If the toplevel_path is empty we end up with doubled leading slash,
which looks weird:
[ 4737.028985] testsuite-74.sh[102]: Inode '//var/lib/machines/mytree.v/mytree_37.0_arm64+2-3' has wrong type, found 'dir'.
[ 4737.028985] testsuite-74.sh[102]: Failed to pick version for '/var/lib/machines/mytree.v': Is a directory
...
[ 4316.957536] testsuite-74.sh[99]: Failed to open '//var/lib/machines/mytree.v/mytree_37.0': No such file or directory
...
Since we might edit the string later on by inserting NULs, which then
leads up to using an invalid dname when opening the potential chosen
directory:
[ 4316.957536] testsuite-74.sh[99]: make_choice: entry: mytree_37.0_arm64+2-3
[ 4316.957536] testsuite-74.sh[99]: make_choice: best_version: 37.0
[ 4316.957536] testsuite-74.sh[99]: make_choice: best_filename: mytree_37.0
[ 4316.957536] testsuite-74.sh[99]: Failed to open '//var/lib/machines/mytree.v/mytree_37.0': No such file or directory
Uncovered by vpick tests from TEST-74-AUX-UTILS when run on aarch64.
In pam_systemd.so and pam_systemd_home.so, we open a bus connection on
session close, which is called after fork. Closing the connection is
harmless, and should not warn about that.
This suppresses the following log message:
===
(sd-pam)[127]: PAM Attempted to close sd-bus after fork, this should not happen.
===
The man page pam_setcred(3) states:
> The credentials should be deleted after the session has been closed
> (with pam_close_session(3)).
Follow-up for 3bb39ea936.
this patch adds the interface name of the interface to be modified
to *details* when verifying dbus calls to the `org.freedesktop.resolve1`
D-Bus interface for all `Set*` and the `Revert` method.
when defining a polkit rule, this allows limiting the access to a specific
interface:
```js
// This rule prevents the user "vpn" to disable DNSoverTLS for any
// other interface than "vpn0". The vpn service should be allowed
// to disable DNSoverTLS on its own as it provides a local DNS
// server with search domains on the interface and this server does
// not support DNSoverTLS.
polkit.addRule(function(action, subject) {
if (action.id == "org.freedesktop.resolve1.set-dns-over-tls" &&
action.lookup("interface") == "vpn0" &&
subject.user == "vpn") {
return polkit.Result.YES;
}
});
```
It's easy to add. Let's do so.
This only covers record lookups, i.e. with the --type= switch.
The higher level lookups are not covered, I opted instead to print a
message there to use --type= instead.
I am a bit reluctant to defining a new JSON format for the high-level
lookups, hence I figured for now a helpful error is good enough, that
points people to the right use.
Fixes: #29755
dnf5 does not download filelists metadata by default anymore as this
consists of a pretty big chunk of the repository metadata. Let's make
sure the filelists metadata doesn't have to be downloaded by dnf5 by
removing any usage of file provides from our package lists.
This returns an FD that can be used to temporarily inhibit the automatic
locking on system suspend behavior of homed. As long as the FD is open,
LockAllHomes() won't lock that home directory on suspend. This allows
desktop environments to implement custom more complicated behavior
20-systemd-ssh-generator.conf expands SSHCONFDIR, which is bogus when we
build with -Dsshconfdir=no. Similarly, avoid expanding SSHDCONFDIR in
20-systemd-userdb.conf when building with -Dsshconfdir=no.
Follow-up 6c7fc5d5f2.
knot v3.2 and later does this by default. knot v3.1 still has the default set to
10, but it also introduced a warning that the default will be changed to 0 in
later versions, so it effectively complains about its own default, which then
fails the config check. Let's just set the value explicitly to zero to avoid
that.
~# knotc --version
knotc (Knot DNS), version 3.1.6
~# grep nsec3-iterations test/knot-data/knot.conf || echo nope
nope
~# knotc -c /build/test/knot-data/knot.conf conf-check
warning: config, policy[auto_rollover_nsec3].nsec3-iterations defaults to 10, since version 3.2 the default becomes 0
Configuration is valid
Follow-up to 0652cf8e7b.
Previously, unit_{start,stop,reload} would call the low-level cgroup
unfreeze function whenever a unit was started, stopped, or reloaded. It
did so with no error checking. This call would ultimately recurse up the
cgroup tree, and unfreeze all the parent cgroups of the unit, unless an
error occurred (in which case I have no idea what would happen...)
After the freeze/thaw rework in a previous commit, this can no longer
work. If we recursively thaw the parent cgroups of the unit, there may
be sibling units marked as PARENT_FROZEN which will no longer actually
have frozen parents. Fixing this is a lot more complicated than simply
disallowing start/stop/reload on a frozen unit
Fixes https://github.com/systemd/systemd/issues/15849
