containerd

package module
Version: v1.4.3 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Nov 30, 2020 License: Apache-2.0 Imports: 82 Imported by: 0

README

containerd banner

GoDoc Build Status Windows Build Status Nightlies FOSSA Status Go Report Card CII Best Practices

containerd is an industry-standard container runtime with an emphasis on simplicity, robustness and portability. It is available as a daemon for Linux and Windows, which can manage the complete container lifecycle of its host system: image transfer and storage, container execution and supervision, low-level storage and network attachments, etc.

containerd is designed to be embedded into a larger system, rather than being used directly by developers or end-users.

architecture

Getting Started

See our documentation on containerd.io:

See how to build containerd from source at BUILDING.

If you are interested in trying out containerd see our example at Getting Started.

Nightly builds

There are nightly builds available for download here. Binaries are generated from master branch every night for Linux and Windows.

Please be aware: nightly builds might have critical bugs, it's not recommended for use in prodution and no support provided.

Runtime Requirements

Runtime requirements for containerd are very minimal. Most interactions with the Linux and Windows container feature sets are handled via runc and/or OS-specific libraries (e.g. hcsshim for Microsoft). The current required version of runc is always listed in RUNC.md.

There are specific features used by containerd core code and snapshotters that will require a minimum kernel version on Linux. With the understood caveat of distro kernel versioning, a reasonable starting point for Linux is a minimum 4.x kernel version.

The overlay filesystem snapshotter, used by default, uses features that were finalized in the 4.x kernel series. If you choose to use btrfs, there may be more flexibility in kernel version (minimum recommended is 3.18), but will require the btrfs kernel module and btrfs tools to be installed on your Linux distribution.

To use Linux checkpoint and restore features, you will need criu installed on your system. See more details in Checkpoint and Restore.

Build requirements for developers are listed in BUILDING.

Features

Client

containerd offers a full client package to help you integrate containerd into your platform.


import (
  "github.com/containerd/containerd"
  "github.com/containerd/containerd/cio"
)


func main() {
	client, err := containerd.New("/run/containerd/containerd.sock")
	defer client.Close()
}

Namespaces

Namespaces allow multiple consumers to use the same containerd without conflicting with each other. It has the benefit of sharing content but still having separation with containers and images.

To set a namespace for requests to the API:

context = context.Background()
// create a context for docker
docker = namespaces.WithNamespace(context, "docker")

containerd, err := client.NewContainer(docker, "id")

To set a default namespace on the client:

client, err := containerd.New(address, containerd.WithDefaultNamespace("docker"))

Distribution

// pull an image
image, err := client.Pull(context, "docker.io/library/redis:latest")

// push an image
err := client.Push(context, "docker.io/library/redis:latest", image.Target())

Containers

In containerd, a container is a metadata object. Resources such as an OCI runtime specification, image, root filesystem, and other metadata can be attached to a container.

redis, err := client.NewContainer(context, "redis-master")
defer redis.Delete(context)

OCI Runtime Specification

containerd fully supports the OCI runtime specification for running containers. We have built in functions to help you generate runtime specifications based on images as well as custom parameters.

You can specify options when creating a container about how to modify the specification.

redis, err := client.NewContainer(context, "redis-master", containerd.WithNewSpec(oci.WithImageConfig(image)))

Root Filesystems

containerd allows you to use overlay or snapshot filesystems with your containers. It comes with builtin support for overlayfs and btrfs.

// pull an image and unpack it into the configured snapshotter
image, err := client.Pull(context, "docker.io/library/redis:latest", containerd.WithPullUnpack)

// allocate a new RW root filesystem for a container based on the image
redis, err := client.NewContainer(context, "redis-master",
	containerd.WithNewSnapshot("redis-rootfs", image),
	containerd.WithNewSpec(oci.WithImageConfig(image)),
)

// use a readonly filesystem with multiple containers
for i := 0; i < 10; i++ {
	id := fmt.Sprintf("id-%s", i)
	container, err := client.NewContainer(ctx, id,
		containerd.WithNewSnapshotView(id, image),
		containerd.WithNewSpec(oci.WithImageConfig(image)),
	)
}

Tasks

Taking a container object and turning it into a runnable process on a system is done by creating a new Task from the container. A task represents the runnable object within containerd.

// create a new task
task, err := redis.NewTask(context, cio.NewCreator(cio.WithStdio))
defer task.Delete(context)

// the task is now running and has a pid that can be use to setup networking
// or other runtime settings outside of containerd
pid := task.Pid()

// start the redis-server process inside the container
err := task.Start(context)

// wait for the task to exit and get the exit status
status, err := task.Wait(context)

Checkpoint and Restore

If you have criu installed on your machine you can checkpoint and restore containers and their tasks. This allow you to clone and/or live migrate containers to other machines.

// checkpoint the task then push it to a registry
checkpoint, err := task.Checkpoint(context)

err := client.Push(context, "myregistry/checkpoints/redis:master", checkpoint)

// on a new machine pull the checkpoint and restore the redis container
checkpoint, err := client.Pull(context, "myregistry/checkpoints/redis:master")

redis, err = client.NewContainer(context, "redis-master", containerd.WithNewSnapshot("redis-rootfs", checkpoint))
defer container.Delete(context)

task, err = redis.NewTask(context, cio.NewCreator(cio.WithStdio), containerd.WithTaskCheckpoint(checkpoint))
defer task.Delete(context)

err := task.Start(context)

Snapshot Plugins

In addition to the built-in Snapshot plugins in containerd, additional external plugins can be configured using GRPC. An external plugin is made available using the configured name and appears as a plugin alongside the built-in ones.

To add an external snapshot plugin, add the plugin to containerd's config file (by default at /etc/containerd/config.toml). The string following proxy_plugin. will be used as the name of the snapshotter and the address should refer to a socket with a GRPC listener serving containerd's Snapshot GRPC API. Remember to restart containerd for any configuration changes to take effect.

[proxy_plugins]
  [proxy_plugins.customsnapshot]
    type = "snapshot"
    address =  "/var/run/mysnapshotter.sock"

See PLUGINS.md for how to create plugins

Releases and API Stability

Please see RELEASES.md for details on versioning and stability of containerd components.

Downloadable 64-bit Intel/AMD binaries of all official releases are available on our releases page, as well as auto-published to the cri-containerd-release storage bucket.

For other architectures and distribution support, you will find that many Linux distributions package their own containerd and provide it across several architectures, such as Canonical's Ubuntu packaging.

Enabling command auto-completion

Starting with containerd 1.4, the urfave client feature for auto-creation of bash and zsh autocompletion data is enabled. To use the autocomplete feature in a bash shell for example, source the autocomplete/ctr file in your .bashrc, or manually like:

$ source ./contrib/autocomplete/ctr
Distribution of ctr autocomplete for bash and zsh

For bash, copy the contrib/autocomplete/ctr script into /etc/bash_completion.d/ and rename it to ctr. The zsh_autocomplete file is also available and can be used similarly for zsh users.

Provide documentation to users to source this file into their shell if you don't place the autocomplete file in a location where it is automatically loaded for the user's shell environment.

Communication

For async communication and long running discussions please use issues and pull requests on the github repo. This will be the best place to discuss design and implementation.

For sync communication we have a community slack with a #containerd channel that everyone is welcome to join and chat about development.

Slack: Catch us in the #containerd and #containerd-dev channels on dockercommunity.slack.com. Click here for an invite to docker community slack.

Security audit

A third party security audit was performed by Cure53 in 4Q2018; the full report is available in our docs/ directory.

Reporting security issues

If you are reporting a security issue, please reach out discreetly at security@containerd.io.

Licenses

The containerd codebase is released under the Apache 2.0 license. The README.md file, and files in the "docs" folder are licensed under the Creative Commons Attribution 4.0 International License. You may obtain a copy of the license, titled CC-BY-4.0, at http://creativecommons.org/licenses/by/4.0/.

Project details

containerd is the primary open source project within the broader containerd GitHub repository. However, all projects within the repo have common maintainership, governance, and contributing guidelines which are stored in a project repository commonly for all containerd projects.

Please find all these core project documents, including the:

information in our containerd/project repository.

Adoption

Interested to see who is using containerd? Are you using containerd in a project? Please add yourself via pull request to our ADOPTERS.md file.

Documentation

Index

Constants

View Source
const (
	// DefaultSnapshotter will set the default snapshotter for the platform.
	// This will be based on the client compilation target, so take that into
	// account when choosing this value.
	DefaultSnapshotter = "overlayfs"
)
View Source
const StopSignalLabel = "io.containerd.image.config.stop-signal"

StopSignalLabel is a well-known containerd label for storing the stop signal specified in the OCI image config

View Source
const UnknownExitStatus = 255

UnknownExitStatus is returned when containerd is unable to determine the exit status of a process. This can happen if the process never starts or if an error was encountered when obtaining the exit status, it is set to 255.

Variables

View Source
var (
	// ErrCheckpointRWUnsupported is returned if the container runtime does not support checkpoint
	ErrCheckpointRWUnsupported = errors.New("rw checkpoint is only supported on v2 runtimes")
	// ErrMediaTypeNotFound returns an error when a media type in the manifest is unknown
	ErrMediaTypeNotFound = errors.New("media type not found")
)
View Source
var (
	// ErrImageNameNotFoundInIndex is returned when the image name is not found in the index
	ErrImageNameNotFoundInIndex = errors.New("image name not found in index")
	// ErrRuntimeNameNotFoundInIndex is returned when the runtime is not found in the index
	ErrRuntimeNameNotFoundInIndex = errors.New("runtime not found in index")
	// ErrSnapshotterNameNotFoundInIndex is returned when the snapshotter is not found in the index
	ErrSnapshotterNameNotFoundInIndex = errors.New("snapshotter not found in index")
)

Functions

func CheckRuntime added in v1.3.0

func CheckRuntime(current, expected string) bool

CheckRuntime returns true if the current runtime matches the expected runtime. Providing various parts of the runtime schema will match those parts of the expected runtime

func GetIndexByMediaType added in v1.3.0

func GetIndexByMediaType(index *imagespec.Index, mt string) (*imagespec.Descriptor, error)

GetIndexByMediaType returns the index in a manifest for the specified media type

func GetOCIStopSignal added in v1.2.0

func GetOCIStopSignal(ctx context.Context, image Image, defaultSignal string) (string, error)

GetOCIStopSignal retrieves the stop signal specified in the OCI image config

func GetStopSignal added in v1.2.0

func GetStopSignal(ctx context.Context, container Container, defaultSignal syscall.Signal) (syscall.Signal, error)

GetStopSignal retrieves the container stop signal, specified by the well-known containerd label (StopSignalLabel)

func NewImageStoreFromClient

func NewImageStoreFromClient(client imagesapi.ImagesClient) images.Store

NewImageStoreFromClient returns a new image store client

func NewNamespaceStoreFromClient

func NewNamespaceStoreFromClient(client api.NamespacesClient) namespaces.Store

NewNamespaceStoreFromClient returns a new namespace store

func NewRemoteContainerStore

func NewRemoteContainerStore(client containersapi.ContainersClient) containers.Store

NewRemoteContainerStore returns the container Store connected with the provided client

func ParseSignal added in v1.2.0

func ParseSignal(rawSignal string) (syscall.Signal, error)

ParseSignal parses a given string into a syscall.Signal the rawSignal can be a string with "SIG" prefix, or a signal number in string format.

func WithCheckpointImage added in v1.3.0

func WithCheckpointImage(ctx context.Context, client *Client, c *containers.Container, index *imagespec.Index, copts *options.CheckpointOptions) error

WithCheckpointImage includes the container image in the checkpoint

func WithCheckpointRW added in v1.3.0

func WithCheckpointRW(ctx context.Context, client *Client, c *containers.Container, index *imagespec.Index, copts *options.CheckpointOptions) error

WithCheckpointRW includes the rw in the checkpoint

func WithCheckpointRuntime added in v1.3.0

func WithCheckpointRuntime(ctx context.Context, client *Client, c *containers.Container, index *imagespec.Index, copts *options.CheckpointOptions) error

WithCheckpointRuntime includes the container runtime info

func WithCheckpointTask added in v1.3.0

func WithCheckpointTask(ctx context.Context, client *Client, c *containers.Container, index *imagespec.Index, copts *options.CheckpointOptions) error

WithCheckpointTask includes the running task

func WithCheckpointTaskExit added in v1.3.0

func WithCheckpointTaskExit(ctx context.Context, client *Client, c *containers.Container, index *imagespec.Index, copts *options.CheckpointOptions) error

WithCheckpointTaskExit causes the task to exit after checkpoint

func WithInstallLibs added in v1.2.0

func WithInstallLibs(c *InstallConfig)

WithInstallLibs installs libs from the image

func WithInstallReplace added in v1.2.0

func WithInstallReplace(c *InstallConfig)

WithInstallReplace will replace existing files

func WithKillAll

func WithKillAll(ctx context.Context, i *KillInfo) error

WithKillAll kills all processes for a task

func WithNoNewKeyring added in v1.2.0

func WithNoNewKeyring(ctx context.Context, c *Client, ti *TaskInfo) error

WithNoNewKeyring causes tasks not to be created with a new keyring for secret storage. There is an upper limit on the number of keyrings in a linux system

func WithNoPivotRoot

func WithNoPivotRoot(_ context.Context, _ *Client, ti *TaskInfo) error

WithNoPivotRoot instructs the runtime not to you pivot_root

func WithProcessKill

func WithProcessKill(ctx context.Context, p Process) error

WithProcessKill will forcefully kill and delete a process

func WithPullUnpack

func WithPullUnpack(_ *Client, c *RemoteContext) error

WithPullUnpack is used to unpack an image after pull. This uses the snapshotter, content store, and diff service configured for the client.

func WithRemapperLabels added in v1.4.0

func WithRemapperLabels(ctrUID, hostUID, ctrGID, hostGID, length uint32) snapshots.Opt

WithRemapperLabels creates the labels used by any supporting snapshotter to shift the filesystem ownership (user namespace mapping) automatically; currently supported by the fuse-overlayfs snapshotter

func WithSchema1Conversion

func WithSchema1Conversion(client *Client, c *RemoteContext) error

WithSchema1Conversion is used to convert Docker registry schema 1 manifests to oci manifests on pull. Without this option schema 1 manifests will return a not supported error.

func WithSnapshotCleanup

func WithSnapshotCleanup(ctx context.Context, client *Client, c containers.Container) error

WithSnapshotCleanup deletes the rootfs snapshot allocated for the container

func WithStdinCloser

func WithStdinCloser(r *IOCloseInfo)

WithStdinCloser closes the stdin of a process

func WithoutRefreshedMetadata added in v1.3.0

func WithoutRefreshedMetadata(i *InfoConfig)

WithoutRefreshedMetadata will use the current metadata attached to the container object

Types

type CheckpointOpts added in v1.3.0

CheckpointOpts are options to manage the checkpoint operation

type CheckpointTaskInfo

type CheckpointTaskInfo struct {
	Name string
	// ParentCheckpoint is the digest of a parent checkpoint
	ParentCheckpoint digest.Digest
	// Options hold runtime specific settings for checkpointing a task
	Options interface{}
	// contains filtered or unexported fields
}

CheckpointTaskInfo allows specific checkpoint information to be set for the task

func (*CheckpointTaskInfo) Runtime added in v1.3.0

func (i *CheckpointTaskInfo) Runtime() string

Runtime name for the container

type CheckpointTaskOpts

type CheckpointTaskOpts func(*CheckpointTaskInfo) error

CheckpointTaskOpts allows the caller to set checkpoint options

func WithCheckpointImagePath added in v1.3.0

func WithCheckpointImagePath(path string) CheckpointTaskOpts

WithCheckpointImagePath sets image path for checkpoint option

func WithCheckpointName

func WithCheckpointName(name string) CheckpointTaskOpts

WithCheckpointName sets the image name for the checkpoint

type Client

type Client struct {
	// contains filtered or unexported fields
}

Client is the client to interact with containerd and its various services using a uniform interface

func New

func New(address string, opts ...ClientOpt) (*Client, error)

New returns a new containerd client that is connected to the containerd instance provided by address

func NewWithConn

func NewWithConn(conn *grpc.ClientConn, opts ...ClientOpt) (*Client, error)

NewWithConn returns a new containerd client that is connected to the containerd instance provided by the connection

func (*Client) Close

func (c *Client) Close() error

Close closes the clients connection to containerd

func (*Client) Conn added in v1.3.0

func (c *Client) Conn() *grpc.ClientConn

Conn returns the underlying GRPC connection object

func (*Client) ContainerService

func (c *Client) ContainerService() containers.Store

ContainerService returns the underlying container Store

func (*Client) Containers

func (c *Client) Containers(ctx context.Context, filters ...string) ([]Container, error)

Containers returns all containers created in containerd

func (*Client) ContentStore

func (c *Client) ContentStore() content.Store

ContentStore returns the underlying content Store

func (*Client) DiffService

func (c *Client) DiffService() DiffService

DiffService returns the underlying Differ

func (*Client) EventService

func (c *Client) EventService() EventService

EventService returns the underlying event service

func (*Client) Export

func (c *Client) Export(ctx context.Context, w io.Writer, opts ...archive.ExportOpt) error

Export exports images to a Tar stream. The tar archive is in OCI format with a Docker compatible manifest when a single target platform is given.

func (*Client) Fetch added in v1.2.0

func (c *Client) Fetch(ctx context.Context, ref string, opts ...RemoteOpt) (images.Image, error)

Fetch downloads the provided content into containerd's content store and returns a non-platform specific image reference

func (*Client) GetImage

func (c *Client) GetImage(ctx context.Context, ref string) (Image, error)

GetImage returns an existing image

func (*Client) GetLabel added in v1.3.0

func (c *Client) GetLabel(ctx context.Context, label string) (string, error)

GetLabel gets a label value from namespace store If there is no default label, an empty string returned with nil error

func (*Client) HealthService

func (c *Client) HealthService() grpc_health_v1.HealthClient

HealthService returns the underlying GRPC HealthClient

func (*Client) ImageService

func (c *Client) ImageService() images.Store

ImageService returns the underlying image Store

func (*Client) Import

func (c *Client) Import(ctx context.Context, reader io.Reader, opts ...ImportOpt) ([]images.Image, error)

Import imports an image from a Tar stream using reader. Caller needs to specify importer. Future version may use oci.v1 as the default. Note that unreferenced blobs may be imported to the content store as well.

func (*Client) Install added in v1.2.0

func (c *Client) Install(ctx context.Context, image Image, opts ...InstallOpts) error

Install a binary image into the opt service

func (*Client) IntrospectionService

func (c *Client) IntrospectionService() introspection.Service

IntrospectionService returns the underlying Introspection Client

func (*Client) IsServing

func (c *Client) IsServing(ctx context.Context) (bool, error)

IsServing returns true if the client can successfully connect to the containerd daemon and the healthcheck service returns the SERVING response. This call will block if a transient error is encountered during connection. A timeout can be set in the context to ensure it returns early.

func (*Client) LeasesService

func (c *Client) LeasesService() leases.Manager

LeasesService returns the underlying Leases Client

func (*Client) ListImages

func (c *Client) ListImages(ctx context.Context, filters ...string) ([]Image, error)

ListImages returns all existing images

func (*Client) LoadContainer

func (c *Client) LoadContainer(ctx context.Context, id string) (Container, error)

LoadContainer loads an existing container from metadata

func (*Client) NamespaceService

func (c *Client) NamespaceService() namespaces.Store

NamespaceService returns the underlying Namespaces Store

func (*Client) NewContainer

func (c *Client) NewContainer(ctx context.Context, id string, opts ...NewContainerOpts) (Container, error)

NewContainer will create a new container in container with the provided id the id must be unique within the namespace

func (*Client) Pull

func (c *Client) Pull(ctx context.Context, ref string, opts ...RemoteOpt) (_ Image, retErr error)

Pull downloads the provided content into containerd's content store and returns a platform specific image object

func (*Client) Push

func (c *Client) Push(ctx context.Context, ref string, desc ocispec.Descriptor, opts ...RemoteOpt) error

Push uploads the provided content to a remote resource

func (*Client) Reconnect

func (c *Client) Reconnect() error

Reconnect re-establishes the GRPC connection to the containerd daemon

func (*Client) Restore added in v1.3.0

func (c *Client) Restore(ctx context.Context, id string, checkpoint Image, opts ...RestoreOpts) (Container, error)

Restore restores a container from a checkpoint

func (*Client) Server added in v1.3.0

func (c *Client) Server(ctx context.Context) (ServerInfo, error)

func (*Client) SnapshotService

func (c *Client) SnapshotService(snapshotterName string) snapshots.Snapshotter

SnapshotService returns the underlying snapshotter for the provided snapshotter name

func (*Client) Subscribe

func (c *Client) Subscribe(ctx context.Context, filters ...string) (ch <-chan *events.Envelope, errs <-chan error)

Subscribe to events that match one or more of the provided filters.

Callers should listen on both the envelope and errs channels. If the errs channel returns nil or an error, the subscriber should terminate.

The subscriber can stop receiving events by canceling the provided context. The errs channel will be closed and return a nil error.

func (*Client) TaskService

func (c *Client) TaskService() tasks.TasksClient

TaskService returns the underlying TasksClient

func (*Client) Version

func (c *Client) Version(ctx context.Context) (Version, error)

Version returns the version of containerd that the client is connected to

func (*Client) VersionService

func (c *Client) VersionService() versionservice.VersionClient

VersionService returns the underlying VersionClient

func (*Client) WithLease

func (c *Client) WithLease(ctx context.Context, opts ...leases.Opt) (context.Context, func(context.Context) error, error)

WithLease attaches a lease on the context

type ClientOpt

type ClientOpt func(c *clientOpts) error

ClientOpt allows callers to set options on the containerd client

func WithDefaultNamespace

func WithDefaultNamespace(ns string) ClientOpt

WithDefaultNamespace sets the default namespace on the client

Any operation that does not have a namespace set on the context will be provided the default namespace

func WithDefaultPlatform added in v1.3.0

func WithDefaultPlatform(platform platforms.MatchComparer) ClientOpt

WithDefaultPlatform sets the default platform matcher on the client

func WithDefaultRuntime added in v1.2.0

func WithDefaultRuntime(rt string) ClientOpt

WithDefaultRuntime sets the default runtime on the client

func WithDialOpts

func WithDialOpts(opts []grpc.DialOption) ClientOpt

WithDialOpts allows grpc.DialOptions to be set on the connection

func WithServices

func WithServices(opts ...ServicesOpt) ClientOpt

WithServices sets services used by the client.

func WithTimeout added in v1.2.0

func WithTimeout(d time.Duration) ClientOpt

WithTimeout sets the connection timeout for the client

type Container

type Container interface {
	// ID identifies the container
	ID() string
	// Info returns the underlying container record type
	Info(context.Context, ...InfoOpts) (containers.Container, error)
	// Delete removes the container
	Delete(context.Context, ...DeleteOpts) error
	// NewTask creates a new task based on the container metadata
	NewTask(context.Context, cio.Creator, ...NewTaskOpts) (Task, error)
	// Spec returns the OCI runtime specification
	Spec(context.Context) (*oci.Spec, error)
	// Task returns the current task for the container
	//
	// If cio.Attach options are passed the client will reattach to the IO for the running
	// task. If no task exists for the container a NotFound error is returned
	//
	// Clients must make sure that only one reader is attached to the task and consuming
	// the output from the task's fifos
	Task(context.Context, cio.Attach) (Task, error)
	// Image returns the image that the container is based on
	Image(context.Context) (Image, error)
	// Labels returns the labels set on the container
	Labels(context.Context) (map[string]string, error)
	// SetLabels sets the provided labels for the container and returns the final label set
	SetLabels(context.Context, map[string]string) (map[string]string, error)
	// Extensions returns the extensions set on the container
	Extensions(context.Context) (map[string]prototypes.Any, error)
	// Update a container
	Update(context.Context, ...UpdateContainerOpts) error
	// Checkpoint creates a checkpoint image of the current container
	Checkpoint(context.Context, string, ...CheckpointOpts) (Image, error)
}

Container is a metadata object for container resources and task creation

type DeleteOpts

type DeleteOpts func(ctx context.Context, client *Client, c containers.Container) error

DeleteOpts allows the caller to set options for the deletion of a container

type DiffService

type DiffService interface {
	diff.Comparer
	diff.Applier
}

DiffService handles the computation and application of diffs

func NewDiffServiceFromClient

func NewDiffServiceFromClient(client diffapi.DiffClient) DiffService

NewDiffServiceFromClient returns a new diff service which communicates over a GRPC connection.

type EventService

type EventService interface {
	events.Publisher
	events.Forwarder
	events.Subscriber
}

EventService handles the publish, forward and subscribe of events.

func NewEventServiceFromClient

func NewEventServiceFromClient(client eventsapi.EventsClient) EventService

NewEventServiceFromClient returns a new event service which communicates over a GRPC connection.

type ExitStatus

type ExitStatus struct {
	// contains filtered or unexported fields
}

ExitStatus encapsulates a process's exit status. It is used by `Wait()` to return either a process exit code or an error

func NewExitStatus added in v1.3.0

func NewExitStatus(code uint32, t time.Time, err error) *ExitStatus

NewExitStatus populates an ExitStatus

func (ExitStatus) Error

func (s ExitStatus) Error() error

Error returns the error, if any, that occurred while waiting for the process.

func (ExitStatus) ExitCode

func (s ExitStatus) ExitCode() uint32

ExitCode returns the exit code of the process. This is only valid is Error() returns nil

func (ExitStatus) ExitTime

func (s ExitStatus) ExitTime() time.Time

ExitTime returns the exit time of the process This is only valid is Error() returns nil

func (ExitStatus) Result

func (s ExitStatus) Result() (uint32, time.Time, error)

Result returns the exit code and time of the exit status. An error may be returned here to which indicates there was an error

at some point while waiting for the exit status. It does not signify
an error with the process itself.

If an error is returned, the process may still be running.

type IOCloseInfo

type IOCloseInfo struct {
	Stdin bool
}

IOCloseInfo allows specific io pipes to be closed on a process

type IOCloserOpts

type IOCloserOpts func(*IOCloseInfo)

IOCloserOpts allows the caller to set specific pipes as closed on a process

type Image

type Image interface {
	// Name of the image
	Name() string
	// Target descriptor for the image content
	Target() ocispec.Descriptor
	// Labels of the image
	Labels() map[string]string
	// Unpack unpacks the image's content into a snapshot
	Unpack(context.Context, string, ...UnpackOpt) error
	// RootFS returns the unpacked diffids that make up images rootfs.
	RootFS(ctx context.Context) ([]digest.Digest, error)
	// Size returns the total size of the image's packed resources.
	Size(ctx context.Context) (int64, error)
	// Usage returns a usage calculation for the image.
	Usage(context.Context, ...UsageOpt) (int64, error)
	// Config descriptor for the image.
	Config(ctx context.Context) (ocispec.Descriptor, error)
	// IsUnpacked returns whether or not an image is unpacked.
	IsUnpacked(context.Context, string) (bool, error)
	// ContentStore provides a content store which contains image blob data
	ContentStore() content.Store
	// Metadata returns the underlying image metadata
	Metadata() images.Image
}

Image describes an image used by containers

func NewImage

func NewImage(client *Client, i images.Image) Image

NewImage returns a client image object from the metadata image

func NewImageWithPlatform added in v1.2.0

func NewImageWithPlatform(client *Client, i images.Image, platform platforms.MatchComparer) Image

NewImageWithPlatform returns a client image object from the metadata image

type ImportOpt

type ImportOpt func(*importOpts) error

ImportOpt allows the caller to specify import specific options

func WithAllPlatforms added in v1.3.0

func WithAllPlatforms(allPlatforms bool) ImportOpt

WithAllPlatforms is used to import content for all platforms.

func WithDigestRef added in v1.2.0

func WithDigestRef(f func(digest.Digest) string) ImportOpt

WithDigestRef is used to create digest images for each manifest in the index.

func WithImageRefTranslator added in v1.2.0

func WithImageRefTranslator(f func(string) string) ImportOpt

WithImageRefTranslator is used to translate the index reference to an image reference for the image store.

func WithImportCompression added in v1.3.0

func WithImportCompression() ImportOpt

WithImportCompression compresses uncompressed layers on import. This is used for import formats which do not include the manifest.

func WithIndexName added in v1.2.0

func WithIndexName(name string) ImportOpt

WithIndexName creates a tag pointing to the imported index

type InfoConfig added in v1.3.0

type InfoConfig struct {
	// Refresh will to a fetch of the latest container metadata
	Refresh bool
}

InfoConfig specifies how container metadata is fetched

type InfoOpts added in v1.3.0

type InfoOpts func(*InfoConfig)

InfoOpts controls how container metadata is fetched and returned

type InstallConfig added in v1.2.0

type InstallConfig struct {
	// Libs installs libs from the image
	Libs bool
	// Replace will overwrite existing binaries or libs in the opt directory
	Replace bool
	// Path to install libs and binaries to
	Path string
}

InstallConfig sets the binary install configuration

type InstallOpts added in v1.2.0

type InstallOpts func(*InstallConfig)

InstallOpts configures binary installs

func WithInstallPath added in v1.2.0

func WithInstallPath(path string) InstallOpts

WithInstallPath sets the optional install path

type KillInfo

type KillInfo struct {
	// All kills all processes inside the task
	// only valid on tasks, ignored on processes
	All bool
	// ExecID is the ID of a process to kill
	ExecID string
}

KillInfo contains information on how to process a Kill action

type KillOpts

type KillOpts func(context.Context, *KillInfo) error

KillOpts allows options to be set for the killing of a process

func WithKillExecID

func WithKillExecID(execID string) KillOpts

WithKillExecID specifies the process ID

type NewContainerOpts

type NewContainerOpts func(ctx context.Context, client *Client, c *containers.Container) error

NewContainerOpts allows the caller to set additional options when creating a container

func WithContainerExtension

func WithContainerExtension(name string, extension interface{}) NewContainerOpts

WithContainerExtension appends extension data to the container object. Use this to decorate the container object with additional data for the client integration.

Make sure to register the type of `extension` in the typeurl package via `typeurl.Register` or container creation may fail.

func WithContainerLabels

func WithContainerLabels(labels map[string]string) NewContainerOpts

WithContainerLabels adds the provided labels to the container

func WithImage

func WithImage(i Image) NewContainerOpts

WithImage sets the provided image as the base for the container

func WithImageName added in v1.3.0

func WithImageName(n string) NewContainerOpts

WithImageName allows setting the image name as the base for the container

func WithImageStopSignal added in v1.2.0

func WithImageStopSignal(image Image, defaultSignal string) NewContainerOpts

WithImageStopSignal sets a well-known containerd label (StopSignalLabel) on the container for storing the stop signal specified in the OCI image config

func WithNewSnapshot

func WithNewSnapshot(id string, i Image, opts ...snapshots.Opt) NewContainerOpts

WithNewSnapshot allocates a new snapshot to be used by the container as the root filesystem in read-write mode

func WithNewSnapshotView

func WithNewSnapshotView(id string, i Image, opts ...snapshots.Opt) NewContainerOpts

WithNewSnapshotView allocates a new snapshot to be used by the container as the root filesystem in read-only mode

func WithNewSpec

func WithNewSpec(opts ...oci.SpecOpts) NewContainerOpts

WithNewSpec generates a new spec for a new container

func WithRemappedSnapshot

func WithRemappedSnapshot(id string, i Image, uid, gid uint32) NewContainerOpts

WithRemappedSnapshot creates a new snapshot and remaps the uid/gid for the filesystem to be used by a container with user namespaces

func WithRemappedSnapshotView

func WithRemappedSnapshotView(id string, i Image, uid, gid uint32) NewContainerOpts

WithRemappedSnapshotView is similar to WithRemappedSnapshot but rootfs is mounted as read-only.

func WithRestoreImage added in v1.3.0

func WithRestoreImage(ctx context.Context, id string, client *Client, checkpoint Image, index *imagespec.Index) NewContainerOpts

WithRestoreImage restores the image for the container

func WithRestoreRW added in v1.3.0

func WithRestoreRW(ctx context.Context, id string, client *Client, checkpoint Image, index *imagespec.Index) NewContainerOpts

WithRestoreRW restores the rw layer from the checkpoint for the container

func WithRestoreRuntime added in v1.3.0

func WithRestoreRuntime(ctx context.Context, id string, client *Client, checkpoint Image, index *imagespec.Index) NewContainerOpts

WithRestoreRuntime restores the runtime for the container

func WithRestoreSpec added in v1.3.0

func WithRestoreSpec(ctx context.Context, id string, client *Client, checkpoint Image, index *imagespec.Index) NewContainerOpts

WithRestoreSpec restores the spec from the checkpoint for the container

func WithRuntime

func WithRuntime(name string, options interface{}) NewContainerOpts

WithRuntime allows a user to specify the runtime name and additional options that should be used to create tasks for the container

func WithSnapshot

func WithSnapshot(id string) NewContainerOpts

WithSnapshot uses an existing root filesystem for the container

func WithSnapshotter

func WithSnapshotter(name string) NewContainerOpts

WithSnapshotter sets the provided snapshotter for use by the container

This option must appear before other snapshotter options to have an effect.

func WithSpec

func WithSpec(s *oci.Spec, opts ...oci.SpecOpts) NewContainerOpts

WithSpec sets the provided spec on the container

type NewTaskOpts

type NewTaskOpts func(context.Context, *Client, *TaskInfo) error

NewTaskOpts allows the caller to set options on a new task

func WithGIDOwner added in v1.4.0

func WithGIDOwner(gid uint32) NewTaskOpts

WithGIDOwner allows console I/O to work with the remapped GID in user namespace

func WithRestoreImagePath added in v1.3.0

func WithRestoreImagePath(path string) NewTaskOpts

WithRestoreImagePath sets image path for create option

func WithRootFS

func WithRootFS(mounts []mount.Mount) NewTaskOpts

WithRootFS allows a task to be created without a snapshot being allocated to its container

func WithShimCgroup added in v1.3.0

func WithShimCgroup(path string) NewTaskOpts

WithShimCgroup sets the existing cgroup for the shim

func WithTaskCheckpoint

func WithTaskCheckpoint(im Image) NewTaskOpts

WithTaskCheckpoint allows a task to be created with live runtime and memory data from a previous checkpoint. Additional software such as CRIU may be required to restore a task from a checkpoint

func WithUIDOwner added in v1.4.0

func WithUIDOwner(uid uint32) NewTaskOpts

WithUIDOwner allows console I/O to work with the remapped UID in user namespace

type Process

type Process interface {
	// ID of the process
	ID() string
	// Pid is the system specific process id
	Pid() uint32
	// Start starts the process executing the user's defined binary
	Start(context.Context) error
	// Delete removes the process and any resources allocated returning the exit status
	Delete(context.Context, ...ProcessDeleteOpts) (*ExitStatus, error)
	// Kill sends the provided signal to the process
	Kill(context.Context, syscall.Signal, ...KillOpts) error
	// Wait asynchronously waits for the process to exit, and sends the exit code to the returned channel
	Wait(context.Context) (<-chan ExitStatus, error)
	// CloseIO allows various pipes to be closed on the process
	CloseIO(context.Context, ...IOCloserOpts) error
	// Resize changes the width and height of the process's terminal
	Resize(ctx context.Context, w, h uint32) error
	// IO returns the io set for the process
	IO() cio.IO
	// Status returns the executing status of the process
	Status(context.Context) (Status, error)
}

Process represents a system process

type ProcessDeleteOpts

type ProcessDeleteOpts func(context.Context, Process) error

ProcessDeleteOpts allows the caller to set options for the deletion of a task

type ProcessInfo

type ProcessInfo struct {
	// Pid is the process ID
	Pid uint32
	// Info includes additional process information
	// Info varies by platform
	Info *google_protobuf.Any
}

ProcessInfo provides platform specific process information

type ProcessStatus

type ProcessStatus string

ProcessStatus returns a human readable status for the Process representing its current status

const (
	// Running indicates the process is currently executing
	Running ProcessStatus = "running"
	// Created indicates the process has been created within containerd but the
	// user's defined process has not started
	Created ProcessStatus = "created"
	// Stopped indicates that the process has ran and exited
	Stopped ProcessStatus = "stopped"
	// Paused indicates that the process is currently paused
	Paused ProcessStatus = "paused"
	// Pausing indicates that the process is currently switching from a
	// running state into a paused state
	Pausing ProcessStatus = "pausing"
	// Unknown indicates that we could not determine the status from the runtime
	Unknown ProcessStatus = "unknown"
)

type RemoteContext

type RemoteContext struct {
	// Resolver is used to resolve names to objects, fetchers, and pushers.
	// If no resolver is provided, defaults to Docker registry resolver.
	Resolver remotes.Resolver

	// PlatformMatcher is used to match the platforms for an image
	// operation and define the preference when a single match is required
	// from multiple platforms.
	PlatformMatcher platforms.MatchComparer

	// Unpack is done after an image is pulled to extract into a snapshotter.
	// It is done simultaneously for schema 2 images when they are pulled.
	// If an image is not unpacked on pull, it can be unpacked any time
	// afterwards. Unpacking is required to run an image.
	Unpack bool

	// UnpackOpts handles options to the unpack call.
	UnpackOpts []UnpackOpt

	// Snapshotter used for unpacking
	Snapshotter string

	// SnapshotterOpts are additional options to be passed to a snapshotter during pull
	SnapshotterOpts []snapshots.Opt

	// Labels to be applied to the created image
	Labels map[string]string

	// BaseHandlers are a set of handlers which get are called on dispatch.
	// These handlers always get called before any operation specific
	// handlers.
	BaseHandlers []images.Handler

	// HandlerWrapper wraps the handler which gets sent to dispatch.
	// Unlike BaseHandlers, this can run before and after the built
	// in handlers, allowing operations to run on the descriptor
	// after it has completed transferring.
	HandlerWrapper func(images.Handler) images.Handler

	// ConvertSchema1 is whether to convert Docker registry schema 1
	// manifests. If this option is false then any image which resolves
	// to schema 1 will return an error since schema 1 is not supported.
	ConvertSchema1 bool

	// Platforms defines which platforms to handle when doing the image operation.
	// Platforms is ignored when a PlatformMatcher is set, otherwise the
	// platforms will be used to create a PlatformMatcher with no ordering
	// preference.
	Platforms []string

	// MaxConcurrentDownloads is the max concurrent content downloads for each pull.
	MaxConcurrentDownloads int

	// AllMetadata downloads all manifests and known-configuration files
	AllMetadata bool

	// ChildLabelMap sets the labels used to reference child objects in the content
	// store. By default, all GC reference labels will be set for all fetched content.
	ChildLabelMap func(ocispec.Descriptor) []string
}

RemoteContext is used to configure object resolutions and transfers with remote content stores and image providers.

type RemoteOpt

type RemoteOpt func(*Client, *RemoteContext) error

RemoteOpt allows the caller to set distribution options for a remote

func WithAllMetadata added in v1.3.0

func WithAllMetadata() RemoteOpt

WithAllMetadata downloads all manifests and known-configuration files

func WithChildLabelMap added in v1.4.0

func WithChildLabelMap(fn func(ocispec.Descriptor) []string) RemoteOpt

WithChildLabelMap sets the map function used to define the labels set on referenced child content in the content store. This can be used to overwrite the default GC labels or filter which labels get set for content. The default is `images.ChildGCLabels`.

func WithImageHandler

func WithImageHandler(h images.Handler) RemoteOpt

WithImageHandler adds a base handler to be called on dispatch.

func WithImageHandlerWrapper added in v1.3.0

func WithImageHandlerWrapper(w func(images.Handler) images.Handler) RemoteOpt

WithImageHandlerWrapper wraps the handlers to be called on dispatch.

func WithMaxConcurrentDownloads added in v1.3.0

func WithMaxConcurrentDownloads(max int) RemoteOpt

WithMaxConcurrentDownloads sets max concurrent download limit.

func WithPlatform

func WithPlatform(platform string) RemoteOpt

WithPlatform allows the caller to specify a platform to retrieve content for

func WithPlatformMatcher added in v1.2.0

func WithPlatformMatcher(m platforms.MatchComparer) RemoteOpt

WithPlatformMatcher specifies the matcher to use for determining which platforms to pull content for. This value supersedes anything set with `WithPlatform`.

func WithPullLabel

func WithPullLabel(key, value string) RemoteOpt

WithPullLabel sets a label to be associated with a pulled reference

func WithPullLabels

func WithPullLabels(labels map[string]string) RemoteOpt

WithPullLabels associates a set of labels to a pulled reference

func WithPullSnapshotter

func WithPullSnapshotter(snapshotterName string, opts ...snapshots.Opt) RemoteOpt

WithPullSnapshotter specifies snapshotter name used for unpacking.

func WithResolver

func WithResolver(resolver remotes.Resolver) RemoteOpt

WithResolver specifies the resolver to use.

func WithUnpackOpts added in v1.4.0

func WithUnpackOpts(opts []UnpackOpt) RemoteOpt

WithUnpackOpts is used to add unpack options to the unpacker.

type RestoreOpts added in v1.3.0

RestoreOpts are options to manage the restore operation

type ServerInfo added in v1.3.0

type ServerInfo struct {
	UUID string
}

type ServicesOpt

type ServicesOpt func(c *services)

ServicesOpt allows callers to set options on the services

func WithContainerService

func WithContainerService(containerService containersapi.ContainersClient) ServicesOpt

WithContainerService sets the container service.

func WithContentStore

func WithContentStore(contentStore content.Store) ServicesOpt

WithContentStore sets the content store.

func WithDiffService

func WithDiffService(diffService diff.DiffClient) ServicesOpt

WithDiffService sets the diff service.

func WithEventService

func WithEventService(eventService EventService) ServicesOpt

WithEventService sets the event service.

func WithImageService

func WithImageService(imageService imagesapi.ImagesClient) ServicesOpt

WithImageService sets the image service.

func WithIntrospectionService added in v1.4.0

func WithIntrospectionService(in introspectionapi.IntrospectionClient) ServicesOpt

WithIntrospectionService sets the introspection service.

func WithLeasesService

func WithLeasesService(leasesService leases.Manager) ServicesOpt

WithLeasesService sets the lease service.

func WithNamespaceService

func WithNamespaceService(namespaceService namespacesapi.NamespacesClient) ServicesOpt

WithNamespaceService sets the namespace service.

func WithSnapshotters

func WithSnapshotters(snapshotters map[string]snapshots.Snapshotter) ServicesOpt

WithSnapshotters sets the snapshotters.

func WithTaskService

func WithTaskService(taskService tasks.TasksClient) ServicesOpt

WithTaskService sets the task service.

type Status

type Status struct {
	// Status of the process
	Status ProcessStatus
	// ExitStatus returned by the process
	ExitStatus uint32
	// ExitedTime is the time at which the process died
	ExitTime time.Time
}

Status returns process status and exit information

type Task

type Task interface {
	Process

	// Pause suspends the execution of the task
	Pause(context.Context) error
	// Resume the execution of the task
	Resume(context.Context) error
	// Exec creates a new process inside the task
	Exec(context.Context, string, *specs.Process, cio.Creator) (Process, error)
	// Pids returns a list of system specific process ids inside the task
	Pids(context.Context) ([]ProcessInfo, error)
	// Checkpoint serializes the runtime and memory information of a task into an
	// OCI Index that can be pushed and pulled from a remote resource.
	//
	// Additional software like CRIU maybe required to checkpoint and restore tasks
	// NOTE: Checkpoint supports to dump task information to a directory, in this way,
	// an empty OCI Index will be returned.
	Checkpoint(context.Context, ...CheckpointTaskOpts) (Image, error)
	// Update modifies executing tasks with updated settings
	Update(context.Context, ...UpdateTaskOpts) error
	// LoadProcess loads a previously created exec'd process
	LoadProcess(context.Context, string, cio.Attach) (Process, error)
	// Metrics returns task metrics for runtime specific metrics
	//
	// The metric types are generic to containerd and change depending on the runtime
	// For the built in Linux runtime, github.com/containerd/cgroups.Metrics
	// are returned in protobuf format
	Metrics(context.Context) (*types.Metric, error)
	// Spec returns the current OCI specification for the task
	Spec(context.Context) (*oci.Spec, error)
}

Task is the executable object within containerd

type TaskInfo

type TaskInfo struct {
	// Checkpoint is the Descriptor for an existing checkpoint that can be used
	// to restore a task's runtime and memory state
	Checkpoint *types.Descriptor
	// RootFS is a list of mounts to use as the task's root filesystem
	RootFS []mount.Mount
	// Options hold runtime specific settings for task creation
	Options interface{}
	// contains filtered or unexported fields
}

TaskInfo sets options for task creation

func (*TaskInfo) Runtime added in v1.3.0

func (i *TaskInfo) Runtime() string

Runtime name for the container

type UnpackConfig added in v1.3.0

type UnpackConfig struct {
	// ApplyOpts for applying a diff to a snapshotter
	ApplyOpts []diff.ApplyOpt
	// SnapshotOpts for configuring a snapshotter
	SnapshotOpts []snapshots.Opt
}

UnpackConfig provides configuration for the unpack of an image

type UnpackOpt added in v1.3.0

type UnpackOpt func(context.Context, *UnpackConfig) error

UnpackOpt provides configuration for unpack

type UpdateContainerOpts

type UpdateContainerOpts func(ctx context.Context, client *Client, c *containers.Container) error

UpdateContainerOpts allows the caller to set additional options when updating a container

type UpdateTaskInfo

type UpdateTaskInfo struct {
	// Resources updates a tasks resource constraints
	Resources interface{}
}

UpdateTaskInfo allows updated specific settings to be changed on a task

type UpdateTaskOpts

type UpdateTaskOpts func(context.Context, *Client, *UpdateTaskInfo) error

UpdateTaskOpts allows a caller to update task settings

func WithResources

func WithResources(resources interface{}) UpdateTaskOpts

WithResources sets the provided resources for task updates. Resources must be either a *specs.LinuxResources or a *specs.WindowsResources

type UsageOpt added in v1.3.0

type UsageOpt func(*usageOptions) error

UsageOpt is used to configure the usage calculation

func WithManifestUsage added in v1.3.0

func WithManifestUsage() UsageOpt

WithManifestUsage is used to get the usage for an image based on what is reported by the manifests rather than what exists in the content store. NOTE: This function is best used with the manifest limit set to get a consistent value, otherwise non-existent manifests will be excluded.

func WithSnapshotUsage added in v1.3.0

func WithSnapshotUsage() UsageOpt

WithSnapshotUsage will check for referenced snapshots from the image objects and include the snapshot size in the total usage.

func WithUsageManifestLimit added in v1.3.0

func WithUsageManifestLimit(i int) UsageOpt

WithUsageManifestLimit sets the limit to the number of manifests which will be walked for usage. Setting this value to 0 will require all manifests to be walked, returning ErrNotFound if manifests are missing. NOTE: By default all manifests which exist will be walked and any non-existent manifests and their subobjects will be ignored.

type Version

type Version struct {
	// Version number
	Version string
	// Revision from git that was built
	Revision string
}

Version of containerd

Directories

Path Synopsis
api
events
Package events has protobuf types for various events that are used in containerd.
Package events has protobuf types for various events that are used in containerd.
services/events/v1
Package events defines the event pushing and subscription service.
Package events defines the event pushing and subscription service.
services/ttrpc/events/v1
Package events defines the event pushing and subscription service.
Package events defines the event pushing and subscription service.
cmd
ctr
contrib
Package defaults provides several common defaults for interacting with containerd.
Package defaults provides several common defaults for interacting with containerd.
Package errdefs defines the common errors used throughout containerd packages.
Package errdefs defines the common errors used throughout containerd packages.
Package filters defines a syntax and parser that can be used for the filtration of items across the containerd API.
Package filters defines a syntax and parser that can be used for the filtration of items across the containerd API.
gc
Package gc experiments with providing central gc tooling to ensure deterministic resource removal within containerd.
Package gc experiments with providing central gc tooling to ensure deterministic resource removal within containerd.
Package identifiers provides common validation for identifiers and keys across containerd.
Package identifiers provides common validation for identifiers and keys across containerd.
archive
Package archive provides a Docker and OCI compatible importer
Package archive provides a Docker and OCI compatible importer
log
Package metadata stores all labels and object specific metadata by namespace.
Package metadata stores all labels and object specific metadata by namespace.
metrics
pkg
oom
progress
Package progress assists in displaying human readable progress information.
Package progress assists in displaying human readable progress information.
Package platforms provides a toolkit for normalizing, matching and specifying container platforms.
Package platforms provides a toolkit for normalizing, matching and specifying container platforms.
protobuf
docker
Package docker provides a general type to represent any way of referencing images within the registry.
Package docker provides a general type to represent any way of referencing images within the registry.
docker/config
config package containers utilities for helping configure the Docker resolver
config package containers utilities for helping configure the Docker resolver
restart
Package restart enables containers to have labels added and monitored to keep the container's task running if it is killed.
Package restart enables containers to have labels added and monitored to keep the container's task running if it is killed.
v1
v2
opt
storage
Package storage provides a metadata storage implementation for snapshot drivers.
Package storage provides a metadata storage implementation for snapshot drivers.
sys

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL