The sys-iface-state "assume" means to gracefully take over a device (for
example, after a restart). The end result is a fully managed interface.
The flag only has meaning while activating, and for most practical
purposes, such devices should be treated the same as fully activated
ones.
Without this, the MTU is not reset until the device reaches fully
activated state, at which point the sys-iface-state switches from
"assume" to "managed". With the previous commit, at that point we also
schedule an idle commit, which ends up also setting the MTU. Before
that, the MTU was only reset some undefined time later, when we happened
to do another NML3Cfg commit. Nonetheless, even waiting until we reach
fully activated state is wrong. Also during activation, commit the MTU.
I guess, what theoretically could happen is that we get our MTU via
ip-config (like DHCP). Then, during assuming we hit _commit_mtu()
without having the DHCP lease yet. This happens after a restart, so it
would be wrong to first reset the MTU, before we re-receive the DHCP
lease. However, if the MTU is really to be set due via
NM_DEVICE_MTU_SOURCE_IP_CONFIG, then all other MTU sources are also not
in effect (because ip-config has a low priority). In that case, we would
not have an MTU to reset and the code would not commit a new MTU. Thus
this should still be fine, also during activation when we didn't yet get
the DHCP lease (or other information to dynamically set the MTU).
When assuming a device, the NMActiveConnection switches the
sys-iface-state from "assume" to "managed" when the device reaches the
activated state.
<debug> [1679353062.8884] active-connection[000055bd310b92e0]: set state activated (was activating)
<debug> [1679353062.8885] active-connection[000055bd310b92e0]: update activation type from assume to managed
Note that the "assume" state is probably a misfeature, and should be
dropped in favor of more appropriate flags. Meaning, "assume" state for
the most part is very similar to sys-iface-state "managed", and the
cases where (during activation) we need to be graceful, may be better
covered with other (more specialized) state flags. Regardless, for most
practical purposes, sys-iface-state "assume" should be treated similar
to "managed" state.
When we fully activated, we should be sure to do yet another idle
commit. Note that scheduling an idle-commit is something that must
always be allowed to any users of NML3Cfg. The users have no knowledge
about each other and coordinate by registering their commit type
handles. Issuing an idle "auto" commit must be therefore allowed to
them at any time. If that were not the case, then there would be a bug
to fix. The only reason to maybe not do it, is when we are sure there is
nothing to commit and we would want to avoid unnecessary work.
You can easily reproduce this and see that we don't in fact schedule a
commit after becoming managed. A commit usually only happens later, for
example when we receive an autoconf6 update.
This affects for example setting the MTU. Currently, _commit_mtu() bails
out for nm_device_sys_iface_state_is_external_or_assume() and thus
during activation the MTU will not be set. Later, once we reach
activated state, due to this it still is not set right away. This patch
fixes that, although we should also change _commit_mtu() to not bail out
for sys-iface-state "assume".
NetworkManager.service is "Type=dbus". Systemd takes that as indication
for declaring the service as started when the D-Bus name is acquired.
Currently, we acquire the name very early. The benefit is, that the
service appears to start very fast. However, most the D-Bus API is not
yet populated or ready to use. So if you order your service
`After=NetworkManager.service`, then there is a race that NetworkManager
might not yet be fully usable.
Another benefit was that requesting a D-Bus name is atomic. That means,
we could take that to ensure only one NetworkManager daemon was running.
If we noticed that NetworkManager is already running, we would quit
without doing anything. In practice, systemd already ensures that the
daemon is not running in parallel. This was still useful for catching
misuse when testing manually. This is now no longer done. We will notice
a concurrent NetworkManager only very late, at which point we might have
already broken things (e.g. rewrite wrong state files).
Fix the race with `After=` by acquiring the name much later.
Note that NetworkManager is pretty slow during initialization. This
easily adds several hundreds of milliseconds to the startup.
Since l3cfg rework, NetworkManager tracks IP routes early, not not only
when IP configuration is ready. That means, with `ipv4.method=auto` and
static `ipv4.routes`, then routes are most likely already configured
before the IP address is obtained via DHCP.
That may be desirable in some cases, but for many cases it's probably
wrong.
Instead, only configure the routes (with an ifindex) when we also have
an IP address.
https://bugzilla.redhat.com/show_bug.cgi?id=2102212https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1574
"wifi.seen-bssids" looks like a regular property, but it is not. Unlike
almost all other properties, it does not contain user configuration,
rather it gets filled by the daemon.
The values are thus stored in "/var/lib/NetworkManager/seen-bssids"
file, and the daemon maintains the values separately from the profile.
Only before exporting the profile on D-Bus, the value gets merged (see
NM_SETTINGS_CONNECTION_GET_PRIVATE(self)->>getsettings_cached and
nm_connection_to_dbus_full().
Hence, looking at nm_setting_wireless_get_num_seen_bssids() is not
working. Fix that.
https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/issues/1253
Fixes: 0f3203338c ('wifi: roam aggressively if we on a multi-AP network')
https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1577
Previously, there was "temporary-not-available" mechanism in NML3Cfg,
which aimed to handle IPv6 routes with prefsrc. Theoretically, that
mechanism may have been extended to other use-cases, like IPv4 routes
with prefsrc. What it attempted to handle, is the inability to configure
such routes, unless the respective prefsrc address is configured and
non-tentative. However, the address that we are waiting for, could also
be on another interface, so that mechanism wasn't applicable. This is
now replaced by _routes_watch_ip_addrs(). It seems there isn't anything
useful left for the "temporary-not-available" mechanism and it can go,
except...
We want to log a warning when we are unable to configure a route. Also,
in the future we might want to know when the IP configuration is
degradated due to inability to configure the desired routes (a condition
that we might want to expose to the user, not only via logging; or we
may want to react on that).
However, with prefsrc routes we don't know right away whether the
inability to configure the route right away indicates an actual problem,
or whether that will resolve itself (e.g. after the address passes
DAD/ACD, after we received an DHCP lease or after the address was
configured on another interface). Consequently, to know whether the
current inability to configure such a route is a problem, we need to
know the larger context. nm_platform_ip_route_sync() does not have that
context.
Instead, nm_platform_ip_route_sync() needs only do debug log about
failure to configure routes. It will now also return all the failed
routes to NML3Cfg, which can decide whether that is a problem.
This reworks the previous "temporary-not-available" mechanism to track
the state of the failed routes, to eventually decide whether there is an
actual problem (and log about it).
Another problem this solves is that since commit ('platform: always
reconfigure IP routes even if removed externally'), we will eagerly
re-try to configure the same route over and over. We cannot just spam
the log with warnings about the same failure on every commit. We need to
remember that we already logged about the problem and rate limit
warnings otherwise. This is what the new mechanism also achieves.
Indeed, all this is mostly for the sole benefit of logging better
warnings (and not duplicated).
It was unused anyway.
But also, what would we do with this? We are in the middle of a commit,
if something goes wrong, we cannot just abort but need to continue on
and make the best of it.
Maybe there are very specific error cases that we need to handle, but
those are not covered by a boolean return value. Instead, we might need
to take specific action.
The boolean success variable was meaningless. Drop it.
Routes with pref_src (RTA_PREFSRC) can only be added when the
corresponding IP address is configured (and non-tentative, in case of
IPv6). Additionally, that address may be on any interface, not only on
the one we want to configure the route on. This means, when we first
activate a profile with a route that has a src attrbute, then that src
address might only be configured later. For example, with IPv6, it takes
a while for the address to become non-tentative. Or the address might
come from DHCP, and not be present initially. Or the address might even
be configured on another interface/profile. That means, while we might
be unable to configure the route now, we may become able any time later.
Solve that by subscribing to NMNetns to get notifications whenever such
an address gets added. In that case, schedule an idle commit, which may
then succeed.
The implementation came with two flavors, where watcher could either
specify a tag or no tag. That resulted in different usage patterns and
behavior.
Handles with tag are indexed by a dictionary and de-duplicated. Also the intended
pattern is to delete them with nm_netns_watcher_remove_all(),
Currently, nm_netns_watcher_remove_handle() was not permissible to tag-full handles,
because of the de-duplication and because handles had no ref-counting
implemented (the latter would be fixable, so
nm_netns_watcher_remove_handle() would be made to work).
On the other hand, handles without tag are never de-duplicated. They are
also not indexed, so nm_netns_watcher_remove_all() doesn't work for
them. They could only be removed via nm_netns_watcher_remove_handle().
Currently, the only user of the API will use tag-full handles. Drop the
unused API. This is done as a separate commit, to potentially revert and
restore tag-less handles (after they were already implemented).
NML3Cfg will want to know when an address changes -- on any interface.
We want to support gazillion of interfaces, a naive approach is not
going to scale. Instead, NMNetns already subscribes to all platform
signals, it should dispatch events for address changes.
Add a mechanism how users (NML3Cfg) can register watches, and get called
back when the event happens.
NML3Cfg is stateful, that means it remembers which address/route it
configured earlier. That is important because the API users of NML3Cfg
only say what the want to configure now, and NML3Cfg needs to remove
addresses/routes that it configured earlier but are no longer to be
present. Also, NetworkManager wants to allow the user to add
addresses/routes externally with `ip addr|route add` and NetworkManager
not removing it. This is a common use case for dispatcher scripts, but
in general, we want to allow other components to add addresses/routes.
We try something similar with the removal of routes/addresses managed by
NetworkManager. When NetworkManager adds a route/address, which later
disappears, then we assume that the user intentionally removed the
address/route and take the hint to not re-add it.
However, it doesn't work. It is problematic for two reasons:
- kernel can automatically remove routes. For example, deleting an IPv4
address that is the prefsrc of a route, will cause kernel to delete
that route. Sure, we may be unable to re-configure the route at this
moment, but we shouldn't remember indefinitely that the route is
supposed to be absent. Rather, we should re-add it when possible.
- kernel is a pain with validating consistencies of routes. For example,
when a route has a nexthop gateway, then the gateway must be onlink
(directly reachable), or kernel refuses to add it with "Nexthop has
invalid gateway". Of course, when removing the onlink route kernel is
fine leaving the gateway route behind, which it would otherwise refuse
to add.
Anyway. Such interdependencies for when kernel rejects adding a route
with "Nexthop has invalid gateway" are non-trivial. We try to work
around that by always adding the necessary onlink routes. See
nm_l3_config_data_add_dependent_onlink_routes(). But if the user
externally removed the dependent onlink route, and NetworkManager
remembers to not re-adding it, then the efforts from
nm_l3_config_data_add_dependent_onlink_routes() are ignored. This
causes ripple effects and NetworkManager will also be unable to add the
nexthop route.
Trying to preserve absence of routes that NetworkManager would like to
configure is not tenable. Don't do it anymore. There was anyway no
guarantee that on the next update NetworkManager wouldn't try to re-add
the route in question. For example, if the route came from DHCP, and the
lease temporarily went away and came back, then NetworkManager probably
would have (correctly) forgotten that the user wished that the route be
absent. This did not work reliably and it just causes problems.
For each connection that corresponds to a software device, we create a
"unrealized" device that then becomes realized just before the
connection starts activating. Currently, in certain conditions NM
creates two devices with the same name and type, one realized and one
not; this is not expected and can lead to other issues especially when
a software device is reactivated.
Avoid that by relaxing the check in system_create_virtual_device(): if
a device exists with the same name and type, we don't want to create
another even if the type-specific parameters differ.
When creating a parent dependent device it can have software device as
parent without an ifindex. In that case, it will fail on an ssertion on
parent being missing.
In order to avoid this, we are handling the situation similar to what we
do for VLAN devices. NetworkManager will raise different error and block
the autoconnection instead of asserting.
This solves the assert error for the following commands:
```
nmcli connection add type macvlan ifname mv1 con-name mv1+ macvlan.parent dummy0 mode vepa
nmcli connection add type dummy ifname dummy0 con-name dummy0+ autoconnect no
```
There were a few places where we did already this but there was one
place where we missed it, in nm-manager.c:do_sleep_wake(). Therefore,
the device end in DISCONNECTED state and did not assume the connection.
When the state is DISCONNECTED is being set from a
configuring/configured state we might want to always DECONFIGURE the
interface (ifindex, ip addresses, routes..) except if the
sys-iface-state is REMOVED in that case we would like to remove it.
This is the IPv6 equivalent of arp_ip_target option. It requires
arp_interval set and allow the user to specify up to 16 IPv6 addresses
as targets. By default, the list is empty.
The valid values for this option are 0 (off) and 1 (on). By default the
value is 1 (on). Please notice that this option is only compatible with
802.3AD mode.
The new arp_missed_max option valid range is 0-255 where value 0 means
not set. Please notice that this option is not compatible with 802.3AD,
balance-tlb and balance-alb modes.
When reloading the configuration and there is a global DNS
configuration without domains, NM crashes in global_dns_equal()
because `old->domains` and `new->domains` are both NULL. Fix that.
Thread 1 "NetworkManager" received signal SIGTRAP, Trace/breakpoint trap.
0 g_logv (log_domain=0x7fe81a2110be "GLib", log_level=G_LOG_LEVEL_CRITICAL, format=<optimized out>, args=<optimized out>) at ../glib/gmessages.c:1433
1 g_log (log_domain=<optimized out>, log_level=<optimized out>, format=<optimized out>) at ../glib/gmessages.c:1471
2 g_hash_table_size (hash_table=<optimized out>) at ../glib/ghash.c:2183
3 g_hash_table_size (hash_table=<optimized out>) at ../glib/ghash.c:2181
4 global_dns_equal (new=0xecc540, old=0xe618e0) at ../src/core/nm-config-data.c:1466
5 nm_config_data_diff (old_data=old_data@entry=0xe60020, new_data=new_data@entry=0xe606a0) at ../src/core/nm-config-data.c:1946
6 _set_config_data (self=0xe45810, new_data=0xe606a0, reload_flags=NM_CONFIG_CHANGE_CAUSE_SIGHUP) at ../src/core/nm-config.c:2923
7 nm_config_reload (self=0xe45810, reload_flags=NM_CONFIG_CHANGE_CAUSE_SIGHUP, emit_warnings=emit_warnings@entry=1) at ../src/core/nm-config.c:2875
8 nm_main_config_reload (signal=<optimized out>) at ../src/core/main.c:141
9 sighup_handler (user_data=<optimized out>) at ../src/core/main-utils.c:26
10 g_main_dispatch (context=0xe619e0) at ../glib/gmain.c:3444
11 g_main_context_dispatch (context=0xe619e0) at ../glib/gmain.c:4162
12 g_main_context_iterate.constprop.0 (context=0xe619e0, block=1, dispatch=1, self=<optimized out>) at ../glib/gmain.c:4238
13 g_main_loop_run (loop=0xe5e310) at ../glib/gmain.c:4438
14 main (argc=<optimized out>, argv=<optimized out>) at ../src/core/main.c:515
Fixes: 1f0d1d78d2 ('dns-manager: always apply options from [global-dns]')
In file included from ./src/libnm-std-aux/nm-default-std.h:102,
from ./src/libnm-glib-aux/nm-default-glib.h:11,
from ./src/libnm-glib-aux/nm-default-glib-i18n-prog.h:13,
from ./src/core/nm-default-daemon.h:11,
from src/core/nm-connectivity.c:8:
src/core/nm-connectivity.c: In function ‘nm_connectivity_check_start’:
./src/libnm-std-aux/nm-std-aux.h:180:12: error: ‘reason’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
if (expr) \
^
src/core/nm-connectivity.c:1055:29: note: ‘reason’ was declared here
const char *reason;
^~~~~~
The compiler may dislike this:
CC src/core/platform/tests/libNetworkManagerTest_la-test-common.lo
In function '_ip_address_add',
inlined from 'nmtstp_ip4_address_add' at ../src/core/platform/tests/test-common.c:1892:5:
../src/core/platform/tests/test-common.c:1807:63: error: array subscript 'NMIPAddr {aka const struct _NMIPAddr}[0]' is partly outside array bounds of 'in_addr_t[1]' {aka 'unsigned int[1]'} [-Werror=array-bounds]
1807 | peer_address->addr4,
| ~~~~~~~~~~~~^~~~~~~
../src/core/platform/tests/test-common.c: In function 'nmtstp_ip4_address_add':
../src/core/platform/tests/test-common.c:1886:36: note: object 'peer_address' of size 4
1886 | in_addr_t peer_address,
| ~~~~~~~~~~~~^~~~~~~~~~~~
...
Fixes: 06aafabf14 ('platform/test: add test adding IPv4 addresses that only differ by their peer-address')
The global configuration now can be valid when there are no global
domains defined. The dnsmasq backend must process it and then, if
there is no global default domain, also process the per-connection
settings.
Fixes: 1f0d1d78d2 ('dns-manager: always apply options from [global-dns]')
The link object is no longer valid after the cache gets updated in
nm_platform_link_change().
Fixes: e02fd76d9f ('platform: support changing link properties')
OVS interfaces are special: the kernel link is created only after the
device is attached to the ovs-port, and as with all ports this happens
during stage3(ip-config). That means that the link doesn't exist
during stage2(config); therefore, explicitly update link properties
once the link appears.
We need to set the ethtool and tc properties for assumed devices,
since they go through a normal activation. External devices should not
be touched by NM.
We are passing to the plugin only 'sett_flags', which is the bitmask
of flags to change and works together with 'sett_mask'; however,
plugins interpret that value as the new flags value. The result is
that if there is no change needed (0/0), the existing flags are lost.
Simple reproducer:
ip link add dummy1 type dummy
ip link set dummy1 up
ip addr add dev dummy1 fd01::12/64
sleep 1
# now, a external connection is created by NM
echo "BEFORE:"
cat /run/NetworkManager/system-connections/dummy1.nmconnection | grep "nm-generated\|volatile\|external"
# just add a new address to the interface to make it lose
# the external flag
ip addr add dev dummy1 172.25.42.1/24
sleep 1
echo "AFTER:"
cat /run/NetworkManager/system-connections/dummy1.nmconnection | grep "nm-generated\|volatile\|external"
Output:
BEFORE:
nm-generated=true
volatile=true
external=true
AFTER:
Fixes: d35d3c468a ('settings: rework tracking settings connections and settings plugins')
https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/merge_requests/1548
Autoconnect retries are not being tracked by connection anymore. Now it
is tracked per Device x Connection. In addition, autoconnect might be
blocked for the connection due to no secrets or user requested.
All the properties tracking the retries and blocked time were move to
DevConData and the functions to manipulate them aswell. In NMPolicy the
logic didn't change very much. Instead of looking into the connection
when the device failed activation it looks for DevConData.
