* use 127.0.1.1 under debian (since it's the default there)
* also set the IPv6 loopback address (::1) since some tools tries to
AAAA resolve the hostname (for example sendmail)
* ensure proper /etc/hosts format through postinst-script (hostname as
last entry)
The ext4 reserved space is necessary for root partitions, but in the
private.img data partition, it is wasted space (accessible only to root
processes), which means losing 100 MB of the default 2GB.
From mkfs.ext4 man page: "-m reserved-blocks-percentage Specify the
percentage of the filesystem blocks reserved for the super-user." ...
"The default percentage is 5%."
Do not reexec systemd when running version is the same as installed
binary. Apparently reexec causes some race condifions, which result in
assertion fail in systemd.
SystemD version can differ from initramfs one (which is build in dom0
build environment), so reexec it at startup.
This fixes systemd-212 archlinux issue.
This time it is for cups, which have socket-based and path-based
activators. When activator tires to start the service which is disabled
by condition file it enters infinite loop (as service wont start, but
will not report an error).
Get rid of underscores in filenames, use dashes instead.
This is first part of cleanup in filenames.
"qubes_rpc" still untouched - will be in separate commit.
No more ugly symlink creation at VM startup, nautilus-actions have system-wide
dir (in opposite to nautilus-scripts).
Currently old symlinks are not cleaned up. Maybe it should, but leaving them
have one advantage: will not break existing users behavior.
This will ensure that /rw/config/rc.local is called after applying default
iptables rules, so it can safely modify it without the risk to be overridden
later by default ones.
When /rw/home/user/.qubes-dispvm-customized is present use /rw/home/user
instead of default /etc/dispvm-dotfiles.tbz. Also make sure that /rw will not
remain mounted during DispVM creation.
The simplest way is just add proxy=... entry to /etc/yum.conf, but sometimes it
is reasonable to bypass the proxy. Some examples:
- usage of non-standard repos with some exotic file layout, which will be
blocked by the proxy
- usage of repos not-accessible via proxy (eg only via VPN stared in VpnVM)
This commit introduces 'yum-proxy-setup' pseudo-service, which can be
controlled via standard qvm-service or qubes-manager. When enabled - yum will
be configured at VM startup to use qubes proxy, otherwise - to connect directly
(proxy setting will be cleared).
On FC>=15 /var/run is on tmpfs, so /var/run/tinyproxy from rpm don't survive
reboot. This is bug in Fedora package (should include config file for tmpfiles
service). For now create dir just before starting service.
Introduce proxy service, which allow only http(s) traffic to yum repos. The
filter rules are based on URL regexp, so it isn't full-featured content
inspection and can be easy bypassed, but should be enough to prevent some
erroneus user actions (like clicking on invalid link).
It is set up to intercept connections to 10.137.255.254:8082, so VM can connect
to this IP regardless of VM in which proxy is running. By default it is
started in every NetVM, but this can be changed using qvm-service or
qubes-manager (as always).
In FC15, NetworkManager by default uses global connections ("Available to all users"). Save them in /rw instead of /etc, to preserve them across reboots.
