Documentation
¶
Overview ¶
Package virtcontainers manages hardware virtualized containers. Each container belongs to a set of containers sharing the same networking namespace and storage, also known as a pod.
Virtcontainers pods are hardware virtualized, i.e. they run on virtual machines. Virtcontainers will create one VM per pod, and containers will be created as processes within the pod VM.
The virtcontainers package manages both pods and containers lifecycles.
Example (CreateAndStartPod) ¶
This example creates and starts a single container pod, using qemu as the hypervisor and hyperstart as the VM agent.
package main
import (
"fmt"
"strings"
vc "github.com/kata-containers/runtime/virtcontainers"
)
const containerRootfs = "/var/lib/container/bundle/"
// This example creates and starts a single container pod,
// using qemu as the hypervisor and hyperstart as the VM agent.
func main() {
envs := []vc.EnvVar{
{
Var: "PATH",
Value: "/bin:/usr/bin:/sbin:/usr/sbin",
},
}
cmd := vc.Cmd{
Args: strings.Split("/bin/sh", " "),
Envs: envs,
WorkDir: "/",
}
// Define the container command and bundle.
container := vc.ContainerConfig{
ID: "1",
RootFs: containerRootfs,
Cmd: cmd,
}
// Sets the hypervisor configuration.
hypervisorConfig := vc.HypervisorConfig{
KernelPath: "/usr/share/kata-containers/vmlinux.container",
ImagePath: "/usr/share/kata-containers/kata-containers.img",
HypervisorPath: "/usr/bin/qemu-system-x86_64",
}
// Use hyperstart default values for the agent.
agConfig := vc.HyperConfig{}
// VM resources
vmConfig := vc.Resources{
Memory: 1024,
}
// The pod configuration:
// - One container
// - Hypervisor is QEMU
// - Agent is hyperstart
podConfig := vc.PodConfig{
VMConfig: vmConfig,
HypervisorType: vc.QemuHypervisor,
HypervisorConfig: hypervisorConfig,
AgentType: vc.HyperstartAgent,
AgentConfig: agConfig,
Containers: []vc.ContainerConfig{container},
}
_, err := vc.RunPod(podConfig)
if err != nil {
fmt.Printf("Could not run pod: %s", err)
}
return
}
Output:
Index ¶
- Constants
- Variables
- func ConstraintsToVCPUs(quota int64, period uint64) uint
- func CreateContainer(podID string, containerConfig ContainerConfig) (VCPod, VCContainer, error)
- func EnterContainer(podID, containerID string, cmd Cmd) (VCPod, VCContainer, *Process, error)
- func GetHostPath(devInfo DeviceInfo) (string, error)
- func KillContainer(podID, containerID string, signal syscall.Signal, all bool) error
- func RunningOnVMM(cpuInfoPath string) (bool, error)
- func SerializeParams(params []Param, delim string) []string
- func SetLogger(logger logrus.FieldLogger)
- type AgentType
- type BlockDevice
- type Bridge
- type CPUDevice
- type Cmd
- type Container
- func (c *Container) GetAnnotations() map[string]string
- func (c *Container) GetPid() int
- func (c *Container) GetToken() string
- func (c *Container) ID() string
- func (c *Container) Logger() *logrus.Entry
- func (c *Container) Pod() VCPod
- func (c *Container) Process() Process
- func (c *Container) SetPid(pid int) error
- type ContainerConfig
- type ContainerResources
- type ContainerStatus
- type ContainerType
- type DNSInfo
- type Device
- type DeviceInfo
- type Drive
- type Endpoint
- type EndpointType
- type EnvVar
- type GenericDevice
- type Hook
- type Hooks
- type HyperAgentState
- type HyperConfig
- type HypervisorConfig
- func (conf *HypervisorConfig) AddKernelParam(p Param) error
- func (conf *HypervisorConfig) CustomFirmwareAsset() bool
- func (conf *HypervisorConfig) CustomHypervisorAsset() bool
- func (conf *HypervisorConfig) CustomImageAsset() bool
- func (conf *HypervisorConfig) CustomInitrdAsset() bool
- func (conf *HypervisorConfig) CustomKernelAsset() bool
- func (conf *HypervisorConfig) FirmwareAssetPath() (string, error)
- func (conf *HypervisorConfig) HypervisorAssetPath() (string, error)
- func (conf *HypervisorConfig) ImageAssetPath() (string, error)
- func (conf *HypervisorConfig) InitrdAssetPath() (string, error)
- func (conf *HypervisorConfig) KernelAssetPath() (string, error)
- type HypervisorType
- type KataAgentConfig
- type KataAgentState
- type KataShimConfig
- type LinuxCapabilities
- type Mount
- type NetInterworkingModel
- type NetlinkIface
- type NetworkConfig
- type NetworkInfo
- type NetworkInterface
- type NetworkInterfacePair
- type NetworkModel
- type NetworkNamespace
- type Param
- type PhysicalEndpoint
- func (endpoint *PhysicalEndpoint) Attach(h hypervisor) error
- func (endpoint *PhysicalEndpoint) Detach() error
- func (endpoint *PhysicalEndpoint) HardwareAddr() string
- func (endpoint *PhysicalEndpoint) Name() string
- func (endpoint *PhysicalEndpoint) Properties() NetworkInfo
- func (endpoint *PhysicalEndpoint) SetProperties(properties NetworkInfo)
- func (endpoint *PhysicalEndpoint) Type() EndpointType
- type Pod
- func (p *Pod) Annotations(key string) (string, error)
- func (p *Pod) GetAllContainers() []VCContainer
- func (p *Pod) GetAnnotations() map[string]string
- func (p *Pod) GetContainer(containerID string) VCContainer
- func (p *Pod) ID() string
- func (p *Pod) Logger() *logrus.Entry
- func (p *Pod) SetAnnotations(annotations map[string]string) error
- type PodConfig
- type PodStatus
- type Process
- type ProcessList
- type ProcessListOptions
- type ProxyConfig
- type ProxyType
- type QemuState
- type Resources
- type ShimConfig
- type ShimParams
- type ShimType
- type Socket
- type Sockets
- type SpawnerType
- type State
- type SystemMountsInfo
- type TypedDevice
- type TypedJSONEndpoint
- type VC
- type VCContainer
- type VCImpl
- func (impl *VCImpl) CreateContainer(podID string, containerConfig ContainerConfig) (VCPod, VCContainer, error)
- func (impl *VCImpl) CreatePod(podConfig PodConfig) (VCPod, error)
- func (impl *VCImpl) DeleteContainer(podID, containerID string) (VCContainer, error)
- func (impl *VCImpl) DeletePod(podID string) (VCPod, error)
- func (impl *VCImpl) EnterContainer(podID, containerID string, cmd Cmd) (VCPod, VCContainer, *Process, error)
- func (impl *VCImpl) KillContainer(podID, containerID string, signal syscall.Signal, all bool) error
- func (impl *VCImpl) ListPod() ([]PodStatus, error)
- func (impl *VCImpl) PausePod(podID string) (VCPod, error)
- func (impl *VCImpl) ProcessListContainer(podID, containerID string, options ProcessListOptions) (ProcessList, error)
- func (impl *VCImpl) ResumePod(podID string) (VCPod, error)
- func (impl *VCImpl) RunPod(podConfig PodConfig) (VCPod, error)
- func (impl *VCImpl) SetLogger(logger logrus.FieldLogger)
- func (impl *VCImpl) StartContainer(podID, containerID string) (VCContainer, error)
- func (impl *VCImpl) StartPod(podID string) (VCPod, error)
- func (impl *VCImpl) StatusContainer(podID, containerID string) (ContainerStatus, error)
- func (impl *VCImpl) StatusPod(podID string) (PodStatus, error)
- func (impl *VCImpl) StopContainer(podID, containerID string) (VCContainer, error)
- func (impl *VCImpl) StopPod(podID string) (VCPod, error)
- type VCPod
- func CreatePod(podConfig PodConfig) (VCPod, error)
- func DeletePod(podID string) (VCPod, error)
- func PausePod(podID string) (VCPod, error)
- func ResumePod(podID string) (VCPod, error)
- func RunPod(podConfig PodConfig) (VCPod, error)
- func StartPod(podID string) (VCPod, error)
- func StopPod(podID string) (VCPod, error)
- type VFIODevice
- type VhostUserBlkDevice
- type VhostUserDevice
- type VhostUserDeviceAttrs
- type VhostUserDeviceType
- type VhostUserEndpoint
- func (endpoint *VhostUserEndpoint) Attach(h hypervisor) error
- func (endpoint *VhostUserEndpoint) Detach() error
- func (endpoint *VhostUserEndpoint) HardwareAddr() string
- func (endpoint *VhostUserEndpoint) Name() string
- func (endpoint *VhostUserEndpoint) Properties() NetworkInfo
- func (endpoint *VhostUserEndpoint) SetProperties(properties NetworkInfo)
- func (endpoint *VhostUserEndpoint) Type() EndpointType
- type VhostUserNetDevice
- type VhostUserSCSIDevice
- type VirtualEndpoint
- func (endpoint *VirtualEndpoint) Attach(h hypervisor) error
- func (endpoint *VirtualEndpoint) Detach() error
- func (endpoint *VirtualEndpoint) HardwareAddr() string
- func (endpoint *VirtualEndpoint) Name() string
- func (endpoint *VirtualEndpoint) Properties() NetworkInfo
- func (endpoint *VirtualEndpoint) SetProperties(properties NetworkInfo)
- func (endpoint *VirtualEndpoint) Type() EndpointType
- type Volume
- type Volumes
Examples ¶
Constants ¶
const ( // DeviceVFIO is the VFIO device type DeviceVFIO = "vfio" // DeviceBlock is the block device type DeviceBlock = "block" // DeviceGeneric is a generic device type DeviceGeneric = "generic" )
const ( //VhostUserSCSI - SCSI based vhost-user type VhostUserSCSI = "vhost-user-scsi-pci" //VhostUserNet - net based vhost-user type VhostUserNet = "virtio-net-pci" //VhostUserBlk represents a block vhostuser device type VhostUserBlk = "vhost-user-blk-pci" )
const ( // StateReady represents a pod/container that's ready to be run StateReady stateString = "ready" // StateRunning represents a pod/container that's currently running. StateRunning stateString = "running" // StatePaused represents a pod/container that has been paused. StatePaused stateString = "paused" // StateStopped represents a pod/container that has been stopped. StateStopped stateString = "stopped" )
const ( // VirtioBlock means use virtio-blk for hotplugging drives VirtioBlock = "virtio-blk" // VirtioSCSI means use virtio-scsi for hotplugging drives VirtioSCSI = "virtio-scsi" )
const ( // QemuPCLite is the QEMU pc-lite machine type for amd64 QemuPCLite = "pc-lite" // QemuPC is the QEMU pc machine type for amd64 QemuPC = "pc" // QemuQ35 is the QEMU Q35 machine type for amd64 QemuQ35 = "q35" // QemuVirt is the QEMU virt machine type for aarch64 QemuVirt = "virt" )
const CniPrimaryInterface = "eth0"
CniPrimaryInterface Name chosen for the primary interface If CNI ever support multiple primary interfaces this should be revisited
Variables ¶
var DefaultNetInterworkingModel = NetXConnectMacVtapModel
DefaultNetInterworkingModel is a package level default that determines how the VM should be connected to the the container network interface
var GetHostPathFunc = GetHostPath
GetHostPathFunc is function pointer used to mock GetHostPath in tests.
Functions ¶
func ConstraintsToVCPUs ¶
ConstraintsToVCPUs converts CPU quota and period to vCPUs
func CreateContainer ¶
func CreateContainer(podID string, containerConfig ContainerConfig) (VCPod, VCContainer, error)
CreateContainer is the virtcontainers container creation entry point. CreateContainer creates a container on a given pod.
func EnterContainer ¶
EnterContainer is the virtcontainers container command execution entry point. EnterContainer enters an already running container and runs a given command.
func GetHostPath ¶
func GetHostPath(devInfo DeviceInfo) (string, error)
GetHostPath is used to fetcg the host path for the device. The path passed in the spec refers to the path that should appear inside the container. We need to find the actual device path on the host based on the major-minor numbers of the device.
func KillContainer ¶
KillContainer is the virtcontainers entry point to send a signal to a container running inside a pod. If all is true, all processes in the container will be sent the signal.
func RunningOnVMM ¶
RunningOnVMM checks if the system is running inside a VM.
func SerializeParams ¶
SerializeParams converts []Param to []string
func SetLogger ¶
func SetLogger(logger logrus.FieldLogger)
SetLogger sets the logger for virtcontainers package.
Types ¶
type AgentType ¶
type AgentType string
AgentType describes the type of guest agent a Pod should run.
const ( // NoopAgentType is the No-Op agent. NoopAgentType AgentType = "noop" // HyperstartAgent is the Hyper hyperstart agent. HyperstartAgent AgentType = "hyperstart" // KataContainersAgent is the Kata Containers agent. KataContainersAgent AgentType = "kata" // SocketTypeVSOCK is a VSOCK socket type for talking to an agent. SocketTypeVSOCK = "vsock" // SocketTypeUNIX is a UNIX socket type for talking to an agent. // It typically means the agent is living behind a host proxy. SocketTypeUNIX = "unix" )
type BlockDevice ¶
type BlockDevice struct {
DeviceType string
DeviceInfo DeviceInfo
// SCSI Address of the block device, in case the device is attached using SCSI driver
// SCSI address is in the format SCSI-Id:LUN
SCSIAddr string
// Path at which the device appears inside the VM, outside of the container mount namespace
VirtPath string
}
BlockDevice refers to a block storage device implementation.
type Bridge ¶
type Bridge struct {
// Address contains information about devices plugged and its address in the bridge
Address map[uint32]string
// Type is the type of the bridge (pci, pcie, etc)
Type bridgeType
//ID is used to identify the bridge in the hypervisor
ID string
}
Bridge is a bridge where devices can be hot plugged
type CPUDevice ¶
type CPUDevice struct {
// ID is used to identify this CPU in the hypervisor options.
ID string
}
CPUDevice represents a CPU device which was hot-added in a running VM
type Cmd ¶
type Cmd struct {
Args []string
Envs []EnvVar
SupplementaryGroups []string
// Note that these fields *MUST* remain as strings.
//
// The reason being that we want runtimes to be able to support CLI
// operations like "exec --user=". That option allows the
// specification of a user (either as a string username or a numeric
// UID), and may optionally also include a group (groupame or GID).
//
// Since this type is the interface to allow the runtime to specify
// the user and group the workload can run as, these user and group
// fields cannot be encoded as integer values since that would imply
// the runtime itself would need to perform a UID/GID lookup on the
// user-specified username/groupname. But that isn't practically
// possible given that to do so would require the runtime to access
// the image to allow it to interrogate the appropriate databases to
// convert the username/groupnames to UID/GID values.
//
// Note that this argument applies solely to the _runtime_ supporting
// a "--user=" option when running in a "standalone mode" - there is
// no issue when the runtime is called by a container manager since
// all the user and group mapping is handled by the container manager
// and specified to the runtime in terms of UID/GID's in the
// configuration file generated by the container manager.
User string
PrimaryGroup string
WorkDir string
Console string
Capabilities LinuxCapabilities
Interactive bool
Detach bool
NoNewPrivileges bool
}
Cmd represents a command to execute in a running container.
type Container ¶
type Container struct {
// contains filtered or unexported fields
}
Container is composed of a set of containers and a runtime environment. A Container can be created, deleted, started, stopped, listed, entered, paused and restored.
func (*Container) GetAnnotations ¶
GetAnnotations returns container's annotations
func (*Container) Logger ¶
Logger returns a logrus logger appropriate for logging Container messages
type ContainerConfig ¶
type ContainerConfig struct {
ID string
// RootFs is the container workload image on the host.
RootFs string
// ReadOnlyRootfs indicates if the rootfs should be mounted readonly
ReadonlyRootfs bool
// Cmd specifies the command to run on a container
Cmd Cmd
// Annotations allow clients to store arbitrary values,
// for example to add additional status values required
// to support particular specifications.
Annotations map[string]string
Mounts []Mount
// Device configuration for devices that must be available within the container.
DeviceInfos []DeviceInfo
// Resources container resources
Resources ContainerResources
}
ContainerConfig describes one container runtime configuration.
type ContainerResources ¶
type ContainerResources struct {
// CPUQuota specifies the total amount of time in microseconds
// The number of microseconds per CPUPeriod that the container is guaranteed CPU access
CPUQuota int64
// CPUPeriod specifies the CPU CFS scheduler period of time in microseconds
CPUPeriod uint64
CPUShares uint64
}
ContainerResources describes container resources
type ContainerStatus ¶
type ContainerStatus struct {
ID string
State State
PID int
StartTime time.Time
RootFs string
// Annotations allow clients to store arbitrary values,
// for example to add additional status values required
// to support particular specifications.
Annotations map[string]string
}
ContainerStatus describes a container status.
func StatusContainer ¶
func StatusContainer(podID, containerID string) (ContainerStatus, error)
StatusContainer is the virtcontainers container status entry point. StatusContainer returns a detailed container status.
type ContainerType ¶
type ContainerType string
ContainerType defines a type of container.
var ( PodContainer ContainerType = "pod_container" PodSandbox ContainerType = "pod_sandbox" UnknownContainerType ContainerType = "unknown_container_type" )
List different types of containers
func (ContainerType) IsPod ¶
func (cType ContainerType) IsPod() bool
IsPod determines if the container type can be considered as a pod. We can consider a pod in case we have a PodSandbox or a RegularContainer.
type Device ¶
type Device interface {
// contains filtered or unexported methods
}
Device is the virtcontainers device interface.
type DeviceInfo ¶
type DeviceInfo struct {
// Device path on host
HostPath string
// Device path inside the container
ContainerPath string
// Type of device: c, b, u or p
// c , u - character(unbuffered)
// p - FIFO
// b - block(buffered) special file
// More info in mknod(1).
DevType string
// Major, minor numbers for device.
Major int64
Minor int64
// FileMode permission bits for the device.
FileMode os.FileMode
// id of the device owner.
UID uint32
// id of the device group.
GID uint32
// Hotplugged is used to store device state indicating if the
// device was hotplugged.
Hotplugged bool
// ID for the device that is passed to the hypervisor.
ID string
}
DeviceInfo is an embedded type that contains device data common to all types of devices.
type Drive ¶
type Drive struct {
// Path to the disk-image/device which will be used with this drive
File string
// Format of the drive
Format string
// ID is used to identify this drive in the hypervisor options.
ID string
// Index assigned to the drive. In case of virtio-scsi, this is used as SCSI LUN index
Index int
}
Drive represents a block storage drive which may be used in case the storage driver has an underlying block storage device.
type Endpoint ¶
type Endpoint interface {
Properties() NetworkInfo
Name() string
HardwareAddr() string
Type() EndpointType
SetProperties(NetworkInfo)
Attach(hypervisor) error
Detach() error
}
Endpoint represents a physical or virtual network interface.
type EndpointType ¶
type EndpointType string
EndpointType identifies the type of the network endpoint.
const ( // PhysicalEndpointType is the physical network interface. PhysicalEndpointType EndpointType = "physical" // VirtualEndpointType is the virtual network interface. VirtualEndpointType EndpointType = "virtual" // VhostUserEndpointType is the vhostuser network interface. VhostUserEndpointType EndpointType = "vhost-user" )
func (*EndpointType) Set ¶
func (endpointType *EndpointType) Set(value string) error
Set sets an endpoint type based on the input string.
func (*EndpointType) String ¶
func (endpointType *EndpointType) String() string
String converts an endpoint type to a string.
type GenericDevice ¶
type GenericDevice struct {
DeviceType string
DeviceInfo DeviceInfo
}
GenericDevice refers to a device that is neither a VFIO device or block device.
type HyperAgentState ¶
HyperAgentState is the structure describing the data stored from this agent implementation.
type HyperConfig ¶
HyperConfig is a structure storing information needed for hyperstart agent initialization.
type HypervisorConfig ¶
type HypervisorConfig struct {
// KernelParams are additional guest kernel parameters.
KernelParams []Param
// HypervisorParams are additional hypervisor parameters.
HypervisorParams []Param
// KernelPath is the guest kernel host path.
KernelPath string
// ImagePath is the guest image host path.
ImagePath string
// InitrdPath is the guest initrd image host path.
// ImagePath and InitrdPath cannot be set at the same time.
InitrdPath string
// FirmwarePath is the bios host path
FirmwarePath string
// MachineAccelerators are machine specific accelerators
MachineAccelerators string
// HypervisorPath is the hypervisor executable host path.
HypervisorPath string
// BlockDeviceDriver specifies the driver to be used for block device
// either VirtioSCSI or VirtioBlock with the default driver being defaultBlockDriver
BlockDeviceDriver string
// HypervisorMachineType specifies the type of machine being
// emulated.
HypervisorMachineType string
// DefaultVCPUs specifies default number of vCPUs for the VM.
DefaultVCPUs uint32
//DefaultMaxVCPUs specifies the maximum number of vCPUs for the VM.
DefaultMaxVCPUs uint32
// DefaultMem specifies default memory size in MiB for the VM.
// Pod configuration VMConfig.Memory overwrites this.
DefaultMemSz uint32
// DefaultBridges specifies default number of bridges for the VM.
// Bridges can be used to hot plug devices
DefaultBridges uint32
// DisableBlockDeviceUse disallows a block device from being used.
DisableBlockDeviceUse bool
// Debug changes the default hypervisor and kernel parameters to
// enable debug output where available.
Debug bool
// MemPrealloc specifies if the memory should be pre-allocated
MemPrealloc bool
// HugePages specifies if the memory should be pre-allocated from huge pages
HugePages bool
// Realtime Used to enable/disable realtime
Realtime bool
// Mlock is used to control memory locking when Realtime is enabled
// Realtime=true and Mlock=false, allows for swapping out of VM memory
// enabling higher density
Mlock bool
// DisableNestingChecks is used to override customizations performed
// when running on top of another VMM.
DisableNestingChecks bool
// contains filtered or unexported fields
}
HypervisorConfig is the hypervisor configuration.
func (*HypervisorConfig) AddKernelParam ¶
func (conf *HypervisorConfig) AddKernelParam(p Param) error
AddKernelParam allows the addition of new kernel parameters to an existing hypervisor configuration.
func (*HypervisorConfig) CustomFirmwareAsset ¶
func (conf *HypervisorConfig) CustomFirmwareAsset() bool
CustomFirmwareAsset returns true if the firmware asset is a custom one, false otherwise.
func (*HypervisorConfig) CustomHypervisorAsset ¶
func (conf *HypervisorConfig) CustomHypervisorAsset() bool
CustomHypervisorAsset returns true if the hypervisor asset is a custom one, false otherwise.
func (*HypervisorConfig) CustomImageAsset ¶
func (conf *HypervisorConfig) CustomImageAsset() bool
CustomImageAsset returns true if the image asset is a custom one, false otherwise.
func (*HypervisorConfig) CustomInitrdAsset ¶
func (conf *HypervisorConfig) CustomInitrdAsset() bool
CustomInitrdAsset returns true if the initrd asset is a custom one, false otherwise.
func (*HypervisorConfig) CustomKernelAsset ¶
func (conf *HypervisorConfig) CustomKernelAsset() bool
CustomKernelAsset returns true if the kernel asset is a custom one, false otherwise.
func (*HypervisorConfig) FirmwareAssetPath ¶
func (conf *HypervisorConfig) FirmwareAssetPath() (string, error)
FirmwareAssetPath returns the guest firmware path
func (*HypervisorConfig) HypervisorAssetPath ¶
func (conf *HypervisorConfig) HypervisorAssetPath() (string, error)
HypervisorAssetPath returns the VM hypervisor path
func (*HypervisorConfig) ImageAssetPath ¶
func (conf *HypervisorConfig) ImageAssetPath() (string, error)
ImageAssetPath returns the guest image path
func (*HypervisorConfig) InitrdAssetPath ¶
func (conf *HypervisorConfig) InitrdAssetPath() (string, error)
InitrdAssetPath returns the guest initrd path
func (*HypervisorConfig) KernelAssetPath ¶
func (conf *HypervisorConfig) KernelAssetPath() (string, error)
KernelAssetPath returns the guest kernel path
type HypervisorType ¶
type HypervisorType string
HypervisorType describes an hypervisor type.
const ( // QemuHypervisor is the QEMU hypervisor. QemuHypervisor HypervisorType = "qemu" // MockHypervisor is a mock hypervisor for testing purposes MockHypervisor HypervisorType = "mock" )
func (*HypervisorType) Set ¶
func (hType *HypervisorType) Set(value string) error
Set sets an hypervisor type based on the input string.
func (*HypervisorType) String ¶
func (hType *HypervisorType) String() string
String converts an hypervisor type to a string.
type KataAgentConfig ¶
type KataAgentConfig struct {
GRPCSocket string
}
KataAgentConfig is a structure storing information needed to reach the Kata Containers agent.
type KataAgentState ¶
KataAgentState is the structure describing the data stored from this agent implementation.
type KataShimConfig ¶
KataShimConfig is the structure providing specific configuration for kataShim implementation.
type LinuxCapabilities ¶
type LinuxCapabilities struct {
// Bounding is the set of capabilities checked by the kernel.
Bounding []string
// Effective is the set of capabilities checked by the kernel.
Effective []string
// Inheritable is the capabilities preserved across execve.
Inheritable []string
// Permitted is the limiting superset for effective capabilities.
Permitted []string
// Ambient is the ambient set of capabilities that are kept.
Ambient []string
}
LinuxCapabilities specify the capabilities to keep when executing the process inside the container.
type Mount ¶
type Mount struct {
Source string
Destination string
// Type specifies the type of filesystem to mount.
Type string
// Options list all the mount options of the filesystem.
Options []string
// HostPath used to store host side bind mount path
HostPath string
// ReadOnly specifies if the mount should be read only or not
ReadOnly bool
}
Mount describes a container mount.
type NetInterworkingModel ¶
type NetInterworkingModel int
NetInterworkingModel defines the network model connecting the network interface to the virtual machine.
const ( // NetXConnectDefaultModel Ask to use DefaultNetInterworkingModel NetXConnectDefaultModel NetInterworkingModel = iota // NetXConnectBridgedModel uses a linux bridge to interconnect // the container interface to the VM. This is the // safe default that works for most cases except // macvlan and ipvlan NetXConnectBridgedModel // NetXConnectMacVtapModel can be used when the Container network // interface can be bridged using macvtap NetXConnectMacVtapModel // NetXConnectEnlightenedModel can be used when the Network plugins // are enlightened to create VM native interfaces // when requested by the runtime // This will be used for vethtap, macvtap, ipvtap NetXConnectEnlightenedModel // NetXConnectInvalidModel is the last item to check valid values by IsValid() NetXConnectInvalidModel )
func (NetInterworkingModel) IsValid ¶
func (n NetInterworkingModel) IsValid() bool
IsValid checks if a model is valid
func (*NetInterworkingModel) SetModel ¶
func (n *NetInterworkingModel) SetModel(modelName string) error
SetModel change the model string value
type NetlinkIface ¶
NetlinkIface describes fully a network interface.
type NetworkConfig ¶
type NetworkConfig struct {
NetNSPath string
NumInterfaces int
InterworkingModel NetInterworkingModel
}
NetworkConfig is the network configuration related to a network.
type NetworkInfo ¶
type NetworkInfo struct {
Iface NetlinkIface
Addrs []netlink.Addr
Routes []netlink.Route
DNS DNSInfo
}
NetworkInfo gathers all information related to a network interface. It can be used to store the description of the underlying network.
type NetworkInterface ¶
NetworkInterface defines a network interface.
type NetworkInterfacePair ¶
type NetworkInterfacePair struct {
ID string
Name string
VirtIface NetworkInterface
TAPIface NetworkInterface
NetInterworkingModel
VMFds []*os.File
VhostFds []*os.File
}
NetworkInterfacePair defines a pair between VM and virtual network interfaces.
type NetworkModel ¶
type NetworkModel string
NetworkModel describes the type of network specification.
const ( // NoopNetworkModel is the No-Op network. NoopNetworkModel NetworkModel = "noop" // CNINetworkModel is the CNI network. CNINetworkModel NetworkModel = "CNI" // CNMNetworkModel is the CNM network. CNMNetworkModel NetworkModel = "CNM" )
func (*NetworkModel) Set ¶
func (networkType *NetworkModel) Set(value string) error
Set sets a network type based on the input string.
func (*NetworkModel) String ¶
func (networkType *NetworkModel) String() string
String converts a network type to a string.
type NetworkNamespace ¶
NetworkNamespace contains all data related to its network namespace.
func (NetworkNamespace) MarshalJSON ¶
func (n NetworkNamespace) MarshalJSON() ([]byte, error)
MarshalJSON is the custom NetworkNamespace JSON marshalling routine. This is needed to properly marshall Endpoints array.
func (*NetworkNamespace) UnmarshalJSON ¶
func (n *NetworkNamespace) UnmarshalJSON(b []byte) error
UnmarshalJSON is the custom NetworkNamespace unmarshalling routine. This is needed for unmarshalling the Endpoints interfaces array.
type Param ¶
Param is a key/value representation for hypervisor and kernel parameters.
func DeserializeParams ¶
DeserializeParams converts []string to []Param
type PhysicalEndpoint ¶
type PhysicalEndpoint struct {
IfaceName string
HardAddr string
EndpointProperties NetworkInfo
EndpointType EndpointType
BDF string
Driver string
VendorDeviceID string
}
PhysicalEndpoint gathers a physical network interface and its properties
func (*PhysicalEndpoint) Attach ¶
func (endpoint *PhysicalEndpoint) Attach(h hypervisor) error
Attach for physical endpoint binds the physical network interface to vfio-pci and adds device to the hypervisor with vfio-passthrough.
func (*PhysicalEndpoint) Detach ¶
func (endpoint *PhysicalEndpoint) Detach() error
Detach for physical endpoint unbinds the physical network interface from vfio-pci and binds it back to the saved host driver.
func (*PhysicalEndpoint) HardwareAddr ¶
func (endpoint *PhysicalEndpoint) HardwareAddr() string
HardwareAddr returns the mac address of the physical network interface.
func (*PhysicalEndpoint) Name ¶
func (endpoint *PhysicalEndpoint) Name() string
Name returns name of the physical interface.
func (*PhysicalEndpoint) Properties ¶
func (endpoint *PhysicalEndpoint) Properties() NetworkInfo
Properties returns the properties of the physical interface.
func (*PhysicalEndpoint) SetProperties ¶
func (endpoint *PhysicalEndpoint) SetProperties(properties NetworkInfo)
SetProperties sets the properties of the physical endpoint.
func (*PhysicalEndpoint) Type ¶
func (endpoint *PhysicalEndpoint) Type() EndpointType
Type indentifies the endpoint as a physical endpoint.
type Pod ¶
type Pod struct {
// contains filtered or unexported fields
}
Pod is composed of a set of containers and a runtime environment. A Pod can be created, deleted, started, paused, stopped, listed, entered, and restored.
func (*Pod) Annotations ¶
Annotations returns any annotation that a user could have stored through the pod.
func (*Pod) GetAllContainers ¶
func (p *Pod) GetAllContainers() []VCContainer
GetAllContainers returns all containers.
func (*Pod) GetAnnotations ¶
GetAnnotations returns pod's annotations
func (*Pod) GetContainer ¶
func (p *Pod) GetContainer(containerID string) VCContainer
GetContainer returns the container named by the containerID.
type PodConfig ¶
type PodConfig struct {
ID string
Hostname string
// Field specific to OCI specs, needed to setup all the hooks
Hooks Hooks
// VMConfig is the VM configuration to set for this pod.
VMConfig Resources
HypervisorType HypervisorType
HypervisorConfig HypervisorConfig
AgentType AgentType
AgentConfig interface{}
ProxyType ProxyType
ProxyConfig ProxyConfig
ShimType ShimType
ShimConfig interface{}
NetworkModel NetworkModel
NetworkConfig NetworkConfig
// Volumes is a list of shared volumes between the host and the Pod.
Volumes []Volume
// Containers describe the list of containers within a Pod.
// This list can be empty and populated by adding containers
// to the Pod a posteriori.
Containers []ContainerConfig
// Annotations keys must be unique strings and must be name-spaced
// with e.g. reverse domain notation (org.clearlinux.key).
Annotations map[string]string
}
PodConfig is a Pod configuration.
type PodStatus ¶
type PodStatus struct {
ID string
State State
Hypervisor HypervisorType
HypervisorConfig HypervisorConfig
Agent AgentType
ContainersStatus []ContainerStatus
// Annotations allow clients to store arbitrary values,
// for example to add additional status values required
// to support particular specifications.
Annotations map[string]string
}
PodStatus describes a pod status.
type Process ¶
type Process struct {
// Token is the process execution context ID. It must be
// unique per pod.
// Token is used to manipulate processes for containers
// that have not started yet, and later identify them
// uniquely within a pod.
Token string
// Pid is the process ID as seen by the host software
// stack, e.g. CRI-O, containerd. This is typically the
// shim PID.
Pid int
StartTime time.Time
}
Process gathers data related to a container process.
type ProcessList ¶
type ProcessList []byte
ProcessList represents the list of running processes inside the container
func ProcessListContainer ¶
func ProcessListContainer(podID, containerID string, options ProcessListOptions) (ProcessList, error)
ProcessListContainer is the virtcontainers entry point to list processes running inside a container
type ProcessListOptions ¶
type ProcessListOptions struct {
// Format describes the output format to list the running processes.
// Formats are unrelated to ps(1) formats, only two formats can be specified:
// "json" and "table"
Format string
// Args contains the list of arguments to run ps(1) command.
// If Args is empty the agent will use "-ef" as options to ps(1).
Args []string
}
ProcessListOptions contains the options used to list running processes inside the container
type ProxyConfig ¶
ProxyConfig is a structure storing information needed from any proxy in order to be properly initialized.
type ProxyType ¶
type ProxyType string
ProxyType describes a proxy type.
type QemuState ¶
type QemuState struct {
Bridges []Bridge
// HotpluggedCPUs is the list of CPUs that were hot-added
HotpluggedVCPUs []CPUDevice
UUID string
}
QemuState keeps Qemu's state
type Resources ¶
type Resources struct {
// Memory is the amount of available memory in MiB.
Memory uint
}
Resources describes VM resources configuration.
type ShimConfig ¶
ShimConfig is the structure providing specific configuration for shim implementations.
type ShimParams ¶
type ShimParams struct {
Container string
Token string
URL string
Console string
Terminal bool
Detach bool
PID int
CreateNS []ns.NSType
EnterNS []ns.Namespace
}
ShimParams is the structure providing specific parameters needed for the execution of the shim binary.
type ShimType ¶
type ShimType string
ShimType describes a shim type.
type Socket ¶
Socket defines a socket to communicate between the host and any process inside the VM.
type Sockets ¶
type Sockets []Socket
Sockets is a Socket list.
type SpawnerType ¶
type SpawnerType string
SpawnerType describes the type of guest agent a Pod should run.
const ( // NsEnter is the nsenter spawner type NsEnter SpawnerType = "nsenter" )
func (*SpawnerType) Set ¶
func (spawnerType *SpawnerType) Set(value string) error
Set sets an agent type based on the input string.
func (*SpawnerType) String ¶
func (spawnerType *SpawnerType) String() string
String converts an agent type to a string.
type State ¶
type State struct {
State stateString `json:"state"`
// Index of the block device passed to hypervisor.
BlockIndex int `json:"blockIndex"`
// File system of the rootfs incase it is block device
Fstype string `json:"fstype"`
// Bool to indicate if the drive for a container was hotplugged.
HotpluggedDrive bool `json:"hotpluggedDrive"`
}
State is a pod state structure.
type SystemMountsInfo ¶
type SystemMountsInfo struct {
// Indicates if /dev has been passed as a bind mount for the host /dev
BindMountDev bool
// Size of /dev/shm assigned on the host.
DevShmSize uint
}
SystemMountsInfo describes additional information for system mounts that the agent needs to handle
type TypedDevice ¶
type TypedDevice struct {
Type string
// Data is assigned the Device object.
// This being declared as RawMessage prevents it from being marshalled/unmarshalled.
// We do that explicitly depending on Type.
Data json.RawMessage
}
TypedDevice is used as an intermediate representation for marshalling and unmarshalling Device implementations.
type TypedJSONEndpoint ¶
type TypedJSONEndpoint struct {
Type EndpointType
Data json.RawMessage
}
TypedJSONEndpoint is used as an intermediate representation for marshalling and unmarshalling Endpoint objects.
type VC ¶
type VC interface {
SetLogger(logger logrus.FieldLogger)
CreatePod(podConfig PodConfig) (VCPod, error)
DeletePod(podID string) (VCPod, error)
ListPod() ([]PodStatus, error)
PausePod(podID string) (VCPod, error)
ResumePod(podID string) (VCPod, error)
RunPod(podConfig PodConfig) (VCPod, error)
StartPod(podID string) (VCPod, error)
StatusPod(podID string) (PodStatus, error)
StopPod(podID string) (VCPod, error)
CreateContainer(podID string, containerConfig ContainerConfig) (VCPod, VCContainer, error)
DeleteContainer(podID, containerID string) (VCContainer, error)
EnterContainer(podID, containerID string, cmd Cmd) (VCPod, VCContainer, *Process, error)
KillContainer(podID, containerID string, signal syscall.Signal, all bool) error
StartContainer(podID, containerID string) (VCContainer, error)
StatusContainer(podID, containerID string) (ContainerStatus, error)
StopContainer(podID, containerID string) (VCContainer, error)
ProcessListContainer(podID, containerID string, options ProcessListOptions) (ProcessList, error)
}
VC is the Virtcontainers interface
type VCContainer ¶
type VCContainer interface {
GetAnnotations() map[string]string
GetPid() int
GetToken() string
ID() string
Pod() VCPod
Process() Process
SetPid(pid int) error
}
VCContainer is the Container interface (required since virtcontainers.Container only contains private fields)
func DeleteContainer ¶
func DeleteContainer(podID, containerID string) (VCContainer, error)
DeleteContainer is the virtcontainers container deletion entry point. DeleteContainer deletes a Container from a Pod. If the container is running, it needs to be stopped first.
func StartContainer ¶
func StartContainer(podID, containerID string) (VCContainer, error)
StartContainer is the virtcontainers container starting entry point. StartContainer starts an already created container.
func StopContainer ¶
func StopContainer(podID, containerID string) (VCContainer, error)
StopContainer is the virtcontainers container stopping entry point. StopContainer stops an already running container.
type VCImpl ¶
type VCImpl struct {
}
VCImpl is the official virtcontainers function of the same name.
func (*VCImpl) CreateContainer ¶
func (impl *VCImpl) CreateContainer(podID string, containerConfig ContainerConfig) (VCPod, VCContainer, error)
CreateContainer implements the VC function of the same name.
func (*VCImpl) DeleteContainer ¶
func (impl *VCImpl) DeleteContainer(podID, containerID string) (VCContainer, error)
DeleteContainer implements the VC function of the same name.
func (*VCImpl) EnterContainer ¶
func (impl *VCImpl) EnterContainer(podID, containerID string, cmd Cmd) (VCPod, VCContainer, *Process, error)
EnterContainer implements the VC function of the same name.
func (*VCImpl) KillContainer ¶
KillContainer implements the VC function of the same name.
func (*VCImpl) ProcessListContainer ¶
func (impl *VCImpl) ProcessListContainer(podID, containerID string, options ProcessListOptions) (ProcessList, error)
ProcessListContainer implements the VC function of the same name.
func (*VCImpl) SetLogger ¶
func (impl *VCImpl) SetLogger(logger logrus.FieldLogger)
SetLogger implements the VC function of the same name.
func (*VCImpl) StartContainer ¶
func (impl *VCImpl) StartContainer(podID, containerID string) (VCContainer, error)
StartContainer implements the VC function of the same name.
func (*VCImpl) StatusContainer ¶
func (impl *VCImpl) StatusContainer(podID, containerID string) (ContainerStatus, error)
StatusContainer implements the VC function of the same name.
func (*VCImpl) StopContainer ¶
func (impl *VCImpl) StopContainer(podID, containerID string) (VCContainer, error)
StopContainer implements the VC function of the same name.
type VCPod ¶
type VCPod interface {
Annotations(key string) (string, error)
GetAllContainers() []VCContainer
GetAnnotations() map[string]string
GetContainer(containerID string) VCContainer
ID() string
SetAnnotations(annotations map[string]string) error
}
VCPod is the Pod interface (required since virtcontainers.Pod only contains private fields)
func CreatePod ¶
CreatePod is the virtcontainers pod creation entry point. CreatePod creates a pod and its containers. It does not start them.
func DeletePod ¶
DeletePod is the virtcontainers pod deletion entry point. DeletePod will stop an already running container and then delete it.
func PausePod ¶
PausePod is the virtcontainers pausing entry point which pauses an already running pod.
func ResumePod ¶
ResumePod is the virtcontainers resuming entry point which resumes (or unpauses) and already paused pod.
func RunPod ¶
RunPod is the virtcontainers pod running entry point. RunPod creates a pod and its containers and then it starts them.
type VFIODevice ¶
type VFIODevice struct {
DeviceType string
DeviceInfo DeviceInfo
BDF string
}
VFIODevice is a vfio device meant to be passed to the hypervisor to be used by the Virtual Machine.
type VhostUserBlkDevice ¶
type VhostUserBlkDevice struct {
VhostUserDeviceAttrs
}
VhostUserBlkDevice is a block vhost-user based device
func (*VhostUserBlkDevice) Attrs ¶
func (vhostUserBlkDevice *VhostUserBlkDevice) Attrs() *VhostUserDeviceAttrs
Attrs returns the VhostUserDeviceAttrs associated with the vhost-user device
func (*VhostUserBlkDevice) Type ¶
func (vhostUserBlkDevice *VhostUserBlkDevice) Type() string
Type returns the type associated with the vhost-user device
type VhostUserDevice ¶
type VhostUserDevice interface {
Attrs() *VhostUserDeviceAttrs
Type() string
}
VhostUserDevice represents a vhost-user device. Shared attributes of a vhost-user device can be retrieved using the Attrs() method. Unique data can be obtained by casting the object to the proper type.
type VhostUserDeviceAttrs ¶
type VhostUserDeviceAttrs struct {
DeviceType string
DeviceInfo DeviceInfo
SocketPath string
ID string
}
VhostUserDeviceAttrs represents data shared by most vhost-user devices
type VhostUserDeviceType ¶
type VhostUserDeviceType string
VhostUserDeviceType - represents a vhost-user device type Currently support just VhostUserNet
type VhostUserEndpoint ¶
type VhostUserEndpoint struct {
// Path to the vhost-user socket on the host system
SocketPath string
// MAC address of the interface
HardAddr string
IfaceName string
EndpointProperties NetworkInfo
EndpointType EndpointType
}
VhostUserEndpoint represents a vhost-user socket based network interface
func (*VhostUserEndpoint) Attach ¶
func (endpoint *VhostUserEndpoint) Attach(h hypervisor) error
Attach for vhostuser endpoint
func (*VhostUserEndpoint) Detach ¶
func (endpoint *VhostUserEndpoint) Detach() error
Detach for vhostuser endpoint
func (*VhostUserEndpoint) HardwareAddr ¶
func (endpoint *VhostUserEndpoint) HardwareAddr() string
HardwareAddr returns the mac address of the vhostuser network interface
func (*VhostUserEndpoint) Name ¶
func (endpoint *VhostUserEndpoint) Name() string
Name returns name of the interface.
func (*VhostUserEndpoint) Properties ¶
func (endpoint *VhostUserEndpoint) Properties() NetworkInfo
Properties returns the properties of the interface.
func (*VhostUserEndpoint) SetProperties ¶
func (endpoint *VhostUserEndpoint) SetProperties(properties NetworkInfo)
SetProperties sets the properties of the endpoint.
func (*VhostUserEndpoint) Type ¶
func (endpoint *VhostUserEndpoint) Type() EndpointType
Type indentifies the endpoint as a vhostuser endpoint.
type VhostUserNetDevice ¶
type VhostUserNetDevice struct {
VhostUserDeviceAttrs
MacAddress string
}
VhostUserNetDevice is a network vhost-user based device
func (*VhostUserNetDevice) Attrs ¶
func (vhostUserNetDevice *VhostUserNetDevice) Attrs() *VhostUserDeviceAttrs
Attrs returns the VhostUserDeviceAttrs associated with the vhost-user device
func (*VhostUserNetDevice) Type ¶
func (vhostUserNetDevice *VhostUserNetDevice) Type() string
Type returns the type associated with the vhost-user device
type VhostUserSCSIDevice ¶
type VhostUserSCSIDevice struct {
VhostUserDeviceAttrs
}
VhostUserSCSIDevice is a SCSI vhost-user based device
func (*VhostUserSCSIDevice) Attrs ¶
func (vhostUserSCSIDevice *VhostUserSCSIDevice) Attrs() *VhostUserDeviceAttrs
Attrs returns the VhostUserDeviceAttrs associated with the vhost-user device
func (*VhostUserSCSIDevice) Type ¶
func (vhostUserSCSIDevice *VhostUserSCSIDevice) Type() string
Type returns the type associated with the vhost-user device
type VirtualEndpoint ¶
type VirtualEndpoint struct {
NetPair NetworkInterfacePair
EndpointProperties NetworkInfo
Physical bool
EndpointType EndpointType
}
VirtualEndpoint gathers a network pair and its properties.
func (*VirtualEndpoint) Attach ¶
func (endpoint *VirtualEndpoint) Attach(h hypervisor) error
Attach for virtual endpoint bridges the network pair and adds the tap interface of the network pair to the hypervisor.
func (*VirtualEndpoint) Detach ¶
func (endpoint *VirtualEndpoint) Detach() error
Detach for the virtual endpoint tears down the tap and bridge created for the veth interface.
func (*VirtualEndpoint) HardwareAddr ¶
func (endpoint *VirtualEndpoint) HardwareAddr() string
HardwareAddr returns the mac address that is assigned to the tap interface in th network pair.
func (*VirtualEndpoint) Name ¶
func (endpoint *VirtualEndpoint) Name() string
Name returns name of the veth interface in the network pair.
func (*VirtualEndpoint) Properties ¶
func (endpoint *VirtualEndpoint) Properties() NetworkInfo
Properties returns properties for the veth interface in the network pair.
func (*VirtualEndpoint) SetProperties ¶
func (endpoint *VirtualEndpoint) SetProperties(properties NetworkInfo)
SetProperties sets the properties for the endpoint.
func (*VirtualEndpoint) Type ¶
func (endpoint *VirtualEndpoint) Type() EndpointType
Type identifies the endpoint as a virtual endpoint.
Source Files
¶
- agent.go
- api.go
- asset.go
- bridge.go
- capabilities.go
- cc_proxy.go
- cc_shim.go
- cni.go
- cnm.go
- container.go
- device.go
- doc.go
- errors.go
- filesystem.go
- hook.go
- hyperstart_agent.go
- hypervisor.go
- implementation.go
- interfaces.go
- kata_agent.go
- kata_proxy.go
- kata_shim.go
- mock_hypervisor.go
- mount.go
- network.go
- no_proxy.go
- noop_agent.go
- noop_network.go
- noop_proxy.go
- noop_shim.go
- nsenter.go
- pod.go
- proxy.go
- qemu.go
- qemu_amd64.go
- qemu_arch_base.go
- shim.go
- spawner.go
- syscall.go
- types.go
- utils.go
Directories
¶
| Path | Synopsis |
|---|---|
|
hack
|
|
|
virtc
command
|
|
|
hook
|
|
|
mock
command
|
|
|
pkg
|
|
|
uuid
Package uuid can be used to generate 128 bit UUIDs compatible with rfc4122.
|
Package uuid can be used to generate 128 bit UUIDs compatible with rfc4122. |
|
shim
|
|
|
mock/cc-shim
command
|
|
|
mock/kata-shim
command
|


