tasks/service/matching/db.go

// The MIT License
//
// Copyright (c) 2020 Temporal Technologies Inc.  All rights reserved.
//
// Copyright (c) 2020 Uber Technologies, Inc.
//
// Permission is hereby granted, free of charge, to any person obtaining a copy
// of this software and associated documentation files (the "Software"), to deal
// in the Software without restriction, including without limitation the rights
// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
// copies of the Software, and to permit persons to whom the Software is
// furnished to do so, subject to the following conditions:
//
// The above copyright notice and this permission notice shall be included in
// all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
// THE SOFTWARE.

package matching

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

	enumspb "go.temporal.io/api/enums/v1"
	"go.temporal.io/api/serviceerror"

	"go.temporal.io/server/api/matchingservice/v1"
	persistencespb "go.temporal.io/server/api/persistence/v1"
	"go.temporal.io/server/common"
	"go.temporal.io/server/common/log"
	"go.temporal.io/server/common/log/tag"
	"go.temporal.io/server/common/namespace"
	"go.temporal.io/server/common/persistence"
	"go.temporal.io/server/common/primitives/timestamp"
)

const (
	initialRangeID     = 1 // Id of the first range of a new task queue
	stickyTaskQueueTTL = 24 * time.Hour

	// userDataEnabled is the default state: user data is enabled.
	userDataEnabled userDataState = iota
	// userDataDisabled means user data is disabled due to the LoadUserData dynamic config
	// being turned off on this node or the parent node. This should cause GetUserData to
	// return a FailedPrecondition error.
	userDataDisabled
	// userDataSpecificVersion means this tqm/db is for a specific version set, which doesn't
	// have its own user data and it should not be used. This should cause GetUserData to
	// return an Internal error (access would indicate a bug).
	userDataSpecificVersion
)

type (
	taskQueueDB struct {
		sync.Mutex
		namespaceID     namespace.ID
		taskQueue       *taskQueueID
		taskQueueKind   enumspb.TaskQueueKind
		rangeID         int64
		ackLevel        int64
		userData        *persistencespb.VersionedTaskQueueUserData
		userDataChanged chan struct{}
		userDataState   userDataState
		store           persistence.TaskManager
		logger          log.Logger
		matchingClient  matchingservice.MatchingServiceClient
	}
	taskQueueState struct {
		rangeID  int64
		ackLevel int64
	}

	userDataState int
)

var (
	errUserDataNoMutateNonRoot = serviceerror.NewInvalidArgument("can only mutate user data on root workflow task queue")

	// This is an internal error when requesting user data on a TQM created for a specific
	// version set. This indicates a bug in the server since nothing should be using this data.
	errNoUserDataOnVersionedTQM = serviceerror.NewInternal("should not get user data on versioned tqm")

	errUserDataDisabled = serviceerror.NewFailedPrecondition("Task queue user data operations are disabled")
)

// newTaskQueueDB returns an instance of an object that represents
// persistence view of a taskQueue. All mutations / reads to taskQueues
// wrt persistence go through this object.
//
// This class will serialize writes to persistence that do condition updates. There are
// two reasons for doing this:
//   - To work around known Cassandra issue where concurrent LWT to the same partition cause timeout errors
//   - To provide the guarantee that there is only writer who updates taskQueue in persistence at any given point in time
//     This guarantee makes some of the other code simpler and there is no impact to perf because updates to taskqueue are
//     spread out and happen in background routines
func newTaskQueueDB(
	store persistence.TaskManager,
	matchingClient matchingservice.MatchingServiceClient,
	namespaceID namespace.ID,
	taskQueue *taskQueueID,
	kind enumspb.TaskQueueKind,
	logger log.Logger,
) *taskQueueDB {
	return &taskQueueDB{
		namespaceID:     namespaceID,
		taskQueue:       taskQueue,
		taskQueueKind:   kind,
		store:           store,
		logger:          logger,
		userDataChanged: make(chan struct{}),
		matchingClient:  matchingClient,
	}
}

// RangeID returns the current persistence view of rangeID
func (db *taskQueueDB) RangeID() int64 {
	db.Lock()
	defer db.Unlock()
	return db.rangeID
}

// RenewLease renews the lease on a taskqueue. If there is no previous lease,
// this method will attempt to steal taskqueue from current owner
func (db *taskQueueDB) RenewLease(
	ctx context.Context,
) (taskQueueState, error) {
	db.Lock()
	defer db.Unlock()

	if db.rangeID == 0 {
		if err := db.takeOverTaskQueueLocked(ctx); err != nil {
			return taskQueueState{}, err
		}
	} else {
		if err := db.renewTaskQueueLocked(ctx, db.rangeID+1); err != nil {
			return taskQueueState{}, err
		}
	}
	return taskQueueState{rangeID: db.rangeID, ackLevel: db.ackLevel}, nil
}

func (db *taskQueueDB) takeOverTaskQueueLocked(
	ctx context.Context,
) error {
	response, err := db.store.GetTaskQueue(ctx, &persistence.GetTaskQueueRequest{
		NamespaceID: db.namespaceID.String(),
		TaskQueue:   db.taskQueue.FullName(),
		TaskType:    db.taskQueue.taskType,
	})
	switch err.(type) {
	case nil:
		response.TaskQueueInfo.Kind = db.taskQueueKind
		response.TaskQueueInfo.ExpiryTime = db.expiryTime()
		response.TaskQueueInfo.LastUpdateTime = timestamp.TimeNowPtrUtc()
		if _, err := db.store.UpdateTaskQueue(ctx, &persistence.UpdateTaskQueueRequest{
			RangeID:       response.RangeID + 1,
			TaskQueueInfo: response.TaskQueueInfo,
			PrevRangeID:   response.RangeID,
		}); err != nil {
			return err
		}
		db.ackLevel = response.TaskQueueInfo.AckLevel
		db.rangeID = response.RangeID + 1
		return nil

	case *serviceerror.NotFound:
		if _, err := db.store.CreateTaskQueue(ctx, &persistence.CreateTaskQueueRequest{
			RangeID:       initialRangeID,
			TaskQueueInfo: db.cachedQueueInfo(),
		}); err != nil {
			return err
		}
		db.rangeID = initialRangeID
		return nil

	default:
		return err
	}
}

func (db *taskQueueDB) renewTaskQueueLocked(
	ctx context.Context,
	rangeID int64,
) error {
	if _, err := db.store.UpdateTaskQueue(ctx, &persistence.UpdateTaskQueueRequest{
		RangeID:       rangeID,
		TaskQueueInfo: db.cachedQueueInfo(),
		PrevRangeID:   db.rangeID,
	}); err != nil {
		return err
	}

	db.rangeID = rangeID
	return nil
}

// UpdateState updates the taskQueue state with the given value
func (db *taskQueueDB) UpdateState(
	ctx context.Context,
	ackLevel int64,
) error {
	db.Lock()
	defer db.Unlock()
	queueInfo := db.cachedQueueInfo()
	queueInfo.AckLevel = ackLevel
	_, err := db.store.UpdateTaskQueue(ctx, &persistence.UpdateTaskQueueRequest{
		RangeID:       db.rangeID,
		TaskQueueInfo: queueInfo,
		PrevRangeID:   db.rangeID,
	})
	if err == nil {
		db.ackLevel = ackLevel
	}
	return err
}

// CreateTasks creates a batch of given tasks for this task queue
func (db *taskQueueDB) CreateTasks(
	ctx context.Context,
	tasks []*persistencespb.AllocatedTaskInfo,
) (*persistence.CreateTasksResponse, error) {
	db.Lock()
	defer db.Unlock()
	return db.store.CreateTasks(
		ctx,
		&persistence.CreateTasksRequest{
			TaskQueueInfo: &persistence.PersistedTaskQueueInfo{
				Data:    db.cachedQueueInfo(),
				RangeID: db.rangeID,
			},
			Tasks: tasks,
		})
}

// GetTasks returns a batch of tasks between the given range
func (db *taskQueueDB) GetTasks(
	ctx context.Context,
	inclusiveMinTaskID int64,
	exclusiveMaxTaskID int64,
	batchSize int,
) (*persistence.GetTasksResponse, error) {
	return db.store.GetTasks(ctx, &persistence.GetTasksRequest{
		NamespaceID:        db.namespaceID.String(),
		TaskQueue:          db.taskQueue.FullName(),
		TaskType:           db.taskQueue.taskType,
		PageSize:           batchSize,
		InclusiveMinTaskID: inclusiveMinTaskID,
		ExclusiveMaxTaskID: exclusiveMaxTaskID,
	})
}

// CompleteTask deletes a single task from this task queue
func (db *taskQueueDB) CompleteTask(
	ctx context.Context,
	taskID int64,
) error {
	err := db.store.CompleteTask(ctx, &persistence.CompleteTaskRequest{
		TaskQueue: &persistence.TaskQueueKey{
			NamespaceID:   db.namespaceID.String(),
			TaskQueueName: db.taskQueue.FullName(),
			TaskQueueType: db.taskQueue.taskType,
		},
		TaskID: taskID,
	})
	if err != nil {
		db.logger.Error("Persistent store operation failure",
			tag.StoreOperationCompleteTask,
			tag.Error(err),
			tag.TaskID(taskID),
			tag.WorkflowTaskQueueType(db.taskQueue.taskType),
			tag.WorkflowTaskQueueName(db.taskQueue.FullName()),
		)
	}
	return err
}

// CompleteTasksLessThan deletes of tasks less than the given taskID. Limit is
// the upper bound of number of tasks that can be deleted by this method. It may
// or may not be honored
func (db *taskQueueDB) CompleteTasksLessThan(
	ctx context.Context,
	exclusiveMaxTaskID int64,
	limit int,
) (int, error) {
	n, err := db.store.CompleteTasksLessThan(ctx, &persistence.CompleteTasksLessThanRequest{
		NamespaceID:        db.namespaceID.String(),
		TaskQueueName:      db.taskQueue.FullName(),
		TaskType:           db.taskQueue.taskType,
		ExclusiveMaxTaskID: exclusiveMaxTaskID,
		Limit:              limit,
	})
	if err != nil {
		db.logger.Error("Persistent store operation failure",
			tag.StoreOperationCompleteTasksLessThan,
			tag.Error(err),
			tag.TaskID(exclusiveMaxTaskID),
			tag.WorkflowTaskQueueType(db.taskQueue.taskType),
			tag.WorkflowTaskQueueName(db.taskQueue.FullName()),
		)
	}
	return n, err
}

// Returns true if we are storing user data in the db. We need to be the root partition,
// workflow type, unversioned, and also a normal queue.
func (db *taskQueueDB) DbStoresUserData() bool {
	return db.taskQueue.OwnsUserData() && db.taskQueueKind == enumspb.TASK_QUEUE_KIND_NORMAL
}

// GetUserData returns the versioning data for this task queue. Do not mutate the returned pointer, as doing so
// will cause cache inconsistency.
func (db *taskQueueDB) GetUserData(
	ctx context.Context,
) (*persistencespb.VersionedTaskQueueUserData, chan struct{}, error) {
	db.Lock()
	defer db.Unlock()
	return db.getUserDataLocked()
}

func (db *taskQueueDB) getUserDataLocked() (*persistencespb.VersionedTaskQueueUserData, chan struct{}, error) {
	switch db.userDataState {
	case userDataEnabled:
		return db.userData, db.userDataChanged, nil
	case userDataDisabled:
		return nil, nil, errUserDataDisabled
	case userDataSpecificVersion:
		return nil, nil, errNoUserDataOnVersionedTQM
	default:
		// shouldn't happen
		return nil, nil, serviceerror.NewInternal("unexpected user data enabled state")
	}
}

func (db *taskQueueDB) setUserDataLocked(userData *persistencespb.VersionedTaskQueueUserData) {
	db.userData = userData
	close(db.userDataChanged)
	db.userDataChanged = make(chan struct{})
}

// Loads user data from db (called only on initialization of taskQueueManager).
func (db *taskQueueDB) loadUserData(ctx context.Context) error {
	if !db.DbStoresUserData() {
		return nil
	}

	response, err := db.store.GetTaskQueueUserData(ctx, &persistence.GetTaskQueueUserDataRequest{
		NamespaceID: db.namespaceID.String(),
		TaskQueue:   db.taskQueue.BaseNameString(),
	})
	if common.IsNotFoundError(err) {
		// not all task queues have user data
		response, err = &persistence.GetTaskQueueUserDataResponse{}, nil
	}
	if err != nil {
		return err
	}

	db.Lock()
	defer db.Unlock()
	db.setUserDataLocked(response.UserData)

	return nil
}

func (db *taskQueueDB) setUserDataState(setUserDataState userDataState) {
	db.Lock()
	defer db.Unlock()
	db.userDataState = setUserDataState
}

// UpdateUserData allows callers to update user data (such as worker build IDs) for this task queue. The pointer passed
// to the update function is guaranteed to be non-nil.
// Note that the user data's clock may be nil and should be initialized externally where there's access to the cluster
// metadata and the cluster ID can be obtained.
//
// If knownVersion is non 0 and not equal to the current version, the update will fail.
//
// The DB write is performed remotely on an owning node for all user data updates in the namespace.
//
// On success returns a pointer to the updated data, which must *not* be mutated, and a boolean indicating whether the
// data should be replicated.
func (db *taskQueueDB) UpdateUserData(
	ctx context.Context,
	updateFn UserDataUpdateFunc,
	knownVersion int64,
	taskQueueLimitPerBuildId int,
) (*persistencespb.VersionedTaskQueueUserData, bool, error) {
	if !db.DbStoresUserData() {
		return nil, false, errUserDataNoMutateNonRoot
	}

	db.Lock()
	defer db.Unlock()

	userData, _, err := db.getUserDataLocked()
	if err != nil {
		return nil, false, err
	}

	preUpdateData := userData.GetData()
	preUpdateVersion := userData.GetVersion()
	if preUpdateData == nil {
		preUpdateData = &persistencespb.TaskQueueUserData{}
	}
	if knownVersion > 0 && preUpdateVersion != knownVersion {
		return nil, false, serviceerror.NewFailedPrecondition(fmt.Sprintf("user data version mismatch: requested: %d, current: %d", knownVersion, preUpdateVersion))
	}
	updatedUserData, shouldReplicate, err := updateFn(preUpdateData)
	if err != nil {
		return nil, false, err
	}
	added, removed := GetBuildIdDeltas(preUpdateData.GetVersioningData(), updatedUserData.GetVersioningData())
	if taskQueueLimitPerBuildId > 0 && len(added) > 0 {
		// We iterate here but in practice there should only be a single build Id added when the limit is enforced.
		// We do not enforce the limit when applying replication events.
		for _, buildId := range added {
			numTaskQueues, err := db.store.CountTaskQueuesByBuildId(ctx, &persistence.CountTaskQueuesByBuildIdRequest{
				NamespaceID: db.namespaceID.String(),
				BuildID:     buildId,
			})
			if err != nil {
				return nil, false, err
			}
			if numTaskQueues >= taskQueueLimitPerBuildId {
				return nil, false, serviceerror.NewFailedPrecondition(fmt.Sprintf("Exceeded max task queues allowed to be mapped to a single build id: %d", taskQueueLimitPerBuildId))
			}
		}
	}

	_, err = db.matchingClient.UpdateTaskQueueUserData(ctx, &matchingservice.UpdateTaskQueueUserDataRequest{
		NamespaceId:     db.namespaceID.String(),
		TaskQueue:       db.cachedQueueInfo().Name,
		UserData:        &persistencespb.VersionedTaskQueueUserData{Version: preUpdateVersion, Data: updatedUserData},
		BuildIdsAdded:   added,
		BuildIdsRemoved: removed,
	})
	var updatedVersionedData *persistencespb.VersionedTaskQueueUserData
	if err == nil {
		updatedVersionedData = &persistencespb.VersionedTaskQueueUserData{Version: preUpdateVersion + 1, Data: updatedUserData}
		db.setUserDataLocked(updatedVersionedData)
	}
	return updatedVersionedData, shouldReplicate, err
}

func (db *taskQueueDB) setUserDataForNonOwningPartition(userData *persistencespb.VersionedTaskQueueUserData) {
	db.Lock()
	defer db.Unlock()
	db.setUserDataLocked(userData)
}

func (db *taskQueueDB) expiryTime() *time.Time {
	switch db.taskQueueKind {
	case enumspb.TASK_QUEUE_KIND_NORMAL:
		return nil
	case enumspb.TASK_QUEUE_KIND_STICKY:
		return timestamp.TimePtr(time.Now().UTC().Add(stickyTaskQueueTTL))
	default:
		panic(fmt.Sprintf("taskQueueDB encountered unknown task kind: %v", db.taskQueueKind))
	}
}

func (db *taskQueueDB) cachedQueueInfo() *persistencespb.TaskQueueInfo {
	return &persistencespb.TaskQueueInfo{
		NamespaceId:    db.namespaceID.String(),
		Name:           db.taskQueue.FullName(),
		TaskType:       db.taskQueue.taskType,
		Kind:           db.taskQueueKind,
		AckLevel:       db.ackLevel,
		ExpiryTime:     db.expiryTime(),
		LastUpdateTime: timestamp.TimeNowPtrUtc(),
	}
}