Add a new config directive called NetClass= to CGroup enabled units.
Allowed values are positive numbers for fix assignments and "auto" for
picking a free value automatically, for which we need to keep track of
dynamically assigned net class IDs of units. Introduce a hash table for
this, and also record the last ID that was given out, so the allocator
can start its search for the next 'hole' from there. This could
eventually be optimized with something like an irb.
The class IDs up to 65536 are considered reserved and won't be
assigned automatically by systemd. This barrier can be made a config
directive in the future.
Values set in unit files are stored in the CGroupContext of the
unit and considered read-only. The actually assigned number (which
may have been chosen dynamically) is stored in the unit itself and
is guaranteed to remain stable as long as the unit is active.
In the CGroup controller, set the configured CGroup net class to
net_cls.classid. Multiple unit may share the same net class ID,
and those which do are linked together.
.nspawn fiels are simple settings files that may accompany container
images and directories and contain settings otherwise passed on the
nspawn command line. This provides an efficient way to attach execution
data directly to containers.
This did not really work out as we had hoped. Trying to do this upstream
introduced several problems that probably makes it better suited as a
downstream patch after all. At any rate, it is not releaseable in the
current state, so we at least need to revert this before the release.
* by adjusting the path to binaries, but not do the same thing to the
search path we end up with inconsistent man-pages. Adjusting the search
path too would be quite messy, and it is not at all obvious that this is
worth the effort, but at any rate it would have to be done before we
could ship this.
* this means that distributed man-pages does not make sense as they depend
on config options, and for better or worse we are still distributing
man pages, so that is something that definitely needs sorting out before
we could ship with this patch.
* we have long held that split-usr is only minimally supported in order
to boot, and something we hope will eventually go away. So before we start
adding even more magic/effort in order to make this work nicely, we should
probably question if it makes sense at all.
In particular, use /lib/systemd instead of /usr/lib/systemd in distributions
like Debian which still have not adopted a /usr merge setup.
Use XML entities from man/custom-entities.ent to replace configured paths while
doing XSLT processing of the original XML files. There was precedent of some
files (such as systemd.generator.xml) which were already using this approach.
This addresses most of the (manual) fixes from this patch:
http://anonscm.debian.org/cgit/pkg-systemd/systemd.git/tree/debian/patches/Fix-paths-in-man-pages.patch?h=experimental-220
The idea of using generic XML entities was presented here:
http://lists.freedesktop.org/archives/systemd-devel/2015-May/032240.html
This patch solves almost all the issues, with the exception of:
- Path to /bin/mount and /bin/umount.
- Generic statements about preference of /lib over /etc.
These will be handled separately by follow up patches.
Tested:
- With default configure settings, ran "make install" to two separate
directories and compared the output to confirm they matched exactly.
- Used a set of configure flags including $CONFFLAGS from Debian:
http://anonscm.debian.org/cgit/pkg-systemd/systemd.git/tree/debian/rules
Installed the tree and confirmed the paths use /lib/systemd instead of
/usr/lib/systemd and that no other unexpected differences exist.
- Confirmed that `make distcheck` still passes.
This patch adds more detail to the description of how path escaping
operates and provides a pointer to the systemd-escape program. Either
would serve to answer the question raised in the bug report, so
hopefully this will allow it to be closed.
https://bugs.freedesktop.org/show_bug.cgi?id=87688
This sentence can be misread to mean that "\x20" is the escape code for
"-" which is the only character explicitly mentioned. This lead to at
least one user loosing hair over why a mount unit for "/foo/bar-baz"
didn't work. The example escape is arbitrary so lets prevent hair loss.
A system that is running on a logical partition (LPAR) provided by
PR/SM has access to physical hardware (except CPU). It is true that
PR/SM abstracts the hardware, but only for sharing purposes.
Details are statet at:
http://publib.boulder.ibm.com/infocenter/eserver/v1r2/topic/eicaz/eicazzlpar.htm
-->--
In other words, PR/SM transforms physical resources into virtual resources so
that many logical partitions can share the same physical resources.
--<--
Still, from the OS point of view, the shared virtual resource is real
hardware. ConditionVirtualization must be set to false if the OS runs
directly on PR/SM (e.g. in an LPAR).
[zj: reorder code so that variables are not allocated when #if-def is
false. Add commit message.]
It is annoying when we have dead links on fd.o.
Add project='man-pages|die-net|archlinux' to <citerefentry>-ies.
In generated html, add external links to
http://man7.org/linux/man-pages/man, http://linux.die.net/man/,
https://www.archlinux.org/.
By default, pages in sections 2 and 4 go to man7, since Michael
Kerrisk is the autorative source on kernel related stuff.
The rest of links goes to linux.die.net, because they have the
manpages.
Except for the pacman stuff, since it seems to be only available from
archlinux.org.
Poor gummiboot gets no link, because gummitboot(8) ain't to be found
on the net. According to common wisdom, that would mean that it does
not exist. But I have seen Kay using it, so I know it does, and
deserves to be found. Can somebody be nice and put it up somewhere?
As Zbigniew pointed out a new ConditionFirstBoot= appears like the nicer
way to hook in systemd-firstboot.service on first boots (those with /etc
unpopulated), so let's do this, and get rid of the generator again.
The DefaultInstance= name is used when enabling template units when only
specifying the template name, but no instance.
Add DefaultInstance=tty1 to getty@.service, so that when the template
itself is enabled an instance for tty1 is created.
This is useful so that we "systemctl preset-all" can work properly,
because we can operate on getty@.service after finding it, and the right
instance is created.
This is a recurring submission and includes corrections to:
- missing words, preposition choice.
- change of /lib to /usr/lib, because that is what most distros are
using as the system-wide location for systemd/udev files.
These specifiers require NSS lookups to work, and we really shouldn't do
them from PID 1 hence. With this change they are now only supported for
user systemd instance, or when the configured user for a unit is root.
itistoday> how do you specify multiple dependencies in a unit file? i've been
googling and can't find this basic thing :-\
itistoday> do you use a comma, or use multiple After= statements?
This is a recurring submission and includes corrections to various
issue spotted. I guess I can just skip over reporting ubiquitous comma
placement fixes…
This is a recurring submission and includes corrections to various
issue spotted. I guess I can just skip over reporting ubiquitous
comma placement fixes…
Highligts in this particular commit:
- the "unsigned" type qualifier is completed to form a full type
"unsigned int"
- alphabetic -> lexicographic (that way we automatically define how
numbers get sorted)
The generator paths are internal implementation details, they should not
be documented explicitly.
We should document where private user units are found however.
In a User-Mode Linux session:
$ systemd-detect-virt
none
Although it is possible to reliably detect virtualization:
$ cat /proc/cpuinfo
processor : 0
vendor_id : User Mode Linux
model name : UML
mode : skas
host : Linux kytes 3.11.0-rc1-00009-ge5fd680 (...)
bogomips : 7007.43
So, grep for the string "\nvendor_id\t: User Mode Linux\n" in
/proc/cpuinfo, and say "uml" when asked.
Especially sentences like "filename ends in .suffix" are easier to
parse if the suffix is surrounded by quotes. In sentences like
"requires a .service unit", where the suffix is used as a class
designation, there is no need to use quotes.
Use proper grammar, word usage, adjective hyphenation, commas,
capitalization, spelling, etc.
To improve readability, some run-on sentences or sentence fragments were
revised.
[zj: remove the space from 'file name', 'host name', and 'time zone'.]
When manpages are displayed on a terminal, <literal>s are indistinguishable
from surrounding text. Add quotes everywhere, remove duplicate quotes,
and tweak a few lists for consistent formatting.
https://bugzilla.redhat.com/show_bug.cgi?id=874631
This change is based on existing usage in systemd and online.
'File-system' may make sense in adjectival form, but man pages
seem to prefer 'file system' even in those situations.
Just as with SMACK, we don't really know if a policy has been
loaded or not, as the policy interface is write-only. Assume
therefore that if ima is present in securityfs that it is
enabled.
Update the man page to reflect that "ima" is a valid option
now as well.
According to Documentation/security/Smack.txt:
In keeping with the intent of Smack, configuration data is minimal
and not strictly required. The most important configuration step is
mounting the smackfs pseudo filesystem.
This means that checking the mount point should be enough.
Everything which is an absolute filename marked with <filename></filename>
lands in the index, unless noindex= attribute is present. Should make
it easier for people to find stuff when they are looking at a file on
disk.
Various formatting errors in manpages are fixed, kernel-install(1) is
restored to formatting sanity.
We generally document the suggested paths, not the paths possible in
weird, non-standard setups. We do this in order to not confuse
administrators/users unnecessarily and to push people to install things
into the same directories on all distributions.
We are PID 1 after all, the really basic building block of the OS.
Unlike for an app there's very little benefit in being entirely
relocatable.
This makes it easier to add substitutions to man pages,
avoiding the separate transformation step.
mkdir -p's are removed from the rule, because xsltproc will
will create directories on it's own.
All in all, two or three forks per man page are avoided,
which should make things marginally faster.
Unfortunately python parsers must too be tweaked to handle
entities. This isn't particularly easy: with lxml a custom
Resolver can be used, but the stdlib etree doesn't support
external entities *at all*. So when running without lxml,
the entities are just removed. Right now it doesn't matter,
since the entities are not indexed anyway. But I intend to
add indexing of filenames in the near future, and then the
index generated without lxml might be missing a few lines.
Oh well.
This allows one templated unit to refer to another templated unit
at installation time.
Examples:
> grep WantedBy ~/.config/systemd/user/mpop@.timer
WantedBy=services@%i.target
> srv disable mpop@iit.timer
rm '/home/alxchk/.config/systemd/user/services@iit.target.wants/mpop@iit.timer'
> srv enable mpop@iit.timer
ln -s '/home/alxchk/.config/systemd/user/mpop@.timer' '/home/alxchk/.config/systemd/user/services@iit.target.wants/mpop@iit.timer'
Based-on-patch-by: Oleksii Shevchuk <alxchk@gmail.com>
New sections are added: PAM options, crypttab options, commandline
options, miscellaneous. The last category will be used for all
untagged <varname> elements.
Commandline options sections is meant to be a developer tool: when
adding an option it is sometimes useful to be able to check if
similarly named options exist elsewhere.
For all unit files foobar.service we will now read
foobar.service.d/*.conf, too. This may be used to override certain unit
settings without having to edit unit files directly.
This makes it really easy to change specific settings for services
without having to edit any unit file:
mkdir /etc/systemd/system/avahi-daemon.service.d/
echo -e '[Service]\nNice=99' > /etc/systemd/system/avahi-daemon.service.d/nice.conf
systemctl daemon-reload
https://bugs.freedesktop.org/show_bug.cgi?id=55890
Fixed typos, serial comma, and removed "either" as there were more
than two options. Also did an extra rename of "system-shutdown"
to "systemd-shutdown" that was forgotten in commit
8bd3b8620c
As you likely know, Arch Linux is in the process of moving to systemd.
So I was reading through the various systemd docs and quickly became
baffled by this new abbreviation "resp.", which I've never seen before
in my English-mother-tongue life.
Some quick Googling turned up a reference:
<http://www.transblawg.eu/index.php?/archives/870-Resp.-and-other-non-existent-English-wordsNicht-existente-englische-Woerter.html>
I guess it's a literal translation of the German "Beziehungsweise", but
English doesn't work the same way. The word "respectively" is used
exclusively to provide an ordering connection between two lists. E.g.
"the prefixes k, M, and G refer to kilo-, mega-, and giga-,
respectively." It is also never abbreviated to "resp." So the sentence
"Sets the default output resp. error output for all services and
sockets" makes no sense to a natural English speaker.
This patch removes all instances of "resp." in the man pages and
replaces them with sentences which are much more clear and, hopefully,
grammatically valid. In almost all instances, it was simply replacing
"resp." with "or," which the original author (Lennart?) could probably
just do in the future.
The only other instances of "resp." are in the src/ subtree, which I
don't feel privileged to correct.
Signed-off-by: Andrew Eikum <aeikum@codeweavers.com>
This should address TODO item "new dependency type to "group" services
in a target". Semantic of new dependency is as follows. Once configured
it creates dependency which will cause that all dependent units get
stopped if unit they all depend on is stopped or restarted. Usual use
case would be configuring PartOf=some.target in template unit file
and WantedBy=some.target in [Install] section and enabling desired
number of instances. In this case starting one instance won't pull in
target but stopping or starting target(in case of WantedBy is properly
configured) will cause stop/start of all instances.
These printf specifiers allow us to refer to $HOME and $USER
in unit files. These are particularly helpful in instanced
units that have "User=" set, and in systemd --user domains.
The specifiers will return the pw_name and pw_dir fields
if the unit file has a User= field.
If the unit file does not have a User= field, the value
substituted is either $USER or $HOME from the environment,
or, if unset, the values from pw_name or pw_dir.
This patch is somewhat after Ran Benita's original patch,
which didn't get merged. I've split up the 2 specifiers
and extended them to do what is logically expected from
these specifiers.
Note that expansion is done at `start` time, not after
the units are parsed. Using `systemctl show` will just
show the specifiers.
all other dependencies are in 3rd person. Change BindTo= accordingly to
BindsTo=.
Of course, the dependency is widely used, hence we parse the old name
too for compatibility.
Names= is a source of errors, simply because alias names specified like
this only become relevant after a unit has been loaded but cannot be
used to load a unit.
Let's get rid of the confusion and drop this field. To establish alias
names peope should use symlinks, which have the the benefit of being
useful as key to load a unit, even though they are not taken into
account if unit names are listed but they haven't been explicitly
referenced before.
UnitPath= is also writable via native units and may be used by generators
to clarify from which file a unit is generated. This patch also hooks up
the cryptsetup and fstab generators to set UnitPath= accordingly.
This should help making the boot process a bit easier to explore and
understand for the administrator. The simple idea is that "systemctl
status" now shows a link to documentation alongside the other status and
decriptionary information of a service.
This patch adds the necessary fields to all our shipped units if we have
proper documentation for them.
RequiresMountsFor= is a shortcut for adding requires and after
dependencies to all mount units neeed for the specified paths.
This solves a couple of issues regarding dep loop cycles for encrypted
swap.
We finally got the OK from all contributors with non-trivial commits to
relicense systemd from GPL2+ to LGPL2.1+.
Some udev bits continue to be GPL2+ for now, but we are looking into
relicensing them too, to allow free copy/paste of all code within
systemd.
The bits that used to be MIT continue to be MIT.
The big benefit of the relicensing is that closed source code may now
link against libsystemd-login.so and friends.
ConditionPathExists and ConditionFileIsExecutable follow symlinks.
ConditionPathIsDirectory does not follow symlinks.
(XXX: is this actually the desired behavior?)
It was noted on IRC that the current layout is not necessarily the
easiest to parse by humans. I personnally struggled understanding it the
first time I read it.
Hopefully, this change makes it easier to understand.
Note: I only reformatted the information that was already present in the
documentation, I didn't add anything.
The virtualization condition and others use lowercase identifiers, so
for the sake of keeping things least surprising, use lowercase
identifiers here too.
The property StopRetroactively= needs to be per-dependency, not
per-unit, in order to properly express dependencies between .mount units
and its .device and fsck .service units. If the .device unit is
unplugged the mount should go away, but if the fsck process terminates
the .mount should stay.
Some unit shall never be start on user request (e.g. shutdown.target)
others never be stopped on user request (e.g. auditd.servce), hence
offer options for both.
Lennart Poettering