深入分析kubelet（8）—— 删除extender resource

上一篇分析了kubelet如何选取挂载哪些gpu，那么对应的就会有删除gpu使用记录的操作。

每次分配的时候都是比较已分配和所有健康的设备

devicesInUse := m.allocatedDevices[resource]
available := m.healthyDevices[resource].Difference(devicesInUse)

删除已分配设备，需要先删除Pod，而Pod和Dev的关系是通过podDevices维护的，所以删除已分配设备不需要管allocatedDevices，而是关注podDevices

// ManagerImpl is the structure in charge of managing Device Plugins.
type ManagerImpl struct {
   // healthyDevices contains all of the registered healthy resourceNames and their exported device IDs.
   healthyDevices map[string]sets.String

   // allocatedDevices contains allocated deviceIds, keyed by resourceName.
   allocatedDevices map[string]sets.String

   // podDevices contains pod to allocated device mapping.
   podDevices        podDevices
}

func (pdev podDevices) delete(pods []string) {
   for _, uid := range pods {
      delete(pdev, uid)
   }
}

// updateAllocatedDevices gets a list of active pods and then frees any Devices that are bound to
// terminated pods. Returns error on failure.
func (m *ManagerImpl) updateAllocatedDevices(activePods []*v1.Pod) {
    activePodUids := sets.NewString()
    for _, pod := range activePods {
        activePodUids.Insert(string(pod.UID))
    }
    allocatedPodUids := m.podDevices.pods()
    podsToBeRemoved := allocatedPodUids.Difference(activePodUids)

    klog.V(3).Infof("pods to be removed: %v", podsToBeRemoved.List())
    m.podDevices.delete(podsToBeRemoved.List())
    // Regenerated allocatedDevices after we update pod allocation information.
    m.allocatedDevices = m.podDevices.devices()
}

以podDevices为切入点，找到对应delete方法，再找到updateAllocatedDevices，这里一切就明了了。

1. 比较active和allocate pod，找到removed pod
2. 删除removed pod devices
3. 根据pod devices更新allocated devices