As with the settings, each device type was defining its own error
type, containing either redundant or non-useful error codes. Drop all
of the subtype-specific errors, and reduce things to just
NM_DEVICE_ERROR_FAILED, NM_DEVICE_ERROR_INCOMPATIBLE_CONNECTION, and
NM_DEVICE_ERROR_INVALID_CONNECTION.
The device-type-specific errors were only returned from their
nm_device_connection_compatible() implementations, so this is also a
good opportunity to simplify those, by moving duplicated functionality
into the base NMDevice implementation, and then allowing the
subclasses to assume that the connection has already been validated in
their own code. Most of the implementations now just check that the
connection has the correct type for the device (which can't be done at
the NMDevice level since some device types (eg, Ethernet) support
multiple connection types.)
Also, make sure that all of the error messages are localized.
NMRemoteConnection used to return
NM_REMOTE_CONNECTION_ERROR_DISCONNECTED if you tried to operate on a
connection that had been disconnected from its D-Bus proxy. But this
disappeared in the gdbus port (since gdbus doesn't emit a signal when
it happens, so it's harder to notice. And it's not clear why
NMRemoteConnection did this when no other class did anyway...).
Register NMConnectionError with D-Bus on both sides, so that, eg,
connection validation failures in the daemon will translate to the
correct error codes in the client.
Add nm-errors.[ch], and move libnm-core's two error domains
(NMConnectionError and NMCryptoError) there.
NMCryptoError wasn't previously visible, but it can be returned from
some public API, so it should be.
Many of NMCryptoError's codes would basically never be useful for
programs to distinguish between. Streamline the codes, and fix the
enumeration member names to start with "NM_CRYPTO_ERROR_" rather than
"NM_CRYPTO_ERR_".
Each setting type was defining its own error type, but most of them
had exactly the same three errors ("unknown", "missing property", and
"invalid property"), and none of the other values was of much use
programmatically anyway.
So, this commit merges NMSettingError, NMSettingAdslError, etc, all
into NMConnectionError. (The reason for merging into NMConnectionError
rather than NMSettingError is that we also already have
"NMSettingsError", for errors related to the settings service, so
"NMConnectionError" is a less-confusable name for settings/connection
errors than "NMSettingError".)
Also, make sure that all of the affected error messages are localized,
and (where appropriate) prefix them with the relevant property name.
Renamed error codes:
NM_SETTING_ERROR_PROPERTY_NOT_FOUND -> NM_CONNECTION_ERROR_PROPERTY_NOT_FOUND
NM_SETTING_ERROR_PROPERTY_NOT_SECRET -> NM_CONNECTION_ERROR_PROPERTY_NOT_SECRET
Remapped error codes:
NM_SETTING_*_ERROR_MISSING_PROPERTY -> NM_CONNECTION_ERROR_MISSING_PROPERTY
NM_SETTING_*_ERROR_INVALID_PROPERTY -> NM_CONNECTION_ERROR_INVALID_PROPERTY
NM_SETTING_ERROR_PROPERTY_TYPE_MISMATCH -> NM_CONNECTION_ERROR_INVALID_PROPERTY
NM_SETTING_BLUETOOTH_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_INVALID_SETTING
NM_SETTING_BOND_ERROR_INVALID_OPTION -> NM_CONNECTION_ERROR_INVALID_PROPERTY
NM_SETTING_BOND_ERROR_MISSING_OPTION -> NM_CONNECTION_ERROR_MISSING_PROPERTY
NM_SETTING_CONNECTION_ERROR_TYPE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING
NM_SETTING_CONNECTION_ERROR_SLAVE_SETTING_NOT_FOUND -> NM_CONNECTION_ERROR_MISSING_SETTING
NM_SETTING_IP4_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY
NM_SETTING_IP6_CONFIG_ERROR_NOT_ALLOWED_FOR_METHOD -> NM_CONNECTION_ERROR_INVALID_PROPERTY
NM_SETTING_VLAN_ERROR_INVALID_PARENT -> NM_CONNECTION_ERROR_INVALID_PROPERTY
NM_SETTING_WIRELESS_SECURITY_ERROR_MISSING_802_1X_SETTING -> NM_CONNECTION_ERROR_MISSING_SETTING
NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_802_1X -> NM_CONNECTION_ERROR_INVALID_PROPERTY
NM_SETTING_WIRELESS_SECURITY_ERROR_LEAP_REQUIRES_USERNAME -> NM_CONNECTION_ERROR_MISSING_PROPERTY
NM_SETTING_WIRELESS_SECURITY_ERROR_SHARED_KEY_REQUIRES_WEP -> NM_CONNECTION_ERROR_INVALID_PROPERTY
NM_SETTING_WIRELESS_ERROR_CHANNEL_REQUIRES_BAND -> NM_CONNECTION_ERROR_MISSING_PROPERTY
Dropped error codes (were previously defined but unused):
NM_SETTING_CDMA_ERROR_MISSING_SERIAL_SETTING
NM_SETTING_CONNECTION_ERROR_IP_CONFIG_NOT_ALLOWED
NM_SETTING_GSM_ERROR_MISSING_SERIAL_SETTING
NM_SETTING_PPP_ERROR_REQUIRE_MPPE_NOT_ALLOWED
NM_SETTING_PPPOE_ERROR_MISSING_PPP_SETTING
NM_SETTING_SERIAL_ERROR_MISSING_PPP_SETTING
NM_SETTING_WIRELESS_ERROR_MISSING_SECURITY_SETTING
Rename NM_CONNECTION_ERROR_UNKNOWN to NM_CONNECTION_ERROR_FAILED,
following GError best practices.
Replace NM_CONNECTION_ERROR_CONNECTION_SETTING_NOT_FOUND ("no
NMSettingConnection") with a more generic
NM_CONNECTION_ERROR_MISSING_SETTING. Use that new code in a few places
that had previously been using NM_CONNECTION_ERROR_SETTING_NOT_FOUND,
which was supposed to mean "the setting that you asked about doesn't
exist", not "the connection is invalid because it's missing a required
setting".
Clarify that NM_CONNECTION_ERROR_INVALID_SETTING can be used for any
"invalid or inappropriate NMSetting", not just a "conflicting" one.
(But fix a case in nm_connection_update_secrets() that was returning
INVALID_SETTING when it should have been return-if-failing instead.)
For both MISSING_SETTING and INVALID_SETTING, always prefix the error
message with "setting-name: ", just like we do with the various
NMSetting MISSING_PROPERTY and INVALID_PROPERTY errors. And make sure
that the error message is marked for localization.
Drop NM_CONNECTION_ERROR_CONNECTION_TYPE_INVALID, which is pretty
pointless; it was only used in the case where connection.type was the
name of a valid setting type that is not a base setting type. Instead,
just return NM_SETTING_CONNECTION_ERROR_INVALID_PROPERTY for
connection.type in this case (which is what the code already did when
connection.type was completely unrecognized).
nm_setting_lookup_type_by_quark() was only ever used in places that
were still mistakenly assuming the old style of nm_connection_verify()
errors, where the error message would contain only a property name and
no further explanation. Fix those places to assume that the error will
contain a real error message, and include both the setting name and
the property name.
Given that, there's no longer any need for
nm_setting_lookup_type_by_quark(), so drop it.
The IPv6LL address handling in userspace patches failed to handle the
case where the IPv6 method was 'ignore'. Previously the kernel would
usually add the IPv6LL address itself, but if NM has turned off kernel
IPv6LL then obviously this wouldn't happen. So when the method is
'ignore', turn off userspace IPv6LL handling and bounce disable_ipv6
to make the kernel add the IPv6LL address if it wants to.
NMActiveConnection:specific-object was renamed to
NMActiveConnection:specific-object-path in 677314c5, but it didn't
actually work, because of assumptions NMObject makes. Fix that.
Later _nmcli_compl_ARGS() wants to check $OPTIONS_TYPE to behave differently
depending on the type. As we accept abbreviations of $OPTIONS_TYPE, let's normalize
the value, so that we don't have to consider abbrevations later on.
Signed-off-by: Thomas Haller <thaller@redhat.com>
The revision number of the RPM (as build by contrib/rpm) should
be increasing so that newer packages can be installed using
`yum install` and older packages can be downgraded using
`yum downgrade`.
By counting only --first-parent, the following example turns
out wrong. Note the duplicate revision numbers.
-- A(100)----------------------------F(101)----G(102)
\ /
B(101)----C(102)----D(103)----E(104)
Just count *all* parent commits
* two users are logged in: user_a and user_b
* user_b creates a connection visible only to him 'user_b_private'
(permissions: user:user_b)
* user_b activates the connection user_b_private
* user_a does not see the connection profile, but he does see the active
connection
* user_a calls
nmcli con
nmcli con show user_b_private
Test NMClient's handling of active connections, and in particular test
that we can correctly resolve the circular reference between an
NMDevice and an NMActiveConnection, both synchronously and
asynchronously.
NMActiveConnections start out in state "unknown", but then quickly
switch to "activating". Unfortunately, it's sometimes possible for
this to be externally visible. Fix this by lying and saying that state
is "activating" during the initial "unknown" stage (though not if the
state changes to "unknown" later on).
(Actually changing the initial state to "activating" breaks things
because some code depends on there being a transition into the
"activating" state.)
In some cases, the nm_client_activate_connection() callback could be
invoked when either the NMActiveConnection or the NMDevice had not
been initialized yet. Fix it to wait for both of them to be fully
initialized before invoking the callback.
3e5b3833 changed various libnm methods to return 0-length arrays
rather than NULL, and changed various other places to assume this
behavior. The guarantee that they didn't return NULL relied on the
assumption that all D-Bus properties would get initialized by NMObject
code before anyone could call their get methods.
However, this assumption was violated in the case where two objects
circularly referenced each other (eg, NMDevice and
NMActiveConnection). f9f9d297 attempted to fix this by slightly
changing the ordering of NMObjectCache operations, but it actually
ended up breaking things and causing infinite loops of object creation
in some cases.
There's no way to guarantee we never return partially-initialized
objects without heavily rewriting the object-creation code, so this
reverts most of f9f9d297 (leaving only the new comment in
_nm_object_create()). The crashes introduced by 3e5b3833 will no
longer happen because we've now fixed the classes to have initialized
their object arrays to non-NULL values even before they get the real
values.
In some cases, code may look at the value of an array-valued property
during object initialization, before NMObject has set it to its actual
initial value. So ensure that we initialize all such properties to an
empty array, rather than leaving them NULL.
Also fix another bug in NMClient that could result in
priv->active_connections being NULL during certain signal emissions,
and fix nm_client_get_active_connections() to not return NULL when NM
was not running.
PPPoE connections involve two different network connections, making it
ambiguous what the "Device" field refers to. Clarify that it refers to
the Ethernet device, not the PPP device.
Routing configuration fails to apply if the device is not IFF_UP, so if
we're going to apply IP configuration to the device, make sure it's IFF_UP
first.
https://bugzilla.gnome.org/show_bug.cgi?id=738479
When a child device is found and an IP configuration already exists
for it even though it is under NM control (like when pppd applies
IP config to a WWAN device before NM gets the IP details from the pppd
plugin), don't deconfigure the child device when removing it from the
device list, because this breaks the device's configuration.
https://bugzilla.gnome.org/show_bug.cgi?id=738479
A generated connection contains a copy of the device's existing
configuration, so it's entirely redundant to merge the connection
back into the device's IP config. But even though that should
result in no changes to the IP config, NMSettingIPxConfig treats a
route metric of '0' as the device priority, while NMIPxConfig
allows 0 as a valid route metric. Since the setting values
are preferred (they are supposed to be user-supplied and thus
override anythign else, but in this case they are generated and
thus not user-supplied) external routes with a metric of 0 are
overwritten with the device priority metric.
https://bugzilla.gnome.org/show_bug.cgi?id=738268
When using a private bus connection, the service is never marked
as running when settings are initialized asynchronously. Successfully
opening a socket in NM's runtime directory should already imply
a running service, so just mark it as such (as we already do in
the synchronous path).
