<vek_tor>
do people generally prefer using home-manager as a module or managing it separately from the main nixos configuration? or is there not a clear preference?
<benkolera>
I think the end game is to want to use it as a nixos module so you can have one declarative conf for the whole system. But it's definitely easier to get started with it standalone.
<benkolera>
If you are pretty familiar with nix and systemd already, skip to the nixos module, but if you are fairly new to at least HM and nix, maybe stick with standalone till you find your feet.
<benkolera>
vek_tor: ^^
<vek_tor>
i've been using nixos without h-m for a few months, and i'm... okay at it, I think
phirsch has quit [Read error: Connection reset by peer]
<benkolera>
Fwiw, i dislike putting hm in as a channel and prefer to fetchgit from my nixos config, but that's just me. :)
boxscape has quit [Remote host closed the connection]
<benkolera>
The main issue with the nixos module is that if your activation script crashes for some reason, you have to go hunting in systemctl and journalctl for info.
<benkolera>
So if you are kinda nix proficient and pretty systemd savvy, then you'll be fine.
<hpfr[m]>
benkolera: is your config public? I’ve had trouble with systemd and paths with stuff like sxhkd
<benkolera>
Yeah, but it's a little weird as it aggressively forces anything that nixos builds to be in git. :)
<vek_tor>
i guess it looks like for some stuff, like i3 config files, home-manager defines way more stuff inside the .nix file than nixpkgs does
mexisme has joined #home-manager
vek_tor has quit [Remote host closed the connection]
mexisme has quit [Ping timeout: 272 seconds]
phirsch has joined #home-manager
mexisme has joined #home-manager
phirsch has quit [Remote host closed the connection]
phirsch has joined #home-manager
cole-h has quit [Ping timeout: 272 seconds]
mexisme has quit [Ping timeout: 248 seconds]
boxscape_ has quit [Remote host closed the connection]
kp125 has joined #home-manager
kp125 has quit [Ping timeout: 272 seconds]
malSet has quit [Ping timeout: 240 seconds]
malSet has joined #home-manager
mexisme has joined #home-manager
sgraf has joined #home-manager
vdemeester has joined #home-manager
davidtwco has joined #home-manager
mexisme has quit [Ping timeout: 245 seconds]
kp125 has joined #home-manager
c0c0 has joined #home-manager
newuser99 has joined #home-manager
<newuser99>
I have `programs.zsh = { plugins = [{name = "starship"; src = pkgs.starship;}]; `. After running home-manager switch, I have `$PATH/home/user/.nix-profile/bin:/nix/var/nix/profiles/default/bin:/sbin:/bin:/usr/sbin:/usr/bin:/home/user/.zsh/plugins/starship`, which points to `-home-manager-files/.zsh/plugins/starship/`, but the actual executable
<newuser99>
is `-home-manager-files/.zsh/plugins/starship/bin/starship*`
<newuser99>
So when I do `eval "$(starship init zsh)"` as in the directions, it doesn't work. Am I supposed to be doing something different?
<newuser99>
(i.e. it's hitting the grandparent directory instead of the actual executable)
newuser99 has quit [Remote host closed the connection]
<phirsch>
Hi! Does anyone know if it is possible to skip a package (possibly including its dependencies) from an update if it fails to build after doing 'nix-channel --update'? As I understand it, it should be possible to have that (those) particular subtree(s) remain on the previous working commit, while still updating everything else - instead of everything else being held up by a single failing package.
tbenst[m] has quit [*.net *.split]
tbenst[m] has joined #home-manager
immae has joined #home-manager
<immae>
hello, I just got bitten by something I didn't expect: latest channel update made hm rebuild tmux, which provoked a restart of tmux.service and thus my whole session without warning. Can I avoid the restart of a specific systemd service when switching config?
mexisme has joined #home-manager
asymmetric has joined #home-manager
asymmetric has quit [Changing host]
mexisme has quit [Ping timeout: 248 seconds]
romanofski has quit [Ping timeout: 240 seconds]
romanofski has joined #home-manager
mexisme has joined #home-manager
mexisme has quit [Ping timeout: 240 seconds]
phirsch has quit [Ping timeout: 240 seconds]
kp125 has quit [Remote host closed the connection]
kp125 has joined #home-manager
mexisme has joined #home-manager
mexisme has quit [Ping timeout: 248 seconds]
cole-h has joined #home-manager
c0c0 has quit [Quit: WeeChat 2.6]
kp125 has quit [Remote host closed the connection]
__monty__ has joined #home-manager
kp125 has joined #home-manager
kp125 has quit [Quit: Leaving]
mexisme has joined #home-manager
boxscape has joined #home-manager
<rycee>
immae: Ah, good point. I guess the tmux service should include `X-RestartIfChanged = false`
<rycee>
Hmm, actually. Is there a tmux service in HM?
<rycee>
newuser99: Try removing the starship plugin from zsh and instead add `programs.starship.enable = true` to your HM configuration.
<cole-h>
Unless you have anything else that needs to be addressed, it's been rebased and has been in working order (at least for me these past couple weeks)
<immae>
rycee: no there is none, but I had one :) X-RestartIfChanged is the solution indeed, thanks :D
<rycee>
Ah, ok. Yeah, putting that in the Unit section should prevent the script from restarting it.
kp125 has quit [Quit: Leaving]
__monty__ has quit [Quit: leaving]
<cole-h>
rycee: Thanks for merging fish plugins. Now I can track master again! ❤
<rycee>
np :-)
<rycee>
Good night!
<cole-h>
o/
boxscape has quit [Remote host closed the connection]