Users who want a user-specific bin directory to override system paths should
configure that in their user-specific ~/.bashrc, not in the system-wide init
file. The global file shouldn't add directories from user homes to $PATH
without knowing whether those actually exist or whether the users even want
them in $PATH. On my system, for example, there is no ~/bin, so I don't want my
$PATH to look for one. Removing an erroneous entry from $PATH is cumbersome,
but adding one is easy, so it feels better to err on the side of caution.
svn path=/nixos/trunk/; revision=31188
xserver is started on start_xserver event, which is emitted by
check_for_xserver_start if there is no "noX11" on the kernel cmdline.
Thanks to viric for the general idea.
svn path=/nixos/trunk/; revision=31166
Upstart won't find a "bash" binary in $PATH when those commands are run, so we
refer to it using an absolute path.
svn path=/nixos/trunk/; revision=31157
pierron recommended the use of types.string over mergeOptionString, as
it is superior but might break things.
For my system the change evaluated to the exactly same.
svn path=/nixos/trunk/; revision=31138
This reverts commit 025f8c40b40fad50086e8761eee61098d8fb2651.
The check was intened for building the initrd of the installer.
svn path=/nixos/trunk/; revision=31137
popt-0.16 and cryptsetup-1.4.1 both generated pkgconfig (in contrast
to older versions). The pkgconfig files (popt.pc and cryptsetup.pc)
contain references into the store that are not removed by patchelf and
stage-1 fails with errors like: "output is not allowed to refer to
path `/nix/store/qccjhn063cfv171rcaxvxh0yk96zf7l2-cryptsetup-1.4.1'".
Now, only the cryptsetup binaries and its dependencies are copied,
determined by ldd. In addition the cryptsetup binary and lvm are
tested after patchelf has adjusted the library paths.
Thanks to Peter Simons and Eelco Dolstra for giving the rights hints.
svn path=/nixos/trunk/; revision=31128
"Permission denied" until I run "restart nfs-kernel-exports". "exportfs -ra" did not help.
I tracked that down to some race condition between loading the module nfsd and
starting the daemons. Therefore, I decided to add nfsd to the boot.kernelModules instead
of using modprove with it.
Now it works for my server. No more Permission denied after reboot.
svn path=/nixos/trunk/; revision=31113
default. See
http://www.codon.org.uk/~mjg59/power/good_practices.html
for the reasoning. (Basically, the ‘performance’ and ‘powersave’
governors don't actually provide extra performance or power savings
in most cases.)
It used to be that desktop environments like KDE were able to set
the governor through HAL (e.g. KDE could be configured to switch to
the powersave governor when the user unplugs his laptop). However,
this is no longer the case with upower — it is now expected that
everybody uses the ondemand governor. See
http://old.nabble.com/-PATCH--powerdevil-remove-cpufreq.patch-td27815354.html
* Rename ‘cpuFreqGovernor’ to ‘powerManagement.cpuFreqGovernor’.
* Include cpufreq-utils in the system path if a governor is set, since
we depend on it anyway.
svn path=/nixos/trunk/; revision=30991
Fix sane-backends to generate udev rules, add a snapshot of sane-backends's unstable repo, and add a SANE nixos module
svn path=/nixos/trunk/; revision=30764
There is room for improvement here. The options in conffile could be broken out into individual options and an extraConfig option added. But I think this looks right.
Patch by mornfall, slightly modified by me
svn path=/nixos/trunk/; revision=30731
For example, I use the following settings to configure T-Mobile Internet
access on my laptop, which is connected to the cell phone by USB:
| environment.wvdial.dialerDefaults = ''
| Init1 = AT+CGDCONT=1,"IP","internet.t-mobile"
| Modem Type = USB Modem
| Phone = *99#
| ISDN = 0
| Username = tm
| Password = tm
| Modem = /dev/ttyACM0
| Baud = 460800
| '';
svn path=/nixos/trunk/; revision=30489
just does what it says (enable a "graphical" configuration).
* Enable KDM in the graphical CD. The "auto" display manager doesn't
properly handle shutdowns etc.
svn path=/nixos/trunk/; revision=30331
all-hardware.nix. This allows base.nix (which should probably be
renamed to something more descriptive) to be reused without getting
the hardware configuration of the installation CD.
svn path=/nixos/trunk/; revision=30327
That's confusing and wrong: nixos-hardware-scan should just enable
support for the detected hardware, not enable lots of software (let
alone KDE).
svn path=/nixos/trunk/; revision=30325
problem in stage 1 should cause a panic. (Actually, the latter
should probably kill the instance since we can't repair it
anyway...)
svn path=/nixos/trunk/; revision=30215
There is really no reason to have a configuration file outside of /etc,
and it's consistent with what the Fedora/Debian packages for Nix use.
svn path=/nixos/trunk/; revision=30212
(e.g. segfaults whn loggin in to the NixOS graphical ISO) and don't
really work very well (e.g. fail to find files). They can be
re-enabled in the KDE System Settings.
svn path=/nixos/trunk/; revision=30155
After the change from revision 30103, nixos-rebuild suddenly consumed
freaky amounts of memory. I had to abort the process after it had
allocated well in excess of 30GB(!) of RAM. I'm not sure what is causing
this behavior, but undoing that assignment fixes the problem. The other
two commits needed to be revoked, too, because they depend on 30103.
svn path=/nixos/trunk/; revision=30127
