AllanEspinosa has quit [(Ping timeout: 255 seconds)]
<NixOS_GitHub>
[nixpkgs] peterhoeg closed pull request #24790: mpv: we package waf, no reason to inline it (master...f/mpv) https://git.io/vSKK9
filterfish has joined #nixos
ryanartecona has quit [(Quit: ryanartecona)]
<gchristensen>
this is wild
rcschm has joined #nixos
dridus has quit [(Quit: znc? wassat?)]
<Dezgeg>
now that's enterprisey
<Dezgeg>
does it do kexec or something?
acowley_away is now known as acowley
<gchristensen>
it must
mtetreault_ has quit [(Quit: Leaving)]
rcschm has quit [(Ping timeout: 252 seconds)]
byteflame has quit [(Ping timeout: 255 seconds)]
byteflam1 has quit [(Ping timeout: 240 seconds)]
<pingveno>
Hmmm, there is a pre-built version of the package (Apache Directory Studio)
<pingveno>
I'm not sure if that's kosher, but for now I'm just using it locally.
<gchristensen>
sounds ok :)
<pingveno>
Does policy allow pre-built versions? There's a GTK component, but most of it's Java-based.
mtetreault has joined #nixos
<gchristensen>
sometimes, we do that for elasticsearch for example
Guest60 has joined #nixos
mudri has quit [(Quit: WeeChat 1.7)]
<mtetreault>
Hi! I'm pretty new to Nixos and I am facing to issues that I cannot find a solution to. First one is probably an easy one, I am trying to enable keepass plugin (keepasshttp), but I cannot figure out how. I tried mirrexagon explanation here: https://github.com/NixOS/nixpkgs/issues/15514 but that doesn't work. It gives me an error the option packageOverrides does not exist..
<tippenein>
mostly, what _is_ reflex-dom in this case? because it's not declared anywhere
takle has quit [(Ping timeout: 255 seconds)]
<gchristensen>
tippenein: ( with p; [reflex-dom]) == ([p.reflex-dom])
eacameron has joined #nixos
spinus has quit [(Ping timeout: 240 seconds)]
<gchristensen>
Dezgeg: http://forum.ipxe.org/showthread.php?tid=7589 "Forgive my injection here, but I believe UEFI booting through iPXE requires the kernel's implicit initrd=<initfilename> here for things to work within the argument's list." "initrd=initrd" trying this
alx741_ has quit [(Quit: alx741_)]
byteflam1 has joined #nixos
byteflame has joined #nixos
alx741 has joined #nixos
johnsonav has joined #nixos
<tippenein>
thanks gchristensen . that's gross, but now I know what to change it to
<ToxicFrog>
I have no idea how to even start debugging when I nixos-rebuild --show-trace and the trace doesn't show any frames in /etc
Guest60 has quit [(Quit: My Mac Pro has gone to sleep. ZZZzzz…)]
<savanni>
Is anyone else having trouble with the heroku CLI? I have version 3.43.16 installed from the 17.03 channel, but I only have auth, build, login, plugins, and update in the list of heroku commands.
johnsonav has quit [(Ping timeout: 240 seconds)]
hamishmack has quit [(Quit: hamishmack)]
Guest60 has joined #nixos
boothead has quit [(Ping timeout: 255 seconds)]
boothead has joined #nixos
hexagoxel has quit [(Ping timeout: 260 seconds)]
hexagoxel has joined #nixos
takle has joined #nixos
johnsonav has joined #nixos
Guest60 has quit [(Quit: My Mac Pro has gone to sleep. ZZZzzz…)]
takle has quit [(Ping timeout: 252 seconds)]
byteflam1 has quit [(Ping timeout: 240 seconds)]
byteflame has quit [(Ping timeout: 240 seconds)]
zeus_ has joined #nixos
zeus_ has quit [(Remote host closed the connection)]
zeus_ has joined #nixos
steshaw has joined #nixos
alunduil has joined #nixos
colescott has joined #nixos
justanotheruser has quit [(Ping timeout: 240 seconds)]
uralbash has joined #nixos
acowley is now known as acowley_away
<alunduil>
Anyone else noticing strange behavior from fcron after the 17.03 release?
jsgrant has joined #nixos
moet has joined #nixos
<NixOS_GitHub>
[nixpkgs] benley opened pull request #24813: nixos: Add nm-openvpn to the networkmanager group (master...nm-openvpn) https://git.io/vSPO5
moet has quit [(Ping timeout: 240 seconds)]
takle has joined #nixos
edvorg has quit [(Ping timeout: 245 seconds)]
mguentner3 has quit [(Quit: WeeChat 1.7)]
takle has quit [(Ping timeout: 255 seconds)]
roygbiv has quit [(Quit: ™)]
edvorg has joined #nixos
mguentner has joined #nixos
RchrdB has quit [(Ping timeout: 260 seconds)]
mguentner2 has joined #nixos
moet has joined #nixos
<NixOS_GitHub>
[nixpkgs] copumpkin pushed 1 new commit to staging: https://git.io/vSP3d
<NixOS_GitHub>
nixpkgs/staging 0b03479 Dan Peebles: bonnie: fix on Darwin
justanotheruser has joined #nixos
moet has quit [(Client Quit)]
mguentner has quit [(Ping timeout: 255 seconds)]
coltfred has quit [(Ping timeout: 252 seconds)]
simendsjo has quit [(Remote host closed the connection)]
sellout- has joined #nixos
colescott has quit [(Quit: WeeChat 1.0.1)]
nh2 has quit [(Quit: Leaving.)]
hyphon81 has quit [(Remote host closed the connection)]
mizu_no_oto has joined #nixos
niklob has joined #nixos
<niklob>
waylang is sooooo good! :)
<niklob>
wayland
schoppenhauer has quit [(Ping timeout: 255 seconds)]
<pi3r>
error: attribute dhall-bash missing, at /home/vagrant/projects/haskell/Haskell-Dhall-Bash-Library/release.nix:26:18
<jluttine>
i'm writing a prestart script for a service. how can i run some commands in it as a different user? i tried `sudo -u otheruser foo bar` but it complains that sudo command isn't found
ebzzry has quit [(Ping timeout: 252 seconds)]
<tippenein>
what is this referring to: variable $src or $srcs should point to the source
<dbrock>
dmj`: python2.7-boto3-1.4.4... it seems to need network access to run its test suite
<dbrock>
I'm running the command nixops-deploy
<dbrock>
`nixops deploy', sorry
<dmj`>
dbrock: can you disable the test suite
[0x4A6F] has quit [(Ping timeout: 260 seconds)]
[0x4A6F] has joined #nixos
<srhb>
niklob: THanks for the link. Unfortunately most of my reason for using XMonad and not one of the many others (probably great) tiling wms is that it's extensible in Haskell. :)
<dbrock>
dmj`: hmm, maybe!?
<dmj`>
dbrock: in haskell packages there is a nix option, dontCheck, maybe there is some python equivalent
<dmj`>
dbrock: yea, as I suspected, doCheck = false;
<dmj`>
put that in your derivation
<dmj`>
and smoke it
calvertvl has quit [(Ping timeout: 240 seconds)]
katyucha has quit [(Ping timeout: 260 seconds)]
derjohn_mob has joined #nixos
Guest48023 has joined #nixos
pi3r has quit [(Ping timeout: 240 seconds)]
phreedom has quit [(Ping timeout: 255 seconds)]
<hyper_ch>
howto find out who is the maintainer of a package?
katyucha has joined #nixos
teknico has joined #nixos
<Havvy>
Find the package's default.nix and look for the meta fields.
systemfault has quit [(Quit: Bye!)]
myndzi has quit [(K-Lined)]
Forkk has quit [(K-Lined)]
<hyper_ch>
Havvy: thx
galen has quit [(Ping timeout: 240 seconds)]
Forkk has joined #nixos
<hyper_ch>
looks like viric is the maintainer
myndzi has joined #nixos
Filystyn has joined #nixos
Filystyn has quit [(Changing host)]
Filystyn has joined #nixos
<NixOS_GitHub>
[nixpkgs] dezgeg pushed 1 new commit to master: https://git.io/vSPBw
<NixOS_GitHub>
nixpkgs/master 9ae8ce8 Tuomas Tynkkynen: dosfstools: Use --enable-compat-symlinks...
<toppler>
gchristensen: I'm just adding systemd.services.rsnapshothourly and systemd.timers.rsnapshothourly elements to configuration.nix. Seems to work.
<NixOS_GitHub>
[nixpkgs] dezgeg pushed 1 new commit to master: https://git.io/vSPE3
<NixOS_GitHub>
nixpkgs/master b002d6a Tuomas Tynkkynen: Revert "btrfs-progs: 4.8.2 -> 4.10.2"...
roberth has joined #nixos
rcschm has quit [(Remote host closed the connection)]
rcschm has joined #nixos
rcschm has quit [(Remote host closed the connection)]
thc202 has joined #nixos
smw has joined #nixos
georges-duperon has joined #nixos
takle has joined #nixos
civodul has joined #nixos
katyucha has quit [(Ping timeout: 255 seconds)]
<NixOS_GitHub>
[nixpkgs] Mic92 pushed 1 new commit to release-17.03: https://git.io/vSPuS
rmrfroot has left #nixos ["ERC (IRC client for Emacs 25.1.1)"]
Dezgeg_ has joined #nixos
vaibhavsagar has quit [(Ping timeout: 240 seconds)]
mudri has joined #nixos
takle has quit [(Remote host closed the connection)]
<jophish>
hahaha, warning: rewriting hashes in ‘/nix/store/...’; cross fingers
filterfish has quit [(Ping timeout: 252 seconds)]
jgertm has quit [(Ping timeout: 240 seconds)]
takle has joined #nixos
takle has quit [(Remote host closed the connection)]
takle has joined #nixos
filterfish has joined #nixos
bastian__ has quit [(Ping timeout: 240 seconds)]
<NixOS_GitHub>
[nixpkgs] Mic92 pushed 1 new commit to release-17.03: https://git.io/vSPa4
<NixOS_GitHub>
nixpkgs/release-17.03 f9ada74 Jon Banafato: keepassx-community: 2.1.3 -> 2.1.4...
<jophish>
Could someone check the validity of /nix/store/qlds7v1anxnl4s7j1vykf4h3a8zr4x6a-unix-compat-0.4.3.1 on cache.nixos.org please
<jophish>
I'm running nix-store --repair-path and still getting a hash mismatch
<jophish>
I have a version on another machine which is correct
takle has quit [(Ping timeout: 260 seconds)]
eacameron has joined #nixos
fabian_a has joined #nixos
filterfish has quit [(Ping timeout: 255 seconds)]
eacameron has quit [(Ping timeout: 240 seconds)]
faffolter has quit [(Ping timeout: 255 seconds)]
eacameron has joined #nixos
eacameron has quit [(Ping timeout: 255 seconds)]
bastian__ has joined #nixos
<jophish>
This might be a problem on my end actually. I tried repair-path with binary-cache set to a machine with a good version of this path and I get a different wrong hash
Filystyn has quit [(Quit: Konversation terminated!)]
Filystyn has joined #nixos
bastian__ has quit [(Ping timeout: 240 seconds)]
stukj has quit [(Remote host closed the connection)]
freusque has joined #nixos
bastian__ has joined #nixos
<unlmtd[m]>
just had a thought; web-facing services with trivial exploits could more easily expose the entire configuration of the machine since the store is world-readable. right?
<jophish>
In fact, `nix-store --verify --check-contents --repair` doesn't seem to fix anything
jgertm has joined #nixos
derjohn_mob has quit [(Ping timeout: 240 seconds)]
jensens has joined #nixos
bennofs has quit [(Ping timeout: 255 seconds)]
<NixOS_GitHub>
[nixpkgs] dbrock opened pull request #24819: ethrun: init at 0.1.0 (master...ethrun) https://git.io/vSP6I
RchrdB has joined #nixos
spinus has joined #nixos
__Sander__ has joined #nixos
goibhniu1 has joined #nixos
eacameron has joined #nixos
goibhniu has quit [(Ping timeout: 245 seconds)]
DutchWolfie has joined #nixos
DutchWolfie has quit [(Changing host)]
DutchWolfie has joined #nixos
<Filystyn>
what package of latex has xelatex on nixos?
<srhb>
pi3r: Are you sure you don't just want to use pin.nix always?
<srhb>
I mean, by default.
<srhb>
Filystyn: It's probably pkgs.texlive.combine in your configuration.nix
<srhb>
pi3r: Anyway, if you leave the issue open (and perhaps add that comment) perhaps someone else can weigh in with a less handwavy non-explanation than mine ;-)
<srhb>
I'd like to read that at least.
jrolfs has quit [(Ping timeout: 245 seconds)]
<Filystyn>
nope pkgs did not fix it
<srhb>
Filystyn: What's the new error?
<srhb>
Filystyn: By the paste it looks like you might be redefining environment.systemPackages rather than adding that one entry to the existing list. That will fail.
<srhb>
(But I'm guessing, so let's see the actual error)
<Filystyn>
seems he is mad on (texlive) now let me apste
aminechikhaoui has quit [(Ping timeout: 240 seconds)]
<srhb>
Oh, yes, same deal
<srhb>
texlive is in pkgs, as we just saw
<srhb>
so it's pkgs.texlive.combine { inherit (pkgs.texlive ...
<steveeJ>
if I use nix-instantiate with --add-root, how can I ever clean the derivation up again?
<srhb>
steveeJ: Delete the root.
<steveeJ>
srhb: which reference do I have to the root?
filterfish has quit [(Ping timeout: 260 seconds)]
<srhb>
doesn't it say when you nix-instantiate it?
<steveeJ>
so there is much potential for wasted space if I ever lose this string?
<steveeJ>
which already happened many, many, times :-D
<srhb>
steveeJ: You can always look through /nix/var/nix/gcroots
<steveeJ>
actually not, I've always used nix-shell as a wrapper so far
<Filystyn>
it worked ;-)
<Filystyn>
thank you
<srhb>
Filystyn: Welcome.
chattered has joined #nixos
<srhb>
steveeJ: As in nix-shell --add-root ?
<srhb>
But that requires you to specify the location of the root, doesn;t it?
proteusguy has quit [(Ping timeout: 240 seconds)]
<steveeJ>
srhb: no, I have looked at the nix-shell code, it adds a root
<steveeJ>
I'm wondering when the environments created by nix-shell are cleaned up
<steveeJ>
and I want to speed instantiate these manually in the future to reduce evaluation costs
<srhb>
steveeJ: Right, so, when you exit the nix shell and nix-collect-garbage, it is gc\ed
<srhb>
If you nix-shell --add-gc-root my-roots, then it won't be collected until you remove those
<srhb>
You may want to keep the roots locally, in which case, use --indirect as well
<Filystyn>
ok it seems to work now guys if i get report package not found i guess i need to put this package to my list right ? (xetex works ;-)))
<srhb>
Filystyn: TeX package? yes.
<Filystyn>
yes it works this is so briliant
<srhb>
Filystyn: It is! :-)
<Filystyn>
I am so glad i gave this system chance.
<jophish>
Can anyone tell me how nix-store --repair-path works?
<srhb>
steveeJ: (--indirect will place a symlink in /nix/var/nix/gcroots/auto TO your locally defined symlink, so that when it becomes dangling, because you removed the local one, gc still occurs.)
<NixOS_GitHub>
[nixpkgs] dezgeg pushed 1 new commit to staging: https://git.io/vSPSW
<NixOS_GitHub>
nixpkgs/staging 52797fc Tuomas Tynkkynen: Merge remote-tracking branch 'upstream/master' into staging
<jophish>
I have some issue where several files in the nix store just contain zero bytes
<srhb>
jophish: eek
<jophish>
nix-store --repair-path downloads a new version from the cache and nothing changes on disk
<jophish>
This is in a VM. I don't think that it's an issue with the underlying storage as I've cloned the disk and have the same results
<jophish>
I'm wondering if it's a nix issue, an ext4 issue or a VirtualBox issue
bkchr has joined #nixos
<jophish>
I'm going to try repairing a path "manually" with rsync
bastian__ has quit [(Remote host closed the connection)]
<steveeJ>
srhb: that's neat, thanks for the explanation. I wasn't aware of the double-symlink
<jophish>
however there are many many files, and I don't know if I can check them all
<steveeJ>
srhb: this is also explained in the nix-store manpage! however, there is one flag "-r" that I can't find in the docs. do you by chance know what it does?
<srhb>
Actually it's totally in the nix-store man page and has been for 12 years. Nevermind me.
<goibhniu1>
steveeJ: you can use `nix-store -r <some-store-path>` to install the thing that belongs at that store path ... e.g. you can grab a store path from hydra and install it using that
goibhniu1 is now known as goibhniu
<Filystyn>
guys i added all packages and got working except one %\usepackage[sfdefault]{AlegreyaSans} had to comment it out for now so how to make it work is this something i need to get just installing fonts?
<Filystyn>
also would like times new roman fonts;)
aminechikhaoui has joined #nixos
fabian_a has joined #nixos
<Filystyn>
ok i see in configuration there are i opts for fonts
<jophish>
Hmm, reading around build.cc hasn't made it clear why delete and copy-closure doesn't work
<jophish>
I'm going to nuke the invalid paths and hope that nix-store --repair does the right thing
vaibhavsagar has joined #nixos
<Filystyn>
ok sooo.. fonts.fonts = [ pkgs.timesnewroman ] am i trying at least in goo direction with this?
wkennington has quit [(Quit: Leaving)]
<srhb>
Filystyn: I don't have pkgs.timesnewroman
<srhb>
Where did you get that from?
<tilpner>
Hi, I'm getting "Qt FATAL: Cannot mix incompatible Qt library (version 0x50602) with this library (version 0x50601)" when starting VirtualBox. Does anyone know how to specify the qt version?
seppellll has joined #nixos
<srhb>
Filystyn: corefonts has a times
<srhb>
Unfree license though.
<Filystyn>
got it thx
<Filystyn>
was not able to enable the algreya package but other stuff works
<Filystyn>
so it's fine
<srhb>
algreya?
bennofs has joined #nixos
<srhb>
Filystyn: works for me: texlive.combine { inherit (texlive) scheme-basic alegreya; }
jrolfs has joined #nixos
<jophish>
That *is* odd. I unlinked /nix/store/ya1kcvz00ycf6w1jq4s355bkig8b5g2f-optparse-applicative-0.13.2.0.tar.gz which was all zeros and then ran nix-store --repair-path ...
<jophish>
still zeros
<jophish>
that does *that* mean
<Filystyn>
okm it works to package broke in kille tho as siad thsi is not crucial
<Filystyn>
tho**
<Filystyn>
thx for huge help anyway srhb
<NixOS_GitHub>
[nixpkgs] dezgeg pushed 1 new commit to master: https://git.io/vSP5t
<NixOS_GitHub>
nixpkgs/master 7f8202b Tuomas Tynkkynen: Fix tarball evaluation...
jrolfs has quit [(Ping timeout: 240 seconds)]
<srhb>
Filystyn: You're welcome.
georges-duperon has joined #nixos
<jophish>
I did that a few more times and it's no longer corrupted
<jophish>
I thought the culprit there was disk failure
<jophish>
perhaps that was on the mailing list
<bennofs>
that sounds like a strange bug :)
<jophish>
it does fit the symptoms
<jophish>
ocharles: Did you ever find out what that was?
Seichi has joined #nixos
<NixOS_GitHub>
[hydra] edolstra pushed 1 new commit to master: https://git.io/vSPdg
<NixOS_GitHub>
hydra/master 585d327 Eelco Dolstra: Better error message if jobset doesn't evaluate to an attrset
davidar has quit [(Quit: Client limit exceeded: 10000)]
bigvalen has quit [(Ping timeout: 240 seconds)]
<gchristensen>
hmm I'm suspecting again that making a squashfs over 96 cores exposes parallelism bugs in mksquashfs
<Dezgeg>
is the squashfs image broken when mounted manually, without the netboot thing?
bigvalen has joined #nixos
vandenoever has joined #nixos
vandenoever has quit [(Changing host)]
vandenoever has joined #nixos
<gchristensen>
not sure how to check, I don't get errors until I try and read a bad section
<gchristensen>
maybe unsquuash it
MinceR has quit [(Ping timeout: 260 seconds)]
nh2 has joined #nixos
<Dezgeg>
mount -o loop -t squashfs ?
<gchristensen>
yeah but that doesn't raise errors
<Dezgeg>
find -type f | xargs cat > /dev/null ?
<gchristensen>
fair
<gchristensen>
this last time I didn't get errors until I was installing grub
MinceR has joined #nixos
vandenoever has quit [(Client Quit)]
katyucha has joined #nixos
<NixOS_GitHub>
[nixpkgs] techhazard opened pull request #24822: add disk type storage to libvirt (master...add-disk-storage-to-libvirt) https://git.io/vSPbX
Seichi has quit [(Read error: Connection reset by peer)]
Seichi has joined #nixos
<gchristensen>
nothing :/
<gchristensen>
good test, Dezgeg
m0rphism has joined #nixos
<Dezgeg>
what about sha*sum of the squashfs file on the netbooted system?
<NixOS_GitHub>
[nixpkgs] kirelagin opened pull request #24826: bonnie: Update Darwin patch (master...bonniepp) https://git.io/vSXf2
Itkovian has joined #nixos
<jophish>
Thanks bennofs
<NixOS_GitHub>
[nixpkgs] garbas pushed 2 new commits to master: https://git.io/vSXfH
<NixOS_GitHub>
nixpkgs/master 8c8942a Rok Garbas: firefox-bin: alsaLib.dev is needed
<NixOS_GitHub>
nixpkgs/master 8fcb927 Rok Garbas: pypi2nix: 1.6.0 -> 1.8.0
fabian_a has quit [(Read error: Connection reset by peer)]
contrapumpkin has joined #nixos
mkoenig has joined #nixos
<NixOS_GitHub>
[nixpkgs] edolstra pushed 9 new commits to staging: https://git.io/vSXJq
<NixOS_GitHub>
nixpkgs/staging c7e1a25 Tuomas Tynkkynen: gawk: 4.1.3 -> 4.1.4
<NixOS_GitHub>
nixpkgs/staging 5de624e Tuomas Tynkkynen: gmp: 6.1.1 -> 6.1.2
<NixOS_GitHub>
nixpkgs/staging 72b2cbe Tuomas Tynkkynen: libsigsegv: 2.10 -> 2.11
jrolfs has joined #nixos
ixxie has joined #nixos
jrolfs has quit [(Ping timeout: 255 seconds)]
ThatDocsLady is now known as ThatDocsLady_afk
<sphalerite>
hyper_ch: is your question still relevant?
<sphalerite>
hyper_ch: all I did was add kvm to systemPackages, and qemu-kvm works fine
<jophish>
Hmm, this nix installation might be completely hosed the database has incorrect hashes in
<jophish>
nix-hash /nix/store/... is the same on two machines, however the database differs
<jophish>
This wouldn't be the end of the world, except that this machine has been suffering lately with a mysterious "replace every byte in a file with zeros" error
rcschm has joined #nixos
bennofs has quit [(Ping timeout: 252 seconds)]
<gchristensen>
Dezgeg: I'm now strongly suspecting a RAM issue
<jophish>
do I have no way to know which paths are correct
<jophish>
so*
<Dezgeg>
you can try including the memtester package and running that, I guess
<gchristensen>
it was happening for 2 different servers, which made me think otherwise, but it turned out there were only 3 servers of this type remaining in the datacenter, making me believe these two had been rejected by other customerss
<gchristensen>
and the problem didn't occur at consistent places
zraexy has quit [(Ping timeout: 260 seconds)]
bkchr has quit [(Ping timeout: 240 seconds)]
bkchr has joined #nixos
<jophish>
domenkozar: if that is the issue is there anything you can suggest?
<domenkozar>
jophish: I'd first check for broken HDD
<domenkozar>
jophish: then verify Nix sqlite is not corrupt
<domenkozar>
then we can start blaming Nix
<jophish>
domenkozar: This is a VM. I've cloned the disk and the issue persists
<jophish>
I think that some hashes in the DB are incorrect
<schoppenhauer>
hi. I have some questions regarding lxc. They are a bit nixos-specific, but not totally. The main question is whether it is possible to run GUIs in an LXC. And the other one is whether there is an (easy) possibility to have a purely user-mode NAT, like qemu and virtualbox have.
marsel has quit [(Ping timeout: 255 seconds)]
<Mic92>
schoppenhauer: there was a very basic web-interface, probably lxd delivers better web-uis.
ixxie has quit [(Ping timeout: 240 seconds)]
<Mic92>
I am not aware of user-mode NAT for containers in general. Maybe it is possible to create a TAP interface.
Seichi has quit [(Quit: Konversation terminated!)]
Seichi has joined #nixos
eacameron has joined #nixos
<schoppenhauer>
Mic92: hm. ok. well, the problem is, I do not want to bridge it to some specific interface, I want it to be behind a virtual NAT, so it is always routed to the interface which I connected to. is this possible?
<Mic92>
schoppenhauer: if you just have one container, you just need to create a single veth pair. If you give the host end an IP too, you can do NAT or Routing as usual
takle has joined #nixos
<Mic92>
you can also create just an internal bridge and route or do natting from their.
<Mic92>
The latter one is how I use LXC.
<schoppenhauer>
Mic92: ok. hm. how do I configure this in nixos? (like, what are the configuration flags whose documentation I would have to read, etc.)
<schoppenhauer>
I am a bit overwhelmed by the many tutorials and stuff
Filystyn has quit [(Quit: Konversation terminated!)]
<schoppenhauer>
and I mainly want to use it to be able to suspend a coq process I am working with when I have to reboot
simukis_ has quit [(Quit: simukis_)]
simukis_ has joined #nixos
<Mic92>
schoppenhauer: I don't think, you that freezing containers surive a reboot.
<schoppenhauer>
Mic92: but lxc-checkpoint should? shouldn't it?
<Mic92>
schoppenhauer: sounds too me like you just want to disable network namespacing and use the host network for the container?
<Mic92>
Then you do not need natting at all.
carlosda1 has joined #nixos
bennofs1 has quit [(Quit: WeeChat 1.6)]
<Mic92>
But I don't know if this is required for CRIU. I have no experience with it.
<domenkozar>
copumpkin: let me know when I should stop playing devils advocate about separate channel :D
<copumpkin>
domenkozar: hah it's not bad at all, I think! niksnut has been talking about removing channels and I do think there needs to be a deep discussion somewhere about what "the essence of a channel is"
<copumpkin>
perhaps a seemingly unrelated PR that's already merged isn't the right place for that discussion, but we can transcribe it later :)
<domenkozar>
copumpkin: ideally there would be a RFC for this
<copumpkin>
yeah
<copumpkin>
although I'd want to make it broader than just the darwin-specific use case
<copumpkin>
we've had implicit rules in place for when channels make sense for a while
<copumpkin>
we have things like the small nixos channel
<domenkozar>
copumpkin: I think a CI testing PRs for all platforms is the way to go
<domenkozar>
then things won't break in master as often
<copumpkin>
definitely :)
<domenkozar>
also a better way to communicate what blocks the channel
newbie90 has quit [(Quit: Bye.)]
georges-duperon has quit [(Ping timeout: 252 seconds)]
jrolfs has joined #nixos
lsix has joined #nixos
earldouglas has joined #nixos
jrolfs has quit [(Ping timeout: 255 seconds)]
AMDmi3 has joined #nixos
AMDmi3 has left #nixos []
<Mic92>
domenkozar: there was a bot repo for this, right?
<schoppenhauer>
Mic92: so this means the container has access to the host interface directly?
<Mic92>
schoppenhauer: if CRIU supports this (I think it does, because it works on process level), this could be easier to handle for your use case.
<Mic92>
lxc.network.type = none in this case
<schoppenhauer>
ok thx
ryanartecona has joined #nixos
<niksnut>
copumpkin: I've talked about removing nix-channel, not removing channels
<niksnut>
slight difference :-)
<copumpkin>
niksnut: sorry, fair enough :)
<copumpkin>
but I do think it'll be good to really spell out what we mean by channel and when it's sensible to use them
<copumpkin>
I also have some thoughts around them for internal deployments
<bitonic>
How does one debug a `error: Multiple exceptions: database disk image is malformed, database disk image is malformed, database disk image is malformed` error when invoking `nixops`?
<bitonic>
Or at least work around it?
georges-duperon has joined #nixos
teknico has left #nixos []
<Mic92>
bitonic: who generate this error?
marsel has joined #nixos
<Mic92>
oh, I did not read nixops
darlan has joined #nixos
fre has quit [(Quit: WeeChat 1.7)]
<Mic92>
bitonic: the cause is probably that ~/.nixops/deployments.nixops is broken: sqlite3 ~/.nixops/deployments.nixops
ixxie has joined #nixos
<bitonic>
Mic92: that works, I think it's some derivation maybe?
<bitonic>
`nixops list` works
MichaelRaskin has joined #nixos
<Mic92>
do you get a stacktrace?
<bitonic>
Mic92: no
<bitonic>
I literally get just that output
boothead has quit [(Ping timeout: 255 seconds)]
boothead has joined #nixos
Filystyn has joined #nixos
Filystyn has quit [(Changing host)]
Filystyn has joined #nixos
<NixOS_GitHub>
[nixpkgs] shlevy pushed 2 new commits to master: https://git.io/vSX0n
<NixOS_GitHub>
[nixpkgs] fpletz opened pull request #24830: treewide: use boolToString function (master...refactor/boolToString) https://git.io/vSXom
eacameron has quit [(Quit: Leaving...)]
katyucha has joined #nixos
<copumpkin>
niksnut, shlevy: I think my tarball nix-channel issue is actually far more subtle than I initially thought
civodul has quit [(Quit: ERC (IRC client for Emacs 25.1.1))]
ryanartecona has joined #nixos
phorse has quit [(Read error: Connection reset by peer)]
jrolfs has joined #nixos
__Sander__ has quit [(Remote host closed the connection)]
acarrico has joined #nixos
byteflame has quit [(Quit: Lost terminal)]
<NixOS_GitHub>
[nixpkgs] fpletz closed pull request #24745: acme: Add "domain" option to separate domain from name (master...add-acme-domain) https://git.io/vSosP
smw has quit [(Ping timeout: 240 seconds)]
<copumpkin>
(I updated the ticket; also dtzWill might be interested)
Seichi_ has joined #nixos
Seichi has quit [(Read error: Connection reset by peer)]
<bennofs>
what's the status of the new nix UI? is there a page/note/tracking issue somewhere about the current design plans?
carlosda1 has joined #nixos
<copumpkin>
bennofs: it exists in 1.12
<ToxicFrog>
Oooo, new UI?
<copumpkin>
in "beta" or something
<copumpkin>
don't think all the commands are in yet, but many of them are
<bennofs>
yeah but is the design fully fleshed out yet? last time I looked it only had some simple commands
<copumpkin>
the most discussion I recall on that was in a gist :)
<copumpkin>
which was eventually transplanted into an issue
<copumpkin>
ToxicFrog: yeah, nix is moving into the 2010s with `nix <subcommand`
<copumpkin>
`nix build`, etc.
<ToxicFrog>
Huzzah
<ToxicFrog>
Although that's less interesting to me, personally, than not getting the raw output of hundreds of curl processes every time I nix
carlosda1 has quit [(Ping timeout: 260 seconds)]
<ToxicFrog>
So I won't be really excited unless this comes with a user-friendly progress indicator :)
eacameron has joined #nixos
stepcut has quit [(Remote host closed the connection)]
smw has joined #nixos
obadz has quit [(Ping timeout: 252 seconds)]
implite has joined #nixos
mekeor` has quit [(Remote host closed the connection)]
noah-nix has joined #nixos
obadz has joined #nixos
<noah-nix>
Hello, I followed the steps (and read the steps carefully) in the NixOS Manual. Still, I'm not able to boot the PC after the installation.
zeus_ has quit [(Read error: Connection reset by peer)]
<noah-nix>
Can that be due to an option required in the NixOS configuration file; boot.initrd.kernelModules
<noah-nix>
I'm sure I've followed the steps correctly.
zeus_ has joined #nixos
<implite>
what happens when you turn on machine after install
<implite>
?
<implite>
no operating system found or?
<noah-nix>
It simply boots to the DVD, even though I select the Hard Drive. Let me eject the DVD and see what happens when I boot. I presume it will say "no OS found" or similar, like it did yesterday. One moment, please!
<noah-nix>
Yep, no OS found.
<noah-nix>
Rather, "BootDevice Not Found. Please install an operating system on your hard disk."
<implite>
ok are you sure the partition you made was found while you where doing install?
<implite>
in the bios does it see your harddrive?
<noah-nix>
What do you mean? I did partition the drive before installing using fdisk.
<noah-nix>
It did find my drive.
<implite>
ok
<Unode>
noah-nix: UEFI mode or legacy?
<noah-nix>
Unode: legacy
<implite>
I wonder if it didnt write to the boot sector of your drive
<Unode>
noah-nix: might just be that the last step of registering grub to the MBR didn't happen.
<noah-nix>
Unode: it did install GRUB successfully
<noah-nix>
It said so, at least.
<Unode>
let me grep my logs, clever saved me with one of these cases once
roberth has joined #nixos
<Unode>
noah-nix: try "nixos-rebuild boot"
<noah-nix>
I saw a quick warning passing by, I didn't have enough time to read it, but it said "install-info, warning: no ..." or something like that.
<noah-nix>
Unode: okay
<noah-nix>
Unode: shall I boot to the DVD and mount the hard drive?
<Unode>
or chroot to the partition in some other way
sellout- has quit [(Quit: Leaving.)]
<Unode>
you'll need to activate the environment so you can run nixos-rebuild.
<noah-nix>
I can boot to the DVD, I have no other way of doing it :-)
<Unode>
/nix/var/nix/profiles/system/activate
codygman_ has joined #nixos
<noah-nix>
What's that you just sent?
<Unode>
^ will get you running if you launch from a chroot
hamishmack has quit [(Quit: hamishmack)]
<noah-nix>
Oh, okay, thank you
<Unode>
if you get errors, it's probably because you are missing some mountpoint
nschoe has quit [(Quit: Program. Terminated.)]
jgertm has joined #nixos
<noah-nix>
I'll do "mount /dev/disk/by-label/nixos /mnt"
<Unode>
"missing mountpoint" == "mount --bind .."
<noah-nix>
Since I have the label nixos on sda2
<Unode>
you'll need to bind /dev /proc /sys and some subdirs in /dev that I don't remember right now. Also /boot needs to be accessible otherwise grub won't find its files.
<Unode>
and I have to leave now. Best of luck
<codygman_>
I'm installing nixos into VMware. To access the internet from VMware I need to put a certificate in the guest nixos. Nixos can't mount VMware shares by default, you have to install it from the Internet. Any ideas on alternative solutions?
<noah-nix>
Unode: thanks, bye.
<Unode>
noah-nix: welcome
mudri has quit [(Ping timeout: 240 seconds)]
<noah-nix>
codygman_: I
<noah-nix>
Ooops.
<Unode>
codygman_: you can setup a port-forward and scp/ssh into the machine.
sigmundv has joined #nixos
<noah-nix>
Is a certificate really required in VMware to access internet? I'm noob though, so I don't know.
<codygman_>
Unode: thanks, I'll give that a try
<noah-nix>
I just know I've used Windows before in a VM without certificate. Altho that's Windows, not Linux. :-)
<Unode>
I'm guessing SSL proxy or something. Otherwise no clue. Usually the answer to that is no
<noah-nix>
Unode: I received a warning in the output of nixos-rebuild boot.
<codygman_>
Yep, ssl proxy
<noah-nix>
oh ok
<noah-nix>
"Warning: do not know how to make this configuration bootable; please enable a boot loader."
<noah-nix>
The only partitions I have are swap and another partition for installation.
<noah-nix>
Wait I might know why it didn't work. Do you have to specify the partition in the configuration.nix file? Because I just uncommented, and the default was "/dev/sda" which is a device, not a partition..
<noah-nix>
That seems not to be the case, so nevermind.
goibhniu has quit [(Ping timeout: 240 seconds)]
noah-nix_ has joined #nixos
<noah-nix_>
Lost connection for a minute
<noah-nix_>
I don't know, shall I ask in GitHub?
<noah-nix_>
About my issue.
<srhb>
noah-nix_: What's the issue again?
noah-nix has quit [(Ping timeout: 260 seconds)]
<noah-nix_>
srhb: oh, it's that I can't boot to the OS. There is no boot loader, the PC says. The output of 'nixos-rebuild boot' is: warning: do not know how to make this configuration bootable; please enable a boot loader
<noah-nix_>
I've followed the steps in the manual properly, I'm quite sure.
byteflam1 has quit [(Ping timeout: 245 seconds)]
<srhb>
noah-nix_: I skimmed, and it was legacy boot?
<noah-nix_>
srhb: yep
<srhb>
noah-nix_: Did you try running grub-install manually?
<noah-nix_>
srhb: I haven't, but I will! What is the command?
<noah-nix_>
grub-install /dev/sda ?
<srhb>
noah-nix_: Uh, yeah. It's been a while, I'm on uefi boot now
<noah-nix_>
okay
<implite>
Google is also a nice friend ;)
<noah-nix_>
Error: "error: failed to get canonical path of 'tmpfs'."
<srhb>
noah-nix_: Regarding an earlier quesiton, boot.loader.grub.device MUST be set, afaik.
<srhb>
noah-nix_: Fun!
<noah-nix_>
srhb: yeah, I've set it. But not to a partition, but to my hard drive device.
<noah-nix_>
i.e. /dev/sda
<srhb>
Wait, is your system currently in /mnt or something?
<noah-nix_>
srhb: it is
<srhb>
noah-nix_: That's correct re boot.loader.grub.device
<noah-nix_>
srhb: it did write the hardware file, altho did not overwrite the configuration.nix
<srhb>
noah-nix_: Ah good. Then rerun nixos-install
<noah-nix_>
srhb: still, that file is empty
<srhb>
Wait, hardware-configuration.nix is empty?
<noah-nix_>
srhb: yep
<srhb>
That sounds like a bug...
<noah-nix_>
Maybe I'm stupid and have spelled wrong, will double check
<NixOS_GitHub>
[nixpkgs] bjornfor pushed 1 new commit to master: https://git.io/vSXD7
<NixOS_GitHub>
nixpkgs/master 6047401 Bjørn Forsman: pythonPackages.boto3: disable tests (fails due to network access)...
jyp[m] has quit [(Quit: Client limit exceeded: 10000)]
<srhb>
noah-nix_: Well if you say it's empty and not nonexistant
byteflame has joined #nixos
<srhb>
It seems unlikely that you mistyped anything
<noah-nix_>
oh I think with nano it opens anyways
<srhb>
ah
<davenull>
Good afternoon everyone. I am trying to set up a systemd unit file, to start PM2, which was installed using node2nix, and it works perfectly. I would like to be able to make it less dependent on the hash for the package though, and have been trying to use the pgks.node-pm2-2.4.4 reference as the PATH, but it complains about the periods in the package name.
<srhb>
ls -l /mnt/etc/nixos ?
<ma9e>
it's a module, right? how do i use modules?
<chattered>
I'm a complete noob to nixops and virtualbox for the first time, and getting "VBoxManage: error: Details: code NS_ERROR_FAILURE (0x80004005), component MachineWrap, interface IMachine". Help?
<srhb>
ma9e: I see, it doesn't appear available as a service.
<davenull>
srhb, I built the package with node2nix, as pm2 needs to be a globally installed package.
byteflame has joined #nixos
DutchWolfie has quit [(Quit: Konversation terminated!)]
<ma9e>
srhb, i noticed there are no options associated.. should they be added? is that the right way to do it?
<srhb>
davenull: But what did you name the attribute in top-level.nix?
<srhb>
davenull: Or what did you import it as in configuration.nix ?
<srhb>
ma9e: I think so, but I'm not sure. Sorry.
<davenull>
Well, I'm not quite sure. This is day 2 with nixos, I'm still finding my way around.
<srhb>
ma9e: I guess you were right in your approach
<SovereignBleak>
Has anyone been able to get thumbnails working without a desktop manager?
<srhb>
davenull: Okay, so, usually you would have... yourPackage = import /path/to/your/nix/expression;
<noah-nix_>
srhb: the UUID is correct in hardware-configuration.nix. I tried to make the part. bootable but it did not help. Shall I make it un-bootable again?
<srhb>
davenull: Then, the value of path that we talked about earlier would be [ yourPackage ]
<noah-nix_>
via fdisk
<SovereignBleak>
I have a barebones i3 config but can't seem to get thumbnails working, despite having xfce.tumbler and ffmpegthumbnailer in my packages list.
<srhb>
davenull: attribute names can't have dots in them, and you don't seem to have given it a name at all, hence the confusion.
<NixOS_GitHub>
[nixpkgs] fpletz pushed 2 new commits to master: https://git.io/vSXQZ
<NixOS_GitHub>
nixpkgs/master 29ed67f Franz Pletz: libressl_2_5: 2.5.1 -> 2.5.3
<NixOS_GitHub>
nixpkgs/master e8351fd Franz Pletz: libressl: 2.5 branch is now stable
luigy has left #nixos []
Seichi_ has quit [(Quit: Konversation terminated!)]
<srhb>
noah-nix_: I doubt that changes anything..
rotaerk has quit [(Ping timeout: 240 seconds)]
<davenull>
srhb, Well, how would I give it a name then? haha
<noah-nix_>
srhb: okay. Shall I change the part. to type Linux?
<srhb>
noah-nix_: uh, of your root partition?
<NixOS_GitHub>
[nixpkgs] bjornfor pushed 1 new commit to master: https://git.io/vSXQ2
<srhb>
davenull: Your configuration.nix looks like this: {...}: {...};
<srhb>
davenull: Make it look like this: {...}: let yourPackage = import ...; in {...}
<srhb>
davenull: That is, your definition goes in a let ... in between the argument list and the body
<noah-nix_>
srhb: speaking to me?
RchrdB has quit [(Ping timeout: 260 seconds)]
<noah-nix_>
oh sorry
<SovereignBleak>
Does everyone here use a desktop environment or am I the only i3 user that can't seem to get thumbnails to work?
<srhb>
noah-nix_: No, I wasn't :)
<noah-nix_>
srhb: I see :-)
<gchristensen>
SovereignBleak: where do you want to have thumbnails work? I don't know what that means, in the context of i3
<srhb>
noah-nix_: I don't see anything wrong with your config, except it seems your swap partition wasn't enabled when you created the hardware config, it's not included.
<SovereignBleak>
gchristensen: pcmanfm or thunar
<simpson>
SovereignBleak: I don't use a DE, but I also don't use i3 nor tools with thumbnails.
<srhb>
SovereignBleak: I use xmonad.
<noah-nix_>
srhb: well it was enabled
<srhb>
SovereignBleak: What's a thumbnail?
<noah-nix_>
srhb: at least with mkswap and swapon
<srhb>
noah-nix_: Your system is weird. Maybe start over with the short list :-P
<SovereignBleak>
srhb: A small picture representation of what's in a file.
<noah-nix_>
srhb: okay, sure, I will :-)
<srhb>
SovereignBleak: Where do you expect to see these where you don;t?
<davenull>
srhb, Ok, I have that in there, and it's happier than it was. When I use path = [ pkgs.pm2 ] I still get the missing attribute error...
<noah-nix_>
srhb: maybe defect DVD medium?
<srhb>
davenull: It's not pkgs. anymore
<SovereignBleak>
srhb: pcmanfm and thunar, two popular file managers.
<srhb>
noah-nix_: Seems unlikely.
<noah-nix_>
srhb: that's good
<srhb>
SovereignBleak: Oh..
<NixOS_GitHub>
[nixpkgs] fpletz pushed 10 new commits to release-16.09: https://git.io/vSXdc
<NixOS_GitHub>
nixpkgs/release-16.09 a21b142 Tim Steinbach: nginxMainline: 1.11.3 -> 1.11.4...
<NixOS_GitHub>
nixpkgs/release-16.09 64cbe89 Franz Pletz: nginx: 1.10.1 -> 1.10.2...
<NixOS_GitHub>
nixpkgs/release-16.09 c9185d6 Franz Pletz: nginxMainline: 1.11.4 -> 1.11.5...
<davenull>
error: cannot coerce a function to a string, at /nix/var/nix/profiles/per-user/root/channels/nixos/nixpkgs/lib/strings.nix:85:38
<srhb>
davenull: What's in the file is a function, we need to call it. Hence the {}
<srhb>
davenull: Sorry, I'm doing this blind. :-P
<davenull>
called without required argument ‘nodeEnv’
<srhb>
SovereignBleak: I see what you mean, they're just white!
<srhb>
davenull: Aaand that's where I can no longer follow...
<SovereignBleak>
I'd be happy to use anything that gives me some thumbnails as long I don't have to give up tiling.
<noah-nix_>
Really nice help here in the IRC, I have to say, being honest. Much better compared to some other distros.
<srhb>
SovereignBleak: Does it work when you use a DE instead of i3?
<chattered>
Ah, nevermind. I had virtualbox as a buildInputs to my shell.nix, and it wasn't happy with that.
<srhb>
davenull: You probably need help from someone who's used to defining node packages here, I'm afraid.
<srhb>
davenull: Actually, perhaps we just need callPackage instead of import...
ambro718 has joined #nixos
ma9e has quit [(Remote host closed the connection)]
sellout- has joined #nixos
<SovereignBleak>
srhb: Like I said, I will use anything but a non-tiling environment to get thumbnails. Reading through Github it looks like Gnome renders thumbnails fine.
<srhb>
davenull: Try pkgs.callPackage instead of import ?
<srhb>
SovereignBleak: Hm...
<srhb>
SovereignBleak: When I launch thunar from a terminal it seems to imply I'm missing some kind of theme
vaibhavsagar has quit [(Ping timeout: 255 seconds)]
<SovereignBleak>
srhb: I'm trying a rebuild with adwaita and hicolor themes installed which pcmanfm complains about but I can't see why they would be at fault for generating image thumbnails. That should be the job of xfce.tumblr and ffmpegthumbnails.
<gchristensen>
anyone in here worked with grpc much?
<srhb>
SovereignBleak: Ah, okay. Sorry, I know too little about this, I'll butt out.
<SovereignBleak>
Seems like everyone here is a coder and has no need for such things as thumbnails haha.
<sphalerite>
SovereignBleak: apparently xfce's thumbnailer is called tumbler
<sphalerite>
maybe if you install that it will help
<gchristensen>
SovereignBleak: I don't really know of many people using i3 + want thumbnails :/
<SovereignBleak>
sphalerite: It’s been installed. No joy. :-(
<sphalerite>
gchristensen: I use i3 and wouldn't mind thumbnails for the occasionas when I use a graphical file manager
<gchristensen>
whoa shlevy
<gchristensen>
sphalerite: yeah same but I've never actually tried to do it
<shlevy>
gchristensen: Cool right?
<gchristensen>
yup
<sphalerite>
neat
<shlevy>
gchristensen: One of those nice convergences where a stupidly simple concept makes a lot of stuff possible
<sphalerite>
SovereignBleak: I've got some time to kill at the airport so I'll see if I can get thm working for myself
<sphalerite>
and let you know
<gchristensen>
shlevy: no kidding, that is awesome!
<sphalerite>
assuming the internet connection cooperates...
ryanartecona has quit [(Quit: ryanartecona)]
blahdodo has quit [(Ping timeout: 255 seconds)]
snikkers has joined #nixos
blahdodo has joined #nixos
<copumpkin>
omg it's shlevy
katyucha has quit [(Ping timeout: 260 seconds)]
<shlevy>
:o where?
HappyEnte has quit [(Ping timeout: 240 seconds)]
<copumpkin>
shlevy: this sounds very much like something I've wanted for a long time, but what I've wanted isn't quite idempotent...
<copumpkin>
shlevy: you know the EC2 AMI upload scripts...
<copumpkin>
and how they're basically simple functions of a nix build :)
jgertm has quit [(Ping timeout: 255 seconds)]
<shlevy>
Yeah
<shlevy>
copumpkin: We're working on an AMI that bootstrapts itself with hail :)
<copumpkin>
could also be used for making official AMIs of unstable nixos releases
<shlevy>
Aah never mind I was thinking about your instance metadata stuff
<copumpkin>
ooh
jgertm has joined #nixos
<shlevy>
You just put a hydra job into your instance metadata
<copumpkin>
in what way?
<shlevy>
And your instance pulls its config down
<copumpkin>
oh okay
<shlevy>
and updates itself
codygman_ has quit [(Ping timeout: 260 seconds)]
<copumpkin>
I welcome improvements to that
<copumpkin>
I have a few ideas but this works well enough fro now
<sphalerite>
Is there a way to avoid having nix-shell pull in docs?
<SovereignBleak>
sphalerite: Thank you. I'd really appreciate it. I'm still getting my bearings with NixOS and some of the linking stuff people do to get functional themes, etc. I still don't quite grasp. I still don't have a proper cursor theme for example haha.
<sphalerite>
I'm on an unreliable connection and every megabyte I can avoid downloading counts
<srhb>
SovereignBleak: Actually I lost that with 17.03 as well, and it's bugging me a bit.
<copumpkin>
shlevy: interesting, I'm curious to see that
<srhb>
SovereignBleak: Not enough to fix it, though... :-P
<copumpkin>
shlevy: while you're here, you have a moment to help me think through a certain PR of mine? :) :) :)
<copumpkin>
(fine if no, I'll crawl back into my hole)
<SovereignBleak>
srhb: Glad I'm not the only one! Makes me feel a little less idiotic.
<srhb>
SovereignBleak: Do you have the little hand when you over oer something clickable too?
<srhb>
SovereignBleak: The hand that *points the wrong bloody way* ?
<shlevy>
copumpkin: Possibly, I'm on a call that currently needs 20% attention but may jump to 100 :)
<copumpkin>
shlevy: I'm trying to get my VM test to pass with sandbox enabled and I'm looking for any way to even get a pessimistic estimate of that build-time closure into the VM's store. Like I'm fine if there's far too much stuff in the store, but right now it's trying to build a ton of junk inside the VM and that fails
<copumpkin>
it almost feels accidental though, that the build-time closure ends up on the target
<copumpkin>
so I'm also tempted to just remove that part of the test, but someone will probably chew me out
<SovereignBleak>
This is what I have, which should do the same, but doesn't work.
<shlevy>
copumpkin: What's the test doing with the build-time closure?
carlosda1 has quit [(Ping timeout: 252 seconds)]
<copumpkin>
shlevy: it tries to make some nixos-rebuild config change inside the installer test, which is in a sandboxed VM, so it can't download anything
<copumpkin>
I guess I could manually add all the dependencies it probably needs
<copumpkin>
but that feels brittle
HappyEnte has joined #nixos
<shlevy>
copumpkin: It's a test, its brittleness detects itself!
<copumpkin>
:P
<copumpkin>
I guess
<NixOS_GitHub>
[nixpkgs] LnL7 pushed 3 new commits to master: https://git.io/vSXpB
<NixOS_GitHub>
nixpkgs/master 08656a7 Robert Helgesson: gpsbabel: fix darwin build
<NixOS_GitHub>
nixpkgs/master 9561fec Daiderd Jordan: gpsbabel: fix build with llvm-4
<noah-nix_>
srhb: indeed as in I can't maybe; or indeed as in I certainly can? ^
<srhb>
noah-nix_: Indeed as in it looks like you cannot right now.
<srhb>
SovereignBleak: Right, that didn't work for me either.
<noah-nix_>
Okay, I'll try with UEFI, but tomorrow. Thanks for the tip, and the help you've given me!
<noah-nix_>
srhb
<srhb>
noah-nix_: You're very welcome, best of luck. Perhaps someone has a fresh idea for you tomorrow.
<noah-nix_>
srhb: thanks
<noah-nix_>
srhb: what is journaling in part.-speaking?
<davenull>
srhb, Is nodenv even on the system? I can't seem to find a package for ti.
<davenull>
it*
<srhb>
noah-nix_: It's a technique to keep track of changes "waiting to be committed to files"
<srhb>
noah-nix_: (That's very, very vague)
<noah-nix_>
srhb: is that required?
<srhb>
noah-nix_: No, but there's no good reason to not use a good journalling fs.
<noah-nix_>
Ok
jensens has quit [(Ping timeout: 260 seconds)]
<srhb>
Or rather, _you_ don't have a good reason ;-)
jgertm has quit [(Ping timeout: 255 seconds)]
<noah-nix_>
srhb: one last question: does this have to do with the fact that when I type 'reboot' after installing, I receive errors in the shutdown log
<noah-nix_>
?
HappyEnte has quit [(Read error: Connection reset by peer)]
<sphalerite>
SovereignBleak: I've not had any luck
<srhb>
noah-nix_: I don't know what those errors are, so hard to say. It's doubtful.
<noah-nix_>
srhb: I think it's like: 'not able to unmount'
<srhb>
noah-nix_: That's probably okay.
<davenull>
srhb, well, the other option I have with this is that I can just manually change the entries for the unit file for each host.
rcschm has quit [()]
<srhb>
davenull: This is fixable, I'm sure. I make these units for my own derivations just fine, it's just that I don't know the node infrastructure.
<srhb>
davenull: You could hardcode it here and now of course.
<noah-nix_>
shrb, okay, thanks again for the help. I'll (probably) come back tomorrow. Bye :-)
<srhb>
noah-nix_: Take care.
<noah-nix_>
srhb: you too
noah-nix_ has quit [(Quit: Page closed)]
<srhb>
davenull: you could check how some of the other node packages refer to nodeEnv
cpennington has quit [(Ping timeout: 240 seconds)]
<Sonarpulse>
pierron: thanks :)
<Sonarpulse>
for reviewing
reinzelmann has joined #nixos
JagaJaga has joined #nixos
AllanEspinosa has joined #nixos
<jophish>
shlevy: Hail looks really neat, I've already thought of a use for it here!
cpennington has joined #nixos
<jophish>
When the time comes, you must let us know how to make Hercules "hail compatible" :)
<shlevy>
jophish: Just need an API endpoint for a) latest build of a given job and b) the outpath of a given build
<shlevy>
jophish: Although "module Hail.Hercules where" is much less fun than the current codebase :P
ThatDocsLady has quit [(Quit: Arma-geddin-outta-here!)]
jgertm has joined #nixos
ryanartecona has joined #nixos
<davenull>
srhb, well, I have made a little bit of progress. I was calling from node-packages.nix, which was not the correct one, so now, using default.nix, I get a new error entirely. cannot coerce a set to a string, at /nix/var/nix/profiles/per-user/root/channels/nixos/nixpkgs/lib/strings.nix:85:38
<Ralith>
davenull: note that npm2nix is dead; use node2nix
alx741 has joined #nixos
<davenull>
Ralith, That's what I used.
<Ralith>
ah, I was confused because the link you just pasted was talking about npm2nix
reinzelmann has quit [(Quit: Leaving)]
<davenull>
It was, but the situation is not really caused by the difference, it's a lack of comprehension on my part as to what it's complaining about.
alx741 has quit [(Client Quit)]
alx741 has joined #nixos
<davenull>
Now that I'm using the correct nix file, I just need to figure out how to fix the set to string issue.
dridus has quit [(Quit: znc? wassat?)]
boothead has quit [(Ping timeout: 240 seconds)]
ixxie has joined #nixos
boothead has joined #nixos
<dmj`>
Ordered a nixos sticker in the mail from sticker mule, it never came :(
<dmj`>
;_;
<davenull>
Email em, they always got me my Arch stickers when they forgot, just had to remind them.
carlosda1 has joined #nixos
derjohn_mob has quit [(Ping timeout: 245 seconds)]
carlosda1 has quit [(Ping timeout: 252 seconds)]
offlinehacker[m] has quit [(Quit: Client limit exceeded: 10000)]
JagaJaga has quit [(Ping timeout: 240 seconds)]
dridus has joined #nixos
dridus has quit [(Client Quit)]
sellout- has quit [(Quit: Leaving.)]
<davenull>
One of the Node people happen to be around?
cfricke has joined #nixos
dridus has joined #nixos
<gchristensen>
dmj`: crappy :(
<gchristensen>
I'd offer to do the mailing myself for people in the community, but typical latency on me mailing stuff is 6-9 months
<mbirkis>
options = "x-systemd.automount,noauto"; <- this is on the bottom of the old wiki https://nixos.org/wiki/NFS, i belive it is incorrect as it threw errors for me when runnning nixos-rebuild switch
<mbirkis>
i fixed it by using ["x-systemd.automount" "noauto"] instead
<dmj`>
gchristensen: that might be more responsive than this current order
<gchristensen>
lol
<gchristensen>
let me reach out to them, you ordered from my sticker right?
ndowens08 has joined #nixos
ma9e has joined #nixos
JagaJaga has joined #nixos
acarrico has quit [(Ping timeout: 252 seconds)]
ma9e has left #nixos ["Leaving"]
darlan has quit [(Remote host closed the connection)]
<dmj`>
sticker mule, yes
jgertm has quit [(Ping timeout: 255 seconds)]
snikkers has quit [(Ping timeout: 240 seconds)]
derjohn_mob has joined #nixos
<gchristensen>
yeah but you ordered one of the stickers I put up?
civodul has joined #nixos
takle has quit [(Remote host closed the connection)]
<gchristensen>
but keep in mind the only way I can help is tell sticker mule I'm upset at their poor support and tell them I won't sell my stickers there anymore
<gchristensen>
or ask people to go elsewhere
byteflame has joined #nixos
carlosda1 has joined #nixos
takle has joined #nixos
ndowens08 has quit [(Quit: WeeChat 1.7)]
carlosda1 has quit [(Ping timeout: 260 seconds)]
cfricke has quit [(Quit: WeeChat 1.7)]
snikkers has joined #nixos
<iclanzan>
davenull, are you still around? I am a node guy, but only a beginner NixOS user. However I managed to install an npm package just now, so maby I can offer some assistance.
<davenull>
I am still around.
roberth has joined #nixos
<davenull>
So, the issue that I am running into at this moment, I am trying to set up a unit file for PM2, and in order to use it without having to update the hash for the package, I need to be able to use the package name.
ma9e has joined #nixos
<iclanzan>
Ok, cool. So can you confirm that you used node2nix as opposed to npm2nix to generate the nix files for your packages? You should have 3 files that have been generated for you: default.nix, node-env.nix and node-packages.nix.
<davenull>
Yup, I have PM2 installed via nix, and it works as expected.
<ma9e>
How would I compile i3 from the "next" branch using nix-shell?
<davenull>
iclanzan, I do have those, yes.
ryanartecona has quit [(Quit: ryanartecona)]
<ma9e>
Wiki article on development environments is really unclear
<monsieurp>
nuclear?
<ma9e>
I tried pointing $src at tar'd git, but that complains during pkg-config
<iclanzan>
davenull, were you able to import the default.nix file into your configuration.nix and have it build then?
<gchristensen>
nixos people, I have very exciting news to share, but before I can, I need to solve a problem. I need to run something on the very next reboot, and never again afterward. any suggestions on how to implement this?
<iclanzan>
gchristensen: reboot, open a terminal and run your command
<gchristensen>
it needs to be automatic without me doing it
<joachifm>
gchristensen: does the script create any files that you could condition the unit on?
<gchristensen>
it doesn't, but I suppose it _could_. it is making a request to a web server. it would be ideal to not repeat the request
<iclanzan>
davenull: Assuming you have your 3 files under a folder called FOO, you would do `let myNodePackages = import ./path/to/FOO {pkgs = pkgs;}`. Then in your code, for example in environment.systemPackages, you can reference packages like this: `environment.systemPackages = [ myNodePackages.pm2 ]`.
<gchristensen>
right now I'm thinking I'll nixos-rebuild with a config.nix file that creates the job, then remove the config.nix for the next rebuild, plus some sort of condition in the job run itself
<joachifm>
gchristensen: I see. well, the hacky way would be to make it create some persistent file and do unitConfig.ConditionPathExists or something, so that it'll run exactly once and then never again
<joachifm>
well, at least so long as the condition holds ...
<gchristensen>
yeah, where would you put such a file?
<joachifm>
/var/lib?
<gchristensen>
alright, cool
<gchristensen>
`/var/lib/.phoned-home` maybe
<joachifm>
per fhs var/lib is *supposed* to persist fwiw
<joachifm>
not sure what you're working with :)
<gchristensen>
the disk should be quite reliable :)
ixxie has quit [(Ping timeout: 240 seconds)]
<joachifm>
anyway, I seem to recall reading about systemd units to perform one-time tasks, such as doing updates at next boot following software upgrades, but I can't find it. maybe I'm imagining things ...
<joachifm>
but as long as you can be reasonably sure the file won't go away, a negated ConditionPathExists seems like it should work just as well
<gchristensen>
cool, thanks for thinking through that with me joachifm :) I'll make sure to put info in the file saying what it is / why it is there / why it isn't spyware
<joachifm>
:)
<davenull>
iclanzan, Holy shit. Thank you.
<Havvy>
We really should remove rustNightly from nixpkgs...
<Havvy>
...in favor of Mozilla's overlay
dridus has joined #nixos
<Yaniel>
イェッス
<Yaniel>
oops
<iclanzan>
davenull, you are welcome
wkennington has joined #nixos
luigy has joined #nixos
Havvy has quit [(Quit: Computer Restarted or Restarting IRC)]
georges-duperon has quit [(Ping timeout: 255 seconds)]
simukis_ has quit [(Ping timeout: 260 seconds)]
Havvy has joined #nixos
ryanartecona has joined #nixos
codygman_ has joined #nixos
marsel has quit [(Ping timeout: 260 seconds)]
codygman_ has quit [(Ping timeout: 245 seconds)]
acarrico has joined #nixos
fnlkj has joined #nixos
<davenull>
iclanzan, Well, now for some reason, it can build the configs, but the start and reload commands are broken.
<iclanzan>
davenull: I’m not sure what you mean by start and reload commands
codygman_ has joined #nixos
hiratara has quit [(Ping timeout: 240 seconds)]
gbbrt has quit [(Ping timeout: 240 seconds)]
hiratara has joined #nixos
codygman_ has quit [(Remote host closed the connection)]
codygman_ has joined #nixos
darthdeus has joined #nixos
<Havvy>
:X kdeinit5 is segfaulting in my virtual machine.
<NixOS_GitHub>
[nixpkgs] Profpatsch closed pull request #24482: /lib/default.nix: Add file docstring & unit test hint (master...lib-main-docstring) https://git.io/vSnD2
<davenull>
iclanzan, pm2 shows up at pm2: /nix/store/31z5zsjaahq7jsn2d87573xd056azfkk-user-environment/bin/pm2 according to whereis, but the systemd unit file is calling /nix/store/96x9qmrn5k7pyrpaczfy0fs4jihpvix5-node-pm2-2.4.4
<davenull>
Which leads to permission denied
<davenull>
I'm retarded.
<NixOS_GitHub>
[nixpkgs] Profpatsch opened pull request #24834: psc-package: init at 0.1.1 (master...psc-package) https://git.io/vS1rl
<davenull>
Oh man. This has been a hell of an 8 hours.
<Profpatsch>
davenull: Try type -p
<davenull>
Profpatsch, What is that?
<Profpatsch>
It’s a builtin in nearly every shell.
<davenull>
I'm on day 2 of nixos.
<Profpatsch>
davenull: Just a normal unix tool.
<davenull>
Ah, gotcha.
Mateon1 has quit [(Ping timeout: 260 seconds)]
<Profpatsch>
Together with ls and readlink you can do nearly everything file and symlink related
Mateon1 has joined #nixos
<davenull>
Welp, fixed the issue, the issue was me.
<Profpatsch>
At least find out what goes where.
ma9e has quit [(Remote host closed the connection)]
<iclanzan>
davenull: Do you have something like this for your custom service `system.user.services."whatever" = {serviceConfig = {execStart = "${myNodePackages.pm2}/bin/pm2";};};`?
ebzzry has joined #nixos
<davenull>
iclanzan, I figured it out, I was doing ExecStart = myPackages.pm2 + " resurrect";
<davenull>
I didn't add the resot of the path (/bin/pm2) and it was trying to execute a directory.
<iclanzan>
That is what I was thinking :)
lucus16 has joined #nixos
<davenull>
I wish I could use nix to clean the house. I'd much rather define a clean house and merge to that...
ryanartecona has quit [(Quit: ryanartecona)]
<lucus16>
warning: Nix search path entry ‘/nix/var/nix/profiles/per-user/root/channels/nixos/nixpkgs’ does not exist, ignoring
<lucus16>
shows up 5 times in a row when I run a nixos-rebuild switch --upgrade, what does that mean and what do I do about it?
aminechikhaoui has quit [(Ping timeout: 240 seconds)]
frankpf has joined #nixos
<NixOS_GitHub>
[nixpkgs] dezgeg pushed 2 new commits to staging: https://git.io/vS1or
<NixOS_GitHub>
nixpkgs/staging 1e1b68b Tuomas Tynkkynen: patchutils: Drop patch applied upstream
<NixOS_GitHub>
nixpkgs/staging 0d86c06 Tuomas Tynkkynen: gawk: Disable tests that don't work in sandbox
Mateon3 has joined #nixos
Mateon1 has quit [(Ping timeout: 255 seconds)]
Mateon3 is now known as Mateon1
Filystyn has quit [(Remote host closed the connection)]
<lucus16>
Figured it out. Had nixpkgs-unstable instead of nixos-unstable as channel
hiratara has joined #nixos
frankpf has quit [(Ping timeout: 252 seconds)]
davenull has quit [(Remote host closed the connection)]
davenull has joined #nixos
sheenobu has joined #nixos
<davenull>
So, funny question, what would I need to do to generate a new hardware-configuration.nix?
<davenull>
I deleted it.
sheenobu has quit [(Client Quit)]
sheenobu has joined #nixos
sheenobu has quit [(Changing host)]
sheenobu has joined #nixos
carlosda1 has joined #nixos
civodul has quit [(Quit: ERC (IRC client for Emacs 25.1.1))]
<davenull>
I fix :)
<Ralith>
rerun nixos-generate-config
lucus16 has left #nixos ["-"]
<davenull>
Wow kids, you have one hell of a Linux system here.
carlosda1 has quit [(Ping timeout: 240 seconds)]
<davenull>
The learning curve has a verticle portion, but not bad at all.
bachp has quit [(Quit: Client limit exceeded: 10000)]
bennofs has quit [(Ping timeout: 240 seconds)]
<dash>
davenull: yes but once you get up here you can see for _miles_
<davenull>
dash, It sure seems that way.
<dash>
davenull: i played with nix for a while but only became really comfortable writing nix exprs last year, now i'm ready to put nixos on everything ;)
Sonarpulse has quit [(Ping timeout: 255 seconds)]
<davenull>
I'm currently using arch for a product I developed, but it's hard to keep it all clean, this seems like a beautiful alternative.
dridus has quit [(Quit: znc? wassat?)]
stew has quit [(Ping timeout: 260 seconds)]
sellout- has joined #nixos
<earldouglas>
Is it feasible to deploy to physical NixOS boxes with NixOps?
mkoenig has quit [(Remote host closed the connection)]
<davenull>
Well, I now have a ready-built NodeJS server with PM2 installed, that takes one command to install, thanks to ansible.
<davenull>
That's pretty darn cool.
<davenull>
Thank you for your time everyone, I am going home.
* davenull
out
erasmas has quit [(Quit: leaving)]
<Profpatsch>
10000 days in the fire is long enough, you’re going home
drasich has quit [(Ping timeout: 240 seconds)]
digitus has quit [(Quit: digitus)]
markus1189 has joined #nixos
markus1199 has joined #nixos
markus1209 has quit [(Ping timeout: 245 seconds)]
davidak has quit [(Quit: Leaving.)]
markus1219 has quit [(Ping timeout: 255 seconds)]
iclanzan has quit [(Ping timeout: 260 seconds)]
ambro718 has quit [(Ping timeout: 240 seconds)]
seagreen has joined #nixos
<jack[m]1>
So, I would like to use fetchgit on a local repository path. Is there something better to use?