Lennart Poettering [Fri, 19 Aug 2016 08:41:26 +0000 (10:41 +0200)]
Merge pull request #3988 from keszybz/journald-dynamic-users
Journald dynamic users
Peter Hutterer [Fri, 19 Aug 2016 08:25:53 +0000 (18:25 +1000)]
hwdb: add resolution for the Dell Precision 5510 (#3989)
https://bugs.freedesktop.org/show_bug.cgi?id=97347
Zbigniew Jędrzejewski-Szmek [Fri, 19 Aug 2016 03:19:10 +0000 (23:19 -0400)]
journald: do not create split journals for dynamic users
Dynamic users should be treated like system users, and their logs
should end up in the main system journal.
Zbigniew Jędrzejewski-Szmek [Fri, 19 Aug 2016 03:09:29 +0000 (23:09 -0400)]
core/dynamic-users: warn when creation of symlinks for dynamic users fails
Also return the first error, since it's most likely to be interesting.
If unlink fails, symlink will usually return EEXIST.
Tejun Heo [Fri, 19 Aug 2016 02:57:53 +0000 (22:57 -0400)]
logind: update empty and "infinity" handling for [User]TasksMax (#3835)
The parsing functions for [User]TasksMax were inconsistent. Empty string and
"infinity" were interpreted as no limit for TasksMax but not accepted for
UserTasksMax. Update them so that they're consistent with other knobs.
* Empty string indicates the default value.
* "infinity" indicates no limit.
While at it, replace opencoded (uint64_t) -1 with CGROUP_LIMIT_MAX in TasksMax
handling.
v2: Update empty string to indicate the default value as suggested by Zbigniew
Jędrzejewski-Szmek.
v3: Fixed empty UserTasksMax handling.
Zbigniew Jędrzejewski-Szmek [Fri, 19 Aug 2016 02:53:12 +0000 (22:53 -0400)]
Merge pull request #3986 from whot/wip/click-wheel-angle-horiz
Add udev property for horizontal wheel click angles
Zbigniew Jędrzejewski-Szmek [Fri, 19 Aug 2016 02:52:13 +0000 (22:52 -0400)]
Merge pull request #3865 from poettering/remove-ipc
add RemoveIPC= service file setting
Peter Hutterer [Tue, 16 Aug 2016 05:23:42 +0000 (15:23 +1000)]
hwdb: add a udev property for a wheel click angle on horiz wheels
The Logitech MX Master has a horizontal scroll wheel with a different click
angle than the vertical one. Add a new property for this case, we can't add
values to the normal one without risking upsetting existing parsers.
Fixes #3947
Peter Hutterer [Mon, 15 Aug 2016 03:22:12 +0000 (13:22 +1000)]
libudev: fix typo in comment
Lennart Poettering [Fri, 19 Aug 2016 01:16:16 +0000 (03:16 +0200)]
hostnamectl: rework pretty hostname validation (#3985)
Rework
17eb9a9ddba3f03fcba33445c1c1eedeb948da04 a bit.
Let's make sure we don't clobber the input parameter args[1], following our
coding style to not clobber parameters unless explicitly indicated. (in
particular, as we don't want to have our changes appear in the command line
shown in "ps"...)
No functional change.
Lennart Poettering [Thu, 18 Aug 2016 22:35:05 +0000 (00:35 +0200)]
nss-mymachines: avoid connecting to dbus from inside dbus-daemon
Inspired from the new logic in nss-systemd let's make sure we don't end up
deadlocking in nss-mymachines either in case dbus-daemon tries to a look up a
name and we want to connect to the bus.
This case is much simpler though, as there's no point in resolving virtual
machine UIDs by dbus-daemon as those should never be able to connect to the
host's busses.
Lennart Poettering [Tue, 2 Aug 2016 10:28:51 +0000 (12:28 +0200)]
core: bypass dynamic user lookups from dbus-daemon
dbus-daemon does NSS name look-ups in order to enforce its bus policy. This
might dead-lock if an NSS module use wants to use D-Bus for the look-up itself,
like our nss-systemd does. Let's work around this by bypassing bus
communication in the NSS module if we run inside of dbus-daemon. To make this
work we keep a bit of extra state in /run/systemd/dynamic-uid/ so that we don't
have to consult the bus, but can still resolve the names.
Note that the normal codepath continues to be via the bus, so that resolving
works from all mount namespaces and is subject to authentication, as before.
This is a bit dirty, but not too dirty, as dbus daemon is kinda special anyway
for PID 1.
Lennart Poettering [Tue, 2 Aug 2016 09:14:56 +0000 (11:14 +0200)]
clean-ipc: debug log about all remove IPC objects
Lennart Poettering [Mon, 1 Aug 2016 17:24:40 +0000 (19:24 +0200)]
core: add RemoveIPC= setting
This adds the boolean RemoveIPC= setting to service, socket, mount and swap
units (i.e. all unit types that may invoke processes). if turned on, and the
unit's user/group is not root, all IPC objects of the user/group are removed
when the service is shut down. The life-cycle of the IPC objects is hence bound
to the unit life-cycle.
This is particularly relevant for units with dynamic users, as it is essential
that no objects owned by the dynamic users survive the service exiting. In
fact, this patch adds code to imply RemoveIPC= if DynamicUser= is set.
In order to communicate the UID/GID of an executed process back to PID 1 this
adds a new "user lookup" socket pair, that is inherited into the forked
processes, and closed before the exec(). This is needed since we cannot do NSS
from PID 1 due to deadlock risks, However need to know the used UID/GID in
order to clean up IPC owned by it if the unit shuts down.
Lennart Poettering [Mon, 1 Aug 2016 15:40:37 +0000 (17:40 +0200)]
core: move obsolete properties to the end of vtables
This makes it easier to discern the relevant and obsolete parts of the vtables,
and in particular helps when comparing introspection data with the actual
vtable definitions.
Lennart Poettering [Fri, 29 Jul 2016 14:22:30 +0000 (16:22 +0200)]
core: make use of uid_is_valid() when checking for UID validity
Lennart Poettering [Thu, 28 Jul 2016 07:51:47 +0000 (09:51 +0200)]
clean-ipc: shorten code a bit
Lennart Poettering [Thu, 28 Jul 2016 07:51:25 +0000 (09:51 +0200)]
clean-ipc: don't filter out '.' and '..' twice
Lennart Poettering [Thu, 28 Jul 2016 07:50:28 +0000 (09:50 +0200)]
unit: remove orphaned cgroup_netclass_id field
Lennart Poettering [Wed, 17 Aug 2016 18:28:45 +0000 (20:28 +0200)]
Merge pull request #3946 from keszybz/open-journal-root
Make journalctl more flexible
Felipe Sateler [Wed, 17 Aug 2016 18:11:27 +0000 (15:11 -0300)]
sysv-generator: better error reporting (#3977)
Currently in the journal you get messages without context like:
systemd-sysv-generator[$pid]: Failed to build name: Invalid argument
When parsing the init script, show the file and line number where the
error was found. At the same time, add more context information if
available.
Thus turning the message into something like:
systemd-sysv-generator[$pid]: [/etc/init.d/root-system-proofd:13] Could not build name for facility $network,: Invalid argument
Vito Caputo [Wed, 17 Aug 2016 12:51:07 +0000 (05:51 -0700)]
journal: ensure open journals from find_journal() (#3973)
If journals get into a closed state like when rotate fails due to
ENOSPC, when space is made available it currently goes unnoticed leaving
the journals in a closed state indefinitely.
By calling system_journal_open() on entry to find_journal() we ensure
the journal has been opened/created if possible.
Also moved system_journal_open() up to after open_journal(), before
find_journal().
Fixes https://github.com/systemd/systemd/issues/3968
Michal Sekletar [Wed, 17 Aug 2016 12:10:28 +0000 (14:10 +0200)]
rules: introduce disk/by-id (model_serial) symlinks for NVMe drives (#3974)
$ ls -l /dev/disk/by-id/nvme*
lrwxrwxrwx. 1 root root 13 Aug 17 04:25 /dev/disk/by-id/nvme-HUSPR3216AHP301_STM0001B6780 -> ../../nvme0n1
lrwxrwxrwx. 1 root root 15 Aug 17 04:25 /dev/disk/by-id/nvme-HUSPR3216AHP301_STM0001B6780-part1 -> ../../nvme0n1p1
https://github.com/systemd/systemd/issues/1453
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=
779ff75617099f4defe14e20443b95019a4c5ae8
Ronny Chevalier [Tue, 16 Aug 2016 19:22:50 +0000 (21:22 +0200)]
Merge pull request #3960 from keszybz/efi-option-description
build-sys: clarify that --disable-efi is about sd-boot and bootctl
Daniel Hahler [Tue, 16 Aug 2016 16:47:39 +0000 (18:47 +0200)]
zsh: _journalctl: also handle --root arg and --key=value style (#3956)
This will now also handle `journalctl --directory=/var/log/journal`
properly.
Daniel Hahler [Tue, 16 Aug 2016 16:42:41 +0000 (18:42 +0200)]
zsh: _journalctl: do not complete exclusive modes (#3957)
After `journalctl -D /var/log/journal` "--directory", "--file",
"--machine" and "--root" should not be available for completion, because
they are exclusive. But multiple `--file` arguments are allowed.
Zbigniew Jędrzejewski-Szmek [Sun, 14 Aug 2016 22:03:35 +0000 (18:03 -0400)]
Merge pull request #3905 from htejun/cgroup-v2-cpu
core: add cgroup CPU controller support on the unified hierarchy
(zj: merging not squashing to make it clear against which upstream this patch was developed.)
Zbigniew Jędrzejewski-Szmek [Sun, 14 Aug 2016 19:24:33 +0000 (15:24 -0400)]
build-sys: clarify that --disable-efi is about sd-boot and bootctl
If wasn't obvious what the effect of --disable-efi was without
going through Makefile.am.
Fixes #3959.
Zbigniew Jędrzejewski-Szmek [Sat, 13 Aug 2016 15:15:19 +0000 (11:15 -0400)]
units: do not start load-random-seed in containers (#3941)
Random numbers are provided by the host kernel, we don't need to do anything.
https://bugzilla.redhat.com/show_bug.cgi?id=1329124
Zbigniew Jędrzejewski-Szmek [Sat, 13 Aug 2016 15:14:58 +0000 (11:14 -0400)]
hwdb: normalize spacing in 60-keyboard.hwdb (#3954)
Daniel Hahler [Sat, 13 Aug 2016 14:42:55 +0000 (16:42 +0200)]
zsh: _systemctl: do not attempt to use "--system" by default (#3951)
In 68c4f6d the following was added:
local -a _modes; _modes=("--user" "--system")
local _sys_service_mgr=${${words:*_modes}[(R)(${(j.|.)_modes})]:---system}
With the following comment:
> If neither are on the line, --system is set; for system services to be
> completed.
But it does not work as documented:
% _modes=(--user --system)
% words=()
% echo ${${words:*_modes}[(R)(${(j.|.)_modes})]:---system}
However, it should not use `--system` in that case anyway, so this patch
removes the part that should cause a default to be used and adds some
comments.
Daniel Hahler [Sat, 13 Aug 2016 14:41:22 +0000 (16:41 +0200)]
zsh: _journalctl: improve support for handling mode args (#3952)
This only completes fields from `journalctl --user` in _journal_fields when `--user`
is used.
It also changes $_sys_service_mgr to include both `--system` and `--user`,
because `journalctl` behaves different from `systemctl` in this regard.
No attempt is made to filter out invalid combinations, e.g. when using both
`--directory` and `--system` (see https://github.com/systemd/systemd/issues/3949).
Zbigniew Jędrzejewski-Szmek [Fri, 12 Aug 2016 04:34:45 +0000 (00:34 -0400)]
journalctl: allow --root argument for journal watching
It is useful to look at a (possibly inactive) container or other os tree
with --root=/path/to/container. This is similar to specifying
--directory=/path/to/container/var/log/journal --directory=/path/to/container/run/systemd/journal
(if using --directory multiple times was allowed), but doesn't require
as much typing.
Zbigniew Jędrzejewski-Szmek [Fri, 12 Aug 2016 04:32:10 +0000 (00:32 -0400)]
sd-journal: fix sd_journal_open_directory with SD_JOURNAL_OS_ROOT
The directory argument that is given to sd_j_o_d was ignored when
SD_JOURNAL_OS_ROOT was given, and directories relative to the root of the host
file system were used. With that flag, sd_j_o_d should do the same as
sd_j_open_container: use the path as "prefix", i.e. the directory relative to
which everything happens.
Instead of touching sd_j_o_d, journal_new is fixed to do what sd_j_o_c
was doing, and treat the specified path as prefix when SD_JOURNAL_OS_ROOT is
specified.
Zbigniew Jędrzejewski-Szmek [Fri, 12 Aug 2016 02:48:58 +0000 (22:48 -0400)]
sd-journal: allow SYSTEM and CURRENT_USER flags with sd_j_open_directory[_fd]
There is no reason not to. This makes journalctl -D ... --system work,
useful for example when viewing files from a deactivated container.
Zbigniew Jędrzejewski-Szmek [Fri, 12 Aug 2016 02:43:07 +0000 (22:43 -0400)]
sd-journal: split out flags into separate defines for legibility
… in preparation for future changes.
Zbigniew Jędrzejewski-Szmek [Thu, 11 Aug 2016 19:39:15 +0000 (15:39 -0400)]
Merge pull request #3942 from blueyed/zsh-respect-user-more
zsh completion: handle --user mode when filtering by state and in journalctl commands
Daniel Hahler [Thu, 11 Aug 2016 16:52:13 +0000 (18:52 +0200)]
zsh: _journalctl: handle --user in _journal_none
This uses the same mechanism from _systemctl to inject `--user` into the
`journalctrl -F _EXE` call to list executables.
Before this patch the "commands" section would list executables from
system units always.
Daniel Hahler [Thu, 11 Aug 2016 16:46:31 +0000 (18:46 +0200)]
zsh: _filter_units_by_property: respect --user
Use `$_sys_service_mgr` to handle `--user`, so that `systemctl --user
stop` will correctly filter the active (user) units. Before this patch,
only user units that also exist as system units and are stoppable there
would be listed.
Zbigniew Jędrzejewski-Szmek [Thu, 11 Aug 2016 08:51:49 +0000 (04:51 -0400)]
man: add "timeout" to status table (#3919)
Zbigniew Jędrzejewski-Szmek [Thu, 11 Aug 2016 08:51:00 +0000 (04:51 -0400)]
coredump: treat RLIMIT_CORE below page size as disabling coredumps (#3932)
The kernel treats values below a certain threshold (minfmt->min_coredump
which is initialized do ELF_EXEC_PAGESIZE, which varies between architectures,
but is usually the same as PAGE_SIZE) as disabling coredumps [1].
Any core image below ELF_EXEC_PAGESIZE will yield an invalid backtrace anyway [2],
so follow the kernel and not try to parse or store such images.
[1] https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/fs/coredump.c#n660
[2] systemd-coredump[16260]: Process 16258 (sleep) of user 1002 dumped core.
Stack trace of thread 16258:
#0 0x00007f1d8b3d3810 n/a (n/a)
https://bugzilla.redhat.com/show_bug.cgi?id=1309172#c19
Lennart Poettering [Thu, 11 Aug 2016 08:49:57 +0000 (10:49 +0200)]
Merge pull request #3933 from keszybz/hwdb-parser
Add hwdb parser to check for inconsistencies
Rhys [Tue, 9 Aug 2016 13:33:46 +0000 (23:33 +1000)]
install: follow config_path symlink (#3362)
Under NixOS, the config_path /etc/systemd/system is a symlink to
/etc/static/systemd/system. Commands such as `systemctl list-unit-files`
and `systemctl is-enabled` did not work as the symlink was not followed.
This does not affect how symlinks are treated within the config_path
directory.
Evgeny Vereshchagin [Tue, 9 Aug 2016 12:49:32 +0000 (15:49 +0300)]
sd-journal: watch logs below container's /{var,run}/log/journal (instead of the /) (#3934)
Fixes #3927.
romanstingler [Tue, 9 Aug 2016 04:51:38 +0000 (06:51 +0200)]
added Zelotes gaming mouse (#3925)
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 19:07:41 +0000 (15:07 -0400)]
build-sys: hook up the hwdb parser to check
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 17:36:26 +0000 (13:36 -0400)]
hwdb: comment out a duplicated key for the XO
It's hard to say which one of the two mappings should stay. But the later
one would win (when both very present), and nobody complained, so let's
assume that that's the one.
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 17:31:37 +0000 (13:31 -0400)]
hwdb: use lowercase hex for key numbers
It seems awkward to have both cases mixes. Note that the real parser
accepts both cases, and this only standarizes the usage in the systemd
database.
Zbigniew Jędrzejewski-Szmek [Thu, 4 Aug 2016 19:31:05 +0000 (15:31 -0400)]
hwdb: remove duplicated matches for old Logitech unifying receiver
Quoting https://github.com/systemd/systemd/pull/3906#discussion_r73828368:
> According to
> http://support.logitech.com/en_us/product/v220-cordless-optical-mouse-for-notebooks
> it seems the mouse is using a pre-version of the small unifying receiver we
> know now. If there are 2 mice with the same receiver, that means that the
> values should both be dropped IMO.
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 14:53:06 +0000 (10:53 -0400)]
tests: add python parser for hwdb grammar
This works for hwdb/[67]0-*.hwdb. I also added code to parse hwdb/20-*, but those
files are huge, and parsing them using this parser is annoyingly slow (about one
minute for the biggest files). So I removed the support for hwdb/20-*, a much simpler
hand-generated parser should suffice for those.
Current output:
hwdb/60-evdev.hwdb: 24 match groups, 35 matches, 88 properties, 0.19323015213012695s to parse
Match 'evdev:input:b0003v05ACp0259*' is duplicated
Match 'evdev:input:b0003v05ACp025A*' is duplicated
Match 'evdev:input:b0003v05ACp025B*' is duplicated
hwdb/60-keyboard.hwdb: 122 match groups, 188 matches, 638 properties, 1.0906572341918945s to parse
Failed to parse: 'KEYBOARD_KEY_8F=switchvideomode'
Failed to parse: 'KEYBOARD_KEY_C0183=media'
Failed to parse: 'KEYBOARD_KEY_C0201=new'
Failed to parse: 'KEYBOARD_KEY_C0289=reply'
Failed to parse: 'KEYBOARD_KEY_C028B=forwardmail'
Failed to parse: 'KEYBOARD_KEY_C028C=send'
Failed to parse: 'KEYBOARD_KEY_C021A=undo'
Failed to parse: 'KEYBOARD_KEY_C0279=redo'
Failed to parse: 'KEYBOARD_KEY_C0208=print'
Failed to parse: 'KEYBOARD_KEY_C0207=save'
Failed to parse: 'KEYBOARD_KEY_C0194=file'
Failed to parse: 'KEYBOARD_KEY_C01A7=documents'
Failed to parse: 'KEYBOARD_KEY_C01B6=images'
Failed to parse: 'KEYBOARD_KEY_C01B7=sound'
Property KEYBOARD_KEY_c7 is duplicated
Failed to parse: 'KEYBOARD_KEY_cF=end'
hwdb/70-mouse.hwdb: 62 match groups, 93 matches, 68 properties, 0.34186625480651855s to parse
Match 'mouse:usb:v046dpc51b:name:Logitech USB Receiver:' is duplicated
hwdb/70-pointingstick.hwdb: 5 match groups, 14 matches, 7 properties, 0.06518816947937012s to parse
hwdb/70-touchpad.hwdb: 3 match groups, 5 matches, 3 properties, 0.039690494537353516s to parse
Subsequest commits will clean those issues up.
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 17:33:22 +0000 (13:33 -0400)]
hwdb: indent commented properties
This way it's clear that the property block does not end at the comment.
The python checker will complain if this is not the case.
We had a few bugs before where two match blocks were merged by mistake,
and this change should help avoid that.
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 14:52:21 +0000 (10:52 -0400)]
hwdb: remove extra spaces
Vito Caputo [Tue, 9 Aug 2016 04:35:07 +0000 (21:35 -0700)]
journal-gatewayd: fix typo; s/Certificat/Certificate/ (#3931)
Peter Hutterer [Mon, 8 Aug 2016 22:58:19 +0000 (08:58 +1000)]
hwdb: fix duplicate entry for Apple touchpads (#3921)
Copy-paste error, correct IDs from the kernel's drivers/input/mouse/bcm5974.c
Fixes: https://github.com/systemd/systemd/pull/3906/
Zbigniew Jędrzejewski-Szmek [Mon, 8 Aug 2016 21:39:16 +0000 (17:39 -0400)]
core: amend policy to open up dynamic user queries (#3920)
romanstingler [Sun, 7 Aug 2016 15:45:44 +0000 (17:45 +0200)]
hwdb: add multimedia keys for Medion Akoya series (#3918)
Tejun Heo [Sun, 7 Aug 2016 13:45:39 +0000 (09:45 -0400)]
core: add cgroup CPU controller support on the unified hierarchy
Unfortunately, due to the disagreements in the kernel development community,
CPU controller cgroup v2 support has not been merged and enabling it requires
applying two small out-of-tree kernel patches. The situation is explained in
the following documentation.
https://git.kernel.org/cgit/linux/kernel/git/tj/cgroup.git/tree/Documentation/cgroup-v2-cpu.txt?h=cgroup-v2-cpu
While it isn't clear what will happen with CPU controller cgroup v2 support,
there are critical features which are possible only on cgroup v2 such as
buffered write control making cgroup v2 essential for a lot of workloads. This
commit implements systemd CPU controller support on the unified hierarchy so
that users who choose to deploy CPU controller cgroup v2 support can easily
take advantage of it.
On the unified hierarchy, "cpu.weight" knob replaces "cpu.shares" and "cpu.max"
replaces "cpu.cfs_period_us" and "cpu.cfs_quota_us". [Startup]CPUWeight config
options are added with the usual compat translation. CPU quota settings remain
unchanged and apply to both legacy and unified hierarchies.
v2: - Error in man page corrected.
- CPU config application in cgroup_context_apply() refactored.
- CPU accounting now works on unified hierarchy.
Susant Sahani [Sun, 7 Aug 2016 09:23:32 +0000 (14:53 +0530)]
networkd: remove duplicate call to manager_dirty (#3917)
since link_dirty itself calls manager_dirty no need to
call it separately .
Lennart Poettering [Sun, 7 Aug 2016 09:17:56 +0000 (11:17 +0200)]
Merge pull request #3914 from keszybz/fix-man-links
Fix man links
Zbigniew Jędrzejewski-Szmek [Sun, 7 Aug 2016 09:14:40 +0000 (05:14 -0400)]
man: add a table of possible exit statuses (#3910)
Zbigniew Jędrzejewski-Szmek [Sat, 6 Aug 2016 20:41:35 +0000 (16:41 -0400)]
Merge pull request #3884 from poettering/private-users
Zbigniew Jędrzejewski-Szmek [Sat, 6 Aug 2016 20:38:09 +0000 (16:38 -0400)]
man: fix some internal man page references
sd_journal-query_enumerate was an early draft, the name was changed
to sd_j_enumerate_fields.
Zbigniew Jędrzejewski-Szmek [Sat, 6 Aug 2016 20:36:51 +0000 (16:36 -0400)]
man: provide html links to a bunch of external man pages
Yi EungJun [Sat, 6 Aug 2016 20:39:13 +0000 (05:39 +0900)]
journal-gatewayd: fix segfault with certain request (#3893)
When client requests to get logs with `follow` and `KEY=match` that
doesn't match any log entry, journal-gatewayd segfaulted.
Make request_reader_entries to return zero in such case to wait for
matching entries.
This fixes https://github.com/systemd/systemd/issues/3873.
Yi EungJun [Sat, 6 Aug 2016 17:00:31 +0000 (02:00 +0900)]
journal-gatewayd: add --directory option (#3913)
Serve journals in the specified directory instead of default journals.
Lennart Poettering [Sat, 6 Aug 2016 15:02:17 +0000 (17:02 +0200)]
Merge pull request #3911 from crrodriguez/asan-fixes
Asan fixes
Cristian Rodríguez [Fri, 5 Aug 2016 16:10:41 +0000 (16:10 +0000)]
buildsys,journal: allow -fsanitize=address without VALGRIND defined
Fixed (master) versions of libtool pass -fsanitize=address correctly
into CFLAGS and LDFLAGS allowing ASAN to be used without any special
configure tricks..however ASAN triggers in lookup3.c for the same
reasons valgrind does. take the alternative codepath if
__SANITIZE_ADDRESS__ is defined as well.
Cristian Rodríguez [Fri, 5 Aug 2016 16:07:49 +0000 (16:07 +0000)]
test: fix stack overflow reported by ASAN
It was meant to write to q instead of t
FAIL: test-id128
================
=================================================================
==125770==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x7ffd4615bd31 at pc 0x7a2f41b1bf33 bp 0x7ffd4615b750 sp 0x7ffd4615b748
WRITE of size 1 at 0x7ffd4615bd31 thread T0
#0 0x7a2f41b1bf32 in id128_to_uuid_string src/libsystemd/sd-id128/id128-util.c:42
#1 0x401f73 in main src/test/test-id128.c:147
#2 0x7a2f41336341 in __libc_start_main (/lib64/libc.so.6+0x20341)
#3 0x401129 in _start (/home/crrodriguez/scm/systemd/.libs/test-id128+0x401129)
Address 0x7ffd4615bd31 is located in stack of thread T0 at offset 1409 in frame
#0 0x401205 in main src/test/test-id128.c:37
This frame has 23 object(s):
[32, 40) 'b'
[96, 112) 'id'
[160, 176) 'id2'
[224, 240) 'a'
[288, 304) 'b'
[352, 368) 'a'
[416, 432) 'b'
[480, 496) 'a'
[544, 560) 'b'
[608, 624) 'a'
[672, 688) 'b'
[736, 752) 'a'
[800, 816) 'b'
[864, 880) 'a'
[928, 944) 'b'
[992, 1008) 'a'
[1056, 1072) 'b'
[1120, 1136) 'a'
[1184, 1200) 'b'
[1248, 1264) 'a'
[1312, 1328) 'b'
[1376, 1409) 't' <== Memory access at offset 1409 overflows this variable
[1472, 1509) 'q'
HINT: this may be a false positive if your program uses some custom stack unwind mechanism or swapcontext
(longjmp and C++ exceptions *are* supported)
SUMMARY: AddressSanitizer: stack-buffer-overflow src/libsystemd/sd-id128/id128-util.c:42 in id128_to_uuid_string
Shadow bytes around the buggy address:
0x100028c23750: f2 f2 00 00 f4 f4 f2 f2 f2 f2 00 00 f4 f4 f2 f2
0x100028c23760: f2 f2 00 00 f4 f4 f2 f2 f2 f2 00 00 f4 f4 f2 f2
0x100028c23770: f2 f2 00 00 f4 f4 f2 f2 f2 f2 00 00 f4 f4 f2 f2
0x100028c23780: f2 f2 00 00 f4 f4 f2 f2 f2 f2 00 00 f4 f4 f2 f2
0x100028c23790: f2 f2 00 00 f4 f4 f2 f2 f2 f2 00 00 f4 f4 f2 f2
=>0x100028c237a0: f2 f2 00 00 00 00[01]f4 f4 f4 f2 f2 f2 f2 00 00
0x100028c237b0: 00 00 05 f4 f4 f4 00 00 00 00 00 00 00 00 00 00
0x100028c237c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x100028c237d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x100028c237e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
0x100028c237f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Shadow byte legend (one shadow byte represents 8 application bytes):
Addressable: 00
Partially addressable: 01 02 03 04 05 06 07
Heap left redzone: fa
Heap right redzone: fb
Freed heap region: fd
Stack left redzone: f1
Stack mid redzone: f2
Stack right redzone: f3
Stack partial redzone: f4
Stack after return: f5
Stack use after scope: f8
Global redzone: f9
Global init order: f6
Poisoned by user: f7
Container overflow: fc
Array cookie: ac
Intra object redzone: bb
ASan internal: fe
Left alloca redzone: ca
Right alloca redzone: cb
==125770==ABORTING
FAIL test-id128 (exit status: 1)
Cristian Rodríguez [Thu, 4 Aug 2016 16:49:19 +0000 (16:49 +0000)]
test: test-sigbus must be skipped when -fsanitize=address is used
ASAN is unable to handle it.
Zbigniew Jędrzejewski-Szmek [Sat, 6 Aug 2016 00:55:08 +0000 (20:55 -0400)]
Merge pull request #3818 from poettering/exit-status-env
beef up /var/tmp and /tmp handling; set $SERVICE_RESULT/$EXIT_CODE/$EXIT_STATUS on ExecStop= and make sure root/nobody are always resolvable
Susant Sahani [Fri, 5 Aug 2016 23:44:57 +0000 (05:14 +0530)]
networkd: add support to set STP (#3903)
fixes #3881
Lennart Poettering [Fri, 5 Aug 2016 15:03:09 +0000 (17:03 +0200)]
Merge pull request #3900 from keszybz/fix-3607
Fix 3607
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 12:24:00 +0000 (08:24 -0400)]
socket: add helper function to remove code duplication
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 03:51:47 +0000 (23:51 -0400)]
core/socket: include remote address in the message when dropping connection
Without the address the message is not very useful.
Aug 04 23:52:21 rawhide systemd[1]: testlimit.socket: Too many incoming connections (4) from source ::1, dropping connection.
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 03:42:27 +0000 (23:42 -0400)]
systemd: do not serialize peer, bump count when deserializing socket instead
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 03:26:31 +0000 (23:26 -0400)]
core/service: serialize and deserialize accept_socket
This fixes an issue during reexec — the count of connections would be lost:
[zbyszek@fedora-rawhide ~]$ systemctl status testlimit.socket | grep Connected
Accepted: 1; Connected: 1
[zbyszek@fedora-rawhide ~]$ sudo systemctl daemon-reexec
[zbyszek@fedora-rawhide ~]$ systemctl status testlimit.socket | grep Connected
Accepted: 1; Connected: 0
With the patch, Connected count is preserved.
Also add "Accept Socket" to the dump output for services.
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 01:42:23 +0000 (21:42 -0400)]
core/socket: rework SocketPeer refcounting
Make functions and definitions that don't need to be shared local to
socket.c.
Ronny Chevalier [Fri, 5 Aug 2016 11:35:52 +0000 (13:35 +0200)]
Merge pull request #3902 from poettering/parse-nice
util-lib: unify parsing of nice level values
Lennart Poettering [Fri, 5 Aug 2016 09:17:08 +0000 (11:17 +0200)]
util-lib: unify parsing of nice level values
This adds parse_nice() that parses a nice level and ensures it is in the right
range, via a new nice_is_valid() helper. It then ports over a number of users
to this.
No functional changes.
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 08:37:10 +0000 (04:37 -0400)]
networkd: do not set NOARP unconditionally (#3891)
Fixes #3890.
Vito Caputo [Fri, 5 Aug 2016 04:09:23 +0000 (21:09 -0700)]
fileio: fix MIN/MAX mixup (#3896)
The intention is to clamp the value to READ_FULL_BYTES_MAX, which
would be the minimum of the two.
Vito Caputo [Fri, 5 Aug 2016 04:05:34 +0000 (21:05 -0700)]
journal-remote: remove unnecessary gnutls includes (#3895)
journal-(gatewayd,remote).c don't actually utilize libgnutls even when
HAVE_GNUTLS is defined.
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 00:55:45 +0000 (20:55 -0400)]
systemd: convert peers_by_address to a set
Zbigniew Jędrzejewski-Szmek [Fri, 5 Aug 2016 00:36:10 +0000 (20:36 -0400)]
basic/set: remove some spurious spaces
Lennart Poettering [Thu, 4 Aug 2016 20:11:29 +0000 (22:11 +0200)]
core: only set the watchdog variables in ExecStart= lines
Lennart Poettering [Thu, 4 Aug 2016 19:14:27 +0000 (21:14 +0200)]
core: remember first unit failure, not last unit failure
Previously, the result value of a unit was overriden with each failure that
took place, so that the result always reported the last failure that took
place.
With this commit this is changed, so that the first failure taking place is
stored instead. This should normally not matter much as multiple failures are
sufficiently uncommon. However, it improves one behaviour: if we send SIGABRT
to a service due to a watchdog timeout, then this currently would be reported
as "coredump" failure, rather than the "watchodg" failure it really is. Hence,
in order to report information about the type of the failure, and not about
the effect of it, let's change this from all unit type to store the first, not
the last failure.
This addresses the issue pointed out here:
https://github.com/systemd/systemd/pull/3818#discussion_r73433520
Lennart Poettering [Wed, 27 Jul 2016 11:30:58 +0000 (13:30 +0200)]
update TODO
Lennart Poettering [Wed, 27 Jul 2016 11:14:01 +0000 (13:14 +0200)]
nss-systemd: resolve root/nobody statically
Let's extend nss-systemd to also synthesize user/group entries for the
UIDs/GIDs 0 and 65534 which have special kernel meaning. Given that nss-systemd
is listed in /etc/nsswitch.conf only very late any explicit listing in
/etc/passwd or /etc/group takes precedence.
This functionality is useful in minimal container-like setups that lack
/etc/passwd files (or only have incompletely populated ones).
Lennart Poettering [Wed, 27 Jul 2016 09:51:11 +0000 (11:51 +0200)]
core: set $SERVICE_RESULT, $EXIT_CODE and $EXIT_STATUS in ExecStop=/ExecStopPost= commands
This should simplify monitoring tools for services, by passing the most basic
information about service result/exit information via environment variables,
thus making it unnecessary to retrieve them explicitly via the bus.
0xAX [Thu, 4 Aug 2016 20:52:24 +0000 (23:52 +0300)]
main: use pager for --dump-configuration-items (#3894)
Vito Caputo [Thu, 4 Aug 2016 20:52:02 +0000 (13:52 -0700)]
fileio: fix read_full_stream() bugs (#3887)
read_full_stream() _always_ allocated twice the memory needed, due to
only breaking the realloc() && fread() loop when fread() returned 0,
requiring another iteration and exponentially enlarged buffer just to
discover the EOF condition.
This also caused file sizes >2MiB && <= 4MiB to erroneously be treated
as E2BIG, due to the inappropriately doubled buffer size exceeding
4*1024*1024.
Also made the 4*1024*1024 magic number a READ_FULL_BYTES_MAX constant.
Lennart Poettering [Thu, 4 Aug 2016 18:39:57 +0000 (20:39 +0200)]
Merge pull request #3892 from toanju/fix-brvlan-error-on-normal-bridge
networkd: apply bridge vlan configuration correct
Tobias Jungel [Thu, 4 Aug 2016 15:11:13 +0000 (17:11 +0200)]
networkd: apply bridge vlan configuration correct
bridge vlan configuration was applied even if it wasn't configured.
fixes #3876
Lennart Poettering [Wed, 27 Jul 2016 09:50:37 +0000 (11:50 +0200)]
core: use the correct APIs to determine whether a dual timestamp is initialized
Lennart Poettering [Tue, 26 Jul 2016 15:53:07 +0000 (17:53 +0200)]
core: move masking of chroot/permission masking into service_spawn()
Let's fix up the flags fields in service_spawn() rather than its callers, in
order to simplify things a bit.
Lennart Poettering [Tue, 26 Jul 2016 15:40:35 +0000 (17:40 +0200)]
core: turn various execution flags into a proper flags parameter
The ExecParameters structure contains a number of bit-flags, that were so far
exposed as bool:1, change this to a proper, single binary bit flag field. This
makes things a bit more expressive, and is helpful as we add more flags, since
these booleans are passed around in various callers, for example
service_spawn(), whose signature can be made much shorter now.
Not all bit booleans from ExecParameters are moved into the flags field for
now, but this can be added later.
Lennart Poettering [Tue, 26 Jul 2016 15:23:28 +0000 (17:23 +0200)]
util-lib: rework /tmp and /var/tmp handling code
Beef up the existing var_tmp() call, rename it to var_tmp_dir() and add a
matching tmp_dir() call (the former looks for the place for /var/tmp, the
latter for /tmp).
Both calls check $TMPDIR, $TEMP, $TMP, following the algorithm Python3 uses.
All dirs are validated before use. secure_getenv() is used in order to limite
exposure in suid binaries.
This also ports a couple of users over to these new APIs.
The var_tmp() return parameter is changed from an allocated buffer the caller
will own to a const string either pointing into environ[], or into a static
const buffer. Given that environ[] is mostly considered constant (and this is
exposed in the very well-known getenv() call), this should be OK behaviour and
allows us to avoid memory allocations in most cases.
Note that $TMPDIR and friends override both /var/tmp and /tmp usage if set.
Lennart Poettering [Thu, 4 Aug 2016 14:16:04 +0000 (16:16 +0200)]
Merge pull request #2471 from michaelolbrich/transient-mounts
allow transient mounts and automounts
Lennart Poettering [Thu, 4 Aug 2016 14:11:38 +0000 (16:11 +0200)]
Merge pull request #3885 from keszybz/help-output
Update help for "short-full" and shorten to 80 columns
Susant Sahani [Thu, 4 Aug 2016 14:00:58 +0000 (19:30 +0530)]
networkd: add support to configure NOARP/ARP for interface (#3854)
https://lists.freedesktop.org/archives/systemd-devel/2016-August/037268.html