In https://github.com/systemd/systemd/issues/28515, multiple people report that
services that have PrivateTmp=yes and DefaultDependencies=no fail to create the
temporary directories under /tmp, when /tmp is e.g. a bind mount or some other
kind of mount that takes more time.
Before PrivateTmp=disconnected was added, we didn't have a nice solution:
DefaultDependencies=no is used to start services very early, so we wouldn't
want to add a dependency on /tmp automatically. With PrivateTmp=disconnected we
have a fairly nice solution. Let's "upgrade" to this mode automatically.
Strictly speaking, it is a small compat break, but in practice it's unlikely to
matter for early-boot services whether their /tmp is private or disconnected.
The dependency on /tmp that is checked is After. I think this is enough,
since any tmp.mount would be pulled in by local-fs.target and the rest of
the transaction anyway, so we don't need to check more than After.
The asserts are relaxed, because now the two settings can now diverge
in either way.
Resolves https://github.com/systemd/systemd/issues/28515.
[yhndnzj: fix unit_add_exec_dependencies() to handle the new
combination; add a comment in exec_needs_sys_admin()]
The only reason why ExecSharedRuntime.(var_)tmp_dir might
be suppressed is when (/var)/tmp/ is listed in InaccessiblePaths=,
in which case mount namespace is needed anyways. Moreover,
the current logic imposes a chicken-and-egg problem:
at various places in pid1 we'd need to check whether
the unit runs in a mountns, but the ExecSharedRuntime
may not be allocated for it just yet.
* Make sure we're not passing NULL to rmdir()
* Remove the inner "tmp" subdir as well, so that callers
can handle this transparently (e.g. exec_shared_runtime_make()
is not aware of the nested rmdir() requirements)
While at it, remove unneeded PROTECT_ERRNO.
These syscalls are part of a newer kernel API to replace interaction
with /proc/self/attr, with the goal of allowing LSM stacking. These are
being used now by e.g. libapparmor, so should be more easily available
to services using seccomp filtering.
test_sd_device_enumerator_filter_subsystem hangs forever in OBS builds,
as it waits for the udev queue to be drained. But this is not a booted
system, with systemd and udev, so nothing will do that.
Skip the test in that case.
The callback might drop the last ref to the slot object, and we still
want to access it. Hence do what we usually do in these cases: keep an
extra reference while processing the callback.
This reverts commit 351b4dd123.
I'm pretty sure we should not create more of these multiplexer
interfaces, which in this specific also leads to a fair bit of
confusion: start/stop/reload-or-restart --marked all enqueue
all job types.
Instead, a dedicated verb will be introduced in later commits.
With v259, resolved can be used to replace Avahi for mDNS service
discovery and announcement. One key difference in the API is that Avahi
by default browses on all available interfaces by passing a special
`AVAHI_IF_UNSPEC` (-1) interface index value. With this change, passing
ifindex value `0` can be used to achieve the same , so this will work:
```
# Browse on all available interfaces
varlinkctl call --more unix:/run/systemd/resolve/io.systemd.Resolve \
io.systemd.Resolve.BrowseServices \
'{"domain":"local","type":"_display._tcp","ifindex":0,"flags":0}'
```
Not having to iterate through all interfaces on the client side then
also makes it more straight forward to build a resolve-avahi-compat
library for painlessly replacing Avahi with systemd-resolved.
A PoC compat lib requiring this change is available at
https://github.com/LorbusChris/resolve-avahi-compat
Currently, there's no way for a desktop environment to detect when a
systemd inhibitor has been added or removed. This means that a desktop
environment that wishes to show current inhibitors to the user will be
either forced to poll systemd or just use stale data.
Now, the desktop can listen for changed events on NCurrentInhibitors and
call GetInhibitors() in response to that.
Desktops cannot currently tell the difference between a power action
being set to challenge permanently (i.e. because the sysadmin wanted to
restrict the given user) and a power action being set to challenge
because it's temporarily inhibited.
Thus, a desktop might take an action that is valid in the first case but
not the second case. For instance: GNOME hides all of its automatic
suspend settings from the UI whenever a sleep inhibitor is active.
This now returns a new state: "inhibited". It communicates to the
desktop that the action is available normally, but at the moment the
desktop isn't allowed to perform the action due to an inhibitor.
Related: https://github.com/systemd/systemd/issues/37311
Now that it’s been replaced by `Acquire()` and `Install()`, there’s no
need to expose it as a separate method. Wrappers around `sysupdated` can
still expose an ‘update’ convenience verb by calling those two methods.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Helps: https://github.com/systemd/systemd/issues/34814
Rather than calling `Update()`. This should allow us to eventually drop
`Update()`.
This should just be an internal change, and should not result in any
user-visible changes to the behaviour of `updatectl update`.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Helps: https://github.com/systemd/systemd/issues/34814
These mirror the `sysupdate acquire` and `sysupdate update --offline`
verbs, which are themselves a split of `sysupdate update` into
downloading and installing stages.
The existing `sysupdated` `Update()` method is kept for now, for
convenience. It might be dropped in future.
Signed-off-by: Philip Withnall <pwithnall@gnome.org>
Helps: https://github.com/systemd/systemd/issues/34814
xfsprogs 6.17.0 added support for populating xfs filesystems from
directories. As this supports extended attributes unlike our current
hack with protofiles. Let's make use of the new feature in mkfs-util.c
As there's no clean way to do feature detection on the mkfs.xfs binary,
we drop support for the old hack with protofiles that we had before.
Useful for packaging scripts, when units are removed.
When multiple markers are assigned without +/-, the last one wins.
When using +/-, the job merging logic is followed to the extent possible.
If we include a verity signature but the corresponding key is not in
the kernel keyring or userspace, try to get polkit auth similar to how
we do the same if the image doesn't pass the policy.
Zbigniew Jędrzejewski-Szmek
Mike Yuan
Nick Rosbrook
Luca Boccassi
Daan De Meyer
Antonio Alvarez Feijoo
Lennart Poettering
Yu Watanabe
Adrian Vovk
Zbigniew Jędrzejewski-Szmek
Philip Withnall
DaanDeMeyer