When at least one of the name, MAC address, udev properties, and so on
for an interface is updated, try to find a matching .network file, and
reconfigure if a new .network file is assigned.
Fixes#24975.
No functional changes, just refactoring and preparation for later
commits.
Note, `link->dev` should always exist when link state is initialized or
later.
Previously, ata_id might not be able to retrieve attributes correctly,
and properties from usb_id were used as a fallback. See issue #24921
and PR #24923. To keep backward compatibility, still we need to create
symlinks based on USB serial.
Fixes#25179.
When installing systemd 252 in debian:
Setting up systemd (252~rc3-2) ...
Installing new version of config file /etc/systemd/logind.conf ...
Installing new version of config file /etc/systemd/system.conf ...
Installing new version of config file /etc/systemd/user.conf ...
/usr/lib/tmpfiles.d/provision.conf:13: Credential 'login.motd' not specified, skipping line.
/usr/lib/tmpfiles.d/provision.conf:14: Credential 'login.issue' not specified, skipping line.
/usr/lib/tmpfiles.d/provision.conf:17: Credential 'network.hosts' not specified, skipping line.
/usr/lib/tmpfiles.d/provision.conf:22: Credential 'ssh.authorized_keys.root' not specified, skipping line.
/usr/lib/tmpfiles.d/systemd.conf:43: Copy source path '/.extra/tpm2-pcr-signature.json' does not exist, skipping line.
/usr/lib/tmpfiles.d/systemd.conf:44: Copy source path '/.extra/tpm2-pcr-public-key.pem' does not exist, skipping line.
Downgrade to debug level
Even if different preference is specified, the kernel merges multiple
routes with the same preference. This is problematic when a network has
multiple routers.
Fixes#25138.
We have already allowed to reconfigure failed interface manually, but
not allowed to do automatically, e.g. on carrier gain.
This makes that failed interfaces also reconfigured automatically.
Note, the condition is inversed to shorten the condition.
Follow-up for 778e3da95e.
These settings are saved only when a .network file is assigned to the
interface. Let's silence noisy logs for unmanaged interfaces.
Note, previously `use_mac` set with `test_mode`. As `dev`, which is set with
`client->dev`, is not set when running test or fuzzer. Hence, the condition
```
if (udev_available() && !use_mac)
```
is effectively equivalent to
```
if (dev)
```
So, this commit mostly does not change behavior. Except for the following
corner case.
The sd_device object assigned from networkd (that is, Link.dev) never
has ID_RENAMING udev property, as sd_device objects which has the property
are filtered out at `link_check_initialized()` or `manager_udev_process_link()`
in networkd-link.c.
However, sd_device object created in `dhcp_identifier_set_iaid()` in the
previous code may have it. Such situation may (at least, theoretically)
happen when the network interface is renamed after initialized, e.g. by
creating the following spurious .link file:
```
[Match]
OriginalName=eno1
[Link]
Name=lan
```
and then trigger uevent for the network interface while systemd-networkd
calling `dhcp_identifier_set_iaid()`.
Otherwise, if a device is removed while switching root, then the
corresponding .device unit will never go to inactive state.
This replaces the code dropped by cf1ac0cfe4.
Fixes#25106.
See parent commit for explanation.
git blame shows Martin Pitt is the author of the code.
(I was considering whether we should instead drop this file, but it's still
useful for *distributions*. Eveen if we discourage people from using sysv
scripts, distributions will have to deal with them for a while yet.)
As the properties loaded from the database are stored to UdevEvent.dev_db_clone,
instead of UdevEvent.dev.
This also makes the removal done unconditionally, for safety.
The CASE param would normally provide the operation for the compiler to
do in this macro magic. But in this case CASE_F_1 was hardcoding the
operation, making the parameter moot.
This just removes the somewhat pointless parameter instead of fixing
the one case. These macros are used for IN_SET case labels only and
not named generically anyways.
This will now catch mistakes like this:
struct s {
int i:2;
} s = { 1 };
assert_se(IN_SET(s.i, ULLONG_MAX));
> warning: implicit conversion from 'unsigned long long' to
> 'typeof (+s.i)' (aka 'int') changes value from 18446744073709551615
> to -1 [-Wconstant-conversion]
Luca Boccassi
Lennart Poettering
Yu Watanabe
Jan Janssen
David Seifert
Zbigniew Jędrzejewski-Szmek
xiaoyang
Jacek Migacz