2018-09-20
21:10
<
clever >
zduch4c: i glanced at the source for the driver, and i think it relies on a second module, for the serio interface
21:05
<
clever >
zduch4c: next thing id check is what dmesg says on nixos, when you rmmod and modprobe again
21:04
<
clever >
zduch4c: and evtest shows xy coords when you touch the screen?
21:03
<
clever >
zduch4c: if you reload it, does it work again?, does evtest list it?
21:02
<
clever >
zduch4c: does the touchscreen stop working?
21:02
<
clever >
zduch4c: what happens if you try to `rmmod wacom_w8001` on ubuntu?
21:01
<
clever >
ah, found it
21:01
<
clever >
zduch4c: dont see the link for the ubuntu modinfo
20:42
<
clever >
so oyu could use a vm or 2nd machine to speed up comparison
20:42
<
clever >
ubuntu doesnt have to actually be booted on the tablet
20:42
<
clever >
zduch4c: next thing i would do then is compare `modinfo wacom_w8001` on both ubuntu and nixos
20:40
<
clever >
zduch4c: what does dmesg say involving wacom?
20:40
<
clever >
zduch4c: what about in evtest?
20:39
<
clever >
zduch4c: does it work once loaded like that?
20:39
<
clever >
zduch4c: and does nixos also show it in lsmod?
20:38
<
clever >
zduch4c: what module did ubuntu show in lsmod?
19:09
<
clever >
LnL: the process was involved in creating backups, so it would also have write to the backups
19:08
<
clever >
LnL: the backups where also mounted over NFS
19:08
<
clever >
LnL: ive heard a horror story of basically that, with rm -rf /${backuppath}
00:27
<
clever >
samueldr: i suspect that the boot firmware is responsible for checking the hat i2c, and injecting it into the DTB, but i'm not entirely sure
2018-09-19
22:35
<
clever >
gchristensen: maybe its been fixed since i saw the bug
22:33
<
clever >
Mic92: dont let NM in!
22:22
<
clever >
rawtaz: if its using fetchFromGitHub, then the name is just "source"
22:22
<
clever >
rawtaz: the storepath ignores the version entirely
22:20
<
clever >
hashing a string that contains the hash, name, and other params
22:20
<
clever >
rawtaz: a fairly complex algo, where the inputs are the hash and derivation name
22:19
<
clever >
rawtaz: you claimed the file hasnt changed, so its not bothering to re-download
22:19
<
clever >
rawtaz: if the hash is identical, nix will use the old download
22:09
<
clever >
it might have been fixed in the years since i memorized how everything works
22:08
<
clever >
at least, last time i messed with containers, it didnt
22:08
<
clever >
gchristensen: nixos-rebuild doesnt understand @ based units
22:01
<
clever >
gchristensen: so if you have a declarative container in a nixops machine, it wont update
22:01
<
clever >
gchristensen: with declarative containers, systemd/nixos-rebuild wont restart them on activation
21:55
<
clever >
gchristensen: havent tried it yet
21:44
<
clever >
catern: fetchgit returns a path, so you could just src = (fetchgit { ... }) + "/foo";
21:43
<
clever >
postUnpack can then append to that var
21:43
<
clever >
and unpackPhase creates a dir named after the tar, and auto-sets sourceRoot to that dir
21:42
<
clever >
catern: sourceRoot is the path relative to $NIX_BUILD_TOP
21:40
<
clever >
append to the sourceRoot in postUnpack
20:51
<
clever >
astronavt: its normal for non-root to lack it
20:51
<
clever >
astronavt: which user is missing .nix-channels ?
20:08
<
clever >
astronavt: run realpath on that path
20:02
<
clever >
nDuff: default location for the file is still /etc/nix/machines
20:01
<
clever >
nDuff: its still configured with the same format, via the same env var
20:00
<
clever >
nDuff: the remote building is now built into nix, and doesnt need that external hook
19:52
<
clever >
anything the profile previously contained will be ignored (if it even existed to begin with)
19:51
<
clever >
Henson: the above command will just set the given profile to point to a certain storepath
19:51
<
clever >
nix-env -p /nix/var/nix/profiles/per-user/clever/profile --set /nix/store/x2f60fy8xc5rkr56ixwg1c2nikjl3v5n-user-environment
19:51
<
clever >
lrwxrwxrwx 1 clever users 60 Sep 12 18:23 /nix/var/nix/profiles/per-user/clever/profile-418-link -> /nix/store/x2f60fy8xc5rkr56ixwg1c2nikjl3v5n-user-environment
19:47
<
clever >
ah right, an example of how to install things
19:45
<
clever >
its now part of nix!
19:39
<
clever >
and :r matches between them as well
19:39
<
clever >
ghci also has :q
19:38
<
clever >
astronavt: it tells you :? for help on startup, and :q is listed in the help
19:37
<
clever >
astronavt: :q
19:33
<
clever >
astronavt: nix-repl is deprecated, you must now use nix repl
19:32
<
clever >
astronavt: what command did you run? are you on nixos or other?
18:42
<
clever >
nix might be doing the same?
18:42
<
clever >
sphalerite: ssh for example, will open /dev/tty, which can bypass all redirection and get the controlling tty of the process
18:33
<
clever >
catern: but with config.nix, you need to manually merge them into a single file
18:33
<
clever >
catern: its easier to use many overlays at once
18:31
<
clever >
you could use mapAttrs to run a function over every package in a set, before putting it into LnL's userPackages
18:29
<
clever >
and you can remove any single package as normal, or nix-env -iA one to just upgrade it
18:29
<
clever >
the same as if you had just ran nix-env -iA nixpkgs.hello against each one
18:29
<
clever >
when you nix-env -iA a set, it will install each attribute of the set, as a seperate entry
18:27
<
clever >
so you dont have to remember the incatation to update it
18:27
<
clever >
catern: yeah, LnL's gist uses an overlay instead, and includes a bash script to run nix-env for you
18:24
<
clever >
catern: and there is a different flag (i forget which one), which will uninstall everything you previously had, so its an atomic swap to the new version of mystuff, and no old stuff left
18:24
<
clever >
catern: if you nix-env -iA nixpkgs.mystuff, it will install everything in the set
18:24
<
clever >
catern: i just create a packageOverride in config.nix, that contains a set, like mystuff = { inherit (pkgs) hello firefox; };
18:16
<
clever >
selfsymmetric-pa: `ls -ltrh /nix/var/nix/profiles/per-user/root` will confirm when you last ran it as root
18:15
<
clever >
selfsymmetric-pa: and you ran `nix-channel --update` as root?
18:14
<
clever >
selfsymmetric-pa: what does `nix-instantiate --find-file unstable` report?
18:14
<
clever >
catern: the stdenv can be modified to add it automatically
18:14
<
clever >
selfsymmetric-pa: and which version are you expecting to get?
18:12
<
clever >
or fix nixpkgs to include man in the right field
18:12
<
clever >
add the buildEnv to config.nix as an override
18:11
<
clever >
sphalerite: which is why i mentioned a buildEnv a few hours ago
18:08
<
clever >
selfsymmetric-pa: what if you just `nix-build '<unstable>' -A kubernetes-helm`
18:07
<
clever >
and then the old version just wont exist
18:07
<
clever >
selfsymmetric-pa: an override would change the package the nixos module is already installing
18:06
<
clever >
selfsymmetric-pa: you usually want to use an override, using systemPackages may lead to 2 versions being in systemPackages, and then you dont always get the right one
18:05
<
clever >
selfsymmetric-pa: did you also enable kurbernetes-helm with a services. option?
18:03
<
clever >
selfsymmetric-pa: `nixos-rebuild --upgrade` only updates the nixos channel, and ignores the others
18:02
<
clever >
selfsymmetric-pa: when did you last `nix-channel --update` ?
18:02
<
clever >
selfsymmetric-pa: what does `nix-channel --list` report?
18:01
<
clever >
selfsymmetric-pa: uninstall wont help any
18:01
<
clever >
selfsymmetric-pa: how did you install it and what channel are you on?
18:00
<
clever >
selfsymmetric-pa: this says the commit exists in nixos-unstable
17:59
<
clever >
selfsymmetric-pa: check the nixos-unstable branch of the nixpkgs-channels repo
17:45
<
clever >
grp: AAAAAASomeThingsFailToEvaluate sorts first, and gives a saner error
17:45
<
clever >
grp: some people will just `nix-build` in the root of nixpkgs, or `nix-env -i` and then nix tries to install literally every single package
17:02
<
clever >
that might be a new error, to make it less confusing then a weird division error
17:02
<
clever >
sphalerite: it also sometimes treats the trailing / as a division operator
17:01
<
clever >
sphalerite: error, trailing slash on path
16:51
<
clever >
but its paths, not pkgs trees
16:50
<
clever >
the same set -iA acts on
16:50
<
clever >
sphalerite: its passing you a set, containing every channel
16:50
<
clever >
sphalerite: nix-env -iE '_: with import <nixpkgs> {}; symlinkJoin {name = "utillinux"; paths = [ utillinux utillinux.man ]; }'
16:50
<
clever >
sphalerite: nix-env supports -E
16:46
<
clever >
catern: it will need an override that can change outputsToInstall, or just buildEnv 2 things together
16:40
<
clever >
i can just omit man pages from everything in my kexec image
16:40
<
clever >
gchristensen: that may fix it then
16:39
<
clever >
systemPackages deal with man pages better
16:39
<
clever >
manpages are in its own output, which nix-env has trouble with
16:38
<
clever >
been there, done that
16:32
<
clever >
maybe i got which one needs help with that mixed up
16:32
<
clever >
vandenoever: alexherbo2 was asking about /boot earlier, thought it was EFI issues
16:31
<
clever >
vandenoever: what does `mount | grep boot` report?
15:52
<
clever >
mpickering: run `nix-store --query --roots` on one of them
15:41
<
clever >
sphalerite: i type too fast and it thinks i'm a spammer!!
15:41
<
clever >
sphalerite: it has warned me before, when i was just typing by hand!
07:00
<
clever >
virtio_pci and virtio_blk depend on it, and nixos will figure it out
07:00
<
clever >
virtio itself doesnt have to be listed
06:59
<
clever >
oh, and virtio_pci
06:59
<
clever >
my simple test probably did sata
06:59
<
clever >
so it cant find any disk
06:59
<
clever >
your initrd likely lacks virtio support
06:59
<
clever >
eeva: add virtio and virtio_blk to boot.initrd.availableKernelModules
06:58
<
clever >
eeva: oh, drivers
06:58
<
clever >
and then there are matching -device entries to create pci nodes for each disk
06:58
<
clever >
both have if=none, so the -drive wont create an interface
06:58
<
clever >
eeva: first disk has format=qcow2, second is format=row, that looks good
06:50
<
clever >
eeva: perhaps kvm isnt treating it as a qcow2, what does the commandline look like in `ps aux` when its running the vm?
06:50
<
clever >
but it hangs bringing up eth0, but thats not your issue
06:50
<
clever >
2nd just boots it
06:49
<
clever >
so all writes go to the 2nd disk.img, and all reads initially come from the nix store
06:49
<
clever >
1st command creates a 2nd qcow2, using the read-only file as its base
06:49
<
clever >
eeva: it boots
06:49
<
clever >
[clever@amd-nixos:~/apps/9bddb5012384d8f16e42fc78109f989c]$ ./qemu/bin/qemu-kvm -drive file=disk.img
06:49
<
clever >
[clever@amd-nixos:~/apps/9bddb5012384d8f16e42fc78109f989c]$ ./qemu/bin/qemu-img create -f qcow2 -o backing_file=result/nixos.qcow2 disk.img
06:45
<
clever >
eeva: testing the image locally...
06:42
<
clever >
eeva: unrelated, lines 13-15, isNormalUser=true; will do most of that
03:31
<
clever >
just /quit your msg
03:31
<
clever >
/quit is part of the irc protocol, it should work the same way in every client
03:30
<
clever >
Henson: /exit
03:02
<
clever >
Henson: you can just delete all of its symlink
02:41
<
clever >
so you cant test the pre-compiled version
02:41
<
clever >
haslersn: it is, but you cant run the script without that fixing
02:40
<
clever >
haslersn: only /bin/sh and /usr/bin/env exist, all else is in /run or /nix
02:40
<
clever >
haslersn: /bin/bash doesnt exist on nixos
02:33
<
clever >
growpotkin: nix's stdenv will even patch that to the absolute path of bash when installing scripts
02:33
<
clever >
growpotkin: #!/usr/bin/env bash
02:14
<
clever >
haslersn: if you run `nix-env -iA` on a set, it will install every attr of the set
01:38
<
clever >
yeah, try just booting the ubuntu livecd and see what happens, and inspect how its workng
01:35
<
clever >
acpi is on by default
01:31
<
clever >
zduch4c: then it could be translated to nixos
01:31
<
clever >
zduch4c: see if you can find directions on how to enable it, or find evidence of it in lspci or lsusb
01:28
<
clever >
zduch4c: anything listed by evtest?
01:17
<
clever >
zduch4c: also try running evtest and see if the wacom device is visible
01:17
<
clever >
zduch4c: and check dmesg for anything wacom related
01:17
<
clever >
zduch4c: check lsmod to see if any wacom things are loaded, and try manually loading things with modprobe
01:14
<
clever >
zduch4c: so wacom would auto-load if a compatible hid device is present
01:14
<
clever >
zduch4c: the wacom one has aliases on hid devices
01:13
<
clever >
zduch4c: this finds 4 modules on my machine, wacom, wacom_i2c, wacom_w8001, and wacom_serial4
01:12
<
clever >
zduch4c: [root@nas:~]# find -L /run/booted-system/kernel-modules | grep wacom
01:00
<
clever >
Henson: you probably want to do python27Packages = super.python27Packages.extend (pysuper: pyself: { ... }); i believe
00:31
<
clever >
tobiasBora: nixos modules are responsible for creating /etc files
00:30
<
clever >
tobiasBora: that would be handled in the nixos module
00:29
<
clever >
tobiasBora: can it accept the path to that config at runtime? via /etc or a cli arg to the daemon?
00:27
<
clever >
tobiasBora: no idea, ive only used exim and postfix
00:14
<
clever >
tobiasBora: and which user is ssmtp being ran as?
00:13
<
clever >
tobiasBora: are you connecting to port 25 on localhost or piping it into the sendmail binary?
00:12
<
clever >
tobiasBora: is a from: header being set when talking to ssmtp?
00:12
<
clever >
tobiasBora: i think thats only for incoming mail
00:11
<
clever >
jasongrossman: i have sometimes answered a question with a pastebin of a convo from 2 hours prior :P
00:10
<
clever >
oops, didnt read it that closely
00:04
<
clever >
and its already 1.7.1 on master and unstable
00:04
<
clever >
so you need to use an overlay to change the znc attribute
00:03
<
clever >
bb010g[m], jasongrossman: it runs whatever pkgs.znc provides
2018-09-18
23:15
<
clever >
modify it to accept patches=? , or to run the patchPhase
23:14
<
clever >
ah, what is that other thing?
23:14
<
clever >
what do you plan to do after patching the src?
23:11
<
clever >
do you even need to build it? heh
23:10
<
clever >
elvishjerricco: what about just using the patches arg of mkDerivation?
17:17
<
clever >
gchristensen: this env var also has nearly max priority, if the file exists
17:17
<
clever >
NIXPKGS_CONFIG=/etc/nix/nixpkgs-config.nix
17:17
<
clever >
gchristensen: add -v and look for any config.nix's
17:11
<
clever >
and then hosts.nix contains { "127.0.0.1" = [ "foo.com" ]; }
17:11
<
clever >
you can just networking.hosts = import ./hosts.nix;
17:10
<
clever >
doesnt have to be a module
17:10
<
clever >
hosts is a set, so it would need imort
17:10
<
clever >
extraHosts is just a string, in the standard /etc/hosts format
17:10
<
clever >
booglewoogle: or import
17:10
<
clever >
booglewoogle: builtins.readFile ./foo
17:00
<
clever >
d1rewolf: i nix-channel --update and nixos-rebuild every now and then
16:59
<
clever >
i dont use that
16:30
<
clever >
norfumpit: yeah
15:08
<
clever >
behind the scenes, `nix-shell -p hello` is identical to `nix-shell -E 'with import <nixpkgs> {}; pkgs.runCommand "shell" { buildInputs = [ (hello) ]; } ""'`
15:06
<
clever >
so you need to either use -E or -I nixpkgs=...
15:06
<
clever >
wpcarro: -p always does import <nixpkgs> {}
14:44
<
clever >
nix-env -e nix-repl to remove it
14:44
<
clever >
you installed nix-repl with nix-env, so it will remain there and ignore all updates until you install another
14:43
<
clever >
what does `which nix-repl` return?
14:43
<
clever >
nix-repl has been removed from nixpkgs
14:41
<
clever >
the tab completion and history also behave differently
14:41
<
clever >
wpcarro: nix-repl only uses 1.11, you want nix repl
04:00
<
clever >
the only other difference is that fetchzip can support zips
03:58
<
clever >
adisbladis: fetchurl uses the hash of the file it downloads, fetchzip uses the hash of the NAR that results from unpacking the thing it downloads
03:57
<
clever >
oldandwise: by default, qemu sets up userland nat and it just works
03:00
<
clever >
fetchzip still works on tar files, the name is a bit misleading
02:58
<
clever >
adisbladis: use fetchzip to get an unpacked copy
02:43
<
clever >
copumpkin: yeah, not seeing any real way to make it build the thing
02:40
<
clever >
so copyPaths(src,dst,pathset) can support drv's
02:39
<
clever >
copumpkin: ok, i can confirm that `nix copy` supports copying drvs over ssh-ng, without building
02:39
<
clever >
[clever@amd-nixos:~]$ nix copy --to ssh-ng://root@system76 /nix/store/v2l3ncp0ypdssjf7b8rxyppn3kyq3w87-hello-2.10.drv
02:33
<
clever >
then you should run `nix-store -r` on that drv first i guess
02:33
<
clever >
the s3 backend doesnt really support drv files
02:33
<
clever >
ah, some may not handle it at all
02:32
<
clever >
they can also be given paths to finished builds
02:32
<
clever >
copumpkin: all of the nix 2.0 commands will try to build the .drv you give it before doing the action you requested
02:14
<
clever >
copumpkin: thats what i feared, and why i would prefer copy
02:13
<
clever >
oldandwise: and how did you install vde2?
02:12
<
clever >
oldandwise: what command did you run?
02:11
<
clever >
copumpkin: you can probably use `--store s3://` but i would prefer to use `nix copy --to s3://`
02:11
<
clever >
,locate vde_switch
02:08
<
clever >
but extending that to everything, 3d models, textures, map layout, music
02:08
<
clever >
and just render at startup
02:08
<
clever >
the basic idea of how they made it so damn small, is to store svg rather then png
02:07
<
clever >
but you are free to modify the build instructions for any asset, and it will just rebuild it locally
02:06
<
clever >
so it would download pre-rendered game assets on startup
02:06
<
clever >
growpotkin: oh god, you could even have a binary cache, lol
02:05
<
clever >
and if the args to it havent changed, nix just pulls the old image out of /nix/store/
02:05
<
clever >
you basically just have a pkgs.runCommand derivation, that does whatever you want, and produces an image with the texture as $out
02:04
<
clever >
nix will hash the build instructions to compute the $out
02:04
<
clever >
yeah, it wold be all of the hashing nix already does
02:03
<
clever >
and yet i didnt write a single line of code, lol
01:58
<
clever >
so you could have a derivation to compile the texture from a set of directions
01:58
<
clever >
and with the right code, it can support building derivations to produce values passed to opengl
01:58
<
clever >
and then run the nix function for any event, and to compute the state changes
01:57
<
clever >
you just need to make a primop that accepts a nix function for the state machine and event handling
01:57
<
clever >
in theory, you can just eval a file with nix-instantiate, and it would fire up opengl and do anything you want
01:56
<
clever >
growpotkin: that value can be a set containing more primops (functions that run c++ code)
01:56
<
clever >
growpotkin: builtins.importNative("${foo}/lib/libfoo.so") will return an artbitary nix Value*
01:54
<
clever >
growpotkin: plan a, haskell :P
01:53
<
clever >
growpotkin: with recent changes, its actually easier
01:52
<
clever >
growpotkin: i have since gotten over that temporary insanity :P
01:51
<
clever >
growpotkin: including using nix derivations to cache computation in the store, with normal GC to clean up the cache
01:51
<
clever >
growpotkin: when i first got into nix, i had ideas on how to add a c ffi to nix, and use it for opengl stuff, lol
01:49
<
clever >
oldandwise: services.timesyncd.enable = false;
00:57
<
clever >
Lisanna: nice
00:44
<
clever >
Lisanna: probably
00:36
<
clever >
Lisanna: yeah, the values in the set are thunks, that run the function once and remember the result
00:30
<
clever >
Lisanna: a branch with that already exists
00:26
<
clever >
and those 2 let you measure the effect and see if your actually improving it
00:25
<
clever >
as long as you are using the same let block instance, and not re-importing that file too much
00:25
<
clever >
a let block may also work
00:24
<
clever >
but if you save that into an attr, it will only call it once
00:24
<
clever >
Lisanna: the master branch of nix isnt capable of remembering what a function returned for a given value
00:23
<
clever >
Lisanna: one min
2018-09-17
23:02
<
clever >
and if the dep's name existed in a certain env var, it didnt bother adding it
23:02
<
clever >
infinisil: i think it searched for the name of a dep to decide if the cflags are already setup or not
23:00
<
clever >
and due to the hashes in store paths, it gets false positives
22:59
<
clever >
infinisil: i just remembered, there is a package that will search $PATH for a given string to auto-detect support for something
18:44
<
clever >
rawtaz: check the journal, systemd failed to mount the fs, and it assumes the fs is mission critical
17:21
<
clever >
rawtaz: systemd is to blame there
16:29
<
clever >
it must have a /, so src = ./.;
16:29
<
clever >
xok: then use ./.; ?
16:28
<
clever >
you also dont need rec on line 1
16:28
<
clever >
remov the / at the end of line 4
16:03
<
clever >
mostly personal preference, i just ditch hardware-configuration.nix entirely
16:02
<
clever >
rawtaz: b: mount it manually, and re-run nixos-generate-config to update hardware-configuration.nix
16:02
<
clever >
rawtaz: a: nixos always merges them
15:53
<
clever >
nix will then copy the entire html_root dir over, and configure nginx to use that copy
15:52
<
clever >
lesh: i would do something like services.nginx.virtualHosts."
www.example.com ".root = ./html_root;
15:51
<
clever >
lesh: depends on where the files need to be, what they are used for, and if they contain any secrets
14:33
<
clever >
last one, plus some quotes
08:35
<
clever >
its /lib64/ld-linux-x86-64.so.2 that doesnt exist, and causes file not found