This is an attempt to improve #8228 a bit, by extending the /run/nologin
a bit, but still keeping it somewhat brief.
On purpose I used the vague wording "unprivileged user" rather than
"non-root user" so that pam_nologin can be updated to disable its
behaviour for members of the "wheel" group one day, and our messages
would still make sense.
See #8228.
This partially reverts 3536f49e8f and
3536f49e8f.
When the user is dynamic, and we are setting up state, cache, or logs dirs,
behaviour is unchanged, we always do a recursive chown. This is necessary
because the user number might change between invocations.
But when setting up a directory for non-dynamic user, or a runtime directory
for a dynamic user, do any ownership or mode changes only when the directory
is initially created. Nothing says that the files under those directories have
to be all recursively owned by our user. This restores behaviour before
3536f49e8f, so modifications to the state of
the runtime directory persist between ExecStartPre's and ExecStart's, and even
longer in case the directory is persistent.
I think it _would_ be a nice property if setting a user would automatically
propagate to ownership of any Runtime/Logs/Cache directories. But this is
incompatible with another nice property, namely preserving changes to those
directories made by an admin, and with allowing change of ownership of files
in those directories by the service (e.g. to allow other users to access them).
Of the two, I think the second property is more important. Also, it's backwards
compatible.
https://bugzilla.redhat.com/show_bug.cgi?id=1508495
There is no need to chmod a directory we just created, so move that step
up into a branch. After that, 'effective' is only used once, so get rid of
it too.
Generally we prefer 'return' from main() over exit() so that automatic
cleanups and such work correct. Let's do that in shutdown.c too, becuase
there's not really any reason not to.
With this we are pretty good in consistently using return from main()
rather than exit() all across the codebase. Yay!
So far, we had two implementations of reboot-with-parameter doing pretty
much the same. Let's unify that in a generic implementation used by
both.
This is particulary nice as it unifies all /run/systemd/reboot-param
handling in a single .c file.
This is primarily preparation for a follow-up commit that adds a common
implementation of the other side of the reboot parameter file, i.e. the
code that reads the file and issues reboot() for it.
This mimics the raw_clone() call we have in place already and
establishes a new syscall wrapper raw_reboot() that wraps the kernel's
reboot() system call in a bit more low-level fashion that glibc's
reboot() wrapper. The main difference is that the extra "arg" argument
is supported.
Ultimately this just replaces the syscall wrapper implementation we
currently have at three places in our codebase by a single one.
With this change this means that all our syscall() invocations are
neatly separated out in static inline system call wrappers in our header
functions.
In a number of occasions we use FORK_CLOSE_ALL_FDS when forking off a
child, since we don't want to pass fds to the processes spawned (either
because we later want to execve() some other process there, or because
our child might hang around for longer than expected, in which case it
shouldn't keep our fd pinned). This also closes any logging fds, and
thus means logging is turned off in the child. If we want to do proper
logging, explicitly reopen the logs hence in the child at the right
time.
This is particularly crucial in the umount/remount children we fork off
the shutdown binary, as otherwise the children can't log, which is
why #8155 is harder to debug than necessary: the log messages we
generate about failing mount() system calls aren't actually visible on
screen, as they done in the child processes where the log fds are
closed.
We used to set this, but this was dropped when shutdown got taught to
get the target passed in from the regular PID 1. Let's readd this to
make things more explanatory, and cover all grounds, since after all the
target passed is in theory an optional part of the protocol between the
regular PID 1 and the shutdown PID 1.
Previously, we'd try to open kmsg on failure of the journal/syslog even
if no automatic fallback to kmsg was requested — and we wouldn't even
use the open connection afterwards...
In meson.build we check that functions are available using:
meson.get_compiler('c').has_function('foo')
which checks the following:
- if __stub_foo or __stub___foo are defined, return false
- if foo is declared (a pointer to the function can be taken), return true
- otherwise check for __builtin_memfd_create
_stub is documented by glibc as
It defines a symbol '__stub_FUNCTION' for each function
in the C library which is a stub, meaning it will fail
every time called, usually setting errno to ENOSYS.
So if __stub is defined, we know we don't want to use the glibc version, but
this doesn't tell us if the name itself is defined or not. If it _is_ defined,
and we define our replacement as an inline static function, we get an error:
In file included from ../src/basic/missing.h:1358:0,
from ../src/basic/util.h:47,
from ../src/basic/calendarspec.h:29,
from ../src/basic/calendarspec.c:34:
../src/basic/missing_syscall.h:65:19: error: static declaration of 'memfd_create' follows non-static declaration
static inline int memfd_create(const char *name, unsigned int flags) {
^~~~~~~~~~~~
.../usr/include/bits/mman-shared.h:46:5: note: previous declaration of 'memfd_create' was here
int memfd_create (const char *__name, unsigned int __flags) __THROW;
^~~~~~~~~~~~
To avoid this problem, call our inline functions different than glibc,
and use a #define to map the official name to our replacement.
Fixes#8099.
v2:
- use "missing_" as the prefix instead of "_"
v3:
- rebase and update for statx()
Unfortunately "statx" is also present in "struct statx", so the define
causes issues. Work around this by using a typedef.
I checked that systemd compiles with current glibc
(glibc-devel-2.26-24.fc27.x86_64) if HAVE_MEMFD_CREATE, HAVE_GETTID,
HAVE_PIVOT_ROOT, HAVE_SETNS, HAVE_RENAMEAT2, HAVE_KCMP, HAVE_KEYCTL,
HAVE_COPY_FILE_RANGE, HAVE_BPF, HAVE_STATX are forced to 0.
Setting HAVE_NAME_TO_HANDLE_AT to 0 causes an issue, but it's not because of
the define, but because of struct file_handle.
When synthetisation is turned off, there's just too many ways those tests can
go wrong. We are not interested in verifying that the db on disk is correct,
let's just skip all checks.
In the first version of this patch, I recorded if we detected a mismatch during
configuration and only skipped tests in that case, but actually it is possible
to change the host configuration between our configuration phase and running
of the tests. It's just more robust to skip always. (This is particularly true
if tests are installed.)
This introduces a new setting TemporaryFileSystem=. This is useful
to hide files not relevant to the processes invoked by unit, while
necessary files or directories can be still accessed by combining
with Bind{,ReadOnly}Paths=.
When running journalctl --user-unit=foo as an unprivileged user we could get
the usual hint:
Hint: You are currently not seeing messages from the system and other users.
Users in groups 'adm', 'systemd-journal', 'wheel' can see all messages.
...
But with --user-unit our filter is:
(((_UID=0 OR _UID=1000) AND OBJECT_SYSTEMD_USER_UNIT=foo.service) OR
((_UID=0 OR _UID=1000) AND COREDUMP_USER_UNIT=foo.service) OR
(_UID=1000 AND USER_UNIT=foo.service) OR
(_UID=1000 AND _SYSTEMD_USER_UNIT=foo.service))
so we would never see messages from other users.
We could still see messages from the system. In fact, on my machine the
only messages with OBJECT_SYSTEMD_USER_UNIT= are from the system:
journalctl $(journalctl -F OBJECT_SYSTEMD_USER_UNIT|sed 's/.*/OBJECT_SYSTEMD_USER_UNIT=\0/')
Thus, a more correct hint is that we cannot see messages from the system.
Make it so.
Fixes#7887.
This makes it easier to see what is going on. Crashes may happen in a
nested test_{uid,gid}_to_name_one() function, and the default backtrace
doesn't show the actual string being tested.
The Linux kernel exposes the birth time now for files through statx()
hence make use of it where available. We keep the xattr logic in place
for this however, since only a subset of file systems on Linux currently
expose the birth time. NFS and tmpfs for example do not support it. OTOH
there are other file systems that do support the birth time but might
not support xattrs (smb…), hence make the best of the two, in particular
in order to deal with journal files copied between file system types and
to maintain compatibility with older file systems that are updated to
newer version of the file system.
Let's make use this at various places we call fsync(), to make things
fully reliable, as the kernel devs suggest to first fsync() files and
then fsync() the directories they are located in.
Let's add a common implementation for regular file checks, that are
careful to return the right error code (EISDIR/EISLNK/EBADFD) when we
are encountering a wrong file node.
Let's make sure we aren't confused if a journal file is replaced by a
different one (for example due to rotation) if we are in a q overflow:
let's compare the inode/device information, and if it changed replace
any open file object as needed.
Fixes: #8198
Let's be more careful with the naming, and indicate that the function
is about *named* journal files, and will validate the name as needed.
(in opposition to add_any_file() which doesn't care about names)
Lennart Poettering
Zbigniew Jędrzejewski-Szmek
Evgeny Vereshchagin
Alan Jenkins
Yu Watanabe