gvisor/pkg/v1/proc/init.go

/*
Copyright The containerd Authors.
Copyright 2018 Google LLC

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    https://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/

package proc

import (
	"context"
	"encoding/json"
	"io"
	"path/filepath"
	"strings"
	"sync"
	"syscall"
	"time"

	"github.com/containerd/console"
	"github.com/containerd/containerd/errdefs"
	"github.com/containerd/containerd/log"
	"github.com/containerd/containerd/mount"
	"github.com/containerd/containerd/runtime/proc"
	"github.com/containerd/fifo"
	runc "github.com/containerd/go-runc"
	specs "github.com/opencontainers/runtime-spec/specs-go"
	"github.com/pkg/errors"

	runsc "github.com/google/gvisor-containerd-shim/pkg/go-runsc"
)

// InitPidFile name of the file that contains the init pid
const InitPidFile = "init.pid"

// Init represents an initial process for a container
type Init struct {
	wg        sync.WaitGroup
	initState initState

	// mu is used to ensure that `Start()` and `Exited()` calls return in
	// the right order when invoked in separate go routines.
	// This is the case within the shim implementation as it makes use of
	// the reaper interface.
	mu sync.Mutex

	waitBlock chan struct{}

	WorkDir string

	id       string
	Bundle   string
	console  console.Console
	Platform proc.Platform
	io       runc.IO
	runtime  *runsc.Runsc
	status   int
	exited   time.Time
	pid      int
	closers  []io.Closer
	stdin    io.Closer
	stdio    proc.Stdio
	Rootfs   string
	IoUID    int
	IoGID    int
	Sandbox  bool
	UserLog  string
	Monitor  ProcessMonitor
}

// NewRunsc returns a new runsc instance for a process
func NewRunsc(root, path, namespace, runtime string, config map[string]string) *runsc.Runsc {
	if root == "" {
		root = RunscRoot
	}
	return &runsc.Runsc{
		Command:      runtime,
		PdeathSignal: syscall.SIGKILL,
		Log:          filepath.Join(path, "log.json"),
		LogFormat:    runc.JSON,
		Root:         filepath.Join(root, namespace),
		Config:       config,
	}
}

// New returns a new init process
func New(id string, runtime *runsc.Runsc, stdio proc.Stdio) *Init {
	p := &Init{
		id:        id,
		runtime:   runtime,
		stdio:     stdio,
		status:    0,
		waitBlock: make(chan struct{}),
	}
	p.initState = &createdState{p: p}
	return p
}

// Create the process with the provided config
func (p *Init) Create(ctx context.Context, r *CreateConfig) (err error) {
	var socket *runc.Socket
	if r.Terminal {
		if socket, err = runc.NewTempConsoleSocket(); err != nil {
			return errors.Wrap(err, "failed to create OCI runtime console socket")
		}
		defer socket.Close()
	} else if hasNoIO(r) {
		if p.io, err = runc.NewNullIO(); err != nil {
			return errors.Wrap(err, "creating new NULL IO")
		}
	} else {
		if p.io, err = runc.NewPipeIO(p.IoUID, p.IoGID, withConditionalIO(p.stdio)); err != nil {
			return errors.Wrap(err, "failed to create OCI runtime io pipes")
		}
	}
	pidFile := filepath.Join(p.Bundle, InitPidFile)
	opts := &runsc.CreateOpts{
		PidFile: pidFile,
	}
	if socket != nil {
		opts.ConsoleSocket = socket
	}
	if p.Sandbox {
		opts.IO = p.io
		// UserLog is only useful for sandbox.
		opts.UserLog = p.UserLog
	}
	if err := p.runtime.Create(ctx, r.ID, r.Bundle, opts); err != nil {
		return p.runtimeError(err, "OCI runtime create failed")
	}
	if r.Stdin != "" {
		sc, err := fifo.OpenFifo(context.Background(), r.Stdin, syscall.O_WRONLY|syscall.O_NONBLOCK, 0)
		if err != nil {
			return errors.Wrapf(err, "failed to open stdin fifo %s", r.Stdin)
		}
		p.stdin = sc
		p.closers = append(p.closers, sc)
	}
	var copyWaitGroup sync.WaitGroup
	ctx, cancel := context.WithTimeout(ctx, 30*time.Second)
	defer cancel()
	if socket != nil {
		console, err := socket.ReceiveMaster()
		if err != nil {
			return errors.Wrap(err, "failed to retrieve console master")
		}
		console, err = p.Platform.CopyConsole(ctx, console, r.Stdin, r.Stdout, r.Stderr, &p.wg, &copyWaitGroup)
		if err != nil {
			return errors.Wrap(err, "failed to start console copy")
		}
		p.console = console
	} else if !hasNoIO(r) {
		if err := copyPipes(ctx, p.io, r.Stdin, r.Stdout, r.Stderr, &p.wg, &copyWaitGroup); err != nil {
			return errors.Wrap(err, "failed to start io pipe copy")
		}
	}

	copyWaitGroup.Wait()
	pid, err := runc.ReadPidFile(pidFile)
	if err != nil {
		return errors.Wrap(err, "failed to retrieve OCI runtime container pid")
	}
	p.pid = pid
	return nil
}

// Wait for the process to exit
func (p *Init) Wait() {
	<-p.waitBlock
}

// ID of the process
func (p *Init) ID() string {
	return p.id
}

// Pid of the process
func (p *Init) Pid() int {
	return p.pid
}

// ExitStatus of the process
func (p *Init) ExitStatus() int {
	p.mu.Lock()
	defer p.mu.Unlock()
	return p.status
}

// ExitedAt at time when the process exited
func (p *Init) ExitedAt() time.Time {
	p.mu.Lock()
	defer p.mu.Unlock()
	return p.exited
}

// Status of the process
func (p *Init) Status(ctx context.Context) (string, error) {
	p.mu.Lock()
	defer p.mu.Unlock()
	c, err := p.runtime.State(ctx, p.id)
	if err != nil {
		if strings.Contains(err.Error(), "does not exist") {
			return "stopped", nil
		}
		return "", p.runtimeError(err, "OCI runtime state failed")
	}
	return p.convertStatus(c.Status), nil
}

// Start the init process
func (p *Init) Start(ctx context.Context) error {
	p.mu.Lock()
	defer p.mu.Unlock()

	return p.initState.Start(ctx)
}

func (p *Init) start(ctx context.Context) error {
	var cio runc.IO
	if !p.Sandbox {
		cio = p.io
	}
	if err := p.runtime.Start(ctx, p.id, cio); err != nil {
		return p.runtimeError(err, "OCI runtime start failed")
	}
	go func() {
		status, err := p.runtime.Wait(context.Background(), p.id)
		if err != nil {
			log.G(ctx).WithError(err).Errorf("Failed to wait for container %q", p.id)
			// TODO(random-liu): Handle runsc kill error.
			if err := p.killAll(ctx); err != nil {
				log.G(ctx).WithError(err).Errorf("Failed to kill container %q", p.id)
			}
			status = internalErrorCode
		}
		ExitCh <- Exit{
			Timestamp: time.Now(),
			ID:        p.id,
			Status:    status,
		}
	}()
	return nil
}

// SetExited of the init process with the next status
func (p *Init) SetExited(status int) {
	p.mu.Lock()
	defer p.mu.Unlock()

	p.initState.SetExited(status)
}

func (p *Init) setExited(status int) {
	p.exited = time.Now()
	p.status = status
	p.Platform.ShutdownConsole(context.Background(), p.console)
	close(p.waitBlock)
}

// Delete the init process
func (p *Init) Delete(ctx context.Context) error {
	p.mu.Lock()
	defer p.mu.Unlock()

	return p.initState.Delete(ctx)
}

func (p *Init) delete(ctx context.Context) error {
	p.killAll(ctx)
	p.wg.Wait()
	err := p.runtime.Delete(ctx, p.id, nil)
	// ignore errors if a runtime has already deleted the process
	// but we still hold metadata and pipes
	//
	// this is common during a checkpoint, runc will delete the container state
	// after a checkpoint and the container will no longer exist within runc
	if err != nil {
		if strings.Contains(err.Error(), "does not exist") {
			err = nil
		} else {
			err = p.runtimeError(err, "failed to delete task")
		}
	}
	if p.io != nil {
		for _, c := range p.closers {
			c.Close()
		}
		p.io.Close()
	}
	if err2 := mount.UnmountAll(p.Rootfs, 0); err2 != nil {
		log.G(ctx).WithError(err2).Warn("failed to cleanup rootfs mount")
		if err == nil {
			err = errors.Wrap(err2, "failed rootfs umount")
		}
	}
	return err
}

// Resize the init processes console
func (p *Init) Resize(ws console.WinSize) error {
	p.mu.Lock()
	defer p.mu.Unlock()

	if p.console == nil {
		return nil
	}
	return p.console.Resize(ws)
}

func (p *Init) resize(ws console.WinSize) error {
	if p.console == nil {
		return nil
	}
	return p.console.Resize(ws)
}

// Kill the init process
func (p *Init) Kill(ctx context.Context, signal uint32, all bool) error {
	p.mu.Lock()
	defer p.mu.Unlock()

	return p.initState.Kill(ctx, signal, all)
}

func (p *Init) kill(context context.Context, signal uint32, all bool) error {
	var (
		killErr error
		backoff = 100 * time.Millisecond
	)
	timeout := 1 * time.Second
	for start := time.Now(); time.Now().Sub(start) < timeout; {
		c, err := p.runtime.State(context, p.id)
		if err != nil {
			if strings.Contains(err.Error(), "does not exist") {
				return errors.Wrapf(errdefs.ErrNotFound, "no such process")
			}
			return p.runtimeError(err, "OCI runtime state failed")
		}
		// For runsc, signal only works when container is running state.
		// If the container is not in running state, directly return
		// "no such process"
		if p.convertStatus(c.Status) == "stopped" {
			return errors.Wrapf(errdefs.ErrNotFound, "no such process")
		}
		killErr = p.runtime.Kill(context, p.id, int(signal), &runsc.KillOpts{
			All: all,
		})
		if killErr == nil {
			return nil
		}
		time.Sleep(backoff)
		backoff *= 2
	}
	return p.runtimeError(killErr, "kill timeout")
}

// KillAll processes belonging to the init process
func (p *Init) KillAll(context context.Context) error {
	p.mu.Lock()
	defer p.mu.Unlock()
	return p.killAll(context)
}

func (p *Init) killAll(context context.Context) error {
	p.runtime.Kill(context, p.id, int(syscall.SIGKILL), &runsc.KillOpts{
		All: true,
	})
	// Ignore error handling for `runsc kill --all` for now.
	// * If it doesn't return error, it is good;
	// * If it returns error, consider the container has already stopped.
	// TODO: Fix `runsc kill --all` error handling.
	return nil
}

// Stdin of the process
func (p *Init) Stdin() io.Closer {
	return p.stdin
}

// Runtime returns the OCI runtime configured for the init process
func (p *Init) Runtime() *runsc.Runsc {
	return p.runtime
}

// Exec returns a new child process
func (p *Init) Exec(ctx context.Context, path string, r *ExecConfig) (proc.Process, error) {
	p.mu.Lock()
	defer p.mu.Unlock()

	return p.initState.Exec(ctx, path, r)
}

// exec returns a new exec'd process
func (p *Init) exec(ctx context.Context, path string, r *ExecConfig) (proc.Process, error) {
	// process exec request
	var spec specs.Process
	if err := json.Unmarshal(r.Spec.Value, &spec); err != nil {
		return nil, err
	}
	spec.Terminal = r.Terminal

	e := &execProcess{
		id:     r.ID,
		path:   path,
		parent: p,
		spec:   spec,
		stdio: proc.Stdio{
			Stdin:    r.Stdin,
			Stdout:   r.Stdout,
			Stderr:   r.Stderr,
			Terminal: r.Terminal,
		},
		waitBlock: make(chan struct{}),
	}
	e.execState = &execCreatedState{p: e}
	return e, nil
}

// Stdio of the process
func (p *Init) Stdio() proc.Stdio {
	return p.stdio
}

func (p *Init) runtimeError(rErr error, msg string) error {
	if rErr == nil {
		return nil
	}

	rMsg, err := getLastRuntimeError(p.runtime)
	switch {
	case err != nil:
		return errors.Wrapf(rErr, "%s: %s (%s)", msg, "unable to retrieve OCI runtime error", err.Error())
	case rMsg == "":
		return errors.Wrap(rErr, msg)
	default:
		return errors.Errorf("%s: %s", msg, rMsg)
	}
}

func (p *Init) convertStatus(status string) string {
	if status == "created" && !p.Sandbox && p.status == internalErrorCode {
		// Treat start failure state for non-root container as stopped.
		return "stopped"
	}
	return status
}

func withConditionalIO(c proc.Stdio) runc.IOOpt {
	return func(o *runc.IOOption) {
		o.OpenStdin = c.Stdin != ""
		o.OpenStdout = c.Stdout != ""
		o.OpenStderr = c.Stderr != ""
	}
}