Currently mount options are handled in such a way that the first
definition for a given partition wins, and documented as such.
Change them so that they behave like other options, and the
last specified wins.
Applies to RootImageOptions=, MountImages= and ExtensionImages=.
Switch from a linked list to an array indexed by the partition
specifier to store them.
Transparent Hugepages (THP) is a Linux kernel feature that manages
memory using larger pages (2MB on x86, compared to the default 4KB).
The main goal is to improve memory management efficiency and system
performance, especially for memory-intensive applications.
However, it can cause drawbacks in some scenarios, such as memory
regression and latency spikes. THP policy is governed for the entire
system via /sys/kernel/mm/transparent_hugepage/enabled.
However, it can be overridden for individual workloads via prctl(2)
call.
MemoryTHP= is used to disable THPs at exec-invoke to stop
providing THPs for workloads where the drawbacks outweigh the advantages.
When set to "disable", MemoryTHP= disables THPs completely for the
process, irrespecitive of global THP controls.
When set to "madvise", MemoryTHP= disables THPs for the process except
when specifically madvised by the process with MADV_HUGEPAGE or MADV_COLLAPSE.
Before doing anything with the root directory file descriptor, let's
make sure we clone it first so that the caller can't mess with mount fd
attributes via mount_setattr() anymore.
We clone during parsing instead of in executor so that the caller can't
mess with the mount fd between invocations.
sys/prctl.h anyway includes linux/prctl.h and actually these .c files
includes sys/prctl.h. Hence, it is not necessary to explicitly include
linux/prctl.h.
This allows a service to reuse the user namespace created for an
existing service, similarly to NetworkNamespacePath=. The configuration
is the initial user namespace (e.g. ID mapping) is preserved.
Let's extract common capability parsing code into a generic function
parse_capability_set() with a comprehensive set of unit tests.
We also replace usages of UINT64_MAX with CAP_MASK_UNSET where
applicable and replace the default value of CapabilityBoundingSet
with CAP_MASK_ALL which more clearly identifies that it is initialized
to all capabilities.
AI (copilot) was used to extract the generic function and write the
unit tests, with manual review and fixing afterwards to make sure
everything was correct.
RootDirectory= but via a open_tree() file descriptor. This allows
setting up the execution environment for a service by the client in
a mount namespace and then starting a transient unit in that execution
environment using the new property.
We also add --root-directory= and --same-root-dir= to systemd-run to
have it run services within the given root directory. As systemd-run
might be invoked from a different mount namespace than what systemd is
running in, systemd-run opens the given path with open_tree() and then
sends it to systemd using the new RootDirectoryFileDescriptor= property.
If socket is used as stdio, we'd currently imply EXEC_PASS_FDS
and dump the whole set of fds to the control processes. This is
pretty much unexpected and unnecessary though, instead let's
pass only the socket fds.
Yes, this is a compat break, but a relatively minor one I'd
argue. And we can always revisit things if users do complain.
The distinction between fdstore and extra fds is only meaningful
to struct Service. As far as executor is concerned they're just
some fds to pass to the service. Let's just merge it hence,
for the sake of simplicity.
Add a new option PrivateBPF= to mount a new instance of bpffs within a
namespace.
PrivateBPF= can be set to "no" to use the host bpffs in readonly mode
and "yes" to do a new mount.
The mount is done with the new fsopen()/fsmount() API because in future
we'll hook some commands between the two calls.
Currently, callers of safe_glob() set an empty glob_t or glob_t with
opendir func, and all other components are always zero.
So, let's introduce safe_glob_full() which optionally takes opendir
function, rather than glob_t, and returns result strv, rather than
storing results in glob_t.
Also, introduce safe_glob() which is a trivial wrapper of
safe_glob_full() without opendir func.
No functional change, just refactoring.
Currently, if we use a cgroup namespace together with DelegateSubgroup=,
the subgroup becomes the root of the cgroup namespace because we move the
service process to the subgroup before we unshare the cgroup namespace, and
the current cgroup becomes the root of the cgroup namespace when we unshare
the cgroup namespace.
Let's fix the problem by not moving the service process to the subgroup until
we've unshared the cgroup namespace. Note that this doesn't break the primary use
case of CLONE_INTO_CGROUP since we still use it to immediately clone into the service
main cgroup, just not anymore into the subgroup, but this shouldn't matter in practice.
Additionally, we need special handling for control processes, as those *do*
need to get spawned into the subcgroup immediately if delegation is configured to
avoid violating the cgroupsv2 "no inner processes" rule.
Effectively, this leaves us with the following logic:
- In exec_spawn(), spawn into subgroup if we're spawning a control process
that needs to be spawned into a subgroup immediately. Otherwise, spawn into
main service cgroup.
- In exec_invoke(), move into subgroup early if we don't need a cgroup namespace.
Otherwise, move into subgroup after we've unshared the cgroup namespace.
`ExtensionImages=` and `ExtensionDirectories=` now let you specify
vpick-named extensions; however, since they just get set up once when
the service is started, you can't see newer versions without restarting
the service entirely. Here, also reload confext extensions when you
reload a service. This allows you to deploy a new version of some
configuration and have it picked up at reload time without interruption
to your workload.
Right now, we would only reload confext extensions and leave the sysext
ones behind, since it didn't seem prudent to swap out what is likely
program code at reload. This is made possible by only going for the
`SYSTEMD_CONFEXT_HIERARCHIES` overlays (which only contains `/etc`).
Implementation wise, this uses the new kernel API and two collaborating
child processes under the host & child namespaces in order to gather the
right FDs needed:
- (1) In child, set up the extension images and directories in a slave
mountns, and obtain their FDs.
- (2) Fork into a grandchild under target process namespace, and do a
"fake" unmount to obtain the FD of the underlying target folder
say /etc).
- (3) In the child again, set up new overlay under host NS rights.
We do not want to do I/O heavy jobs inline in PID1 blocking the state
machine, so add separate async states to handle this case.
Co-authored-by: Luca Boccassi <luca.boccassi@gmail.com>
If DefaultDependencies=no, /var/ may not be mounted yet when the service
is being started. Previously, In such case, if the service has
PrivateTmp=disconnected, the service manager created /var/tmp/ on the
root filesystem and mounted the disconnected private tmpfs there. That
poluted the root filesystem and disturbed gpt-auto-generator on next
boot, as /var/ would not be empty anymore. See issue #37258.
This changes PrivateTmp=disconnected as the following:
- If DefaultDependencies=no and RootDirectory=/RootImage= are not set,
then a private tmpfs is mounted _only_ on /tmp/, and set $TMPDIR=/tmp
environment variable to suggest the service to use /tmp/.
- If DefaultDependencies=yes and RootDirectory=/RootImage= are not set,
then implies RequiresMountsFor=/var/, though that is typically
redundant, but anyway. Hence, we can safely mount /var/tmp/.
- Otherwise, i.e. when one of RootDirectory=/RootImage= is set, behaves
as the same as the previous, as the private root filesystem for the
service is explicitly prepared by the service manager, and we can
safely mount a private tmpfs on /var/tmp/ without any extra
dependencies.
Fixes#37258.
Co-authored-by: Mike Yuan <me@yhndnzj.com>
Currently there are various circular dependencies between headers
in core/. Let's get rid of these by making judicious use of forward
declarations and moving includes into implementation files instead of
having them in header files.
Getting rid of circular header includes simplifies the code and makes
various clang based tooling such as iwyu work much better on our code.
The most important change is getting rid of the manager.h include in
unit.h which is possible thanks to the previous commits. We also move
the OOMPolicy and StatusType enums to unit.h to remove the need for
other unit headers to include manager.h to get access to these enums.
These variables are assigned with set_put_strv(), hence
string_hash_ops_free is used.
This also drops unnecessary call of set_ensure_allocated() with wrong
hash ops.
This introduce LOG_ITEM() macro that checks arbitrary formats in
log_struct().
Then, drop _printf_ attribute from log_struct_internal(), as it does not
help so much, and compiler checked only the first format string.
Hopefully, this silences false-positive warnings by Coverity.
To resolve conflict with sys/mount.h and linux/mount.h or linux/fs.h.
The conflict between sys/mount.h and linux/mount.h is resolved in
glibc-2.37 (774058d72942249f71d74e7f2b639f77184160a6), but our baseline
is still glibc-2.31. Also, even with the version or newer, still
sys/mount.h conflicts with linux/fs.h, which is included by
linux/btrfs.h.
This introduces our own implementation of sys/mount.h, that can be
simultaneously included with linux/mount.h and linux/fs.h. This also
imports linux/fs.h, linux/mount.h, and several other dependent headers.
The introduced sys/mount.h header itself may not be enough simple, but
by using the header, we can drop most of workarounds in other source files.
Mike Yuan
Yu Watanabe
Luca Boccassi
Usama Arif
Daan De Meyer
Lennart Poettering
Quentin Deslandes
Ivan Kruglov
Matteo Croce
Andres Beltran
Jelle van der Waa
maia x.andLuca Boccassi