summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2016-07-29tests: skip process 1 tests if systemd not is runningDavide Cavalca
No point running tests against process 1 if systemd is not running as that process. This is a rework of an unpublished patch by @9muir.
2016-07-29tests: don't test hostname if it looks like an id128Steve Muir
The condition tests for hostname will fail if hostname looks like an id128. The test function attempts to convert hostname to an id128, and if that succeeds compare it to the machine ID (presumably because the 'hostname' condition test is overloaded to also test machine ID). That will typically fail, and unfortunately the 'mock' utility generates a random hostname that happens to have the same format as an id128, thus causing a test failure.
2016-07-26string-util: rework memory_erase() to not use GCC optimize attribute (#3812)Michael Biebl
"#pragma GCC optimize" is merely a convenience to decorate multiple functions with attribute optimize. And the manual has this to say about this attribute: This attribute should be used for debugging purposes only. It is not suitable in production code. Some versions of GCC also seem to have a problem with this pragma in combination with LTO, resulting in ICEs. So use a different approach (indirect the memset call via a volatile function pointer) as implemented in openssl's crypto/mem_clr.c. Closes: #3811
2016-07-26nspawn: add SYSTEMD_NSPAWN_USE_CGNS env variable (#3809)Christian Brauner
SYSTEMD_NSPAWN_USE_CGNS allows to disable the use of cgroup namespaces.
2016-07-26systemctl: allow disable on the unit file path, but warn about it (#3806)Michal Sekletar
systemd now returns an error when it is asked to perform disable on the unit file path. In the past this was allowed, but systemd never really considered an actual content of the [Install] section of the unit file. Instead it performed disable on the unit name, i.e. purged all symlinks pointing to the given unit file (undo of implicit link action done by systemd when enable is called on the unit file path) and all symlinks that have the same basename as the given unit file. However, to notice that [Install] info of the file is not consulted one must create additional symlinks manually. I argue that in most cases users do not create such links. Let's be nice to our users and don't break existing scripts that expect disable to work with the unit file path. Fixes #3706.
2016-07-26journald: deprecate SplitMode=login (#3805)Zbigniew Jędrzejewski-Szmek
In this mode, messages from processes which are not part of the session land in the main journal file, and only output of processes which are properly part of the session land in the user's journal. This is confusing, in particular because systemd-coredump runs outside of the login session. "Deprecate" SplitMode=login by removing it from documentation, to discourage people from using it.
2016-07-25Merge pull request #3728 from poettering/dynamic-usersZbigniew Jędrzejewski-Szmek
2016-07-25Merge pull request #3757 from poettering/efi-searchZbigniew Jędrzejewski-Szmek
2016-07-25Merge pull request #3589 from brauner/cgroup_namespaceLennart Poettering
Cgroup namespace
2016-07-25Merge pull request #3802 from poettering/id128-fixesZbigniew Jędrzejewski-Szmek
Id128 fixes and more
2016-07-25Merge pull request #3800 from keszybz/systemctl-more-cleanupLennart Poettering
Systemctl more cleanup
2016-07-25sd-id128: be more liberal when reading files with 128bit IDsLennart Poettering
Accept both files with and without trailing newlines. Apparently some rkt releases generated them incorrectly, missing the trailing newlines, and we shouldn't break that.
2016-07-25fileio: imply /tmp as directory if passed as NULL to open_tmpfile_unlinkable()Lennart Poettering
We can make this smarter one day, to honour $TMPDIR and friends, but for now, let's just use /tmp.
2016-07-25automount: don't cancel mount/umount request on reload/reexec (#3670)Michael Olbrich
All pending tokens are already serialized correctly and will be handled when the mount unit is done. Without this a 'daemon-reload' cancels all pending tokens. Any process waiting for the mount will continue with EHOSTDOWN. This can happen when the mount unit waits for it's dependencies, e.g. network, devices, fsck, etc.
2016-07-25transaction: don't cancel jobs for units with IgnoreOnIsolate=true (#3671)Michael Olbrich
This is important if a job was queued for a unit but not yet started. Without this, the job will be canceled and is never executed even though IgnoreOnIsolate it set to 'true'.
2016-07-25systemctl: use _cleanup_ for UnitConditionZbigniew Jędrzejewski-Szmek
2016-07-25systemctl: simplify machine_info_clearZbigniew Jędrzejewski-Szmek
It is only used with info allocated on the stack, so the pointer cannot be NULL.
2016-07-25coredump: turn off coredump collection only when PID 1 crashes, not when ↵Lennart Poettering
journald crashes (#3799) As suggested: https://github.com/systemd/systemd/pull/3783/files/5157879b757bffce3da0a68ca207753569e8627d#r71906971
2016-07-25systemctl: avoid "leaking" some strings in UnitStatusInfoZbigniew Jędrzejewski-Szmek
% valgrind --leak-check=full systemctl status multipathd.service --no-pager -n0 ... ==431== 16 bytes in 2 blocks are definitely lost in loss record 1 of 2 ==431== at 0x4C2BBAD: malloc (vg_replace_malloc.c:299) ==431== by 0x534AF19: strdup (in /usr/lib64/libc-2.23.so) ==431== by 0x4E81AEE: free_and_strdup (string-util.c:794) ==431== by 0x4EF66C1: map_basic (bus-util.c:1030) ==431== by 0x4EF6A8E: bus_message_map_all_properties (bus-util.c:1153) ==431== by 0x120487: show_one (systemctl.c:4672) ==431== by 0x1218F3: show (systemctl.c:4990) ==431== by 0x4EC359E: dispatch_verb (verbs.c:92) ==431== by 0x12A3AE: systemctl_main (systemctl.c:7742) ==431== by 0x12B1A8: main (systemctl.c:8011) ==431== ==431== LEAK SUMMARY: ==431== definitely lost: 16 bytes in 2 blocks This happens because map_basic() strdups the strings. Other code in systemctl assigns strings to UnitStatusInfo without copying them, relying on the fact that the message is longer lived than UnitStatusInfo. Add a helper function that is similar to map_basic, but only accepts strings and does not copy them. The alternative of continuing to use map_basic() but adding proper cleanup to free fields in UnitStatusInfo seems less attractive because it'd require changing a lot of code and doing a lot of more allocations for little gain. (I put "leaking" in quotes, because systemctl is short lived anyway.)
2016-07-25systemctl: use cleanup function for UnitStatusInfoZbigniew Jędrzejewski-Szmek
There is no functional change, but clarity of the code is increased by splitting out the cleanup part and putting it next to the structure definition.
2016-07-25Merge pull request #3681 from walyong/systemctl_conditionZbigniew Jędrzejewski-Szmek
2016-07-25systemctl: style tweaks for the new condition codeZbigniew Jędrzejewski-Szmek
2016-07-25core: change ExecStart=! syntax to ExecStart=+ (#3797)Lennart Poettering
As suggested by @mbiebl we already use the "!" special char in unit file assignments for negation, hence we should not use it in a different context for privileged execution. Let's use "+" instead.
2016-07-25shared/install: allow "enable" on linked unit files (#3790)Zbigniew Jędrzejewski-Szmek
User expectations are broken when "systemctl enable /some/path/service.service" behaves differently to "systemctl link ..." followed by "systemctl enable". From user's POV, "enable" with the full path just combines the two steps into one. Fixes #3010.
2016-07-25Merge pull request #3796 from poettering/mailmapMartin Pitt
documentation, NEWS and mailmap fixes
2016-07-25namespace: don't fail on masked mounts (#3794)Alban Crequy
Before this patch, a service file with ReadWriteDirectories=/file... could fail if the file exists but is not a mountpoint, despite being listed in /proc/self/mountinfo. It could happen with masked mounts. Fixes https://github.com/systemd/systemd/issues/3793
2016-07-25man: update systemctl man page for unit file commands, in particular ↵Lennart Poettering
"systemctl enable" Clarify that "systemctl enable" can operate either on unit names or on unit file paths (also, adjust the --help text to clarify this). Say that "systemctl enable" on unit file paths also links the unit into the search path. Many other fixes. This should improve the documentation to avoid further confusion around #3706.
2016-07-22nspawn: don't skip cleanup on locking errorZbigniew Jędrzejewski-Szmek
2016-07-22import: don't log "fake" errno valuesZbigniew Jędrzejewski-Szmek
2016-07-22Use "return log_error_errno" in more places"Zbigniew Jędrzejewski-Szmek
2016-07-22Merge pull request #3777 from poettering/id128-reworkZbigniew Jędrzejewski-Szmek
uuid/id128 code rework
2016-07-22coredump: turn off coredump collection entirely after journald or PID 1 crashedLennart Poettering
Safe is safe, let's turn off the whole logic if we can, after all it is unlikely we'll be able to process further crashes in a reasonable way.
2016-07-22coredump: make sure to handle crashes of PID 1 and journald specialLennart Poettering
Fixes: #3285
2016-07-22Merge pull request #3753 from poettering/tasks-max-scaleLennart Poettering
Add support for relative TasksMax= specifications, and bump default for services
2016-07-22coredump: truncate overly long coredump metadata fields (#3780)Lennart Poettering
Fixes: #3573 Replaces: #3588
2016-07-22systemctl: never check inhibitors if -H or -M are used (#3781)Lennart Poettering
Don't check inhibitors when operating remotely. The interactivity inhibitors imply can#t be provided anyway, and the current code checks for local sessions directly, via various sd_session_xyz() APIs, hence bypass it entirely if we operate on remote systems. Fixes: #3476
2016-07-22cgroup: whitelist inaccessible devices for "auto" and "closed" DevicePolicy.Alessandro Puccetti
https://github.com/systemd/systemd/pull/3685 introduced /run/systemd/inaccessible/{chr,blk} to map inacessible devices, this patch allows systemd running inside a nspawn container to create /run/systemd/inaccessible/{chr,blk}.
2016-07-22nspawn: set DevicesPolicy closed and clean up duplicated devicesAlessandro Puccetti
2016-07-22namespace: ensure to return a valid inaccessible nodes (#3778)Alessandro Puccetti
Because /run/systemd/inaccessible/{chr,blk} are devices with major=0 and minor=0 it might be possible that these devices cannot be created so we use /run/systemd/inaccessible/sock instead to map them.
2016-07-22nss: add new "nss-systemd" NSS module for mapping dynamic usersLennart Poettering
With this NSS module all dynamic service users will be resolvable via NSS like any real user.
2016-07-22core: enforce user/group name validity also when creating transient unitsLennart Poettering
2016-07-22core: add a concept of "dynamic" user ids, that are allocated as long as a ↵Lennart Poettering
service is running This adds a new boolean setting DynamicUser= to service files. If set, a new user will be allocated dynamically when the unit is started, and released when it is stopped. The user ID is allocated from the range 61184..65519. The user will not be added to /etc/passwd (but an NSS module to be added later should make it show up in getent passwd). For now, care should be taken that the service writes no files to disk, since this might result in files owned by UIDs that might get assigned dynamically to a different service later on. Later patches will tighten sandboxing in order to ensure that this cannot happen, except for a few selected directories. A simple way to test this is: systemd-run -p DynamicUser=1 /bin/sleep 99999
2016-07-22core: be stricter when parsing User=/Group= fieldsLennart Poettering
Let's verify the validity of the syntax of the user/group names set.
2016-07-22sysusers: move various user credential validity checks to src/basic/Lennart Poettering
This way we can reuse them for validating User=/Group= settings in unit files (to be added in a later commit). Also, add some tests for them.
2016-07-22core: check for overflow when handling scaled MemoryLimit= settingsLennart Poettering
Just in case...
2016-07-22macros.systemd.in: add %systemd_ordering (#3776)Harald Hoyer
To remove the hard dependency on systemd, for packages, which function without a running systemd the %systemd_ordering macro can be used to ensure ordering in the rpm transaction. %systemd_ordering makes sure, the systemd rpm is installed prior to the package, so the %pre/%post scripts can execute the systemd parts. Installing systemd afterwards though, does not result in the same outcome.
2016-07-22core: change TasksMax= default for system services to 15%Lennart Poettering
As it turns out 512 is max number of tasks per service is hit by too many applications, hence let's bump it a bit, and make it relative to the system's maximum number of PIDs. With this change the new default is 15%. At the kernel's default pids_max value of 32768 this translates to 4915. At machined's default TasksMax= setting of 16384 this translates to 2457. Why 15%? Because it sounds like a round number and is close enough to 4096 which I was going for, i.e. an eight-fold increase over the old 512 Summary: | on the host | in a container old default | 512 | 512 new default | 4915 | 2457
2016-07-22logind: change TasksMax= value for user logins to 33%Lennart Poettering
Let's change from a fixed value of 12288 tasks per user to a relative value of 33%, which with the kernel's default of 32768 translates to 10813. This is a slight decrease of the limit, for no other reason than "33%" sounding like a nice round number that is close enough to 12288 (which would translate to 37.5%). (Well, it also has the nice effect of still leaving a bit of room in the PID space if there are 3 cooperating evil users that try to consume all PIDs... Also, I like my bikesheds blue). Since the new value is taken relative, and machined's TasksMax= setting defaults to 16384, 33% inside of containers is usually equivalent to 5406, which should still be ample space. To summarize: | on the host | in the container old default | 12288 | 12288 new default | 10813 | 5406
2016-07-22main: simplify things a bit by moving container check into fixup_environment()Lennart Poettering
2016-07-22core: rename MemoryLimitByPhysicalMemory transient property to MemoryLimitScaleLennart Poettering
That way, we can neatly keep this in line with the new TasksMaxScale= option. Note that we didn't release a version with MemoryLimitByPhysicalMemory= yet, hence this change should be unproblematic without breaking API.