kubernetes/pkg/kubelet/cm/helpers.go
Francesco Romani 3bcf4220ec kubelet: devices: skip allocation for running pods
When kubelet initializes, runs admission for pods and possibly
allocated requested resources. We need to distinguish between
node reboot (no containers running) versus kubelet restart (containers
potentially running).

Running pods should always survive kubelet restart.
This means that device allocation on admission should not be attempted,
because if a container requires devices and is still running when kubelet
is restarting, that container already has devices allocated and working.

Thus, we need to properly detect this scenario in the allocation step
and handle it explicitely. We need to inform
the devicemanager about which pods are already running.

Note that if container runtime is down when kubelet restarts, the
approach implemented here won't work. In this scenario, so on kubelet
restart containers will again fail admission, hitting
https://github.com/kubernetes/kubernetes/issues/118559 again.
This scenario should however be pretty rare.

Signed-off-by: Francesco Romani <fromani@redhat.com>
2023-07-12 13:25:36 +02:00

79 lines
2.8 KiB
Go

/*
Copyright 2018 The Kubernetes Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package cm
import (
"context"
"k8s.io/api/core/v1"
"k8s.io/apimachinery/pkg/util/sets"
internalapi "k8s.io/cri-api/pkg/apis"
runtimeapi "k8s.io/cri-api/pkg/apis/runtime/v1"
"k8s.io/klog/v2"
"k8s.io/kubernetes/pkg/kubelet/cm/containermap"
evictionapi "k8s.io/kubernetes/pkg/kubelet/eviction/api"
)
// hardEvictionReservation returns a resourcelist that includes reservation of resources based on hard eviction thresholds.
func hardEvictionReservation(thresholds []evictionapi.Threshold, capacity v1.ResourceList) v1.ResourceList {
if len(thresholds) == 0 {
return nil
}
ret := v1.ResourceList{}
for _, threshold := range thresholds {
if threshold.Operator != evictionapi.OpLessThan {
continue
}
switch threshold.Signal {
case evictionapi.SignalMemoryAvailable:
memoryCapacity := capacity[v1.ResourceMemory]
value := evictionapi.GetThresholdQuantity(threshold.Value, &memoryCapacity)
ret[v1.ResourceMemory] = *value
case evictionapi.SignalNodeFsAvailable:
storageCapacity := capacity[v1.ResourceEphemeralStorage]
value := evictionapi.GetThresholdQuantity(threshold.Value, &storageCapacity)
ret[v1.ResourceEphemeralStorage] = *value
}
}
return ret
}
func buildContainerMapAndRunningSetFromRuntime(ctx context.Context, runtimeService internalapi.RuntimeService) (containermap.ContainerMap, sets.String) {
podSandboxMap := make(map[string]string)
podSandboxList, _ := runtimeService.ListPodSandbox(ctx, nil)
for _, p := range podSandboxList {
podSandboxMap[p.Id] = p.Metadata.Uid
}
runningSet := sets.NewString()
containerMap := containermap.NewContainerMap()
containerList, _ := runtimeService.ListContainers(ctx, nil)
for _, c := range containerList {
if _, exists := podSandboxMap[c.PodSandboxId]; !exists {
klog.InfoS("No PodSandBox found for the container", "podSandboxId", c.PodSandboxId, "containerName", c.Metadata.Name, "containerId", c.Id)
continue
}
podUID := podSandboxMap[c.PodSandboxId]
containerMap.Add(podUID, c.Metadata.Name, c.Id)
if c.State == runtimeapi.ContainerState_CONTAINER_RUNNING {
klog.V(4).InfoS("Container reported running", "podSandboxId", c.PodSandboxId, "podUID", podUID, "containerName", c.Metadata.Name, "containerId", c.Id)
runningSet.Insert(c.Id)
}
}
return containerMap, runningSet
}