Am 17.06.2013 15:03, schrieb Dave Reisner:
FYI, for those not subscribed to systemd-devel. Doesn't seem like this will affect too much, unless you've been heavily tweaking your services.
It will affect me a lot. I use custom memory and cpu/cpuacct cgroups outside of systemd, which brings me to my question ...
- The functionality to define orthogonal cgroup trees for the various controllers will be removed. In fact we'll likely remove the entire API for setting abritrary per-controller paths for each unit. Instead we will introduce a new concept of "Slices" which will allow you to partition system resources in a tree and move units, users, and machines to arbitrary places in it. There will only be a single cgroup tree, but the various controllers may be enabled/disabled separately for each group, so that individual controllers might only see a subtree of the full tree, but not orthogonal trees anymore. The ConrolGroup= unit setting will go away, and be replaced by Slice= plus EnableControllers= or so.
Is this only a systemd change, a kernel change or both? Does this mean I will no longer be able to move processes in custom cgroups and instead have to enable controllers in a unified tree? I'm not quite sure I understand this change.
- ControlGroupPersistent= will likely go away, systemd will be the only component of the OS that sets up the cgroup tree.
I don't use ControlGroupPersistent, but I like messing with cgroups manually. These explanations are a bit unclear to me. I guess we'll have to wait and see what gets implemented.