We used MASTER, BRIDGE and TEAM_MASTER keys for a differnet purpose than the
network.service did, confusing the legacy tooling. Let's do our best to write
compatible configuration files:
* Add *_UUID properties that won't clash with initscripts
* Ignore non-*_UUID keys on read if *_UUID is present
* If the connection.master is an UUID of a connection with a
connection.interface-name, write the uuid into the *_UUID key while setting
the non-*_UUID key to the interface name for compatibility
https://bugzilla.redhat.com/show_bug.cgi?id=1369091
Currently ipv4 and ipv6 settings share the ifcfg plugin vars for the
dhcp-hostname and dhcp-send-hostname options (DHCP_HOSTNAME and
DHCP_SEND_HOSTNAME). Add brand new vars DHCPV6_HOSTNAME and
DHCPV6_SEND_HOSTNAME to have var-property one to one mappings.
When those values are missing they are already considered as TRUE.
As they are written also when DHCP is not enabled, write them only
if they have a value different from the default.
If a connection is ACTIVATED and another one is ACTIVATING but there
is no global connectivity, we currently set the manager state to
CONNECTING and start a connectivity check to verify whether the
manager state can be promoted to CONNECTED_GLOBAL.
If this connectivity check fails, we shouldn't promote a CONNECTING
state to CONNECTED_SITE.
Fixes: 084da69a30
The manager state is already computed every time an active connection
changes state, it is not necessary to call nm_manager_update_state()
also when the property is read.
Moreover, nm_manager_update_state() emits a "notify::state" signal
which causes a re-read of the property by the nm-exported-object,
resulting in a nested execution of nm_manager_update_state().
We call nmp_utils_ethtool_get_driver_info() twice when receiving a
netlink message, but we don't need a clone of the string values.
Instead, expose a data structure that should be stack allocated
by the caller.
The ioctl APIs ethtool/mii require an interface ifname. That is inherrently
racy as interfaces can be renamed. This cannot be fixed, we can only
minimize the time between verifying the ifname and calling ioctl.
We already had problems with that when ethtool would access an interface
by name that didn't exists. See commit ab41c13b06 .
Checking for an existing interface only helps avoiding races when an interface
gets deleted. It does not help against renaming.
Go one step further, and instead of checking whether such an ifname
exists, try to get the ifname based on the ifindex immediately before
we need it.
This brings an additional overhead for each ethtool access.
A reimplementation of g_file_get_contents() to overcome two limitations:
- nm_utils_file_get_contents() accepts a @dirfd argument to open the
file relative using openat().
- nm_utils_fd_get_contents() allows to read the content from a file
filedescriptor.
- both support a max_length argument, to fail gracefully if we
get tricked into loading a huge file.
- use nm_auto_close cleanup attribute
- optionally, return the found ifname
- don't stat "phy80211". If such an entity can be opened,
just assume it's a directory.
Since function nmp_utils_open_sysctl() can avoid race condition, use it
in wifi_utils_is_wifi() to open sysfs and correctly check if it's a wifi
device.
https://bugzilla.gnome.org/show_bug.cgi?id=775613
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
A race condition may happen when NetworkManager opens sysfs and udev
renames interface name at the same time. Thomas Haller provides a new
function [1] which can avoid the race condition when opening sysfs.
This patch is a direct copy from [1].
[1] https://mail.gnome.org/archives/networkmanager-list/2016-December/msg00004.html
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
It would be nice that our logging functions are guaranteed to
preserve errno. We are currently not very consistent about handling
errno, let's improve on that.
update_dns() is rather complicated. The part where we collect
resolv-conf data can be moved out. That is nice, because
_collect_resolv_conf_data() has no side-effects (except logging)
and it's only purpose is to return the out-arguments.
It is useful to let clients retrieve the current DNS configuration,
which can be displayed to users or used (together with dns=none) to
implement custom DNS configuration logic through external tools.
Extend the D-Bus API and introduce a new NMDnsManager interface that
contains all the information related to DNS configuration.
At the moment the new DnsManager object only contains basic properties
describing the current DNS mode and resolv.conf manager.
Previously we printed the DNS configuration for all devices including
unmanaged and unconfigured ones:
dns-mgr: config: 100 best v4 ens3 : 192.168.10.1
dns-mgr: config: 100 default v6 ens3 : fd01🔡:3681:c4ff:fec0:fe94
dns-mgr: config: 100 default v4 lo :
dns-mgr: config: 100 default v6 lo :
dns-mgr: config: 100 default v4 ens9 :
dns-mgr: config: 100 default v6 ens9 :
Instead, print only devices that have name servers set.
For new interfaces we don't define a PropertiesChanged signal as we
rely on GDBus to generate one on the standard
org.freedesktop.DBus.Properties interface.
Moving the PPP manager to a separate plugin that is loaded when needed
has the advantage of slightly reducing memory footprint and makes it
possible to install the PPP support only where needed.
https://bugzilla.gnome.org/show_bug.cgi?id=773482
nm_platform_ip4_address_sync() tries to apply the new configuration
with the minimum effort and doesn't delete addresses if they are
already present on the interface. This can break the ordering, as an
existing address would be promoted by kernel to primary, even if it
was last in our configuration.
Add some logic to ensure the correct order of addresses is always
enforced. This fixes situations like:
# nmcli connection add type ethernet ifname eth0 con-name t \
ipv4.method manual \
ipv4.addresses "1.1.1.1/24,1.1.1.2/24,1.1.1.5/24"
# nmcli connection up t
=> addresses are applied in the right order:
inet 1.1.1.1/24 brd 1.1.1.255 scope global eth0
inet 1.1.1.2/24 brd 1.1.1.255 scope global secondary eth0
inet 1.1.1.5/24 brd 1.1.1.255 scope global secondary eth0
# nmcli connection mod t ipv4.addresses "1.1.1.5/24,1.1.1.2/24,1.1.1.1/24"
# nmcli device reapply eth0
=> order is wrong:
inet 1.1.1.2/24 brd 1.1.1.255 scope global eth0
inet 1.1.1.5/24 brd 1.1.1.255 scope global secondary eth0
inet 1.1.1.1/24 brd 1.1.1.255 scope global secondary eth0
Co-Authored-By: Thomas Haller <thaller@redhat.com>
