Admin: misc improvements on admin server and workers. EC now works. (#7055)

* initial design

* added simulation as tests

* reorganized the codebase to move the simulation framework and tests into their own dedicated package

* integration test. ec worker task

* remove "enhanced" reference

* start master, volume servers, filer

Current Status
 Master: Healthy and running (port 9333)
 Filer: Healthy and running (port 8888)
 Volume Servers: All 6 servers running (ports 8080-8085)
🔄 Admin/Workers: Will start when dependencies are ready

* generate write load

* tasks are assigned

* admin start wtih grpc port. worker has its own working directory

* Update .gitignore

* working worker and admin. Task detection is not working yet.

* compiles, detection uses volumeSizeLimitMB from master

* compiles

* worker retries connecting to admin

* build and restart

* rendering pending tasks

* skip task ID column

* sticky worker id

* test canScheduleTaskNow

* worker reconnect to admin

* clean up logs

* worker register itself first

* worker can run ec work and report status

but:
1. one volume should not be repeatedly worked on.
2. ec shards needs to be distributed and source data should be deleted.

* move ec task logic

* listing ec shards

* local copy, ec. Need to distribute.

* ec is mostly working now

* distribution of ec shards needs improvement
* need configuration to enable ec

* show ec volumes

* interval field UI component

* rename

* integration test with vauuming

* garbage percentage threshold

* fix warning

* display ec shard sizes

* fix ec volumes list

* Update ui.go

* show default values

* ensure correct default value

* MaintenanceConfig use ConfigField

* use schema defined defaults

* config

* reduce duplication

* refactor to use BaseUIProvider

* each task register its schema

* checkECEncodingCandidate use ecDetector

* use vacuumDetector

* use volumeSizeLimitMB

* remove

remove

* remove unused

* refactor

* use new framework

* remove v2 reference

* refactor

* left menu can scroll now

* The maintenance manager was not being initialized when no data directory was configured for persistent storage.

* saving config

* Update task_config_schema_templ.go

* enable/disable tasks

* protobuf encoded task configurations

* fix system settings

* use ui component

* remove logs

* interface{} Reduction

* reduce interface{}

* reduce interface{}

* avoid from/to map

* reduce interface{}

* refactor

* keep it DRY

* added logging

* debug messages

* debug level

* debug

* show the log caller line

* use configured task policy

* log level

* handle admin heartbeat response

* Update worker.go

* fix EC rack and dc count

* Report task status to admin server

* fix task logging, simplify interface checking, use erasure_coding constants

* factor in empty volume server during task planning

* volume.list adds disk id

* track disk id also

* fix locking scheduled and manual scanning

* add active topology

* simplify task detector

* ec task completed, but shards are not showing up

* implement ec in ec_typed.go

* adjust log level

* dedup

* implementing ec copying shards and only ecx files

* use disk id when distributing ec shards

🎯 Planning: ActiveTopology creates DestinationPlan with specific TargetDisk
📦 Task Creation: maintenance_integration.go creates ECDestination with DiskId
🚀 Task Execution: EC task passes DiskId in VolumeEcShardsCopyRequest
💾 Volume Server: Receives disk_id and stores shards on specific disk (vs.store.Locations[req.DiskId])
📂 File System: EC shards and metadata land in the exact disk directory planned

* Delete original volume from all locations

* clean up existing shard locations

* local encoding and distributing

* Update docker/admin_integration/EC-TESTING-README.md

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* check volume id range

* simplify

* fix tests

* fix types

* clean up logs and tests

---------

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
Chris Lu
2025-07-30 12:38:03 -07:00
committed by GitHub
parent 64198dad83
commit 891a2fb6eb
130 changed files with 27737 additions and 4429 deletions

View File

@@ -2,31 +2,71 @@ package tasks
import (
"context"
"fmt"
"sync"
"time"
"github.com/seaweedfs/seaweedfs/weed/glog"
"github.com/seaweedfs/seaweedfs/weed/worker/types"
)
// BaseTask provides common functionality for all tasks
type BaseTask struct {
taskType types.TaskType
taskID string
progress float64
cancelled bool
mutex sync.RWMutex
startTime time.Time
estimatedDuration time.Duration
logger TaskLogger
loggerConfig TaskLoggerConfig
progressCallback func(float64) // Callback function for progress updates
}
// NewBaseTask creates a new base task
func NewBaseTask(taskType types.TaskType) *BaseTask {
return &BaseTask{
taskType: taskType,
progress: 0.0,
cancelled: false,
taskType: taskType,
progress: 0.0,
cancelled: false,
loggerConfig: DefaultTaskLoggerConfig(),
}
}
// NewBaseTaskWithLogger creates a new base task with custom logger configuration
func NewBaseTaskWithLogger(taskType types.TaskType, loggerConfig TaskLoggerConfig) *BaseTask {
return &BaseTask{
taskType: taskType,
progress: 0.0,
cancelled: false,
loggerConfig: loggerConfig,
}
}
// InitializeLogger initializes the task logger with task details
func (t *BaseTask) InitializeLogger(taskID string, workerID string, params types.TaskParams) error {
return t.InitializeTaskLogger(taskID, workerID, params)
}
// InitializeTaskLogger initializes the task logger with task details (LoggerProvider interface)
func (t *BaseTask) InitializeTaskLogger(taskID string, workerID string, params types.TaskParams) error {
t.mutex.Lock()
defer t.mutex.Unlock()
t.taskID = taskID
logger, err := NewTaskLogger(taskID, t.taskType, workerID, params, t.loggerConfig)
if err != nil {
return fmt.Errorf("failed to initialize task logger: %w", err)
}
t.logger = logger
t.logger.Info("BaseTask initialized for task %s (type: %s)", taskID, t.taskType)
return nil
}
// Type returns the task type
func (t *BaseTask) Type() types.TaskType {
return t.taskType
@@ -39,24 +79,47 @@ func (t *BaseTask) GetProgress() float64 {
return t.progress
}
// SetProgress sets the current progress
// SetProgress sets the current progress and logs it
func (t *BaseTask) SetProgress(progress float64) {
t.mutex.Lock()
defer t.mutex.Unlock()
if progress < 0 {
progress = 0
}
if progress > 100 {
progress = 100
}
oldProgress := t.progress
callback := t.progressCallback
t.progress = progress
t.mutex.Unlock()
// Log progress change
if t.logger != nil && progress != oldProgress {
t.logger.LogProgress(progress, fmt.Sprintf("Progress updated from %.1f%% to %.1f%%", oldProgress, progress))
}
// Call progress callback if set
if callback != nil && progress != oldProgress {
callback(progress)
}
}
// Cancel cancels the task
func (t *BaseTask) Cancel() error {
t.mutex.Lock()
defer t.mutex.Unlock()
if t.cancelled {
return nil
}
t.cancelled = true
if t.logger != nil {
t.logger.LogStatus("cancelled", "Task cancelled by request")
t.logger.Warning("Task %s was cancelled", t.taskID)
}
return nil
}
@@ -72,6 +135,10 @@ func (t *BaseTask) SetStartTime(startTime time.Time) {
t.mutex.Lock()
defer t.mutex.Unlock()
t.startTime = startTime
if t.logger != nil {
t.logger.LogStatus("running", fmt.Sprintf("Task started at %s", startTime.Format(time.RFC3339)))
}
}
// GetStartTime returns the task start time
@@ -86,6 +153,13 @@ func (t *BaseTask) SetEstimatedDuration(duration time.Duration) {
t.mutex.Lock()
defer t.mutex.Unlock()
t.estimatedDuration = duration
if t.logger != nil {
t.logger.LogWithFields("INFO", "Estimated duration set", map[string]interface{}{
"estimated_duration": duration.String(),
"estimated_seconds": duration.Seconds(),
})
}
}
// GetEstimatedDuration returns the estimated duration
@@ -95,11 +169,115 @@ func (t *BaseTask) GetEstimatedDuration() time.Duration {
return t.estimatedDuration
}
// ExecuteTask is a wrapper that handles common task execution logic
// SetProgressCallback sets the progress callback function
func (t *BaseTask) SetProgressCallback(callback func(float64)) {
t.mutex.Lock()
defer t.mutex.Unlock()
t.progressCallback = callback
}
// SetLoggerConfig sets the logger configuration for this task
func (t *BaseTask) SetLoggerConfig(config TaskLoggerConfig) {
t.mutex.Lock()
defer t.mutex.Unlock()
t.loggerConfig = config
}
// GetLogger returns the task logger
func (t *BaseTask) GetLogger() TaskLogger {
t.mutex.RLock()
defer t.mutex.RUnlock()
return t.logger
}
// GetTaskLogger returns the task logger (LoggerProvider interface)
func (t *BaseTask) GetTaskLogger() TaskLogger {
t.mutex.RLock()
defer t.mutex.RUnlock()
return t.logger
}
// LogInfo logs an info message
func (t *BaseTask) LogInfo(message string, args ...interface{}) {
if t.logger != nil {
t.logger.Info(message, args...)
}
}
// LogWarning logs a warning message
func (t *BaseTask) LogWarning(message string, args ...interface{}) {
if t.logger != nil {
t.logger.Warning(message, args...)
}
}
// LogError logs an error message
func (t *BaseTask) LogError(message string, args ...interface{}) {
if t.logger != nil {
t.logger.Error(message, args...)
}
}
// LogDebug logs a debug message
func (t *BaseTask) LogDebug(message string, args ...interface{}) {
if t.logger != nil {
t.logger.Debug(message, args...)
}
}
// LogWithFields logs a message with structured fields
func (t *BaseTask) LogWithFields(level string, message string, fields map[string]interface{}) {
if t.logger != nil {
t.logger.LogWithFields(level, message, fields)
}
}
// FinishTask finalizes the task and closes the logger
func (t *BaseTask) FinishTask(success bool, errorMsg string) error {
if t.logger != nil {
if success {
t.logger.LogStatus("completed", "Task completed successfully")
t.logger.Info("Task %s finished successfully", t.taskID)
} else {
t.logger.LogStatus("failed", fmt.Sprintf("Task failed: %s", errorMsg))
t.logger.Error("Task %s failed: %s", t.taskID, errorMsg)
}
// Close logger
if err := t.logger.Close(); err != nil {
glog.Errorf("Failed to close task logger: %v", err)
}
}
return nil
}
// ExecuteTask is a wrapper that handles common task execution logic with logging
func (t *BaseTask) ExecuteTask(ctx context.Context, params types.TaskParams, executor func(context.Context, types.TaskParams) error) error {
// Initialize logger if not already done
if t.logger == nil {
// Generate a temporary task ID if none provided
if t.taskID == "" {
t.taskID = fmt.Sprintf("task_%d", time.Now().UnixNano())
}
workerID := "unknown"
if err := t.InitializeLogger(t.taskID, workerID, params); err != nil {
glog.Warningf("Failed to initialize task logger: %v", err)
}
}
t.SetStartTime(time.Now())
t.SetProgress(0)
if t.logger != nil {
t.logger.LogWithFields("INFO", "Task execution started", map[string]interface{}{
"volume_id": params.VolumeID,
"server": params.Server,
"collection": params.Collection,
})
}
// Create a context that can be cancelled
ctx, cancel := context.WithCancel(ctx)
defer cancel()
@@ -114,21 +292,29 @@ func (t *BaseTask) ExecuteTask(ctx context.Context, params types.TaskParams, exe
// Check cancellation every second
}
}
t.LogWarning("Task cancellation detected, cancelling context")
cancel()
}()
// Execute the actual task
t.LogInfo("Starting task executor")
err := executor(ctx, params)
if err != nil {
t.LogError("Task executor failed: %v", err)
t.FinishTask(false, err.Error())
return err
}
if t.IsCancelled() {
t.LogWarning("Task was cancelled during execution")
t.FinishTask(false, "cancelled")
return context.Canceled
}
t.SetProgress(100)
t.LogInfo("Task executor completed successfully")
t.FinishTask(true, "")
return nil
}