quit: reallymemorable]
mexisme_ has joined #nixos
chagra has joined #nixos
<avn> clever: looks like something mystic, now systemd.services.nix-daemon.environment.TMPDIR works again (although it not worked 3 weeks ago, giving me some fun by unpacking chromium to /tmp)
mexisme_ has quit [Ping timeout: 260 seconds]
<Avaq> curl: (60) SSL certificate problem: certificate is not yet valid
<Avaq> Actually, I think my problem is of a different nature. I thought it was https://github.com/NixOS/nix/issues/3155, because that one leads to the same error message. But I just tried the following command:
<Avaq> # CURL_CA_BUNDLE=/etc/ssl/certs/ca-certificates.crt curl https://cache.nixos.org/gy514zy7v3ax5bkrymyvvk9ljn4g1sd7.narinfo > /dev/null
<{^_^}> nix#3155 (by xbreak, 22 weeks ago, open): Nix-daemon systemd unit missing `NIX_SSL_CERT_FILE`
<hio> isnt it annoying to constantly have to nix-rebuild switch and reboot everytime I want to install something
jakobrs has joined #nixos
<jakobrs> What does builtins.storePath do? It's not documented in the manual.
<Avaq> hio: you don't have to reboot, `switch` rewires your runtime and restarts modified services
<hio> ok
<clever> jakobrs: it turns a string into a path
<clever> jakobrs: you usually dont need it
avn has quit [Quit: Lost terminal]
hc has quit [Remote host closed the connection]
armin has quit [Remote host closed the connection]
<jakobrs> also, what does recurseIntoAttrs do? It's not documented either.
<clever> jakobrs: controls nix-env recursion when searching
<jakobrs> ah, that makes sense
jakobrs has quit [Quit: WeeChat 2.7.1]
avn has joined #nixos
<{^_^}> Channel nixpkgs-19.09-darwin advanced to https://github.com/NixOS/nixpkgs/commit/db12da3abc3 (from 23 hours ago, history: https://channels.nix.gsc.io/nixpkgs-19.09-darwin)
<avn> clever: well, not it works. User profile have empty TMPDIR, and nix-daemon have TMPDIR=/var/buildroot again
<clever> avn: how are you checking nix-daemon's env vars?
<avn> systemctl show nix-daemon | grep Environment | tr ' ' '\n'
<avn> bit innacurate, but ok for troubleshoot fast
<clever> avn: thats showing what is in the .service file, but not the final env vars
<clever> [root@amd-nixos:~]# xargs -n1 -0 echo < /proc/17117/environ
<clever> avn: try this on the pid of nix-daemon
<avn> clever: I try to build recent master, and see where it unpacks as well ;)
<avn> clever: btw, is a good idea to make option for service.nix-daemon like `buildLocation`?
<clever> avn: yeah, i could see that being a useful option
kfound has joined #nixos
<avn> clever: will do it then on next cycle ;)
<hio> I export PATH=/myfolder in my .bashrc but in the terminal it doesnt find it. Why?
<clever> hio: it can be confusing as to when .bashrc is ran and when .bash_profile is ran
<avn> And I finally would like to push my pulse changes tomorrow, I boring to rebuild browsers ;)
<hio> something is weird, it pretends that the file isnt there but it is
<hio> I cant even execute it when I specify the full path
<clever> hio: is it an ELF binary? a shell script?
<clever> hio: dotnet is already in nixpkgs
bvdw has quit [Read error: Connection reset by peer]
<hio> I want to do it manually
<clever> hio: then you need to patchelf all of the binaries
<hio> what??
<clever> hio: and those binaries will break next time nix upgrades things
bvdw has joined #nixos
<clever> hio: that is why you should write packages for things
<clever> hio: /lib doesnt exist, so all binaries must be patched to work
<hio> no, I cant constantly write a new package for everything I want to run
<hio> what is this patchelf thing you mentioned, why cant I run a binary
<clever> hio: patchelf is explained in the nixpkgs manual
<clever> hio: normal binaries will fail, because both /usr/lib and /lib are missing
<hio> why does the error say "./dotnet no such file or directory". It doesnt say anything about /usr
<avn> well, abusing ld-linux,si is other way to launch binaries
<clever> hio: the interpreter of dotnet is /lib/ld.so or similar
<avn> (I used it for launching kerbal, and few other games)
<clever> hio: and that is what does not exist
waleee-cl has joined #nixos
<clever> [nix-shell:~]# patchelf --help
<clever> syntax: patchelf [--set-interpreter FILENAME]
<clever> hio: you fix that by using patchelf --set-interpreter
<hio> to what?
chagra has quit [Ping timeout: 240 seconds]
<clever> hio: the real location of ld.so
chagra has joined #nixos
kleisli has quit [Remote host closed the connection]
hodapp1 is now known as hodapp
kleisli has joined #nixos
<Avaq> I am confused. I have two machines (laptop & pi) in the same network running NixOS. One can happily download stuff over https (laptop) using /etc/ssl/certs/ca-bundle.crt as a certificate verify location, the other (pi) can't.
<Avaq> So I thought it must have to do with an outdated/corrupted ca-bundle on the pi. So I moved the bundle from laptop to the pi, and tried to use curl with that one instead. Curl reports successfully using the new bundle file, but it still fails with "SSL certificate problem: certificate is not yet valid". Any ideas on what could be causing this? I understand now it's no longer really a nix related question, but I have great faith in the tech
<Avaq> nical prowess of the nix community in general. :P
<clever> Avaq: what does `date` say on the pi ?
<Avaq> clever: Tue 20 Jan 1970 05:30:11 PM UTC haha
<clever> Avaq: then the SSL certs arent valid yet, or your clock is wrong
<Avaq> The clock is in 1970. So I think it's the clock.
<clever> Avaq: thats a common problem on pi's, since they lack an RTC
<Avaq> I saw earlier that the time sync server couldn't start, and I figured that's because it tries to sync the time using SSL
<Avaq> So I thought it was a sympton, not a cause.
<Avaq> > the time sync server couldn't start
<Avaq> service*
<{^_^}> undefined variable 'the' at (string):289:1
<hio> ok I did this: `patchelf --interpreter /nix/store/wx1vk75bpdr65g6xwxbj4rw0pk04v5j3-glibc-2.27/lib/ld-linux-x86-64.so.2 --set-rpath /nix/store/784rh7jrfhagbkydjfrv68h9x3g4gqmk-gcc-8.3.0-lib/lib/ ./dotnet`
<hio> but I feel like it wont work forever!
<clever> hio: yeah, next time things get updated, that will stop working, and you have to find the new path
<clever> hio: thats why its better to write a nix package
<hio> on any update or on an update that removes gcc-8.3?
<clever> hio: any update that removes gcc-8.3.0
<clever> hio: your better off with a nix package, that runs patchelf for you
leotaku has quit [Ping timeout: 264 seconds]
<clever> hio: and the one i linked, is identical to what you have manually created, so why not just use dotnet-sdk?
<hio> well in theory I like this explicitness more but what about binaries created by golang or rust. Will they run without problems?
<hio> actually binaries created by anybody on nix
<hio> wont they link to weird store locations on the builder's pc?
<avn> or make package with script doing `LD_LIBRARY_PATH=${pkgs.gcc.libs}/lib ${pkgs.glibc}/lib/ld-linux-x86-64.so.2 ./your/binary`
<clever> hio: the compilers used by nix, patch things automatically
<clever> hio: and if its build by `nix-build`, then nix is aware of what its using
<avn> this way commonly used for games who do self-check/self-update
<das-g[m]> When I use a package (derivation) with `nix run` and that package provides bash completions, are those supposed to be available in the spawned shell?
<hio> I assume that it works on my own PC but what if I copy the binary over to another PC?
<clever> hio: you must use nix-copy-closure, to copy everything the binary depends on
cosimone has joined #nixos
<clever> avn: also, you want to use "${stdenv.cc.bintools.dynamicLinker}" or "$(cat $NIX_CC/nix-support/dynamic-linker)"
<avn> (I found install whole game to store is bit contrproductive sometimes, due sizes, attempts to write-in etc)
<clever> avn: ${pkgs.glibc}/lib/ld-linux-x86-64.so.2 will fail on other arches
cjpbirkbeck has joined #nixos
<avn> clever: that almost correct, but I not sure I play kerbal on other arches. (Although if I generalize it, I will use stdenv.cc.bintools.dynamicLinker
iyzsong has quit [Ping timeout: 246 seconds]
reallymemorable has joined #nixos
<das-g[m]> When I use a package (derivation) with "nix run" and that package provides bash completions, are those supposed to be available in the spawned shell?
<clever> das-g[m]: that might be a limit of `nix run`, try `nix-shell` instead?
leotaku has joined #nixos
cjpbirkbeck has quit [Quit: cjpbirkbeck]
andymandias has quit [Ping timeout: 264 seconds]
andymandias has joined #nixos
<jluttine> hmm.. i have mutableUsers=false but i was able to add users.extraUsers.foobar without defining uid.. how is this possible?
rardiol has joined #nixos
kfound has quit [Ping timeout: 246 seconds]
kfound has joined #nixos
<jluttine> also, where are system user ids defined? i remember i once saw a file containing a list of system usernames like "nextcloud" and their ids
greizgh has quit [Quit: greizgh]
<clever> jluttine: mutableUsers=false just means that all users must be defined in the config
<clever> jluttine: you can never use adduser/useradd to mutate things
<jluttine> clever: ah, ok.. i thought one must also provide UID so those would be reproducible too
greizgh has joined #nixos
<das-g[m]> I'd really like something with the features of nix-shell but with a command line interface like the new nix subcommands.
<clever> jluttine: nixpkgs/nixos/modules/misc/ids.nix
<clever> das-g[m]: also, `nix run` only adds things to PATH, while `nix-shell` will deal with the stdenv and running with setup hooks
<clever> das-g[m]: so compilers wont really work under `nix run`
<das-g[m]> Ah, good to know
<das-g[m]> clever++
<avn> hehe. mkShellFor is a thing, I need diving in... I still use myEnvFun for some stuff
<{^_^}> clever's karma got increased to 353
<clever> avn: id just ignore mkShell
<clever> avn: just use stdenv.mkDerivation
<Avaq> And then I have to clear space off my boot sector by removing kernels and whatnot, which I prefer not doing. Do you experience this issue as well?
<Avaq> clever: Thank you for pointing out the date problem to me! After updating the date, everything works fine. I have a pi-specific question to you. All "install nixos on pi" guides out there have you create a 120M boot sector, but it's somehow never enough space for me. Whenever I run a `nixos-rebuild` (or `nixops deploy` in my case), I get this "cp: error writing '/boot/nixos/...-linux-5.5.9-Image.tmp.23409': No space left on device" error.
<das-g[m]> I know (and for stuff I don't want to package use) mkShell. What's mkShellFor, though?
<clever> Avaq: which bootloader do you have enabled?
<clever> Avaq: which `boot.loader.*.enable` do you have set?
<avn> clever: I have few custom load-env-xxx for unclutter language specific environments (c++, go, haskell, rust) and just some custom like wine (load-evn-wine-{unstable,stable,staging})
<avn> and I definitely want to build these environments at once with all other system
<clever> avn: for wine, you just need to append to PATH and set WINEPREFIX
<clever> avn: for c++/go/haskell/rust, you want to stick to nix-shell
<Avaq> clever: grub.enable = false; raspberryPi.enable = true; raspberryPi.version = 3; raspberryPi.uboot.enable = true; raspberryPi.firmwareConfig = "gpu_mem=256";
<avn> clever: well, so I mainly use myEnvFun to generate wrappers to add extra tools to paths. And I found it useable when I need switche between stable/unstable fast
gustavderdrache has joined #nixos
<clever> Avaq: https://nixos.org/nixos/options.html#boot.loader.ras and look at the configurationLimit
<clever> Avaq: that limits how many generations get copied to /boot/
<avn> clever: I use nix-shell mainly, when I hae established project, but sometimes I need just compiler/repl/etc)
<avn> (well honestly not nix-shell, but direnv+shell.nix ;))
ambro718 has joined #nixos
<avn> Think I need switch to lorri at some point, but I am lazy
Diogo9 has quit [Quit: The Lounge - https://thelounge.chat]
<Avaq> clever: Ahh, nice! :D So it's the older generations filling up the space. How many generations you reckon would 120M support?
<clever> Avaq: depends on how often your kernel and initrd are changing
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<Avaq> Makes sense. Okay, thanks. I probably don't need many generations anyway. :)
Diogo has joined #nixos
<clever> Avaq: try 5 or 10
<hio> I just installed zlib and it pulled emscripten. Wow
<Avaq> Would lowering that number have an immediate effect? I lowered it to 5 and I still get the error. Do I need to manually clean up space one more time?
<jluttine> if i have an attrsOf option with a default value and a user sets that in configuration.nix, does it replace the default or are they merged?
<clever> hio: also, libraries you "install" wont be visible to any program, nix doesnt work like that
<clever> Avaq: it deletes the extra ones after it copies the required 5 in
<clever> Avaq: so you may need to manually free up some space firt
gustavderdrache has left #nixos [#nixos]
orivej has quit [Ping timeout: 264 seconds]
<clever> jluttine: i think it will overwrite all of the defaults
<jluttine> clever: ok.. :/
<clever> jluttine: there is similar problems with the nix binary cache config, adding one binary cache causes cache.nixos.org to vanish
abathur has joined #nixos
cosimone has quit [Remote host closed the connection]
cosimone has joined #nixos
<clever> Avaq: what are you using your rpi for?
<Avaq> clever: Currently just using it to explore nixops. Once I have a setup going where I can reliably update my pi using nixops, I intend to run wireguard, and possibly if I can manage run a pihole server as well.
<Avaq> clever: Is there a way to free up spae on /boot formally, with some garbage collection tool, like: `nix-collect-garbage --delete-old-generations-from-boot`?
<Avaq> I'm afraid to remove things there manually, because I might break something.
<clever> Avaq: nixos-rebuild switch will delete things from /boot after it has copied new things in, based on what generations exist
<clever> Avaq: so you have to delete generations, then switch, and successfully copy all kernels in
<clever> Avaq: you can usually delete all of the initrd's, and it will recover
<infinisil> Avaq: I've been doing `sudo rm /boot/kernels/*` for a while when I couldn't switch anymore, it's regenerated automatically, so no problem
<hio> the zlib package doesnt seem to work: ./java: error while loading shared libraries: /nix/store/vh16dj8qssk6xvw007y12pq0lxf68l99-zlib-1.2.11/lib/libz.so.1: invalid ELF header
<Avaq> Okay :)
<infinisil> (regenerated when you switch that is)
orivej has joined #nixos
<clever> Avaq: ive been experimenting with custom firmware and nixops on my rpi's
rail has quit [Ping timeout: 250 seconds]
glittershark has quit [Ping timeout: 256 seconds]
<clever> Avaq: first, i have a pretty standard nixops deployment file
glittershark has joined #nixos
<clever> Avaq: then my nixos config, line 27 forces a cross-compile to arm, 59 enables my custom bootloader, the rest is pretty standard stuff
<clever> Avaq: and my custom bootloader, will compile all of the rpi firmware from source, and copy it to /boot, along with the kernel and initrd
gustavderdrache has joined #nixos
gustavderdrache has left #nixos [#nixos]
philr has quit [Ping timeout: 250 seconds]
<Avaq> Woah. Let me try to get this. rpi-open-firmware is a project that provides a non-proprietary firmware for rpi, and your fork allows you to deploy it with nixops? :o
<clever> Avaq: yeah
<clever> Avaq: i just `nixops deploy` and `nixops reboot`, and the pi boots nixos with custom firmware
<clever> nixos.nix defines everything about the nixos that boots
<clever> and the source in firmware/ and arm_chainloader/ is how it boots
<hio> this page says that my choice should be jdk11 but nix-env -i jdk11 doesnt work https://nixos.wiki/wiki/Java
<clever> hio: you want ni-env -iA nixos.jdk11
<clever> ,-A hio
<{^_^}> hio: You'll usually want to use nix-env -i with -A. It's faster and more precise. See https://nixos.wiki/wiki/FAQ/nix-env_-iA for details.
<Avaq> clever: That's really neat!
<clever> Avaq: currently, its only confirmed to work on the rpi2 and rpi3, and no video modes work
<Avaq> "no video modes work" I don't know what a "video mode" is. Does it mean they can only run headless?
<clever> Avaq: yeah
<clever> Avaq: hdmi and composite video dont work currently
<Avaq> That's fine though
mekeor has joined #nixos
<Avaq> For many usecases
<clever> Avaq: ive also not tried to enable SMP, so its limited to 1 core, but that is likely an easy problem
<clever> Avaq: 3 of the cores just sit here and never recover
<clever> Avaq: i think i need to copy and adapt this part of the official code
CRTified has quit [Ping timeout: 250 seconds]
bhipple has joined #nixos
chloekek has joined #nixos
<{^_^}> Channel nixpkgs-unstable advanced to https://github.com/NixOS/nixpkgs/commit/b7bdf48e3fa (from 3 days ago, history: https://channels.nix.gsc.io/nixpkgs-unstable)
relaxmax has joined #nixos
cosimone has quit [Remote host closed the connection]
cosimone has joined #nixos
mekeor has quit [Ping timeout: 246 seconds]
<Avaq> clever: You've lost me. :P Sorry.
<clever> Avaq: yeah, its fairly low-level arm stuff
jordandoyle has joined #nixos
<jordandoyle> hey guys, i'm trying to create an overlays/default.nix that i'm importing in my configuration.nix using nixpkgs.overlays = (import ./overlays);
<jordandoyle> however nixos-rebuild is giving me "attempt to call something which is not a function but a list", any ideas?
<clever> jordandoyle: overlays/default.nix must return a list
<clever> jordandoyle: or you must do `nixpkgs.overlays = [ (import ./overlays) ];`
<jordandoyle> ah right, makes sense
<jordandoyle> my overlays/default.nix pretty much just contains [(import ./firefox-overlays.nix)] (trying to return a list) so i might be missing something syntax-wise here
<clever> jordandoyle: what about firefox-overlays.nix ?
<jordandoyle> that's just a symlink of the firefox-overlay.nix from mozilla's repo
<jordandoyle> it works if i inline my overlays/default.nix in my configuration.nix
<clever> jordandoyle: and is firefox-overlay.nix a list or a function?
Rusty1 has joined #nixos
<jordandoyle> ah right yeah sorry, believe it's a function
oever has joined #nixos
vandenoever has quit [Ping timeout: 264 seconds]
mexisme_ has joined #nixos
wavirc22 has quit [Ping timeout: 256 seconds]
wavirc22 has joined #nixos
kleisli has quit [Ping timeout: 264 seconds]
mexisme_ has quit [Ping timeout: 246 seconds]
oever is now known as vandenoever
kenshinCH has quit [Quit: WeeChat 2.7.1]
ardumont has quit [Ping timeout: 246 seconds]
reallymemorable has quit [Quit: reallymemorable]
knupfer has quit [Ping timeout: 246 seconds]
ArchFeh has joined #nixos
roconnor has joined #nixos
ardumont has joined #nixos
<roconnor> Hi, somewhere between about Feb 7th and Feb 17th my KDE window borders disappeard on both my laptops.
<roconnor> I was hoping if I waited long enough, it would fix itself, but that hasn't happened.
vesper11 has quit [Read error: Connection reset by peer]
<roconnor> I'm not sure how to debug this.
mekeor has joined #nixos
vesper11 has joined #nixos
leonardp has joined #nixos
<evils> roconnor: you can check the history of your channel here, https://channels.nix.gsc.io/ maybe you can find an approximate range of commits that could have introduced the change, then look into the git history for KDE related stuff in that range?
alexherbo2 has joined #nixos
oever has joined #nixos
vandenoever has quit [Ping timeout: 256 seconds]
rail has joined #nixos
zupo has joined #nixos
<roconnor> maybe QT was changed from 5.12.6 to 5.16.7?
<roconnor> maybe QT was changed from 5.12.6 to 5.12.7?
<leonardp> hi, does someone have a clue why llvm_7 disappeared from from nixos-19.09 and nixpkgs-unstable and what could/should be done to get it back?
mekeor has quit [Ping timeout: 246 seconds]
<sleeping-> trying to install nixos on my thinkpad x1e gen2. i cant get wifi to work so i cant install. running wpa_supplicant -B -i enp0s31f5 -c CONFIG gives me nl80211 driver does not support authentication
cosimone has quit [Remote host closed the connection]
cosimone has joined #nixos
relaxmax has quit [Ping timeout: 250 seconds]
Jackneill has quit [Ping timeout: 256 seconds]
ardumont has quit [Ping timeout: 250 seconds]
CMCDragonkai1 has quit [Ping timeout: 264 seconds]
jordandoyle has quit [Remote host closed the connection]
rail has quit [Ping timeout: 240 seconds]
knupfer has joined #nixos
bvdw has quit [Quit: bvdw]
rogue_koder has joined #nixos
bvdw has joined #nixos
<leonardp> sleeping- what is the wifi card called?
<sleeping-> leonardp: i dont think its found.
<sleeping-> ip link shows only loopback and enp0s31f6
<sleeping-> i think i should see a wXXX interface if the wifi driver was installed
<leonardp> what does rfkill show? is it disabled by hardware switch?
<cetjs2> why youtube don't normal work in Firefox? https://imgur.com/a/BAbHYXz
<sleeping-> leonardp: rfkill shows two entries for bluetooth
<sleeping-> i dont see a hardware switch for wifi
<clever> sleeping-: does `lspci` show a wifi card?
<sleeping-> Ethernet controller and a network controller
<cetjs2> why youtube don't normal work in Firefox? https://imgur.com/a/BAbHYXz
Jackneill has joined #nixos
<evils> cetjs2: have you tried asking again? s/asking again/reloading the page/
<ar> sleeping-: enp0s31f5 is the ethernet (wired) NIC
<ar> sleeping-: do you have a wlp* interface?
shibboleth has joined #nixos
<sleeping-> ar: nope
rogue_koder has quit [Ping timeout: 256 seconds]
<ar> sleeping-: what does this show? dmesg | grep -i firmwa | grep -v Bluetooth
<cetjs2> I reloaded this page and clean cache
<cetjs2> equals bug
rogue_koder has joined #nixos
<ar> sleeping-: the network controller in lspci, does it show the model number?
<ar> sleeping-: like 04:00.0 Network controller: Intel Corporation Wireless 8260 (rev 3a)
justanotheruser has quit [Ping timeout: 256 seconds]
<sleeping-> ar: First command returns 6 lines, one is Direct firmware load for regulatory.db failed with error -2, then other ones about Spectre V2, ACPI, trackpoint, and finished loading DMC firmware i915
<sleeping-> ar: Network controller: Interl Corporation Device 2723 rev 1a
<leonardp> sleeping-: modprobe iwlwifi ?
<ar> sleeping-: that would be the intel ax200 or ax201 card
<sleeping-> i ran it, nothing returns.
<ar> sleeping-: that's exepected. now: dmesg | tail
<leonardp> lsmod grep | iwlwifi
<leonardp> ?
<sleeping-> iwlwifi is loaded
<ar> sleeping-: uname -r?
<leonardp> kernel version?
<sleeping-> 4.19.109
<sleeping-> I downloaded the minimal installer from the nixos site yesterday
<ar> that might be too old for this particular NIC
<ar> sleeping-: you downloaded the 19.09 version, right?
<sleeping-> yeah
reallymemorable has joined #nixos
drakonis_ has joined #nixos
<leonardp> too old kernel.. your adapter shoudl be supportet from 5.1 onwards
<sleeping-> so does nixos installers come with 5.1+ kernels? sorry, im new to nixos. I've been eyeing this OS up for a few months and decided to install it on my new laptop
<ar> sleeping-: 20.03 should have a new enough kernel
<{^_^}> Channel nixos-20.03-small advanced to https://github.com/NixOS/nixpkgs/commit/b323c010d7a (from 4 hours ago, history: https://channels.nix.gsc.io/nixos-20.03-small)
klntsky has quit [Remote host closed the connection]
Recoil has joined #nixos
drakonis has quit [Ping timeout: 250 seconds]
klntsky has joined #nixos
Recoil has left #nixos ["Leaving"]
drakonis_ has quit [Read error: Connection reset by peer]
<sleeping-> ar: 20.03 isnt stable though right?
drakonis has joined #nixos
<ar> sleeping-: it isn't yet, but it's supposed to be released within a month or so
drakonis_ has joined #nixos
<leonardp> sleeping-: putting `boot.kernelPackages = pkgs.linuxPackages_latest;` in your configuration.nix should do it
<leonardp> or use that iso
<leonardp> :)
<ar> leonardp: well, he'd have to boot first
<ar> to be able to do that
<sleeping-> and have networking
<ar> and get network to be able to update channels and fetch packages
<leonardp> use ethernet cable?
<leonardp> if you don't have that, that would actually really suck :/
<sleeping-> no port, i can check for an adapter in boxes full of wires and other tech neglect
justanotheruser has joined #nixos
<leonardp> can't you use the iso to boot from usbstick?
<leonardp> how are you online right now?
drakonis has quit [Ping timeout: 246 seconds]
<sleeping-> other laptop
ericsagnes has quit [Ping timeout: 246 seconds]
endformationage has joined #nixos
ArchFeh has quit [Remote host closed the connection]
cole-h has joined #nixos
zupo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ericsagnes has joined #nixos
doyougnu has joined #nixos
aveltras has joined #nixos
teto has quit [Ping timeout: 246 seconds]
c0c0 has quit [Ping timeout: 264 seconds]
<DigitalKiwi> through sheer willpower
oever has quit [Read error: Connection reset by peer]
oever has joined #nixos
captjakk has joined #nixos
s1341 has joined #nixos
<samueldr> we also build an unpublished "latest kernel" iso variant
<samueldr> new_kernel*
<evils> if it's reachable it's published, you may as well link to it
davidak has quit [Quit: Leaving]
<samueldr> evils: I say "unpublished" like "unpublicised", it's not part of the isos next to the channel release :)
teto has joined #nixos
<evils> samueldr: :) i just recall being annoyed that it's available but tricky to find
c0c0 has joined #nixos
leotaku has quit [Quit: ZNC 1.7.4 - https://znc.in]
zupo has joined #nixos
leotaku has joined #nixos
mexisme_ has joined #nixos
zupo has quit [Client Quit]
cosimone has quit [Remote host closed the connection]
cosimone has joined #nixos
drakonis has joined #nixos
wildtrees has joined #nixos
drakonis_ has quit [Ping timeout: 246 seconds]
mexisme_ has quit [Ping timeout: 246 seconds]
shidima has joined #nixos
<shidima> Hello, I have installd zsh-syntax-highlighting through my configuration.nix, but how do I source it in my zshrc?
drakonis1 has joined #nixos
drakonis_ has joined #nixos
linarcx has quit [Ping timeout: 250 seconds]
drakonis has quit [Ping timeout: 246 seconds]
vs has joined #nixos
<vs> hi there
teto has quit [Ping timeout: 246 seconds]
vs has quit [Remote host closed the connection]
drakonis_ has quit [Ping timeout: 240 seconds]
teto has joined #nixos
<{^_^}> Channel nixos-unstable-small advanced to https://github.com/NixOS/nixpkgs/commit/e7d2655efbe (from 5 hours ago, history: https://channels.nix.gsc.io/nixos-unstable-small)
<{^_^}> Channel nixpkgs-unstable advanced to https://github.com/NixOS/nixpkgs/commit/9b3515eb95d (from 2 days ago, history: https://channels.nix.gsc.io/nixpkgs-unstable)
xkapastel has quit [Quit: Connection closed for inactivity]
<avn> clever: btw, how is possible to embed derivation (from mkShell) to script? (If I'd like to build few shells with whole system)
linarcx has joined #nixos
<clever> avn: you basically just want to `echo ${toString foo.buildInputs} > $out` under `runCommand`
<clever> avn: and then depend on that runCommand somewhere in /etc
<clever> then they are rooted, and normal nix-shell can reuse them
<clever> bbl
drakonis_ has joined #nixos
rogue_koder has quit [Ping timeout: 250 seconds]
jakobrs has joined #nixos
aleph- has quit [Ping timeout: 268 seconds]
ddellacosta has joined #nixos
<jakobrs> Is it possible to build your profile without adding it to the profiles folder or switching to it? Like `nixos-rebuild build`
Heirlung has quit [Ping timeout: 240 seconds]
<jakobrs> S you'd do something like nix-env -iA nixpkgs.vim --build and it'd create a symlink to a new profile
shibboleth has quit [Quit: shibboleth]
teto has quit [Ping timeout: 246 seconds]
Heirlung has joined #nixos
<avn> clever: actually I want to reproduce `load-env-xxx` semantic probably with/without stdenv inclusion on build time (and in more mainstream way than myEnvFun)
codygman has quit [Read error: Connection reset by peer]
codygman has joined #nixos
aleph- has joined #nixos
aleph- has quit [Client Quit]
teto has joined #nixos
<ottidmes> jakobrs: I believe nix-env is just some tooling around mostly buildEnv, so you could probably just use that, although that is not exactly what you want, should be close enough
<jakobrs> right
jmeredith has quit [Quit: Connection closed for inactivity]
rogue_koder has joined #nixos
relaxmax has joined #nixos
shidima has quit [Ping timeout: 250 seconds]
<bgamari> what dhcp client do people use when debugging network issues?
<pistache> bgamari: I use dhclient (from dhcp)
<bgamari> thanks
<pistache> I recommend option '-d' to make it run in foreground
<pistache> otherwise it will detach and keep renewing the lease, as soon it acquires an address
cosimone has quit [Quit: Terminated!]
cosimone has joined #nixos
organixpear has joined #nixos
Avaq has quit [Ping timeout: 264 seconds]
<jakobrs> Is it possible to use nix-index without having 8 GB of ram?
Avaq has joined #nixos
<jakobrs> My system becomes unusable after a few seconds
<tilpner> jakobrs: curl https://index.nix.tx0.co > .cache/nix-index/files
ardumont has joined #nixos
<jakobrs> ty
energizer has joined #nixos
chagra has quit [Ping timeout: 250 seconds]
codygman has quit [Read error: Connection reset by peer]
aleph- has joined #nixos
alexherbo2 has quit [Read error: Connection reset by peer]
codygman has joined #nixos
alexherbo23 has joined #nixos
mojjo has joined #nixos
codygman has quit [Read error: Connection reset by peer]
codygman has joined #nixos
aleph- has quit [Read error: Connection reset by peer]
aleph- has joined #nixos
zeta_0 has joined #nixos
<T_S_> I have updated the nix expressions for *all* the R packages, adding new ones and removing dead ones. My expressions evaluate but I want to test building. Tested about a thousand by building a shell environment with lots of packages. I have to wait a long time between failures to get something to fix. Need to up my game and build an expression that will run in parallel and let me find a log when there's an error to f
<T_S_> ix. Any on the standard way to do thatwould be welcome.
<T_S_> *Any pointers on*
organixpear has quit [Ping timeout: 264 seconds]
xkapastel has joined #nixos
Avaq has quit [Quit: Leaving.]
<nilsirl[m]> I'd like to change my root partition from ext4 to btrs. How would I do that? My problem is that I need to run the equivalent of `nixos-rebuild` and I don't know how to do that on a liveusb
<jakobrs> nilsirl[m]: I recently did that
chagra has joined #nixos
<jakobrs> Boot from the live usb, run btrfs-convert (or whatever it is called),
codygman has quit [Read error: Connection reset by peer]
<jakobrs> mount your root partition on /mnt, and your boot partition on /mnt/boot,
kolbycrouch has quit [Remote host closed the connection]
<jakobrs> `sudo nixos-generate-config --root /mnt` to regenerate the hardware-config.nix file,
kolbycrouch has joined #nixos
<jakobrs> nixos-enter and then nixos-rebuild switch.
<jakobrs> You might need to copy over /etc/resolv.conf from liveusb to /mnt/etc/.
<jakobrs> That *should* be it (unless I missed anything)
<nilsirl[m]> jakobrs: thank you so much
aleph- has quit [Read error: Connection reset by peer]
<mojjo> does some have a recipe to build mongodb 4.x on nixos? Currently there's only 3.4. just overriding `version` and `src` did not build :(
jumper149 has joined #nixos
<mojjo> *someone
leonardp has quit [Ping timeout: 240 seconds]
codygman has joined #nixos
sleeping- has quit [Quit: bye!]
<drakonis1> did 4.0 change how the builds work?
<drakonis1> and did you override the hash?
lejonet has quit [Ping timeout: 272 seconds]
<mojjo> yes, hash is overwritten. the build starts but fails while trying to patch `site_scons/mongo_scons_utils.py`as this file does not exist anymore
sleepingforest has joined #nixos
<mojjo> nice, I can also override `patches`...
<mojjo> so I set this to `[]` hoping that this one patch is not needed anymore. now I get: `Could not find <curl/curl.h> and curl lib`
<aveltras> how can i get the dependencies from a haskell package benchmark suite in nix-shell ? i've tried the doBenchmark function but that doesn't seem to work (or i have used it incorrectly), is there anything more to do ?
o1lo01ol1o has quit [Remote host closed the connection]
<cole-h> mojjo: Did you add curl to buildInputs?
s1341 has quit [Quit: Connection closed for inactivity]
o1lo01ol1o has joined #nixos
aleph- has joined #nixos
<mojjo> yes, added it now. the error is gone. Now it fails while applying this patch: https://github.com/NixOS/nixpkgs/blob/staging/pkgs/servers/nosql/mongodb/default.nix#L44 (just removing does not work)
o1lo01ol1o has quit [Ping timeout: 246 seconds]
mehlon has joined #nixos
jakobrs has quit [Quit: WeeChat 2.7.1]
mexisme_ has joined #nixos
aleph- has quit [Read error: Connection reset by peer]
aleph- has joined #nixos
aleph- has quit [Client Quit]
aleph- has joined #nixos
mehlon has quit [Quit: Leaving]
<cole-h> What do you mean, removing it doesn't work? How can a nonexistent patch fail to apply? :P
sleepingforest has quit [Quit: bye!]
mexisme_ has quit [Ping timeout: 246 seconds]
sleepingforest has joined #nixos
<mojjo> good deduction... what I did not tell is that when removing the (boost) patch I get an error clearly related to it: 'can't find boost headers'
sadmin has joined #nixos
relaxmax has quit [Ping timeout: 246 seconds]
captjakk_ has joined #nixos
<ottidmes> mojjo: Simplest solution is to go the old PR for it, copy the file over and use that, since someone already put in the effort to make 4.0 work under NixOS, AFAIK (might still be something why they haven't merged, not sure)
captjakk_ has quit [Remote host closed the connection]
<ottidmes> How do symlinks get resolved in Nix? I have a /cfg -> /wheel/proj/nixcfg symlink but it normally it would just keep the path as /cfg/whatever, but when I query the path with nix eval, I get returned the real path
energizer has quit [Remote host closed the connection]
gxt has quit [Ping timeout: 240 seconds]
ehmry has quit [Remote host closed the connection]
<mojjo> mhh, I did not find any PR for a 4.0 version... do you maybe have a link?
snajpa- has joined #nixos
<ottidmes> Hmm, seems to happen when nix eval prints
<ottidmes> mojjo: Let me check
mikky has quit [Ping timeout: 256 seconds]
snajpa has quit [Read error: Connection reset by peer]
snajpa- is now known as snajpa
aither has quit [Read error: Connection reset by peer]
<{^_^}> #40082 (by jarjee, 1 year ago, open): mongodb: 3.4.10 -> 4.0.12
lukash|away has quit [Ping timeout: 256 seconds]
gxt has joined #nixos
mikky has joined #nixos
ehmry has joined #nixos
aither has joined #nixos
<mojjo> ok, I give it a try.. thx!
Neo-- has quit [Ping timeout: 246 seconds]
<mojjo> the only failed bullet point is 'build on other Linux distributions' so it might work on NixOS
<ottidmes> mojjo: to just get it to work, it might be easiest to fetchTarball from https://github.com/jarjee/nixpkgs/tree/update/mongodb and add an overlay for it
zeta_0 has left #nixos ["rcirc on GNU Emacs 26.3"]
bvdw has quit [Read error: Connection reset by peer]
bvdw has joined #nixos
joebobjoe has joined #nixos
<mojjo> I copied 2 two files over and added an overlay... it compiles........
<joebobjoe> ,channels
<{^_^}> Largest Nix-related channels: #nixos, #nixos-dev, #nix-darwin, #nixos-chat, #nixos-aarch64, #nixos-wiki, #nixos-borg, #nixos-security, #nix-core
drakonis1 is now known as drakonis
<joebobjoe> hi
shyim has quit [Ping timeout: 268 seconds]
sleepingforest has quit [Quit: bye!]
sleepingforest has joined #nixos
<joebobjoe> I installed nix on my mac but the nix command always disppears after reboot
<tilpner> joebobjoe: What do you do to make it reappear?
<joebobjoe> reinstall
<tilpner> Oh
<joebobjoe> now the install isn't working hold on
Neo-- has joined #nixos
<tilpner> Is it actually uninstalled, or is it just not available to your shell?
<ottidmes> joebobjoe: Is it not a matter of missing some sourcing of a script in your shell that makes them accessible?
<joebobjoe> I guess
<joebobjoe> but I did a multiuser install I think it made path changes system-wide so it's not in my home folder shell config files
<joebobjoe> I think I found the problem the installer isn't working hold on
<joebobjoe> also, when I run the installer and pipe errout to a file it puts a bunch of ^M in the log
<joebobjoe> and interactive terminal output
<joebobjoe> it's obviously not using isatty
mehlon has joined #nixos
<joebobjoe> oh the installer isn't working because I didn't give it the multiuser flags ah
<joebobjoe> would be nice for a prompt
detran has joined #nixos
sleepingforest has quit [Quit: bye!]
sleepingforest has joined #nixos
shyim has joined #nixos
<joebobjoe> ok I see nix did change /etc/zshrc I just don't know why it isn't working
<joebobjoe> when I source /etc/zshrc it works now yay
zgrep_ has joined #nixos
o1lo01ol1o has joined #nixos
zgrep has quit [Ping timeout: 240 seconds]
zgrep_ is now known as zgrep
hmpffff has joined #nixos
o1lo01ol1o has quit [Ping timeout: 250 seconds]
<hpfr[m]> I'm trying to override the src attribute of a derivation with a local path to a repo I don't want to keep fetching. what am I supposed to set src to? a path? I tried reading through fetchzip, fetchurl, and fetchurl's builder.sh but I couldn't figure it out
sadmin has quit [Remote host closed the connection]
ixxie has joined #nixos
nerdmaxx has joined #nixos
o1lo01ol1o has joined #nixos
hmpffff_ has joined #nixos
plp_ has quit [Ping timeout: 260 seconds]
hmpffff has quit [Ping timeout: 246 seconds]
cosimone has quit [Quit: Terminated!]
lord| has quit [Read error: Connection reset by peer]
plp_ has joined #nixos
lord| has joined #nixos
<Valodim[m]> is there a straightforward way to do assertions per submodule?
chagra has quit [Ping timeout: 256 seconds]
chagra has joined #nixos
energizer_ is now known as energizer
<infinisil> Valodim[m]: There are ways, but no straightforward ones
relaxmax has joined #nixos
chimay has quit [Ping timeout: 256 seconds]
<Valodim[m]> dang :)
<Valodim[m]> I've been picking up this PR https://github.com/NixOS/nixpkgs/pull/47636
<{^_^}> #47636 (by Ericson2314, 1 year ago, open): WIP nixos/nix-daemon: Organize buildMachine options with a submodule
<Valodim[m]> I'm mostly done, but can't figure out how to nicely do the system / systems assertions for each submodule
<Valodim[m]> I mean I guess I could just make an assertions that checks all submodules
<infinisil> That's a decent way yeah
shafox has quit [Remote host closed the connection]
<mdash> lewo: do you have an example of getting nixos-generators to work with openstack? i'm having problems figuring out the boot config
<Valodim[m]> ah. I was wondering how that would work, would've mentioned it as a doubt :)
<joebobjoe> ok I just installed nix again but now I get this error when running nix-shell -p nix-info --run "nix-info -m"
<joebobjoe> error: could not set permissions on '/nix/var/nix/profiles/per-user' to 755: Operation not permitted
<cole-h> hpfr[m]: Yes, you should be able to just set `src = /path/to/local/repo;`
inkbottle has quit [Quit: Konversation terminated!]
inkbottle has joined #nixos
<mdash> (has anyone else built nixos images for openstack? feel like i'm missing something crucial)
doyougnu has quit [Ping timeout: 246 seconds]
<hpfr[m]> cole-h: thanks! I tried it already, but the path I'm using gives a warning about how large it is and that memory might run out every time I rebuild. Is there a better approach that takes care of this?
sleepingforest has quit [Quit: bye!]
<cole-h> It's probably because you have build artifacts in there.
sleepingforest has joined #nixos
<cole-h> You can either remove the artifacts, or filter them using something like `lib.cleanSource` and related functions on that dir
<clever> avn: you can run nix-shell on a .drv file
<clever> > hello.drvPath
<{^_^}> "/nix/store/gq5fbhbwhmjxkv6a870kw14v7f1ywmm6-hello-2.10.drv"
<clever> avn: so you can generate a shell script that runs nix-shell ${hello.drvPath}
<ottidmes> I purified my NixOS config, outside <nixos-config> and <nixpkgs> (a special one, that only got 5 files), it should have no external influences. I also implemented my own pure evaluation, which is less strict, not having to copy all my NixOS config repos to /nix/store
<ottidmes> However I was surprised to see builtins.currentSystem still being called
<ottidmes> even though I explicitly assigned system = "x86_64-linux";
sogatori has quit [Remote host closed the connection]
<hpfr[m]> cole-h: unfortunately, I don't think there are any unecessary files there. this is the nerd fonts repo, so it has several GB of font images or something
<ottidmes> Ah, figured it out after thoroughly looking through the trace, its my other nixpkgs checkouts (i.e. unstable) in the overlay that are imported with impure.nix
<tilpner> ottidmes: You can pass --option pure-eval true to disallow currentSystem
<tilpner> (And a bunch of other things)
<joebobjoe> hi I just did a multi-user install but now running nix-shell fails
<clever> tilpner: you can also use `--option system ...` to change builtins.currentSystem, but that will lie to nix a bit too much, and then it thinks you can run the given arch, which will then fail
<tilpner> Just equip all your build machines with all the emulatedSystems c.c
oever has quit [Read error: Connection reset by peer]
<tilpner> (No, don't)
oever has joined #nixos
<clever> tilpner: :D
lassulus has quit [Ping timeout: 272 seconds]
lassulus has joined #nixos
troydm has quit [Ping timeout: 246 seconds]
cosimone has joined #nixos
mjrosenb has quit [Ping timeout: 250 seconds]
codygman has quit [Ping timeout: 240 seconds]
<ottidmes> clever: tilpner: I am already doing that with scopedImport, throwing an exception on use
reallymemorable has quit [Quit: reallymemorable]
<ottidmes> Which means I definitely keep the pure evaluation seperate, cause it will be noticable slower due to no caching caused by scopedImport
<ottidmes> tilpner: I knew about pure-eval, it was why I mentioned I implemented my own, cause "No file system access is allowed outside of the paths returned by fetch{Git,Mercurial,url,Tarball}. Thus 'nix build -f ./foo.nix' is not allowed."
<ottidmes> I dont want to copy all my config to /nix/store on every build
codygman has joined #nixos
sleepingforest has quit [Quit: bye!]
sleepingforest has joined #nixos
lukash|away has joined #nixos
<joebobjoe> hi, why does my multiuser install of nix try to change the permissions of a directory owned by the superuser? https://github.com/NixOS/nix/issues/3435
<{^_^}> nix#3435 (by amckinlay, 17 minutes ago, open): error: could not set permissions on '/nix/var/nix/profiles/per-user' to 755: Operation not permitted
mehlon has quit [Quit: Leaving]
<ottidmes> I know work is being done/proposed to improve error messages within Nix(OS), but I would also love to see debugging becoming easier. Maybe I should put some work in trying to take what is produced by --show-trace and improve the UX
alexherbo23 has quit [Ping timeout: 256 seconds]
jmeredith has joined #nixos
<joebobjoe> I can't even get -vvvv to give me any useful information
<joebobjoe> nix-shell and nix-env just immediately error out with permissions error
mexisme_ has joined #nixos
lejonet has joined #nixos
troydm has joined #nixos
mexisme_ has quit [Ping timeout: 246 seconds]
sleepingforest has quit [Quit: bye!]
sleepingforest has joined #nixos
hmpffff_ has quit [Quit: nchrrrr…]
ajp has quit [Quit: No Ping reply in 180 seconds.]
xelxebar has quit [Ping timeout: 240 seconds]
captjakk has quit [Remote host closed the connection]
captjakk has joined #nixos
justanotheruser has quit [Ping timeout: 250 seconds]
<cole-h> joebobjoe: Are you on Catalina, but any chance?
<cole-h> s/but/by
<joebobjoe> cole-h: yes
<ottidmes> joebobjoe: Although I can't really help you much (I am on NixOS, no experience with Mac or Nix outside of NixOS), but I did saw people say that if you use sudo you have to be careful for the $HOME variable not being changed
<{^_^}> nix#2925 (by mroi, 41 weeks ago, open): /nix will not be writable on macOS Catalina
<joebobjoe> yea I followed the directions there a while ago to create a separate volume for /nix
<joebobjoe> but there are so many comments in the thread I can't find my error
<joebobjoe> github collapses the comments
<cole-h> joebobjoe: I think this is the "best" solution: https://github.com/NixOS/nix/issues/2925#issuecomment-539570232
<cole-h> (I don't use Apple anything so I can't help more than this)
LysergicDreams has quit [Ping timeout: 256 seconds]
captjakk has quit [Ping timeout: 240 seconds]
<cole-h> Might also be interested in https://github.com/NixOS/nix/pull/3212
<{^_^}> nix#3212 (by LnL7, 19 weeks ago, open): install: configure and bootstrap synthetic.conf on darwin
LysergicDreams has joined #nixos
mexisme_ has joined #nixos
orivej has quit [Ping timeout: 240 seconds]
<joebobjoe> yup that's what I did
<joebobjoe> hm for some reason my /nix is owned by me
<joebobjoe> maybe I should make it owned by super user
teto has quit [Ping timeout: 260 seconds]
<joebobjoe> hm either way still doesn't fix the weird permissions error
teto has joined #nixos
justanotheruser has joined #nixos
joebobjoe has quit [Quit: Lost terminal]
vk3wtf has quit [Quit: WeeChat 2.7.1]
vk3wtf has joined #nixos
jgeerds has joined #nixos
<hpfr[m]> I can't find the definition for buildLinux in nixpkgs, what am I doing wrong? running it in a repl said https://github.com/NixOS/nixpkgs/blob/master/pkgs/os-specific/linux/kernel/generic.nix but there's no buildLinux in there?
jb55 has joined #nixos
ajp has joined #nixos
Avaq has joined #nixos
<cole-h> hpfr[m]: buildLinux is a function that takes in attrs and passes them to that file
<hpfr[m]> where is buildLinux defined though?
<cole-h> In a way, that file /is/ `buildLinux`, when combined with the specified attrs and called by `callPackage`
<cole-h> This is the `buildLinux` definition: https://paste.rs/mZ7.jpg
<cole-h> (which is in top-level.nix)
<cole-h> `top-level/all-packages.nix` I mean
orivej has joined #nixos
<linarcx> Guys, i've added `clang-manpages` to shell.nix. How can i use it now?
acarrico has quit [Ping timeout: 256 seconds]
<hpfr[m]> cole-h: oh, sorry. that was the first place I looked, but I'm searching GitHub because I'm on my laptop, and now I see GitHub only renders the first 15k lines of the file. thanks
mojjo has quit [Ping timeout: 240 seconds]
tobi has joined #nixos
tobi is now known as Guest13590
m1cr0m4n has joined #nixos
<linarcx> How use man pages when i'm using nix-shell?
mexisme_ has quit [Ping timeout: 246 seconds]
captjakk has joined #nixos
<Guest13590> hello, the processing package is currently broken because a zip file which is downloaded in the build process has changed and the hash doesn't match anymore (there already is a ticket for this). while this is being fixed, I tried to manually override the hash via an overlay. in order to do this, I created ~/.config/nixpkgs/overlays/processing.nix which contains "self: super: { processing =
<Guest13590> super.processing.override { buildPhase=''...''; }; } (properly formatted of course. the hash is defined and used in the buildPhase). this didn't work however, since now I am getting the error message "anonymous function at ... processing/default.nix:1:1 called with unexpected argument 'buildPhase' at...". what am I doing wrong?
<linarcx> I've add `clang-manpages`, but when i want to use it like this: `man clang`, it says: `man: No entry for clang in the manual.`
captjakk has quit [Ping timeout: 250 seconds]
cosimone has quit [Quit: Quit.]
maddo has quit [Quit: See ya]
sleepingforest has quit [Quit: bye!]
sleepingforest has joined #nixos
chloekek has quit [Ping timeout: 240 seconds]
relaxmax has quit [Quit: Leaving]
albakham has joined #nixos
<albakham> Hello
mjrosenb has joined #nixos
<albakham> How can I allow polybar to access python,ls,grep in a shell script?
<symphorien> Guest13590: you need overrideAttrs, not override
<albakham> If I run the script in a console, no problem, but if I pass this script to polybar, problem because polybar can't access the required programs
maddo has joined #nixos
<Guest13590> symphorien: ahh ok, override is used to change the arguments to the function. it has worked now, thanks a lot
<symphorien> yes exactly
ddellacosta has quit [Ping timeout: 250 seconds]
<lewo> mdash: I unfortunately no longer have an access to an openstack cloud. But IIRC, nixos-generator was working out of the box. However, maybe there is an issue with the NixOS openstack module.
mjrosenb has quit [Ping timeout: 240 seconds]
<roconnor> evils: Okay I think the change is plasma 5.16 -> plasma 5.17: https://kde.org/announcements/plasma-5.17.0.php
<roconnor> `` window borders are turned off by default.''
<roconnor> I guess this is working as intended
<roconnor> Though without window shadows, it looks pretty odd.
codezero has quit [Quit: Ping timeout (120 seconds)]
sleepingforest has quit [Quit: bye!]
codezero has joined #nixos
sleepingforest has joined #nixos
meck has quit [Remote host closed the connection]
meck has joined #nixos
<roconnor> okay I've enabled compositor. Now I have shadows.
vesper11 has quit [Ping timeout: 250 seconds]
fusion809 has quit [Remote host closed the connection]
Tucky has quit [Quit: WeeChat 2.7.1]
m1cr0m4n is now known as m1cr0man
vesper11 has joined #nixos
teto has quit [Ping timeout: 246 seconds]
sleepingforest has quit [Quit: bye!]
sleepingforest has joined #nixos
dingenskirchen has quit [Quit: dingenskirchen]
dingenskirchen has joined #nixos
__monty__ has quit [Quit: leaving]
ixxie has quit [Ping timeout: 240 seconds]
oever has quit [Read error: Connection reset by peer]
captjakk has joined #nixos
oever has joined #nixos
aleph- has quit [Read error: Connection reset by peer]
albakham has quit [Quit: WeeChat 2.7.1]
teto has joined #nixos
reallymemorable has joined #nixos
mjrosenb has joined #nixos
aleph- has joined #nixos
sleepingforest has quit [Quit: bye!]
captjakk has quit [Ping timeout: 240 seconds]
jason` has joined #nixos
jason` has quit [Remote host closed the connection]
aleph- has quit [Read error: Connection reset by peer]
jasongrossman has joined #nixos
aleph- has joined #nixos
sleepingforest has joined #nixos
mjrosenb has quit [Ping timeout: 256 seconds]
aleph- has quit [Read error: Connection reset by peer]
ArchFeh has joined #nixos
aleph- has joined #nixos
obadz has quit [Quit: WeeChat 2.7.1]
cge has joined #nixos
linarcx has quit [Ping timeout: 264 seconds]
swapgs has quit [Ping timeout: 258 seconds]
Rusty1 has quit [Remote host closed the connection]
fendor has quit [Read error: Connection reset by peer]
aleph- has quit [Read error: Connection reset by peer]
aleph- has joined #nixos
captjakk has joined #nixos
obadz has joined #nixos
captjakk has quit [Remote host closed the connection]
captjakk has joined #nixos
sleepingforest has quit [Quit: bye!]
ArchFeh has quit [Remote host closed the connection]
swapgs has joined #nixos
swapgs has joined #nixos
swapgs has quit [Changing host]
<jackdk> I am trying to build the VM at https://github.com/qfpl/reflex-workshop/blob/ce6beb8d27a0f96faf733a41cd5a4cc9c8dd81dc/nix/vm.nix . It successfully built every derivation except the actual VM, where what looks like a `nix copy` fails with "preallocating file of 3346208 bytes: No space left on device". I have >25GiB free on the volume that holds /nix, and I have raised the baseImageSize in vm.nix but still get this problem.
<jackdk> Where else should I be looking?
sleepingforest has joined #nixos
Guest13590 has quit [Quit: leaving]
gustavderdrache has joined #nixos
jasongrossman has quit [Quit: ERC (IRC client for Emacs 26.3)]
jasongrossman has joined #nixos
sleepingforest has quit [Quit: bye!]
sleepingforest has joined #nixos
<atemu12[m]> jackdk: What size is /tmp/?
<atemu12[m]> Just a hunch
jasongrossman has left #nixos [#nixos]
<jackdk> it is a 7.7 GiB tmpfs
<jackdk> I will try setting TMPDIR and get back to you, as nix-build does use /tmp by default
<jackdk> it is still failing, same error. I see nix-build-PID-0 directories created in my new tmpdir but not files in there, and /tmp is still filling up and dying
sleepingforest has quit [Quit: bye!]
<jackdk> what else can I do to get this under control?
<ottidmes> remount /tmp for the time being?
aleph- has quit [Ping timeout: 240 seconds]
sleepingforest has joined #nixos
chocopuff has joined #nixos
aleph- has joined #nixos
<chocopuff> I think I have an issue where multiple times, my computer stops accepting USB input, the mouse and keyboard is unresponsive, and I think it has to do with the kernel
<chocopuff> so I rolled back, and so far, fingers crossed, it seems to hvae fixed the issue, idk how to reproduce it
<chocopuff> how do I find out what package the kernel comes from?
philr has joined #nixos
detran has quit [Ping timeout: 264 seconds]
<chocopuff> as in, is there a way to know what I can put into my configuration.nix to manually select the kernel I am running right now? I don't know how to find that information
<atemu12[m]> -> `boot.kernelPackages`
<ottidmes> chocopuff: boot.kernelPackages = pkgs.linuxPackages_latest; (for example)
<chocopuff> ottidmes: ye, the question is more specifically
<chocopuff> how do I know what to set bootkernelPackages with the kernel that I am currently running in the rolled-back version
oever has quit [Ping timeout: 250 seconds]
<ottidmes> chocopuff: uname -a?
<chocopuff> yeah
<chocopuff> I've done that already
<atemu12[m]> chocopuff: Which is the working and which the broken kernel version?
<chocopuff> I can't say without rebooting, brb, but the one that is working at the moment is 4.19
<chocopuff> the kernel's are pre-compiled correct?
<chocopuff> I was going by the date of compilation in uname -a atm
<chocopuff> will be right back
chocopuff has quit [Quit: WeeChat 2.7.1]
chocopuff has joined #nixos
<chocopuff> so the new one that appears to hang is 4.19.109, the old one that works is 4.19.105
<chocopuff> boot.kernelPackages = pkgs.linuxPackages_4_19_105; does not work
<ottidmes> chocopuff: you could pin a nixpkgs for that version and use the rest with latest version, or you could try maybe the 5.x kernels
drewr has quit [Ping timeout: 240 seconds]
bhipple has quit [Ping timeout: 250 seconds]
<chocopuff> ok let me look into pinning
Avaq has quit [Quit: Leaving.]
bhipple has joined #nixos
<chocopuff> ottidmes: the problem is, they're both 4.19 kernels
<chocopuff> i don't see an option for .105 vs .109
<chocopuff> maybe i'll just jump to the 5.x
<chocopuff> https://nixos.wiki/wiki/Linux_kernel <- I tried listing them all as suggested, but it only shows a _4_19 package
<ottidmes> chocopuff: (import (fetchTarball { url = "https://github.com/NixOS/nixpkgs/tarball/49b4266ad2c779b11c0c492418a3e28be4822320"; sha256 = "1pw5xi7jqrpxj4yvvgkvbkx8dav00zqzvjvs9skrvc9yhcsmk77y"; }) { overlays = []; }).kernelPackages_4_19
civodul has quit [Quit: ERC (IRC client for Emacs 26.3)]
mexisme_ has joined #nixos
<ottidmes> that should in theory (untested) use a nixpkgs pinned to the version it bumped to 4.19.105 (I looked it up and determined the sha)
<ottidmes> jumping to 5.x would always be better than sticking with some old kernel (although it could still be fixed soonish, who knows)
<jackdk> ottidmes: seems promising
mjrosenb has joined #nixos
mjrosenb has quit [Client Quit]
hio has quit [Quit: Connection closed for inactivity]
mexisme_ has quit [Ping timeout: 260 seconds]
cjpbirkbeck has joined #nixos
iqubic has quit [Remote host closed the connection]
maddo has quit [Quit: See ya]
<jackdk> well it got further but still fell over after eating 25GiB. time to break out the USB-C SSD...
<chocopuff> ottidmes: thanks, I will keep both in mind and jump to 5.x for now
o1lo01ol1o has quit [Ping timeout: 246 seconds]
knupfer1 has joined #nixos
<bqv[m]> hey, any particular reason nixos is currently downloading thousands of steam .debs?
ris has quit []
pikajude has quit [Quit: ZNC 1.7.5 - https://znc.in]
knupfer has quit [Ping timeout: 246 seconds]
knupfer1 is now known as knupfer
<bqv[m]> oh nevermind, it's done
<bqv[m]> with those at least
o1lo01ol1o has joined #nixos
knupfer has quit [Remote host closed the connection]
o1lo01ol1o has quit [Remote host closed the connection]
o1lo01ol1o has joined #nixos
<manveru> bqv[m]: it basically replicates SteamOS in a chroot
<ottidmes> makes sense, patching it all would be nigh impossible
<bqv[m]> i know, i was just wondering why it wasn't just copying from hydra
swapgs has quit [Ping timeout: 250 seconds]
<manveru> because steam is marked as non-free
<manveru> so hydra won't build it
<NobbZ[m]> Probably because of redistribution issues in the license of steam?
o1lo01ol1o has quit [Ping timeout: 250 seconds]
Makaveli7 has joined #nixos
jmeredith has quit [Quit: Connection closed for inactivity]
swapgs has joined #nixos
swapgs has joined #nixos
swapgs has quit [Changing host]