github/kata-containers
1
0
Fork 1
mirror of https://github.com/kata-containers/kata-containers.git synced 2025-07-15 16:13:20 +00:00
Code Issues Projects Releases Wiki Activity

runtime: cpuset: when creating container, don't pass cpuset details

Browse Source 
Today we only clear out the cpuset details when doing an update call on
existing container/pods. This works in the case of Kubernetes, but not
in the case where we are explicitly setting the cpuset details at boot
time. For example, if you are running a single container via docker ala:

docker run --cpuset-cpus 0-3 -it alpine sh

What would happen is the cpuset info would be passed in with the
container spec for create container request to the agent. At that point
in time, there'd only be the defualt number of CPUs available in the
guest (1), so you'd be left with cpusets set to 0. Next, we'd hotplug
the vCPUs, providing 0-4 CPUs in the guest, but the cpuset would never
be updated, leaving the application tied to CPU 0.

Ouch.

Until the day we support cpusets in the guest, let's make sure that we
start off clearing the cpuset fields.

Fixes: #1405

Signed-off-by: Eric Ernst <eric.g.ernst@gmail.com>
This commit is contained in:
Eric Ernst 2021-02-11 17:35:11 -08:00 committed by Eric Ernst
parent 96196e102e
commit 3721351324
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/runtime/virtcontainers/kata_agent.go
View File

@ -964,6 +964,10 @@ func (k *kataAgent) constraintGRPCSpec(grpcSpec *grpc.Spec, passSeccomp bool) {
grpcSpec.Linux.Resources.BlockIO = nil
grpcSpec.Linux.Resources.HugepageLimits = nil
grpcSpec.Linux.Resources.Network = nil
if grpcSpec.Linux.Resources.CPU != nil {
grpcSpec.Linux.Resources.CPU.Cpus = ""
grpcSpec.Linux.Resources.CPU.Mems = ""
}
// There are three main reasons to do not apply systemd cgroups in the VM
// - Initrd image doesn't have systemd.