Documentation ¶
Index ¶
- Variables
- func ClearTaskExecutors()
- func GetTaskExecutorFactory(taskType proto.TaskType) taskExecutorFactoryFn
- func RegisterTaskType(taskType proto.TaskType, factory taskExecutorFactoryFn, opts ...TaskTypeOption)
- type BaseTaskExecutor
- func (e *BaseTaskExecutor) Cancel()
- func (e *BaseTaskExecutor) CancelRunningSubtask()
- func (e *BaseTaskExecutor) Close()
- func (e *BaseTaskExecutor) Ctx() context.Context
- func (e *BaseTaskExecutor) GetTaskBase() *proto.TaskBase
- func (*BaseTaskExecutor) Init(_ context.Context) error
- func (e *BaseTaskExecutor) Run(resource *proto.StepResource)
- func (e *BaseTaskExecutor) RunStep(resource *proto.StepResource) (err error)
- type EmptyStepExecutor
- func (*EmptyStepExecutor) Cleanup(context.Context) error
- func (*EmptyStepExecutor) Init(context.Context) error
- func (*EmptyStepExecutor) OnFinished(_ context.Context, _ *proto.Subtask) error
- func (*EmptyStepExecutor) RealtimeSummary() *execute.SubtaskSummary
- func (*EmptyStepExecutor) RunSubtask(context.Context, *proto.Subtask) error
- type Extension
- type Manager
- type Pool
- type TaskExecutor
- type TaskTable
- type TaskTypeOption
Constants ¶
This section is empty.
Variables ¶
var ( // TaskCheckInterval is the interval to check whether there are tasks to run. // TODO maybe change this interval larger for performance. TaskCheckInterval = 300 * time.Millisecond // SubtaskCheckInterval is the interval to check whether there are subtasks to run. // exported for testing. SubtaskCheckInterval = 300 * time.Millisecond // MaxSubtaskCheckInterval is the max interval to check whether there are subtasks to run. // exported for testing. MaxSubtaskCheckInterval = 2 * time.Second )
var ( // ErrCancelSubtask is the cancel cause when cancelling subtasks. ErrCancelSubtask = errors.New("cancel subtasks") // ErrFinishSubtask is the cancel cause when TaskExecutor successfully processed subtasks. ErrFinishSubtask = errors.New("finish subtasks") // ErrNonIdempotentSubtask means the subtask is left in running state and is not idempotent, // so cannot be run again. ErrNonIdempotentSubtask = errors.New("subtask in running state and is not idempotent") // MockTiDBDown is used to mock TiDB node down, return true if it's chosen. MockTiDBDown func(execID string, task *proto.TaskBase) bool )
Functions ¶
func GetTaskExecutorFactory ¶
GetTaskExecutorFactory gets taskExecutorFactory by task type.
func RegisterTaskType ¶
func RegisterTaskType(taskType proto.TaskType, factory taskExecutorFactoryFn, opts ...TaskTypeOption)
RegisterTaskType registers the task type.
Types ¶
type BaseTaskExecutor ¶
type BaseTaskExecutor struct { Extension // contains filtered or unexported fields }
BaseTaskExecutor is the base implementation of TaskExecutor.
func NewBaseTaskExecutor ¶
func NewBaseTaskExecutor(ctx context.Context, id string, task *proto.Task, taskTable TaskTable) *BaseTaskExecutor
NewBaseTaskExecutor creates a new BaseTaskExecutor. see TaskExecutor.Init for why we want to use task-base to create TaskExecutor. TODO: we can refactor this part to pass task base only, but currently ADD-INDEX depends on it to init, so we keep it for now.
func (*BaseTaskExecutor) Cancel ¶
func (e *BaseTaskExecutor) Cancel()
Cancel implements TaskExecutor.Cancel.
func (*BaseTaskExecutor) CancelRunningSubtask ¶
func (e *BaseTaskExecutor) CancelRunningSubtask()
CancelRunningSubtask implements TaskExecutor.CancelRunningSubtask.
func (*BaseTaskExecutor) Close ¶
func (e *BaseTaskExecutor) Close()
Close closes the TaskExecutor when all the subtasks are complete.
func (*BaseTaskExecutor) Ctx ¶
func (e *BaseTaskExecutor) Ctx() context.Context
Ctx returns the context of the task executor. TODO: remove it when add-index.taskexecutor.Init don't depends on it.
func (*BaseTaskExecutor) GetTaskBase ¶
func (e *BaseTaskExecutor) GetTaskBase() *proto.TaskBase
GetTaskBase implements TaskExecutor.GetTaskBase.
func (*BaseTaskExecutor) Init ¶
func (*BaseTaskExecutor) Init(_ context.Context) error
Init implements the TaskExecutor interface.
func (*BaseTaskExecutor) Run ¶
func (e *BaseTaskExecutor) Run(resource *proto.StepResource)
Run implements the TaskExecutor interface.
func (*BaseTaskExecutor) RunStep ¶
func (e *BaseTaskExecutor) RunStep(resource *proto.StepResource) (err error)
RunStep start to fetch and run all subtasks for the step of task on the node. return if there's no subtask to run.
type EmptyStepExecutor ¶
type EmptyStepExecutor struct {
execute.StepExecFrameworkInfo
}
EmptyStepExecutor is an empty Executor. it can be used for the task that does not need to split into subtasks.
func (*EmptyStepExecutor) Cleanup ¶
func (*EmptyStepExecutor) Cleanup(context.Context) error
Cleanup implements the StepExecutor interface.
func (*EmptyStepExecutor) Init ¶
func (*EmptyStepExecutor) Init(context.Context) error
Init implements the StepExecutor interface.
func (*EmptyStepExecutor) OnFinished ¶
OnFinished implements the StepExecutor interface.
func (*EmptyStepExecutor) RealtimeSummary ¶
func (*EmptyStepExecutor) RealtimeSummary() *execute.SubtaskSummary
RealtimeSummary implements the StepExecutor interface.
func (*EmptyStepExecutor) RunSubtask ¶
RunSubtask implements the StepExecutor interface.
type Extension ¶
type Extension interface { // IsIdempotent returns whether the subtask is idempotent. // when tidb restart, the subtask might be left in the running state. // if it's idempotent, the Executor can rerun the subtask, else // the Executor will mark the subtask as failed. IsIdempotent(subtask *proto.Subtask) bool // GetStepExecutor returns the subtask executor for the subtask. // Note: // 1. summary is the summary manager of all subtask of the same type now. // 2. should not retry the error from it. GetStepExecutor(task *proto.Task) (execute.StepExecutor, error) // IsRetryableError returns whether the error is transient. // When error is transient, the framework won't mark subtasks as failed, // then the TaskExecutor can load the subtask again and redo it. IsRetryableError(err error) bool }
Extension extends the TaskExecutor. each task type should implement this interface.
type Manager ¶
type Manager struct {
// contains filtered or unexported fields
}
Manager monitors the task table and manages the taskExecutors.
func NewManager ¶
NewManager creates a new task executor Manager.
func (*Manager) Cancel ¶
func (m *Manager) Cancel()
Cancel cancels the executor manager. used in test to simulate tidb node shutdown.
type Pool ¶
type Pool interface { Run(func()) error RunWithConcurrency(chan func(), uint32) error ReleaseAndWait() }
Pool defines the interface of a pool.
type TaskExecutor ¶
type TaskExecutor interface { // Init initializes the TaskExecutor, the returned error is fatal, it will fail // the task directly, so be careful what to put into it. // The context passing in is Manager.ctx, don't use it to init long-running routines, // as it will NOT be cancelled when the task is finished. // NOTE: do NOT depend on task meta to do initialization, as we plan to pass // task-base to the TaskExecutor in the future, if you need to do some initialization // based on task meta, do it in GetStepExecutor, as execute.StepExecutor is // where subtasks are actually executed. Init(context.Context) error // Run runs the task with given resource, it will try to run each step one by // one, if it cannot find any subtask to run for a while(10s now), it will exit, // so manager can free and reuse the resource. // we assume that all steps will have same resource usage now, will change it // when we support different resource usage for different steps. Run(resource *proto.StepResource) // GetTaskBase returns the task, returned value is for read only, don't change it. GetTaskBase() *proto.TaskBase // CancelRunningSubtask cancels the running subtask and change its state to `cancelled`, // the task executor will keep running, so we can have a context to update the // subtask state or keep handling revert logic. CancelRunningSubtask() // Cancel cancels the task executor, the state of running subtask is not changed. // it's separated with Close as Close mostly mean will wait all resource released // before return, but we only want its context cancelled and check whether it's // closed later. Cancel() // Close closes the TaskExecutor. Close() IsRetryableError(err error) bool }
TaskExecutor is the executor for a task. Each task type should implement this interface. context tree of task execution:
Manager.ctx └── TaskExecutor.ctx: Cancel cancels this one └── RunStep.ctx: CancelRunningSubtask cancels this one
type TaskTable ¶
type TaskTable interface { // GetTaskExecInfoByExecID gets all task exec infos by given execID, if there's // no executable subtask on the execID for some task, it's not returned. GetTaskExecInfoByExecID(ctx context.Context, execID string) ([]*storage.TaskExecInfo, error) GetTasksInStates(ctx context.Context, states ...any) (task []*proto.Task, err error) GetTaskByID(ctx context.Context, taskID int64) (task *proto.Task, err error) GetTaskBaseByID(ctx context.Context, taskID int64) (task *proto.TaskBase, err error) // GetSubtasksByExecIDAndStepAndStates gets all subtasks by given states and execID. GetSubtasksByExecIDAndStepAndStates(ctx context.Context, execID string, taskID int64, step proto.Step, states ...proto.SubtaskState) ([]*proto.Subtask, error) GetFirstSubtaskInStates(ctx context.Context, instanceID string, taskID int64, step proto.Step, states ...proto.SubtaskState) (*proto.Subtask, error) // InitMeta insert the manager information into dist_framework_meta. // Call it when starting task executor or in set variable operation. InitMeta(ctx context.Context, execID string, role string) error // RecoverMeta recover the manager information into dist_framework_meta. // Call it periodically to recover deleted meta. RecoverMeta(ctx context.Context, execID string, role string) error // StartSubtask try to update the subtask's state to running if the subtask is owned by execID. // If the update success, it means the execID's related task executor own the subtask. StartSubtask(ctx context.Context, subtaskID int64, execID string) error // UpdateSubtaskStateAndError update the subtask's state and error. UpdateSubtaskStateAndError(ctx context.Context, execID string, subtaskID int64, state proto.SubtaskState, err error) error // FailSubtask update the task's subtask state to failed and set the err. FailSubtask(ctx context.Context, execID string, taskID int64, err error) error // CancelSubtask update the task's subtasks' state to canceled. CancelSubtask(ctx context.Context, exe string, taskID int64) error // FinishSubtask updates the subtask meta and mark state to succeed. FinishSubtask(ctx context.Context, execID string, subtaskID int64, meta []byte) error // PauseSubtasks update subtasks state to paused. PauseSubtasks(ctx context.Context, execID string, taskID int64) error HasSubtasksInStates(ctx context.Context, execID string, taskID int64, step proto.Step, states ...proto.SubtaskState) (bool, error) // RunningSubtasksBack2Pending update the state of subtask which belongs to this // node from running to pending. // see subtask state machine for more detail. RunningSubtasksBack2Pending(ctx context.Context, subtasks []*proto.SubtaskBase) error }
TaskTable defines the interface to access the task table.
type TaskTypeOption ¶
type TaskTypeOption func(opts *taskTypeOptions)
TaskTypeOption is the option of TaskType.