Cgroups are handled by systemd now. Systemd's cgroup support does not
do all the things that cgrulesengd does, but they're likely to
interact poorly with each other.
So instead of:
boot.systemd.services."foo".serviceConfig =
''
StartLimitInterval=10
CPUShare=500
'';
you can say:
boot.systemd.services."foo".serviceConfig.StartLimitInterval = 10;
boot.systemd.services."foo".serviceConfig.CPUShare = 500;
This way all unit options are available and users can set/override
options in configuration.nix.
I had some problems with LDAP user lookups not working properly
at boot. I found that invalidating passwd and group on the
ip-up event (when nscd-invalidate starts) helped a bit.
Subtle: dhcpcd.service would call resolvconf during shutdown, which in
turn would start invalidate-nscd.service, causing the shutdown to be
cancelled. Instead, give nscd.service a proper reload action, and do
"systemctl reload --no-block nscd.service". The --no-block is
necessary to prevent that command from waiting until a timeout occurs
(bug in systemd?).
Enabled a bunch of units that ship with systemd. Also added an option
‘boot.systemd.units’ that can be used to define additional units
(e.g. ‘sshd.service’).
modprobe.
* Move the implementation of boot.kernelModules from the udev job to
the activation script. This prevents races with the udev job.
* Drop references to the "capability" kernel module, which no longer
exists.
svn path=/nixos/trunk/; revision=33208
first user job I put in ~/.init caused Upstart to crash with an
assertion failure, taking down the system. Given that Upstart has a
non-trivial attack surface with this feature, it seems best to
disable it.
svn path=/nixos/trunk/; revision=32779
event is emitted by dhclient and by the network-interfaces job in
case of statically configured interfaces. Invalidating the cache is
necessary to get rid of negative queries.
svn path=/nixos/trunk/; revision=31779
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
of DBus configuration directories. Thus packages with DBus
configuration files installed in the default profile will be found
automatically. (It would be nice to include $HOME/.nix-profile in
the session bus search path, but I don't think that's possible.)
svn path=/nixos/trunk/; revision=22737
${configDir}/system.conf. This is necessary to enable dbus to
reload its configuration on SIGHUP after a nixos-rebuild. (The
daemon cannot be restarted because then console-kit-daemon loses its
state.)
* Generate a proper /etc/dbus-1/session.conf.
svn path=/nixos/trunk/; revision=22572
* dbus: don't signal Upstart, since that seems to make it forget about
pending events or something. In any case starting dbus after acpid
was running wouldn't trigger hal to be started (but the other way
around did work).
svn path=/nixos/branches/upstart-0.6/; revision=18251
is "ready". This prevents ugly race conditions, e.g. HAL failing to
start because dbus hasn't finished starting yet.
* Support post-start scripts. These are executed after the job's main
process has started but before the job's "started" event is
emitted. For instance, the udev job uses this to perform "udevadm
trigger / settle" to create all devices. Previously this had to be
done in the pre-start script, so the daemon had to started in the
pre-start script as well.
svn path=/nixos/branches/upstart-0.6/; revision=18211
recognise X logins as active sessions (i.e. ck-list-sessions will
show "active = FALSE", and the user doesn't get ownership of
devices). I guess console-kit-daemon (which is started by dbus)
needs some device in /dev to exist.
svn path=/nixos/trunk/; revision=17428
