| Age | Commit message (Collapse) | Author |
|---|
|
|
|
After talking to the cgroup kernel folks at LPC we came to the
conclusion that it is probably a good idea to mount all CPU related
resp. all network related cgroup controllers together, both because they
are good defaults for admins and because this might prepare
for eventual kernel cleanups where the ability to mount them separately
is removed.
|
|
|
|
|
|
|
|
In the initrd we don't need the flush service hence don't attempt to
pull it in.
|
|
boot phase
|
|
|
|
|
|
|
|
|
|
|
|
__secure_getenv() rather than getenv()
It's better to be safe than sorry.
|
|
|
|
Without this, build fail with --disable-xz or xz not installed.
|
|
https://launchpad.net/bugs/1009836
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
persistant logging is off
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
udev has supported this since 172, so it should be a safe (and welcome)
addition for users of GPT partitioned disks.
|
|
Since this is purely duplicated logic, separate it out into a small
static function.
|
|
|
|
|
|
This creates /dev/fd, /dev/stdin, /dev/stdout, /dev/stderr, and
/dev/core as symlinks to /proc on container creation. Except for
/dev/core, these are needed for shells like bash to be fully functional.
|
|
With this adjustment, we can reuse this code elsewhere, such as in
nspawn.
|
|
|
|
|
|
Spotted by Dave Reisner
|
|
|
|
|
|
|
