| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
| |
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This is the first step of generalizing the configurations into a big
repository where people can track their NixOS configurations and write
modules.
In the next few steps I'm going to gradually refactor everything in
labernix/ into the main tree and factor out personal stuff into separate
namespaces.
Also, this will involve a license change from APACHE-2.0 to GPL-3.0.
|
| |
| |
| |
| |
| |
| |
| |
| | |
This currently is just a dummy module for now with a big fat TODO, but
at least we now have a homogenous module namespace starting with
"labernix." and categorized by services/config/tasks/whatnot.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
I shouldn't be wondering why it doesn't find the module options if we
prepare the override but don't actually use it. So this also fixes a few
other errors related to that we're now actually using our own modules.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| | |
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
We're going not only to modify all VM test nodes, but also inject the
updated pkgs into the tests own arguments, so we can refer to them even
outside of testing machine configuration modules.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
We now have "pkgs.labernix" available for every machine. Also, we're
deactivating the packet filter by default, so we can use it whenever we
_really_ need it.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We're going to add this as common modules for all machines in the
repository, so we can define own options easily and without the need to
refer to the modules directly.
Also, we can easily create a manual for labernix specific modules.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| | |
This includes pull request muesli/beehive#49 from @Profpatsch.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| | |
We want to be easily able to fetch the store paths and to monitor build
results.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Roughly generated using go2nix, but patched a few of the examples out
and cleaned up the dependencies expression. Somehowe go2nix fetched
appstats, which did not build but somehow wasn't required in order to
build beehive.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| | |
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| | |
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
We don't have a Nix package for pronterface yet and slic3r is broken
right now, so we need to investigate it after the first build from
Hydra.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Unfortunately, using builtins.readDir isn't very good idea here, because
all file references from expression files found via readDir will have
imports relative to the Nix store.
We can only work around it if we copy the whole source tree into the
store, but for the sake of simplicity it's also better if we use
explicit imports.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
As these machines are not deployed yet, we need to put in dummy values,
because if we're not building VMs those attributes are expected to be
set and end up with an evaluation error.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
At the moment this is also the release Nix expression we're going to use
for Hydra. Right now, the machine expressions don't evaluate because
they're lacking hardware information. But apart from that we should get
out full system builds afterwards.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
At least, this is after I asked around who wanted to have which software
on the LabTop by default.
For repetierhost, we don't have a package yet and we're going to package
it first. Unfortunately it's proprietary software, so we may need to
patchelf :-/
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| | |
The hardware has a US keyboard layout, so let's match it with the
configuration.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
A still unfinished version of @waaaaargh's ansible deployment:
https://github.com/waaaaargh/mailserver-ansible
The idea here is to create a NixOS profile specific to large-scole mail
server deployments. All with a 100% fleshed out Postfix configuration of
course :-)
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
I found this to be a better idea rather than leaving them unlicensed, so
in the event someone likes to distribute the repository it shouldn't
raise questions about whether it's legal.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Right now, the license is only for all Nix expression files, which
means, that patches and other files are unlicensed for now. If someone
would like to redistribute them, feel free to open an issue.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently, this one is unfinished and I was working on it while I've
been at the Lab and don't have a clue right now where I left off.
But it's still a good idea to add it here, so someone[TM] can finish it
someday.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| | |
If we want to make tests or modules for those machines, it makes sense
to properly namespace them.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Posted this at openlab-aux/labnetz-doku#2 already, so it's more or less
the same but with minor enhancements. For example the routes are now
specified using an attribute set and we have option declarations now.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently nothing too fance, it only describes what it is for and why
the name starts with "laber".
Thanks to @grindhold for coming up with this name.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
| |
It's not nearly complete and I have a more complete version on my own
Laptop, but it's a start to get all Lab-related Nix expressions
together.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
| |
I'm including the whole file here, because it's easier to move it to
<nixpkgs> once version 0.9.0 is finally released.
This finally gets rid of the damn phonon dependency which cased most of
my playback issues so far.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
| |
No idea, why I used kernel 3.12 here, but if it has a real reason I'm
going to find it out very soon.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
|
| |
This was in since the initial commit (6c224e4) of the bfqsched Nix
expression and I didn't notice it for all those upgrades.
Basically, this is just a no-op, because nowadays the sed expression
doesn't rewrite anything anymore, but it might be harmful if it should
match something in a future version of the bfqsched patch set.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
| |
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I just tripped over a bug which might have been fixed by one of the
following changes:
* Btrfs:__add_inode_ref: out of bounds memory read when looking for
extended ref.
* Btrfs: fix data loss in the fast fsync path
* Btrfs: remove extra run_delayed_refs in update_cowonly_root
* Btrfs: incremental send, don't rename a directory too soon
* btrfs: fix lost return value due to variable shadowing
* Btrfs: do not ignore errors from btrfs_lookup_xattr in do_setxattr
* Btrfs: fix off-by-one logic error in btrfs_realloc_node
* Btrfs: add missing inode update when punching hole
* Btrfs: abort the transaction if we fail to update the free space
cache inode
* Btrfs: fix fsync race leading to ordered extent memory leaks
So it might be a good idea to use netconsole for mmrnmhrm and dnyarri,
so I get the kernel panic next time.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
| |
Especially while being in an environment with unstable connectivity, it
helps a lot to prevent SSH sessions from being dropped.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
| |
Properly distinguish between patched-in dotfiles and real dotfiles.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
| |
Geesh, the kernel is too OLD! Two fucking days!
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
| |
Plus a few more commits after the tag was created.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
| |
It is no longer needed as it has been applied upstream in version 4.9.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
| |
I prefer networkd over the scripted configuration, because it's able to
properly track the interface states.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
| |
This reverts commit d64ed6fb4fd211f9690b359275371926246f4521.
Since I have moved back to my new home, this is no longer needed.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
| |
Let's get rid of that buggy old 3.19-rc6 and get to the new buggy
4.0-rc1 + a few more commits (so it's more like -rc1.5).
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I can't emphasize this enough: *REALLY*
Whenever the newest stable kernel is released, the version in
linux_testing will *still* the latest -rc even though it's older than
the current stable.
Strictly speaking, it's true that even though the stable version is
newer, the label "testing" is correct. However, and in my case I want to
really have the latest mainline version, no matter whether it's stable
or testing.
So, we're just comparing the major and minor version numbers and if they
match, we're using the latest stable, otherwise we're using testing.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
|
| |
This makes sense especially on tishtushi where disk space is sparse, but
also helps especially when testing/working on Nix expressions.
Motivated by a message from @wmertens:
http://lists.science.uu.nl/pipermail/nix-dev/2015-February/016180.html
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
| |
And as usual it's not -rc6, but a few commits ahead :-)
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Today seems to be a "bad luck day" because my Sheeva Plug just died. So
I'm back to some weird stupid old Netgear thingy that seems to be so old
that it doesn't even know what IPv6 "might be". And it's already the
"latest and gro^Heatest" firmware that's available.
One of the ugly things is that our internal DNS configuration can't be
done using that stupid thing, because it uses DHCP to assign external
DNS servers rather than having an internal DNS forwarder where you can
assign your own[TM] zones.
So let's hope that we can revert this in the foreseeable future.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
| |
Thanks to @abbradar for packaging the TLP nixpkg and NixOS service
module. The upside is that the APM level of the disk is 254 by default,
so we don't need to explicitly set that during bootup and it is done
automatically when we're on AC (when the battery is the power source
it's 128).
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
| |
Quite useful to get power management statistics like battery info.
Of course, in a better world we would have a completly modular status
bar that is able to display such information *only* if we're *not* on
AC.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
| |
For the primary network interface that is going to be displayed in the
status bar, it's a good idea to have a timeout value. I'm sitting in the
train right now and it's annoying to *not* have a working status bar.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
| |
Also now uses dashes for the names instead of the camelCased versions we
had in old haskellPackages.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
|
|
|
|
|
|
| |
We also now use fetchFromGitHub and no longer use "git rev-parse" in
order to truncate the commit hash. Instead, we're just trimming down the
revision we already have in the Nix expression to 7 characters (like per
default if done by "git rev-parse --short").
This should also mitigate the evaluation problems with latest nixpkgs,
where leaveDotGit changed so it results in a different hash.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|
|
|
|
|
|
| |
... actually a few commits ahead of -rc5 ;-)
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
|