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 https://github.com/systemd/systemd/commit/c5de7b14ae2e08d267d8d75bc88934ac6aa7dcd6
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 https://github.com/systemd/systemd/commit/c5de7b14ae2e08d267d8d75bc88934ac6aa7dcd6
Fixes https://github.com/systemd/systemd/issues/38565
Fixes the following error message (the last line):
```
[FAILED] Failed to start TEST-60-MOUNT-RATELIMIT.service.
Sending SIGTERM to remaining processes...
Sending SIGKILL to remaining processes...
All filesystems, swaps, loop devices, MD devices and DM devices detached.
Exiting container.
Failed to read from pty input fd: Bad file descriptor
```
Follow-up for b823809bca and
cf89e48028.
Then, also make nss modules parse $SYSTEMD_ASSERT_RETURN_IS_CRITICAL
environment variable.
This also moves nss-util.c and nss-util.h from src/basic/ to src/shared/,
as they are not used by libsystemd.
gethostname_full() is used in nss-myhostname, and hence random
application may indirectly call it. When an application with a too strict
seccomp filter loads the nss module, the application may trigger the
assertion.
Partially fixes#38582.
For safety, though typically Esys_Free() is just a simple wrapper of
free(), but let's do unconditionally. See the comment in the code.
While at it, this makes it store the result into struct iovec.
../src/shared/seccomp-util.c: In function ‘seccomp_restrict_sxid’:
../src/shared/seccomp-util.c:2228:25: error: ‘__NR_fchmodat2’ undeclared (first use in this function); did you mean ‘fchmodat2’?
2228 | __NR_fchmodat2,
| ^~~~~~~~~~~~~~
| fchmodat2
The override/sys/syscalls.h needs to be included before the seccomp
headers, otherwise the internal seccomp preprocessor machinery will
not see the local definitions, so the local ifdef will be true but
the seccomp own definitions will be empty
It is not easy to understand what happens to a journal file
even with debug logs enabled. Add more dbg messages around operations
started by users to make it possible to follow the flow of operations.
drained() checks PTYForward.master_readable flag, but it may be
tentatively unset due to a tentative error like EAGAIN in the previous
IO event. Let's try to call shovel() one more time, which re-read the
master and call drained() at the end. Otherwise, we may lost some data.
When PTYForward.done is set, the PTYForward.master is already
disconnected. Let's not try to read the already closed file descriptor.
Also, if we previously received vhangup, then it is not necessary to
re-read the device to check vhangup, as we already know.
This also make the check slightly delayed, and use a defer event source,
to make the function can be called safely in another event source.
Currently, pty_forward_set_ignore_vhangup() is only used for disabling
the flag. To make the function also disable PTY_FORWARD_IGNORE_INITIAL_VHANGUP
flag, this renames it to pty_forward_honor_vhangup().
Also, for consistency, pty_forward_get_ignore_vhangup() and
ignore_vhangup() are replaced with pty_forward_vhangup_honored().
Previously, do_shovel() sometimes call pty_forward_done(), and
its caller shovel() also call pty_forward_done(). Let's move all
pty_forward_done() calls to shovel(), and do_shovel() not call it.
No functional change, just refactoring.
We had errno_to_name() which works for "known" errnos, and returns NULL for
unknown ones, and then ERRNO_NAME which always returns an answer, possibly just
a number as a string, but requires a helper buffer.
It is possible for the kernel to add a new errno. We recently learned that some
architectures define custom errno names. Or for some function to unexpectedly
return a bogus errno value. In almost all cases it's better to print that value
rather than "n/a" or "(null)". So let's use ERRNO_NAME is most error handling
code. Noteably, our code wasn't very good in handling the potential NULL, so
in various places we could print "(null)". Since this is supposed to be used
most of the time, let's shorten the names to ERRNO_NAME/errno_name.
There are a few places where we don't want to use the fallback path, in
particular for D-Bus error names or when saving the error name. Let's rename
errno_to_name() to errno_name_no_fallback() to make the distinction clearer.
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
Fixes the following error when running with sanitizers:
```
TEST-87-AUX-UTILS-VM.sh[670]: + bootctl install --make-entry-directory=yes
TEST-87-AUX-UTILS-VM.sh[695]: Copied "/usr/lib/systemd/boot/efi/systemd-bootx64.efi.signed" to "/boot/EFI/systemd/systemd-bootx64.efi".
TEST-87-AUX-UTILS-VM.sh[695]: Copied "/usr/lib/systemd/boot/efi/systemd-bootx64.efi.signed" to "/boot/EFI/BOOT/BOOTX64.EFI".
TEST-87-AUX-UTILS-VM.sh[695]: Created "/boot/fedora".
TEST-87-AUX-UTILS-VM.sh[695]: Random seed file /boot/loader/random-seed successfully refreshed (32 bytes).
TEST-87-AUX-UTILS-VM.sh[695]: ../src/shared/efi-api.c:618:38: runtime error: left shift of 243 by 24 places cannot be represented in type 'int'
```
cpu_set_add_range() is used in parse_cpu_set(), hence already tested.
But it is better to test these functions explicitly.
For CID#1611787 and CID#1611788, that should be false-positive.
We went back and forth between 'prog.sh', files('prog.sh'), and
find_program('prog.sh'). We want to use files() or find_program() so that we
get a good error message if the file is missing. Behaviour of meson changed
over time, and in the past not all forms could be used in all places. For
example 0f4c4f3824 added find_program() in many
places to avoid repeated messages. But it seems that all recent meson versions
work fine with files().
find_program prints silly messages:
Program tools/make-man-index.py found: YES
(/home/zbyszek/src/systemd/tools/make-man-index.py)
Program tools/meson-render-jinja2.py found: YES
(/home/zbyszek/src/systemd/tools/meson-render-jinja2.py)
...
We know that those files will be found, they are part of the git checkout.
With files() this is gone and the meson output is easier to read.
All those files are standalone programs that can be executed directly.
Some .py files were marked executable, others weren't, probably accidentally.
Mark them all as executable in preparation for subsequent changes.
Before 7d247d3cb8, we needed the file name
twice. But now we only need it once, so no need to overcomplicate things
by defining a variable with a single use.
When a variable is used, it's also easier to make a mistake and e.g.
accidentally reuse the variable later.
We require at least crypt_r() exists, and it is provided since glibc-2.0
(and dropped in glibc-2.39) or by libxcrypt, and the function is
provided in crypt.h regardless it is provided by glibc or libxcrypt.
Hence, we cannot fallback to unistd.h.
This makes the condition about crypt.h more strict, and stop compilation
earlier when crypt.h does not exist.
Allow for mDNS service/domain/types browsing.
A client can connect to the backend via varlink and receive updates as
the requested service becomes available.
The interval between the first two queries MUST be at least one second,
the intervals between successive queries MUST increase by at least a
factor of two.
When the interval between queries reaches or exceeds 60 minutes, a
querier MAY cap the interval to a maximum of 60 minutes, and perform
subsequent queries at a steady-state rate of one query per hour.
Cache maintenance performed by issuing queries at 80, 85, 90, 95% of the
TTL, and updating/removing the records depending on replies.
TODO:
Improve the DNS transaction logic when multiple clients subscribe to the
same service, ensuring that continuous queries are optimized.
<!-- devel-freezer =
{"comment-id":"1625154850","freezing-tag":"v256-rc2"} -->
Allow for mDNS service/domain/types browsing.
A client can connect to the backend via varlink and receive updates as the
requested service becomes available.
The interval between the first two queries MUST be at least one second,
the intervals between successive queries MUST increase by at least a factor of two.
When the interval between queries reaches or exceeds 60 minutes, a querier MAY cap
the interval to a maximum of 60 minutes, and perform subsequent queries at a
steady-state rate of one query per hour.
Delete expired cache entries one second after goodbye packet received
as per RFC6762 Section 10.1
Cache maintenance:
The querier should plan to issue a query at 80% of the record lifetime, and
then if no answer is received, at 85%, 90%, and 95%.
If an answer is received, then the remaining TTL is reset to the value given
in the answer, and this process repeats for as long as the Multicast DNS querier
has an ongoing interest in the record.
If no answer is received after four queries, the record is deleted when it
reaches 100% of its lifetime.
TODO:
Improve the DNS transaction logic when multiple clients subscribe to
the same service, ensuring that continuous queries are optimized.
Co-authored-by: Vishwanath Chandapur <vishwanath.chandapur@philips.com>
Yu WatanabeandLuca Boccassi
Lennart Poettering
Christian Hesse
Luca Boccassi
Zbigniew Jędrzejewski-Szmek
Erin Shepherd
Vishal Chillara SrinivasandVishwanath Chandapur