containerd/runtime/restart/monitor/monitor.go

/*
   Copyright The containerd Authors.

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

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

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

package monitor

import (
	"context"
	"fmt"
	"strconv"
	"sync"
	"time"

	"github.com/containerd/containerd"
	"github.com/containerd/containerd/namespaces"
	"github.com/containerd/containerd/plugin"
	"github.com/containerd/containerd/runtime/restart"
	"github.com/sirupsen/logrus"
)

type duration struct {
	time.Duration
}

func (d *duration) UnmarshalText(text []byte) error {
	var err error
	d.Duration, err = time.ParseDuration(string(text))
	return err
}

func (d duration) MarshalText() ([]byte, error) {
	return []byte(d.Duration.String()), nil
}

// Config for the restart monitor
type Config struct {
	// Interval for how long to wait to check for state changes
	Interval duration `toml:"interval"`
}

func init() {
	plugin.Register(&plugin.Registration{
		Type: plugin.InternalPlugin,
		Requires: []plugin.Type{
			plugin.EventPlugin,
			plugin.ServicePlugin,
		},
		ID: "restart",
		Config: &Config{
			Interval: duration{
				Duration: 10 * time.Second,
			},
		},
		InitFn: func(ic *plugin.InitContext) (interface{}, error) {
			ic.Meta.Capabilities = []string{"no", "always", "on-failure", "unless-stopped"}
			client, err := containerd.New("", containerd.WithInMemoryServices(ic))
			if err != nil {
				return nil, err
			}
			m := &monitor{
				client: client,
			}
			go m.run(ic.Config.(*Config).Interval.Duration)
			return m, nil
		},
	})
}

type change interface {
	apply(context.Context, *containerd.Client) error
}

type monitor struct {
	client *containerd.Client
}

func (m *monitor) run(interval time.Duration) {
	if interval == 0 {
		interval = 10 * time.Second
	}
	for {
		if err := m.reconcile(context.Background()); err != nil {
			logrus.WithError(err).Error("reconcile")
		}
		time.Sleep(interval)
	}
}

func (m *monitor) reconcile(ctx context.Context) error {
	ns, err := m.client.NamespaceService().List(ctx)
	if err != nil {
		return err
	}
	var wgNSLoop sync.WaitGroup
	for _, name := range ns {
		name := name
		wgNSLoop.Add(1)
		go func() {
			defer wgNSLoop.Done()
			ctx := namespaces.WithNamespace(ctx, name)
			changes, err := m.monitor(ctx)
			if err != nil {
				logrus.WithError(err).Error("monitor for changes")
				return
			}
			var wgChangesLoop sync.WaitGroup
			for _, c := range changes {
				c := c
				wgChangesLoop.Add(1)
				go func() {
					defer wgChangesLoop.Done()
					if err := c.apply(ctx, m.client); err != nil {
						logrus.WithError(err).Error("apply change")
					}
				}()
			}
			wgChangesLoop.Wait()
		}()
	}
	wgNSLoop.Wait()
	return nil
}

func (m *monitor) monitor(ctx context.Context) ([]change, error) {
	containers, err := m.client.Containers(ctx, fmt.Sprintf("labels.%q", restart.StatusLabel))
	if err != nil {
		return nil, err
	}
	var changes []change
	for _, c := range containers {
		var (
			task   containerd.Task
			status containerd.Status
			err    error
		)
		labels, err := c.Labels(ctx)
		if err != nil {
			return nil, err
		}
		desiredStatus := containerd.ProcessStatus(labels[restart.StatusLabel])
		if task, err = c.Task(ctx, nil); err == nil {
			if status, err = task.Status(ctx); err == nil {
				if desiredStatus == status.Status {
					continue
				}
			}
		}

		// Task or Status return error, only desired to running
		if err != nil {
			logrus.WithError(err).Error("monitor")
			if desiredStatus == containerd.Stopped {
				continue
			}
		}

		// Known issue:
		// The status may be empty when task failed but was deleted,
		// which will result in an `on-failure` restart policy reconcile error.
		switch desiredStatus {
		case containerd.Running:
			if !restart.Reconcile(status, labels) {
				continue
			}
			restartCount, _ := strconv.Atoi(labels[restart.CountLabel])
			changes = append(changes, &startChange{
				container: c,
				logPath:   labels[restart.LogPathLabel],
				logURI:    labels[restart.LogURILabel],
				count:     restartCount + 1,
			})
		case containerd.Stopped:
			changes = append(changes, &stopChange{
				container: c,
			})
		}
	}
	return changes, nil
}