When using networkd we currently have no way of ensuring that static
neighbor entries are set when our link comes up. This change adds a new
section to the network definition that allows multiple static neighbors
to be set on a link.
Patches are shown on github with a fixed width (no matter how wide the window
is). When line numbers are high (we have some files with 5 digit line numbers),
the diff does not fit, and horizontal scrolling must be used when viewing the
patch. This is super annoying. Let's reduce the width a bit. I think 109 is
still very wide, but at least the github issue should be alleviated.
The current use of literal + replaceable is pretty ugly as it usually
ends up with cgroup_disable= rendered in quotes, which looks really
weird, and this doesn't conform with others of a similar type (for
example, the earlier `DefaultDependencies=no` discussion in the same
file.
Some controllers (like the CPU controller) have a performance cost that
is non-trivial on certain workloads. While this can be mitigated and
improved to an extent, there will for some controllers always be some
overheads associated with the benefits gained from the controller.
Inside Facebook, the fix applied has been to disable the CPU controller
forcibly with `cgroup_disable=cpu` on the kernel command line.
This presents a problem: to disable or reenable the controller, a reboot
is required, but this is quite cumbersome and slow to do for many
thousands of machines, especially machines where disabling/enabling a
stateful service on a machine is a matter of several minutes.
Currently systemd provides some configuration knobs for these in the
form of `[Default]CPUAccounting`, `[Default]MemoryAccounting`, and the
like. The limitation of these is that Default*Accounting is overrideable
by individual services, of which any one could decide to reenable a
controller within the hierarchy at any point just by using a controller
feature implicitly (eg. `CPUWeight`), even if the use of that CPU
feature could just be opportunistic. Since many services are provided by
the distribution, or by upstream teams at a particular organisation,
it's not a sustainable solution to simply try to find and remove
offending directives from these units.
This commit presents a more direct solution -- a DisableControllers=
directive that forcibly disallows a controller from being enabled within
a subtree.
add new "systemd-run-generator" for running arbitrary commands from the kernel command line as system services using the "systemd.run=" kernel command line switch
Let's simplify things and drop the logic that /var/lib/machines is setup
as auto-growing btrfs loopback file /var/lib/machines.raw.
THis was done in order to make quota available for machine management,
but quite frankly never really worked properly, as we couldn't grow the
file system in sync with its use properly. Moreover philosophically it's
problematic overriding the admin's choice of file system like this.
Let's hence drop this, and simplify things. Deleting code is a good
feeling.
Now that regular file systems provide project quota we could probably
add per-machine quota support based on that, hence the btrfs quota
argument is not that interesting anymore (though btrfs quota is a bit
more powerful as it allows recursive quota, i.e. that the machine pool
gets an overall quota in addition to per-machine quota).
Fixes: #2728
This is also supposed to be preparation for doing #10234 eventually,
where a very similar operation is requested: instead of importing a tree
to /var/lib/machines it would need to be imported into
/var/lib/portables/.
I keep running "systemd-run -t /bin/bash" to quickly get a shell running
in service context. I suspect I am not the only one, hence let's add a
shortcut for it. While we are at it, let's make it smarter, and
automatically inherit the $SHELL of the invoking user as well as the
working directory, and let's imply --pty. --shell (or -S) is hence
equivalent to "-t -d $SHELL".
I find myself testing service management quite often with "systemd-run
-t /bin/bash". For that it is handy if the invoked shell would use the
working directory I am currently in. Hence introduce a shorthand for
that:
$ systemd-run -dt /bin/bash
This will automatically insert a WorkingDirectory= property into the
transient service, pointing to the working directory of the caller.
The way this is used drifted a bit from the original intent. Let's update
the description and add some examples to inspire people to texts that look
less bad during initial boot.
We now don't enable the CPU controller just for CPU accounting if we are
on 4.15+ and using pure unified hierarchy, as this is provided
externally to the CPU controller. This makes CPUAccounting=yes
essentially free, so enabling it by default when it's cheap seems like a
good idea.
We keep adding new features, let's advertise to the host OS what these
are in a new variable LoaderFeatures.
It works a bit like OsIndicationsSupported, but is about Loader
features.
This removes the ability to configure which cgroup controllers to mount
together. Instead, we'll now hardcode that "cpu" and "cpuacct" are
mounted together as well as "net_cls" and "net_prio".
The concept of mounting controllers together has no future as it does
not exist to cgroupsv2. Moreover, the current logic is systematically
broken, as revealed by the discussions in #10507. Also, we surveyed Red
Hat customers and couldn't find a single user of the concept (which
isn't particularly surprising, as it is broken...)
This reduced the (already way too complex) cgroup handling for us, since
we now know whenever we make a change to a cgroup for one controller to
which other controllers it applies.
The concept is redundant and predates the special chars that do the same
in ExecStar=. Let's settle on advertising just the latter, and hide
PermissionsStartOnly= from the docs (even if we continue supporting it).
