Let's not leak details from src/shared and src/libsystemd into
src/basic, even though you can't actually do anything useful with
just forward declarations from src/shared.
The sd-forward.h header is put in src/libsystemd/sd-common as we
don't have a directory for shared internal headers for libsystemd
yet.
Let's also rename forward.h to basic-forward.h to keep things
self-explanatory.
Commit e2a3d56218 (as part of #36314)
makes sd-boot recognize a 'uki' stanza in a boot loader entry and
uapi-group/specifications@3f2bd8236d adds
it to the BLS, but bootctl and other components parsing said config do
not know about it, leading to the error message
`Unknown line 'uki', ignoring.` when attempting to parse the same entry.
This commit makes it get parsed the same way that that 'efi' is.
We update BOOTX64.EFI explicitly once (because we know that it's the
main entry point of UEFI) and then a second time when we update
everything in $ESP/EFI/*.EFI. That's redundant and pretty ugly/confusing
in the log output. Hence exclude the file we already updated explicitly
from the 2nd run.
Given that we iterate through $ESP/EFI/BOOT/*.EFI these days this is a
pretty common case, hence it's not really noteworthy, hence downgrade
these log messages from LOG_NOTICE to LOG_INFO.
Following the rule that we should always prefer the secure flavour over
the regular one unless there's a clear reason for the regular one, let's
switch this over. Better safe than sorry.
A boot loader entry for a unified kernel image has
BootEntry.kernel : path to the image relative to ESP or XBOOTLDR,
BootEntry.path : path to the image.
Hence, these two effectively point to the same file.
Hence, by unlink command, the image is removed by
```
deref_unlink_file(&known_files, e->kernel, e->root);
```
then later tried again by
```
r = chase_and_unlink(e->path, root, ...);
```
and of course it fails with -ENOENT.
Let's ignore the failure there. We already ignore ENOENT on removal
at various places, especially in deref_unlink_file().
Fixes#38706.
Follow-ups for 8702496bfb.
Since c5de7b14ae
file searching implies a new mount api syscall by default,
to trigger automounts.
But, this is not necessary in most cases, e.g. when chasing
syspath in sd-device (actually this causes regression in umockdev,
see https://github.com/martinpitt/umockdev/issues/271).
Another example is reading unit files, especially .network files,
as automount may trigger mounting network filesystems...
Also, when this is used in NSS plugins, programs that load the
plugins may fail because of spuriously configured seccomp. See #38565.
Let's not trigger automount by default, and do only when explicitly
requested.
This introduces CHASE_TRIGGER_AUTOFS, and use it in
- service manager,
- bootctl and finding ESP/xbootldr,
- sysupdate,
- mountfsd,
- systemd-mount.
There may be several more places we should trigger automount, but let's
do that later.
Follow-up for c5de7b14ae.
Fixes#38565.
Replaces #38569.
Co-authored-by: Luca Boccassi <luca.boccassi@gmail.com>
This reverts commit 490aa05ca1.
As commented https://github.com/systemd/systemd/pull/38569#discussion_r2284978273,
the commit makes autofs check bypassed. Before the commit, when
CHASE_NO_AUTOFS is set, we did not shortcut chasing paths, and refused
any autofs mount points in the path. However, with the commit, the flag
was swapped but even when CHASE_AUTOFS is unset, the autofs check may be
skipped.
To fix the issue, rather than swapping the flag, we should introduce
another flag, say CHASE_TRIGGER_AUTOFS. This revert the commit, and in a
later commit, the new flag will be introduced.
Since c5de7b14ae
file searching implies a new mount api syscall by default,
to trigger automounts.
This is problematic in NSS plugins, as they are dlopen'ed inside
processes by glibc, for two reasons.
First of all, potentially searching on a networked filesystem
automount could lead to nasty surprises, such as the process
responsible for setting up the network filesystem trying to
search on that same filesystem.
More importantly, the new mount api syscall was never part of
the filesystem seccomp filter that we provide by default, and
given mounting/remounting/bind mounting is one of the possible
ways to bypass sandboxing it is very likely not allowed when
custom filters are used in sandboxed processes, if they don't
need to do these operations otherwise.
The filesystem seccomp mask we provide has been updated, however
this only takes effect on the next restart of a service. When
systemd is upgraded via a package upgrade, the new nss plugin is
installed and will be immediately dlopen'ed by glibc when needed,
without waiting for the process to restart, which means the existing
seccomp filter applies, causing the filter to trigger.
Given it's not really possible for any arbitrary program to
predict which NSS modules glibc will load, given programs do not
configure that and instead nsswitch is set up by the sysadmin,
it's impossible to handle at each process level. It's also not
possible to know when it will be triggered, given the plugin
is not linked in each binary tools like need-restart cannot
even pre-emptively restart services that may be affected.
This means in practice, upgrading from systemd << v258 to >= v258
requires a reboot to avoid either subtle or catastrophic system
failures.
By avoiding to trigger automounts in nss-systemd we can avoid
both issues.
userdb drop-ins are searched for in:
/etc/userdb/
/run/userdb/
/run/host/userdb/
/usr/local/lib/userdb/
/usr/lib/userdb/
none of which are supported as automounts anyway.
Note that this happens only when the userdbd service is not running,
as otherwise nss-systemd will go through the varlink IPC, rather than
doing the searches in-process.
So invert CHASE_NO_AUTOFS to CHASE_AUTOFS and set it in the places where
we do want to trigger automounts, like looking for the ESP.
Follow-up for c5de7b14ae
Fixes https://github.com/systemd/systemd/issues/38565
The usual pattern of using colors to distinguish the mount path (/efi/)
and the rest is used. If the file cannot be read for reasons other than
-ENOENT, the error message is highlighted.
I considered a few places where to add this, but this section seems the
most reaosonable. We already print the 'token' there, which is also part of
the configuration.
Boot Loader Entry Locations:
ESP: /efi (/dev/disk/by-partuuid/31659406-5a17-46ec-8195-0dea1667db58)
config: /efi//loader/loader.conf
XBOOTLDR: /boot (/dev/disk/by-partuuid/4f8a8fe9-4b45-4070-9e9b-a681be51c902, $BOOT)
token: fedora
Installing stuff in a chroot should not fail because efivars are
not available. When running in a container touching efivars is
completely disabled, but there are some cases (recovery) where
it is needed to touch them in a chroot, so don't disable them but
avoid failing the run instead.
In the second or later trial, copy_file_with_version_check() -> version_check()
fails with -ESRCH. Let's ignore the failure.
This also adds missing assertions in update_efi_boot_binaries(), and
drop redundant version check in update_efi_boot_binaries(), as version
will be anyway checked later.
Fixes a regression caused by 929f41c652.
Fixes#33392.
It turns out checking sysfs is not 100% reliable to figure out whether
the firmware had TPM2 support enabled or not. For example with EDK2 arm64, the
default upstream build config bundles TPM2 support with SecureBoot support,
so if the latter is disabled, TPM2 is also unavailable. But still, the ACPI
TPM2 table is created just as if it was enabled. So /sys/firmware/acpi/tables/TPM2
exists and looks correct, but there are no measurements, neither the firmware
nor the loader/stub can do them, and /sys/kernel/security/tpm0/binary_bios_measurements
does not exist.
The loader can use the apposite UEFI protocol to check, which is a more
definitive answer. Given userspace can also make use of this information, export
the bitmask with the list of active banks as-is. If it's not 0, then we can be
sure a working TPM2 was available in EFI mode.
Partially fixes https://github.com/systemd/systemd/issues/38071
These source files uses symbols provided by sys/stat.h, e.g. struct stat,
S_IFREG, S_IFBLK, and so on. Let's explicitly include sys/stat.h where
necessary.
Glibc's fcntl.h includes bits/stat.h, which provides these symbols, so
these symbols can be used without explicitly including sys/stat.h. But,
based on the discussion in #37922, we should explicitly include relevant
headers, and should not rely on the indirect inclusion.
Similar to 4f18ff2e29, but for sys/stat.h.
When bootctl is called by an unprivileged user, then previously we got
```
Failed to read "/boot/EFI/systemd": Permission denied
Failed to open '/boot//loader/loader.conf': Permission denied
```
Now, with this patch, we get
```
Failed to read "/boot/EFI/systemd": Permission denied
Failed to open '/boot/loader/loader.conf': Permission denied
```
Continuation of 7851732262 and
5a65d2e535. Absolute paths are always
nicer for the user because it's much easier to figure out where the
file is located. And printing some paths with the prefix and some
without was quite confusing.
Kernels are loaded from two partitions (XBOOTLDR and ESP). In the past this
wasn't always so, but a while ago the docs and the code were updated to load
from both locations. The $BOOT location specifies where to install only. So
rework the status output to print both locations and indicate which one is
$BOOT by appending ", $BOOT".
Also change the section title to "… Locations". We don't show *entries* here,
so the old title could be a bit confusing.
Example:
Boot Loader Entry Locations:
ESP: /efi (/dev/disk/by-partuuid/31659406-5a17-46ec-8195-0dea1667db58)
XBOOTLDR: /boot (/dev/disk/by-partuuid/4f8a8fe9-4b45-4070-9e9b-a681be51c902, $BOOT)
token: fedora
You can configure the sysfail boot entry using the bootctl command:
$ bootctl set-sysfail sysfail.conf
The value will be stored in the `LoaderEntrySysFail` EFI variable.
The `LoaderEntrySysFail` EFI variable would be unset automatically
during next boot by `systemd-boot-clear-sysfail.service` if no
system failure occured, otherwise it would be kept as it is and a system
failure reason will be saved to `LoaderSysFailReason` EFI variable.
Signed-off-by: Igor Opaniuk <igor.opaniuk@foundries.io>
Now that the necessary functions from log.h have been moved to macro.h,
we can stop including log.h in macro.h. This requires modifying source
files all over the tree to include log.h instead.
I think the current behaviour of not doing EFI variables when we are run
in a container makes a ton of sense, but in some cases it's useful to
do EFI var setup even when a set of namespaces is set up for us, for
example to recover a hosed installation from a rescue disk.
While we are at it, let's remove some duplicate checks, and
systematically output information why we skip various operations.
Fixes: #36174#35005
The log line looked like this:
bootctl[1457]: ! Mount point '/efi' which backs the random seed file is world accessible, which is a security hole! !
which doesn't look nice.
Also upgrade the message to error. This is something to fix.
Admittedly, some of our glyphs _are_ special, e.g. "O=" for SPECIAL_GLYPH_TOUCH ;)
But we don't need this in the name. The very long names make some invocations
very wordy, e.g. special_glyph(SPECIAL_GLYPH_SLIGHTLY_UNHAPPY_SMILEY).
Also, I want to add GLYPH_SPACE, which is not special at all.
