pod_task.go

/*
Copyright 2015 The Kubernetes Authors All rights reserved.

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 podtask

import (
	"fmt"
	"strings"
	"time"

	"code.google.com/p/go-uuid/uuid"
	"github.com/GoogleCloudPlatform/kubernetes/contrib/mesos/pkg/offers"
	annotation "github.com/GoogleCloudPlatform/kubernetes/contrib/mesos/pkg/scheduler/meta"
	"github.com/GoogleCloudPlatform/kubernetes/contrib/mesos/pkg/scheduler/metrics"
	"github.com/GoogleCloudPlatform/kubernetes/pkg/api"
	"github.com/gogo/protobuf/proto"
	log "github.com/golang/glog"
	mesos "github.com/mesos/mesos-go/mesosproto"
	mutil "github.com/mesos/mesos-go/mesosutil"
)

const (
	DefaultContainerCpus = 0.25 // initial CPU allocated for executor
	DefaultContainerMem  = 64   // initial MB of memory allocated for executor
)

type StateType int

const (
	StatePending StateType = iota
	StateRunning
	StateFinished
	StateUnknown
)

type FlagType string

const (
	Launched = FlagType("launched")
	Bound    = FlagType("bound")
	Deleted  = FlagType("deleted")
)

// A struct that describes a pod task.
type T struct {
	ID          string
	Pod         api.Pod
	Spec        Spec
	Offer       offers.Perishable // thread-safe
	State       StateType
	Flags       map[FlagType]struct{}
	CreateTime  time.Time
	UpdatedTime time.Time // time of the most recent StatusUpdate we've seen from the mesos master

	podStatus  api.PodStatus
	executor   *mesos.ExecutorInfo // readonly
	podKey     string
	launchTime time.Time
	bindTime   time.Time
	mapper     HostPortMappingType
}

type Spec struct {
	SlaveID string
	CPU     float64
	Memory  float64
	PortMap []HostPortMapping
	Ports   []uint64
	Data    []byte
}

// mostly-clone this pod task. the clone will actually share the some fields:
//   - executor    // OK because it's read only
//   - Offer       // OK because it's guarantees safe concurrent access
func (t *T) Clone() *T {
	if t == nil {
		return nil
	}

	// shallow-copy
	clone := *t

	// deep copy
	(&t.Spec).copyTo(&clone.Spec)
	clone.Flags = map[FlagType]struct{}{}
	for k := range t.Flags {
		clone.Flags[k] = struct{}{}
	}
	return &clone
}

func (old *Spec) copyTo(new *Spec) {
	if len(old.PortMap) > 0 {
		new.PortMap = append(([]HostPortMapping)(nil), old.PortMap...)
	}
	if len(old.Ports) > 0 {
		new.Ports = append(([]uint64)(nil), old.Ports...)
	}
	if len(old.Data) > 0 {
		new.Data = append(([]byte)(nil), old.Data...)
	}
}

func (t *T) HasAcceptedOffer() bool {
	return t.Spec.SlaveID != ""
}

func (t *T) GetOfferId() string {
	if t.Offer == nil {
		return ""
	}
	return t.Offer.Details().Id.GetValue()
}

func generateTaskName(pod *api.Pod) string {
	ns := pod.Namespace
	if ns == "" {
		ns = api.NamespaceDefault
	}
	return fmt.Sprintf("%s.%s.pods", pod.Name, ns)
}

func (t *T) BuildTaskInfo() *mesos.TaskInfo {
	info := &mesos.TaskInfo{
		Name:     proto.String(generateTaskName(&t.Pod)),
		TaskId:   mutil.NewTaskID(t.ID),
		SlaveId:  mutil.NewSlaveID(t.Spec.SlaveID),
		Executor: t.executor,
		Data:     t.Spec.Data,
		Resources: []*mesos.Resource{
			mutil.NewScalarResource("cpus", t.Spec.CPU),
			mutil.NewScalarResource("mem", t.Spec.Memory),
		},
	}
	if portsResource := rangeResource("ports", t.Spec.Ports); portsResource != nil {
		info.Resources = append(info.Resources, portsResource)
	}
	return info
}

// Fill the Spec in the T, should be called during k8s scheduling,
// before binding.
// TODO(jdef): remove hardcoded values and make use of actual pod resource settings
func (t *T) FillFromDetails(details *mesos.Offer) error {
	if details == nil {
		//programming error
		panic("offer details are nil")
	}

	log.V(3).Infof("Recording offer(s) %v against pod %v", details.Id, t.Pod.Name)

	t.Spec = Spec{
		SlaveID: details.GetSlaveId().GetValue(),
		CPU:     DefaultContainerCpus,
		Memory:  DefaultContainerMem,
	}

	if mapping, err := t.mapper.Generate(t, details); err != nil {
		t.Reset()
		return err
	} else {
		ports := []uint64{}
		for _, entry := range mapping {
			ports = append(ports, entry.OfferPort)
		}
		t.Spec.PortMap = mapping
		t.Spec.Ports = ports
	}

	// hostname needs of the executor needs to match that of the offer, otherwise
	// the kubelet node status checker/updater is very unhappy
	const HOSTNAME_OVERRIDE_FLAG = "--hostname-override="
	hostname := details.GetHostname() // required field, non-empty
	hostnameOverride := HOSTNAME_OVERRIDE_FLAG + hostname

	argv := t.executor.Command.Arguments
	overwrite := false
	for i, arg := range argv {
		if strings.HasPrefix(arg, HOSTNAME_OVERRIDE_FLAG) {
			overwrite = true
			argv[i] = hostnameOverride
			break
		}
	}
	if !overwrite {
		t.executor.Command.Arguments = append(argv, hostnameOverride)
	}
	return nil
}

// Clear offer-related details from the task, should be called if/when an offer
// has already been assigned to a task but for some reason is no longer valid.
func (t *T) Reset() {
	log.V(3).Infof("Clearing offer(s) from pod %v", t.Pod.Name)
	t.Offer = nil
	t.Spec = Spec{}
}

func (t *T) AcceptOffer(offer *mesos.Offer) bool {
	if offer == nil {
		return false
	}
	var (
		cpus float64 = 0
		mem  float64 = 0
	)
	for _, resource := range offer.Resources {
		if resource.GetName() == "cpus" {
			cpus = *resource.GetScalar().Value
		}

		if resource.GetName() == "mem" {
			mem = *resource.GetScalar().Value
		}
	}
	if _, err := t.mapper.Generate(t, offer); err != nil {
		log.V(3).Info(err)
		return false
	}

	// for now hard-coded, constant values are used for cpus and mem. This is necessary
	// until parent-cgroup integration is finished for mesos and k8sm. Then the k8sm
	// executor can become the parent of pods and subsume their resource usage and
	// therefore be compliant with expectations of mesos executors w/ respect to
	// resource allocation and management.
	//
	// TODO(jdef): remove hardcoded values and make use of actual pod resource settings
	if (cpus < DefaultContainerCpus) || (mem < DefaultContainerMem) {
		log.V(3).Infof("not enough resources: cpus: %f mem: %f", cpus, mem)
		return false
	}
	return true
}

func (t *T) Set(f FlagType) {
	t.Flags[f] = struct{}{}
	if Launched == f {
		t.launchTime = time.Now()
		queueWaitTime := t.launchTime.Sub(t.CreateTime)
		metrics.QueueWaitTime.Observe(metrics.InMicroseconds(queueWaitTime))
	}
}

func (t *T) Has(f FlagType) (exists bool) {
	_, exists = t.Flags[f]
	return
}

func New(ctx api.Context, id string, pod api.Pod, executor *mesos.ExecutorInfo) (*T, error) {
	if executor == nil {
		return nil, fmt.Errorf("illegal argument: executor was nil")
	}
	key, err := MakePodKey(ctx, pod.Name)
	if err != nil {
		return nil, err
	}
	if id == "" {
		id = "pod." + uuid.NewUUID().String()
	}
	task := &T{
		ID:       id,
		Pod:      pod,
		State:    StatePending,
		podKey:   key,
		mapper:   MappingTypeForPod(&pod),
		Flags:    make(map[FlagType]struct{}),
		executor: proto.Clone(executor).(*mesos.ExecutorInfo),
	}
	task.CreateTime = time.Now()
	return task, nil
}

func (t *T) SaveRecoveryInfo(dict map[string]string) {
	dict[annotation.TaskIdKey] = t.ID
	dict[annotation.SlaveIdKey] = t.Spec.SlaveID
	dict[annotation.OfferIdKey] = t.Offer.Details().Id.GetValue()
	dict[annotation.ExecutorIdKey] = t.executor.ExecutorId.GetValue()
}

// reconstruct a task from metadata stashed in a pod entry. there are limited pod states that
// support reconstruction. if we expect to be able to reconstruct state but encounter errors
// in the process then those errors are returned. if the pod is in a seemingly valid state but
// otherwise does not support task reconstruction return false. if we're able to reconstruct
// state then return a reconstructed task and true.
//
// at this time task reconstruction is only supported for pods that have been annotated with
// binding metadata, which implies that they've previously been associated with a task and
// that mesos knows about it.
//
// assumes that the pod data comes from the k8s registry and reflects the desired state.
//
func RecoverFrom(pod api.Pod) (*T, bool, error) {
	// we only expect annotations if pod has been bound, which implies that it has already
	// been scheduled and launched
	if pod.Spec.NodeName == "" && len(pod.Annotations) == 0 {
		log.V(1).Infof("skipping recovery for unbound pod %v/%v", pod.Namespace, pod.Name)
		return nil, false, nil
	}

	// only process pods that are not in a terminal state
	switch pod.Status.Phase {
	case api.PodPending, api.PodRunning, api.PodUnknown: // continue
	default:
		log.V(1).Infof("skipping recovery for terminal pod %v/%v", pod.Namespace, pod.Name)
		return nil, false, nil
	}

	ctx := api.WithNamespace(api.NewDefaultContext(), pod.Namespace)
	key, err := MakePodKey(ctx, pod.Name)
	if err != nil {
		return nil, false, err
	}

	//TODO(jdef) recover ports (and other resource requirements?) from the pod spec as well

	now := time.Now()
	t := &T{
		Pod:        pod,
		CreateTime: now,
		podKey:     key,
		State:      StatePending, // possibly running? mesos will tell us during reconciliation
		Flags:      make(map[FlagType]struct{}),
		mapper:     MappingTypeForPod(&pod),
		launchTime: now,
		bindTime:   now,
	}
	var (
		offerId  string
		hostname string
	)
	for _, k := range []string{
		annotation.BindingHostKey,
		annotation.TaskIdKey,
		annotation.SlaveIdKey,
		annotation.OfferIdKey,
		annotation.ExecutorIdKey,
	} {
		v, found := pod.Annotations[k]
		if !found {
			return nil, false, fmt.Errorf("incomplete metadata: missing value for pod annotation: %v", k)
		}
		switch k {
		case annotation.BindingHostKey:
			hostname = v
		case annotation.SlaveIdKey:
			t.Spec.SlaveID = v
		case annotation.OfferIdKey:
			offerId = v
		case annotation.TaskIdKey:
			t.ID = v
		case annotation.ExecutorIdKey:
			// this is nowhere near sufficient to re-launch a task, but we really just
			// want this for tracking
			t.executor = &mesos.ExecutorInfo{ExecutorId: mutil.NewExecutorID(v)}
		}
	}
	t.Offer = offers.Expired(offerId, hostname, 0)
	t.Flags[Launched] = struct{}{}
	t.Flags[Bound] = struct{}{}
	return t, true, nil
}