| Age | Commit message (Collapse) | Author |
|---|
|
core/failure-action: set job-modes to replace-irreversibly
|
|
Fixes #2798
|
|
Up until now, the failure action has launched reboot.target and
poweroff.target with a less aggressive job mode than
"systemctl reboot" does. This has meant that the reboot and power-
off operations can stall if there are any conflicts with the target
during rebooting.
|
|
|
|
Selinux loading fix and warning cleanups
|
|
Fixes #2734
|
|
It seems that it is signed both on i386 and arm.
Avoids a stupid gcc warning on arm:
src/udev/udevadm-monitor.c: In function ‘print_device’:
src/udev/udevadm-monitor.c:44:16: warning: format ‘%u’ expects argument of type ‘unsigned int’, but argument 3 has type ‘__time_t {aka long int}’ [-Wformat=]
printf("%-6s[%"PRI_TIME".%06ld] %-8s %s (%s)\n",
^
|
|
This helps to understand misleading gcc warnings about type mismatches.
|
|
results and timings
|
|
As suggested by Evgeny Vereshchagin as a follow up for
https://github.com/systemd/systemd/pull/2781#issuecomment-191043402.
|
|
tree-wide: no need to pass excess flags to open()/openat() if O_PATH …
|
|
https://github.com/systemd/systemd/pull/2508#issuecomment-190901170
Maybe fixes https://bugzilla.redhat.com/show_bug.cgi?id=1308771.
|
|
As described in the documentation:
When O_PATH is specified in flags, flag bits other than O_CLOEXEC,
O_DIRECTORY, and O_NOFOLLOW are ignored.
So, we can remove unnecessary flags in a case when O_PATH is passed
to the open() or openat().
|
|
systemd-ask-password can store passwords in kernel keyring. However it
uses to print the passwords to standard output nevertheless. Depending
on where systemd-ask-password is called passwords may end on display
or in log, leaking sensitive information.
This allows to make systemd-ask-password quiet, effectively disabling
printing passwords to standard output.
|
|
Use `PRIu64` to print `uint64_t` in log msgs
|
|
resolved iteration fix
|
|
systemctl: Fix warn: action_to_runlevel() is not used
|
|
|
|
core: use DUAL_TIMESTAMP_NULL to reset kernel_timestamp
|
|
Fixes for adjtime parsing
|
|
Close the FILE* instead of the underlying fd.
|
|
shorten hostname before checking for trailing dot
|
|
core: avoid compiler warning when compiling with -fexceptions
|
|
Use assert_se() to ensure that they don't get optimized away.
|
|
Don't use {} for single-line "then" blocks.
|
|
If HAVE_SYSV_COMPAT is not defined: ‘action_to_runlevel’ defined but not used
|
|
CID #1352301.
|
|
instead of direct reset of kernel_timestamp fields.
|
|
Return EXIT_FAILURE and print the correct errno code if sd_bus_get_fd() or sd_bus_get_events() fail
|
|
Shortening can lead to a hostname that has a trailing dot.
Therefore it should be done before checking from trailing dots.
|
|
Trivial fixes
|
|
|
|
Initialize auto variables with cleanup attribute, otherwise we
get a compiler warning with -fexceptions.
./configure CFLAGS='-Wmaybe-uninitialized -fexceptions -O2'
|
|
|
|
|
|
Similarly to the previous commit, make context_write_data_local_rtc()
understand /etc/adjtime files with just one or two lines, with or without a
final newline.
Normalize the file to the current definition in hwclock(8), in the spirit of
"be liberal what you accept and strict what you produce": Add line terminators,
and set the second line to "0" if missing.
Fixes: #2638
|
|
As we default to "hardware clock is in UTC" if /etc/adjtime is not present, it
also makes sense to have that default if /etc/adjtime contains only one or two
lines.
Drop the "gibberish" test case, as this was just EIO because of not containing
three lines, which is already contained in other tests. clock_is_localtime()
never actually validated the format of the first two lines, and there is little
point in doing that.
This addresses the reading half of issue #2638.
|
|
Add path argument to clock_is_localtime() and default to "/etc/adjtime" if it's
NULL. This makes the function testable.
Add test-clock: initial test cases for some scenarios, using a temporary file.
This also checks the behaviour with a NULL (i. e. the system's /etc/adjtime)
file.
|
|
tree-wide: merge pager_open_if_enabled() to the pager_open()
|
|
Many subsystems define own pager_open_if_enabled() function which
checks '--no-pager' command line argument and open pager depends
on its value. All implementations of pager_open_if_enabled() are
the same. Let's merger this function with pager_open() from the
shared/pager.c and remove pager_open_if_enabled() from all subsytems
to prevent code duplication.
|
|
RFC 2131 Section 4.1 says that
"If the ’giaddr’ field in a DHCP message from a client is non-zero,
the server sends any return messages to the ’DHCP server’ port on the
BOOTP relay agent whose address appears in ’giaddr’."
Fix this by adding a destination port when sending unicast UDP packets
and provide the server port when a BOOTP relay agent is being used.
|
|
|
|
tests: various fixes
|
|
|
|
tree-wide: minor formatting inconsistency cleanups
|
|
|
|
Bugfix set virt=none when running on XEN Dom0
|
|
When running in XEN Dom0 the virtualization check:
1) detect_xen returns HYPERVISOR_NONE so next checks are executed
2) /proc/sys/hypervisor detects a XEN hypervisor
it is lacking the special Dom0 detection as in detect_xen
With this patch, at the end of all virtualization checks we double-check if running in XEN Dom0 or DomU.
|
|
print out every single detection executed and its result.
|
|
|
