adding snapshot function

2025-12-14 23:17:26 +07:00
parent 461edbc970
commit ed96137bad
8 changed files with 1075 additions and 20 deletions
--- a/docs/BACKGROUND_JOBS.md
+++ b/docs/BACKGROUND_JOBS.md
@@ -0,0 +1,125 @@
+# Background Job System
+
+The atlasOS API includes a background job system that automatically executes snapshot policies and manages long-running operations.
+
+## Architecture
+
+### Components
+
+1. **Job Manager** (`internal/job/manager.go`)
+   - Tracks job lifecycle (pending, running, completed, failed, cancelled)
+   - Stores job metadata and progress
+   - Thread-safe job operations
+
+2. **Snapshot Scheduler** (`internal/snapshot/scheduler.go`)
+   - Automatically creates snapshots based on policies
+   - Prunes old snapshots based on retention rules
+   - Runs every 15 minutes by default
+
+3. **Integration**
+   - Scheduler starts automatically when API server starts
+   - Gracefully stops on server shutdown
+   - Jobs are accessible via API endpoints
+
+## How It Works
+
+### Snapshot Creation
+
+The scheduler checks all enabled snapshot policies every 15 minutes and:
+
+1. **Frequent snapshots**: Creates every 15 minutes if `frequent > 0`
+2. **Hourly snapshots**: Creates every hour if `hourly > 0`
+3. **Daily snapshots**: Creates daily if `daily > 0`
+4. **Weekly snapshots**: Creates weekly if `weekly > 0`
+5. **Monthly snapshots**: Creates monthly if `monthly > 0`
+6. **Yearly snapshots**: Creates yearly if `yearly > 0`
+
+Snapshot names follow the pattern: `{type}-{timestamp}` (e.g., `hourly-20241214-143000`)
+
+### Snapshot Pruning
+
+When `autoprune` is enabled, the scheduler:
+
+1. Groups snapshots by type (frequent, hourly, daily, etc.)
+2. Sorts by creation time (newest first)
+3. Keeps only the number specified in the policy
+4. Deletes older snapshots that exceed the retention count
+
+### Job Tracking
+
+Every snapshot operation creates a job that tracks:
+- Status (pending → running → completed/failed)
+- Progress (0-100%)
+- Error messages (if failed)
+- Timestamps (created, started, completed)
+
+## API Endpoints
+
+### List Jobs
+```bash
+GET /api/v1/jobs
+GET /api/v1/jobs?status=running
+```
+
+### Get Job
+```bash
+GET /api/v1/jobs/{id}
+```
+
+### Cancel Job
+```bash
+POST /api/v1/jobs/{id}/cancel
+```
+
+## Configuration
+
+The scheduler interval is hardcoded to 15 minutes. To change it, modify:
+
+```go
+// In internal/httpapp/app.go
+scheduler.Start(15 * time.Minute)  // Change interval here
+```
+
+## Example Workflow
+
+1. **Create a snapshot policy:**
+```bash
+curl -X POST http://localhost:8080/api/v1/snapshot-policies \
+  -H "Content-Type: application/json" \
+  -d '{
+    "dataset": "pool/dataset",
+    "hourly": 24,
+    "daily": 7,
+    "autosnap": true,
+    "autoprune": true
+  }'
+```
+
+2. **Scheduler automatically:**
+   - Creates hourly snapshots (keeps 24)
+   - Creates daily snapshots (keeps 7)
+   - Prunes old snapshots beyond retention
+
+3. **Monitor jobs:**
+```bash
+curl http://localhost:8080/api/v1/jobs
+```
+
+## Job Statuses
+
+- `pending`: Job created but not started
+- `running`: Job is currently executing
+- `completed`: Job finished successfully
+- `failed`: Job encountered an error
+- `cancelled`: Job was cancelled by user
+
+## Notes
+
+- Jobs are stored in-memory (will be lost on restart)
+- Scheduler runs in a background goroutine
+- Snapshot operations are synchronous (blocking)
+- For production, consider:
+  - Database persistence for jobs
+  - Async job execution with worker pool
+  - Job history retention policies
+  - Metrics/alerting for failed jobs
--- a/internal/httpapp/api_handlers.go
+++ b/internal/httpapp/api_handlers.go
@@ -2,6 +2,7 @@ package httpapp

 import (
 	"encoding/json"
+	"fmt"
 	"log"
 	"net/http"

@@ -287,47 +288,167 @@ func (a *App) handleDeleteZVOL(w http.ResponseWriter, r *http.Request) {

 // Snapshot Handlers
 func (a *App) handleListSnapshots(w http.ResponseWriter, r *http.Request) {
-	snapshots := []models.Snapshot{} // Stub
+	dataset := r.URL.Query().Get("dataset")
+	snapshots, err := a.zfs.ListSnapshots(dataset)
+	if err != nil {
+		log.Printf("list snapshots error: %v", err)
+		writeJSON(w, http.StatusInternalServerError, map[string]string{"error": err.Error()})
+		return
+	}
 	writeJSON(w, http.StatusOK, snapshots)
 }

 func (a *App) handleCreateSnapshot(w http.ResponseWriter, r *http.Request) {
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented"})
+	var req struct {
+		Dataset   string `json:"dataset"`
+		Name      string `json:"name"`
+		Recursive bool   `json:"recursive,omitempty"`
+	}
+
+	if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "invalid request body"})
+		return
+	}
+
+	if req.Dataset == "" || req.Name == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "dataset and name are required"})
+		return
+	}
+
+	if err := a.zfs.CreateSnapshot(req.Dataset, req.Name, req.Recursive); err != nil {
+		log.Printf("create snapshot error: %v", err)
+		writeJSON(w, http.StatusInternalServerError, map[string]string{"error": err.Error()})
+		return
+	}
+
+	snapshotName := fmt.Sprintf("%s@%s", req.Dataset, req.Name)
+	snap, err := a.zfs.GetSnapshot(snapshotName)
+	if err != nil {
+		writeJSON(w, http.StatusCreated, map[string]string{"message": "snapshot created", "name": snapshotName})
+		return
+	}
+
+	writeJSON(w, http.StatusCreated, snap)
 }

 func (a *App) handleGetSnapshot(w http.ResponseWriter, r *http.Request) {
 	name := pathParam(r, "/api/v1/snapshots/")
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented", "name": name})
+	if name == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "snapshot name required"})
+		return
+	}
+
+	snap, err := a.zfs.GetSnapshot(name)
+	if err != nil {
+		writeJSON(w, http.StatusNotFound, map[string]string{"error": err.Error()})
+		return
+	}
+
+	writeJSON(w, http.StatusOK, snap)
 }

 func (a *App) handleDeleteSnapshot(w http.ResponseWriter, r *http.Request) {
 	name := pathParam(r, "/api/v1/snapshots/")
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented", "name": name})
+	if name == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "snapshot name required"})
+		return
+	}
+
+	recursive := r.URL.Query().Get("recursive") == "true"
+
+	if err := a.zfs.DestroySnapshot(name, recursive); err != nil {
+		log.Printf("destroy snapshot error: %v", err)
+		writeJSON(w, http.StatusInternalServerError, map[string]string{"error": err.Error()})
+		return
+	}
+
+	writeJSON(w, http.StatusOK, map[string]string{"message": "snapshot destroyed", "name": name})
 }

 // Snapshot Policy Handlers
 func (a *App) handleListSnapshotPolicies(w http.ResponseWriter, r *http.Request) {
-	policies := []models.SnapshotPolicy{} // Stub
+	dataset := r.URL.Query().Get("dataset")
+	var policies []models.SnapshotPolicy
+	if dataset != "" {
+		policies = a.snapshotPolicy.ListForDataset(dataset)
+	} else {
+		policies = a.snapshotPolicy.List()
+	}
 	writeJSON(w, http.StatusOK, policies)
 }

 func (a *App) handleCreateSnapshotPolicy(w http.ResponseWriter, r *http.Request) {
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented"})
+	var policy models.SnapshotPolicy
+
+	if err := json.NewDecoder(r.Body).Decode(&policy); err != nil {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "invalid request body"})
+		return
+	}
+
+	if policy.Dataset == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "dataset is required"})
+		return
+	}
+
+	a.snapshotPolicy.Set(&policy)
+	writeJSON(w, http.StatusCreated, policy)
 }

 func (a *App) handleGetSnapshotPolicy(w http.ResponseWriter, r *http.Request) {
 	dataset := pathParam(r, "/api/v1/snapshot-policies/")
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented", "dataset": dataset})
+	if dataset == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "dataset name required"})
+		return
+	}
+
+	policy, err := a.snapshotPolicy.Get(dataset)
+	if err != nil {
+		writeJSON(w, http.StatusInternalServerError, map[string]string{"error": err.Error()})
+		return
+	}
+
+	if policy == nil {
+		writeJSON(w, http.StatusNotFound, map[string]string{"error": "policy not found"})
+		return
+	}
+
+	writeJSON(w, http.StatusOK, policy)
 }

 func (a *App) handleUpdateSnapshotPolicy(w http.ResponseWriter, r *http.Request) {
 	dataset := pathParam(r, "/api/v1/snapshot-policies/")
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented", "dataset": dataset})
+	if dataset == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "dataset name required"})
+		return
+	}
+
+	var policy models.SnapshotPolicy
+	if err := json.NewDecoder(r.Body).Decode(&policy); err != nil {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "invalid request body"})
+		return
+	}
+
+	// Ensure dataset matches URL parameter
+	policy.Dataset = dataset
+
+	a.snapshotPolicy.Set(&policy)
+	writeJSON(w, http.StatusOK, policy)
 }

 func (a *App) handleDeleteSnapshotPolicy(w http.ResponseWriter, r *http.Request) {
 	dataset := pathParam(r, "/api/v1/snapshot-policies/")
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented", "dataset": dataset})
+	if dataset == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "dataset name required"})
+		return
+	}
+
+	if err := a.snapshotPolicy.Delete(dataset); err != nil {
+		log.Printf("delete snapshot policy error: %v", err)
+		writeJSON(w, http.StatusInternalServerError, map[string]string{"error": err.Error()})
+		return
+	}
+
+	writeJSON(w, http.StatusOK, map[string]string{"message": "policy deleted", "dataset": dataset})
 }

 // SMB Share Handlers
@@ -417,18 +538,41 @@ func (a *App) handleRemoveLUN(w http.ResponseWriter, r *http.Request) {

 // Job Handlers
 func (a *App) handleListJobs(w http.ResponseWriter, r *http.Request) {
-	jobs := []models.Job{} // Stub
+	status := models.JobStatus(r.URL.Query().Get("status"))
+	jobs := a.jobManager.List(status)
 	writeJSON(w, http.StatusOK, jobs)
 }

 func (a *App) handleGetJob(w http.ResponseWriter, r *http.Request) {
 	id := pathParam(r, "/api/v1/jobs/")
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented", "id": id})
+	if id == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "job id required"})
+		return
+	}
+
+	job, err := a.jobManager.Get(id)
+	if err != nil {
+		writeJSON(w, http.StatusNotFound, map[string]string{"error": err.Error()})
+		return
+	}
+
+	writeJSON(w, http.StatusOK, job)
 }

 func (a *App) handleCancelJob(w http.ResponseWriter, r *http.Request) {
 	id := pathParam(r, "/api/v1/jobs/")
-	writeJSON(w, http.StatusNotImplemented, map[string]string{"error": "not implemented", "id": id})
+	if id == "" {
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": "job id required"})
+		return
+	}
+
+	if err := a.jobManager.Cancel(id); err != nil {
+		log.Printf("cancel job error: %v", err)
+		writeJSON(w, http.StatusBadRequest, map[string]string{"error": err.Error()})
+		return
+	}
+
+	writeJSON(w, http.StatusOK, map[string]string{"message": "job cancelled", "id": id})
 }

 // Auth Handlers (stubs)
--- a/internal/httpapp/app.go
+++ b/internal/httpapp/app.go
@@ -7,6 +7,8 @@ import (
 	"path/filepath"
 	"time"

+	"gitea.avt.data-center.id/othman.suseno/atlas/internal/job"
+	"gitea.avt.data-center.id/othman.suseno/atlas/internal/snapshot"
 	"gitea.avt.data-center.id/othman.suseno/atlas/internal/zfs"
 )

@@ -21,6 +23,9 @@ type App struct {
 	tmpl           *template.Template
 	mux            *http.ServeMux
 	zfs            *zfs.Service
+	snapshotPolicy *snapshot.PolicyStore
+	jobManager     *job.Manager
+	scheduler      *snapshot.Scheduler
 }

 func New(cfg Config) (*App, error) {
@@ -36,13 +41,24 @@ func New(cfg Config) (*App, error) {
 		return nil, err
 	}

+	zfsService := zfs.New()
+	policyStore := snapshot.NewPolicyStore()
+	jobMgr := job.NewManager()
+	scheduler := snapshot.NewScheduler(policyStore, zfsService, jobMgr)
+
 	a := &App{
 		cfg:            cfg,
 		tmpl:           tmpl,
 		mux:            http.NewServeMux(),
-		zfs:  zfs.New(),
+		zfs:            zfsService,
+		snapshotPolicy: policyStore,
+		jobManager:     jobMgr,
+		scheduler:      scheduler,
 	}

+	// Start snapshot scheduler (runs every 15 minutes)
+	scheduler.Start(15 * time.Minute)
+
 	a.routes()
 	return a, nil
 }
@@ -52,6 +68,13 @@ func (a *App) Router() http.Handler {
 	return requestID(logging(a.mux))
 }

+// StopScheduler stops the snapshot scheduler (for graceful shutdown)
+func (a *App) StopScheduler() {
+	if a.scheduler != nil {
+		a.scheduler.Stop()
+	}
+}
+
 // routes() is now in routes.go

 func parseTemplates(dir string) (*template.Template, error) {
--- a/internal/job/manager.go
+++ b/internal/job/manager.go
@@ -0,0 +1,163 @@
+package job
+
+import (
+	"fmt"
+	"sync"
+	"time"
+
+	"gitea.avt.data-center.id/othman.suseno/atlas/internal/models"
+)
+
+// Manager handles job lifecycle and tracking
+type Manager struct {
+	mu     sync.RWMutex
+	jobs   map[string]*models.Job
+	nextID int64
+}
+
+// NewManager creates a new job manager
+func NewManager() *Manager {
+	return &Manager{
+		jobs:   make(map[string]*models.Job),
+		nextID: 1,
+	}
+}
+
+// Create creates a new job
+func (m *Manager) Create(jobType string, metadata map[string]interface{}) *models.Job {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+
+	id := fmt.Sprintf("job-%d", m.nextID)
+	m.nextID++
+
+	job := &models.Job{
+		ID:        id,
+		Type:      jobType,
+		Status:    models.JobStatusPending,
+		Progress:  0,
+		Message:   "Job created",
+		Metadata:  metadata,
+		CreatedAt: time.Now(),
+	}
+
+	m.jobs[id] = job
+	return job
+}
+
+// Get returns a job by ID
+func (m *Manager) Get(id string) (*models.Job, error) {
+	m.mu.RLock()
+	defer m.mu.RUnlock()
+
+	job, exists := m.jobs[id]
+	if !exists {
+		return nil, fmt.Errorf("job %s not found", id)
+	}
+	return job, nil
+}
+
+// List returns all jobs, optionally filtered by status
+func (m *Manager) List(status models.JobStatus) []models.Job {
+	m.mu.RLock()
+	defer m.mu.RUnlock()
+
+	var jobs []models.Job
+	for _, job := range m.jobs {
+		if status == "" || job.Status == status {
+			jobs = append(jobs, *job)
+		}
+	}
+	return jobs
+}
+
+// UpdateStatus updates job status
+func (m *Manager) UpdateStatus(id string, status models.JobStatus, message string) error {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+
+	job, exists := m.jobs[id]
+	if !exists {
+		return fmt.Errorf("job %s not found", id)
+	}
+
+	job.Status = status
+	job.Message = message
+
+	now := time.Now()
+	switch status {
+	case models.JobStatusRunning:
+		if job.StartedAt == nil {
+			job.StartedAt = &now
+		}
+	case models.JobStatusCompleted, models.JobStatusFailed, models.JobStatusCancelled:
+		job.CompletedAt = &now
+	}
+
+	return nil
+}
+
+// UpdateProgress updates job progress (0-100)
+func (m *Manager) UpdateProgress(id string, progress int, message string) error {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+
+	job, exists := m.jobs[id]
+	if !exists {
+		return fmt.Errorf("job %s not found", id)
+	}
+
+	if progress < 0 {
+		progress = 0
+	}
+	if progress > 100 {
+		progress = 100
+	}
+
+	job.Progress = progress
+	if message != "" {
+		job.Message = message
+	}
+
+	return nil
+}
+
+// SetError sets job error and marks as failed
+func (m *Manager) SetError(id string, err error) error {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+
+	job, exists := m.jobs[id]
+	if !exists {
+		return fmt.Errorf("job %s not found", id)
+	}
+
+	job.Status = models.JobStatusFailed
+	job.Error = err.Error()
+	now := time.Now()
+	job.CompletedAt = &now
+
+	return nil
+}
+
+// Cancel cancels a job
+func (m *Manager) Cancel(id string) error {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+
+	job, exists := m.jobs[id]
+	if !exists {
+		return fmt.Errorf("job %s not found", id)
+	}
+
+	if job.Status == models.JobStatusCompleted || job.Status == models.JobStatusFailed {
+		return fmt.Errorf("cannot cancel job in status %s", job.Status)
+	}
+
+	job.Status = models.JobStatusCancelled
+	job.Message = "Job cancelled by user"
+	now := time.Now()
+	job.CompletedAt = &now
+
+	return nil
+}
--- a/internal/snapshot/policy.go
+++ b/internal/snapshot/policy.go
@@ -0,0 +1,76 @@
+package snapshot
+
+import (
+	"sync"
+
+	"gitea.avt.data-center.id/othman.suseno/atlas/internal/models"
+)
+
+// PolicyStore manages snapshot policies
+type PolicyStore struct {
+	mu       sync.RWMutex
+	policies map[string]*models.SnapshotPolicy
+}
+
+// NewPolicyStore creates a new policy store
+func NewPolicyStore() *PolicyStore {
+	return &PolicyStore{
+		policies: make(map[string]*models.SnapshotPolicy),
+	}
+}
+
+// List returns all snapshot policies
+func (s *PolicyStore) List() []models.SnapshotPolicy {
+	s.mu.RLock()
+	defer s.mu.RUnlock()
+
+	policies := make([]models.SnapshotPolicy, 0, len(s.policies))
+	for _, p := range s.policies {
+		policies = append(policies, *p)
+	}
+	return policies
+}
+
+// Get returns a policy for a dataset
+func (s *PolicyStore) Get(dataset string) (*models.SnapshotPolicy, error) {
+	s.mu.RLock()
+	defer s.mu.RUnlock()
+
+	policy, exists := s.policies[dataset]
+	if !exists {
+		return nil, nil // Return nil if not found (not an error)
+	}
+	return policy, nil
+}
+
+// Set creates or updates a policy
+func (s *PolicyStore) Set(policy *models.SnapshotPolicy) {
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	s.policies[policy.Dataset] = policy
+}
+
+// Delete removes a policy
+func (s *PolicyStore) Delete(dataset string) error {
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	delete(s.policies, dataset)
+	return nil
+}
+
+// ListForDataset returns all policies (for future filtering by dataset prefix)
+func (s *PolicyStore) ListForDataset(datasetPrefix string) []models.SnapshotPolicy {
+	s.mu.RLock()
+	defer s.mu.RUnlock()
+
+	var policies []models.SnapshotPolicy
+	for _, p := range s.policies {
+		if datasetPrefix == "" || p.Dataset == datasetPrefix ||
+			(len(p.Dataset) > len(datasetPrefix) && p.Dataset[:len(datasetPrefix)] == datasetPrefix) {
+			policies = append(policies, *p)
+		}
+	}
+	return policies
+}
--- a/internal/snapshot/scheduler.go
+++ b/internal/snapshot/scheduler.go
@@ -0,0 +1,261 @@
+package snapshot
+
+import (
+	"fmt"
+	"log"
+	"sort"
+	"strings"
+	"time"
+
+	"gitea.avt.data-center.id/othman.suseno/atlas/internal/job"
+	"gitea.avt.data-center.id/othman.suseno/atlas/internal/models"
+	"gitea.avt.data-center.id/othman.suseno/atlas/internal/zfs"
+)
+
+// Scheduler manages automatic snapshot creation and pruning
+type Scheduler struct {
+	policyStore *PolicyStore
+	zfsService  *zfs.Service
+	jobManager  *job.Manager
+	stopChan    chan struct{}
+	ticker      *time.Ticker
+}
+
+// NewScheduler creates a new snapshot scheduler
+func NewScheduler(policyStore *PolicyStore, zfsService *zfs.Service, jobManager *job.Manager) *Scheduler {
+	return &Scheduler{
+		policyStore: policyStore,
+		zfsService:  zfsService,
+		jobManager:  jobManager,
+		stopChan:    make(chan struct{}),
+	}
+}
+
+// Start starts the scheduler with the given interval
+func (s *Scheduler) Start(interval time.Duration) {
+	s.ticker = time.NewTicker(interval)
+	log.Printf("[scheduler] started with interval %v", interval)
+
+	go s.run()
+}
+
+// Stop stops the scheduler
+func (s *Scheduler) Stop() {
+	if s.ticker != nil {
+		s.ticker.Stop()
+	}
+	close(s.stopChan)
+	log.Printf("[scheduler] stopped")
+}
+
+// run executes the scheduler loop
+func (s *Scheduler) run() {
+	// Run immediately on start
+	s.execute()
+
+	for {
+		select {
+		case <-s.ticker.C:
+			s.execute()
+		case <-s.stopChan:
+			return
+		}
+	}
+}
+
+// execute checks policies and creates/prunes snapshots
+func (s *Scheduler) execute() {
+	policies := s.policyStore.List()
+	log.Printf("[scheduler] checking %d snapshot policies", len(policies))
+
+	for _, policy := range policies {
+		if !policy.Autosnap {
+			continue
+		}
+
+		// Check if we need to create a snapshot based on schedule
+		s.checkAndCreateSnapshot(policy)
+
+		// Prune old snapshots if enabled
+		if policy.Autoprune {
+			s.pruneSnapshots(policy)
+		}
+	}
+}
+
+// checkAndCreateSnapshot checks if a snapshot should be created
+func (s *Scheduler) checkAndCreateSnapshot(policy models.SnapshotPolicy) {
+	now := time.Now()
+	snapshots, err := s.zfsService.ListSnapshots(policy.Dataset)
+	if err != nil {
+		log.Printf("[scheduler] error listing snapshots for %s: %v", policy.Dataset, err)
+		return
+	}
+
+	// Check if we need frequent snapshots (every 15 minutes)
+	if policy.Frequent > 0 {
+		if s.shouldCreateSnapshot(snapshots, 15*time.Minute, policy.Frequent) {
+			s.createSnapshot(policy.Dataset, "frequent", now)
+		}
+	}
+
+	// Check if we need hourly snapshots
+	if policy.Hourly > 0 {
+		if s.shouldCreateSnapshot(snapshots, time.Hour, policy.Hourly) {
+			s.createSnapshot(policy.Dataset, "hourly", now)
+		}
+	}
+
+	// Check if we need daily snapshots
+	if policy.Daily > 0 {
+		if s.shouldCreateSnapshot(snapshots, 24*time.Hour, policy.Daily) {
+			s.createSnapshot(policy.Dataset, "daily", now)
+		}
+	}
+
+	// Check if we need weekly snapshots
+	if policy.Weekly > 0 {
+		if s.shouldCreateSnapshot(snapshots, 7*24*time.Hour, policy.Weekly) {
+			s.createSnapshot(policy.Dataset, "weekly", now)
+		}
+	}
+
+	// Check if we need monthly snapshots
+	if policy.Monthly > 0 {
+		if s.shouldCreateSnapshot(snapshots, 30*24*time.Hour, policy.Monthly) {
+			s.createSnapshot(policy.Dataset, "monthly", now)
+		}
+	}
+
+	// Check if we need yearly snapshots
+	if policy.Yearly > 0 {
+		if s.shouldCreateSnapshot(snapshots, 365*24*time.Hour, policy.Yearly) {
+			s.createSnapshot(policy.Dataset, "yearly", now)
+		}
+	}
+}
+
+// shouldCreateSnapshot checks if a new snapshot should be created
+func (s *Scheduler) shouldCreateSnapshot(snapshots []models.Snapshot, interval time.Duration, keepCount int) bool {
+	now := time.Now()
+	cutoff := now.Add(-interval)
+
+	// Count snapshots in the interval
+	count := 0
+	for _, snap := range snapshots {
+		if snap.CreatedAt.After(cutoff) {
+			count++
+		}
+	}
+
+	// If we have fewer than keepCount, we should create one
+	return count < keepCount
+}
+
+// createSnapshot creates a snapshot with a timestamped name
+func (s *Scheduler) createSnapshot(dataset, prefix string, t time.Time) {
+	timestamp := t.Format("20060102-150405")
+	name := fmt.Sprintf("%s-%s", prefix, timestamp)
+
+	job := s.jobManager.Create("snapshot_create", map[string]interface{}{
+		"dataset": dataset,
+		"name":    name,
+		"type":    prefix,
+	})
+
+	s.jobManager.UpdateStatus(job.ID, models.JobStatusRunning, fmt.Sprintf("Creating snapshot %s@%s", dataset, name))
+
+	if err := s.zfsService.CreateSnapshot(dataset, name, false); err != nil {
+		log.Printf("[scheduler] error creating snapshot %s@%s: %v", dataset, name, err)
+		s.jobManager.SetError(job.ID, err)
+		return
+	}
+
+	s.jobManager.UpdateProgress(job.ID, 100, "Snapshot created successfully")
+	s.jobManager.UpdateStatus(job.ID, models.JobStatusCompleted, "Snapshot created")
+	log.Printf("[scheduler] created snapshot %s@%s", dataset, name)
+}
+
+// pruneSnapshots removes old snapshots based on retention policy
+func (s *Scheduler) pruneSnapshots(policy models.SnapshotPolicy) {
+	snapshots, err := s.zfsService.ListSnapshots(policy.Dataset)
+	if err != nil {
+		log.Printf("[scheduler] error listing snapshots for pruning %s: %v", policy.Dataset, err)
+		return
+	}
+
+	now := time.Now()
+	pruned := 0
+
+	// Group snapshots by type
+	frequent := []models.Snapshot{}
+	hourly := []models.Snapshot{}
+	daily := []models.Snapshot{}
+	weekly := []models.Snapshot{}
+	monthly := []models.Snapshot{}
+	yearly := []models.Snapshot{}
+
+	for _, snap := range snapshots {
+		// Parse snapshot name to determine type
+		parts := strings.Split(snap.Name, "@")
+		if len(parts) != 2 {
+			continue
+		}
+		snapName := parts[1]
+
+		if strings.HasPrefix(snapName, "frequent-") {
+			frequent = append(frequent, snap)
+		} else if strings.HasPrefix(snapName, "hourly-") {
+			hourly = append(hourly, snap)
+		} else if strings.HasPrefix(snapName, "daily-") {
+			daily = append(daily, snap)
+		} else if strings.HasPrefix(snapName, "weekly-") {
+			weekly = append(weekly, snap)
+		} else if strings.HasPrefix(snapName, "monthly-") {
+			monthly = append(monthly, snap)
+		} else if strings.HasPrefix(snapName, "yearly-") {
+			yearly = append(yearly, snap)
+		}
+	}
+
+	// Prune each type
+	pruned += s.pruneByType(frequent, policy.Frequent, 15*time.Minute, now, policy.Dataset)
+	pruned += s.pruneByType(hourly, policy.Hourly, time.Hour, now, policy.Dataset)
+	pruned += s.pruneByType(daily, policy.Daily, 24*time.Hour, now, policy.Dataset)
+	pruned += s.pruneByType(weekly, policy.Weekly, 7*24*time.Hour, now, policy.Dataset)
+	pruned += s.pruneByType(monthly, policy.Monthly, 30*24*time.Hour, now, policy.Dataset)
+	pruned += s.pruneByType(yearly, policy.Yearly, 365*24*time.Hour, now, policy.Dataset)
+
+	if pruned > 0 {
+		log.Printf("[scheduler] pruned %d snapshots for %s", pruned, policy.Dataset)
+	}
+}
+
+// pruneByType prunes snapshots of a specific type
+func (s *Scheduler) pruneByType(snapshots []models.Snapshot, keepCount int, interval time.Duration, now time.Time, dataset string) int {
+	if keepCount == 0 || len(snapshots) <= keepCount {
+		return 0
+	}
+
+	// Sort by creation time (newest first)
+	sort.Slice(snapshots, func(i, j int) bool {
+		return snapshots[i].CreatedAt.After(snapshots[j].CreatedAt)
+	})
+
+	// Keep the newest keepCount snapshots, delete the rest
+	toDelete := snapshots[keepCount:]
+
+	pruned := 0
+	for _, snap := range toDelete {
+		// Only delete if it's older than the interval
+		if now.Sub(snap.CreatedAt) > interval {
+			if err := s.zfsService.DestroySnapshot(snap.Name, false); err != nil {
+				log.Printf("[scheduler] error pruning snapshot %s: %v", snap.Name, err)
+				continue
+			}
+			pruned++
+		}
+	}
+
+	return pruned
+}
--- a/internal/zfs/service.go
+++ b/internal/zfs/service.go
@@ -398,3 +398,107 @@ func parseSize(s string) (uint64, error) {

 	return val * multiplier, nil
 }
+
+// ListSnapshots returns all snapshots for a dataset (or all if dataset is empty)
+func (s *Service) ListSnapshots(dataset string) ([]models.Snapshot, error) {
+	args := []string{"list", "-H", "-o", "name,used,creation", "-t", "snapshot", "-s", "creation"}
+	if dataset != "" {
+		args = append(args, "-r", dataset)
+	} else {
+		args = append(args, "-r")
+	}
+
+	output, err := s.execCommand(s.zfsPath, args...)
+	if err != nil {
+		return nil, err
+	}
+
+	var snapshots []models.Snapshot
+	lines := strings.Split(output, "\n")
+	for _, line := range lines {
+		if line == "" {
+			continue
+		}
+
+		fields := strings.Fields(line)
+		if len(fields) < 3 {
+			continue
+		}
+
+		fullName := fields[0]
+		// Snapshot name format: dataset@snapshot
+		parts := strings.Split(fullName, "@")
+		if len(parts) != 2 {
+			continue
+		}
+
+		datasetName := parts[0]
+
+		snapshot := models.Snapshot{
+			Name:    fullName,
+			Dataset: datasetName,
+		}
+
+		// Parse size
+		if used, err := parseSize(fields[1]); err == nil {
+			snapshot.Size = used
+		}
+
+		// Parse creation time
+		// ZFS creation format: "Mon Jan 2 15:04:05 2006"
+		createdStr := strings.Join(fields[2:], " ")
+		if t, err := time.Parse("Mon Jan 2 15:04:05 2006", createdStr); err == nil {
+			snapshot.CreatedAt = t
+		} else {
+			// Try RFC3339 format if available
+			if t, err := time.Parse(time.RFC3339, createdStr); err == nil {
+				snapshot.CreatedAt = t
+			}
+		}
+
+		snapshots = append(snapshots, snapshot)
+	}
+
+	return snapshots, nil
+}
+
+// CreateSnapshot creates a new snapshot
+func (s *Service) CreateSnapshot(dataset, name string, recursive bool) error {
+	args := []string{"snapshot"}
+	if recursive {
+		args = append(args, "-r")
+	}
+
+	snapshotName := fmt.Sprintf("%s@%s", dataset, name)
+	args = append(args, snapshotName)
+
+	_, err := s.execCommand(s.zfsPath, args...)
+	return err
+}
+
+// DestroySnapshot destroys a snapshot
+func (s *Service) DestroySnapshot(name string, recursive bool) error {
+	args := []string{"destroy"}
+	if recursive {
+		args = append(args, "-r")
+	}
+	args = append(args, name)
+	_, err := s.execCommand(s.zfsPath, args...)
+	return err
+}
+
+// GetSnapshot returns snapshot details
+func (s *Service) GetSnapshot(name string) (*models.Snapshot, error) {
+	snapshots, err := s.ListSnapshots("")
+	if err != nil {
+		return nil, err
+	}
+
+	for _, snap := range snapshots {
+		if snap.Name == name {
+			return &snap, nil
+		}
+	}
+
+	return nil, fmt.Errorf("snapshot %s not found", name)
+}
--- a/test_api.sh
+++ b/test_api.sh
@@ -0,0 +1,159 @@
+#!/bin/bash
+# Atlas API Test Script
+# This script tests the API endpoints
+
+BASE_URL="${ATLAS_URL:-http://localhost:8080}"
+API_URL="${BASE_URL}/api/v1"
+
+echo "=========================================="
+echo "Atlas API Test Suite"
+echo "=========================================="
+echo "Base URL: $BASE_URL"
+echo "API URL: $API_URL"
+echo ""
+
+# Colors for output
+GREEN='\033[0;32m'
+RED='\033[0;31m'
+YELLOW='\033[1;33m'
+NC='\033[0m' # No Color
+
+# Test helper function
+test_endpoint() {
+    local method=$1
+    local endpoint=$2
+    local data=$3
+    local description=$4
+    
+    echo -e "${YELLOW}Testing: $description${NC}"
+    echo "  $method $endpoint"
+    
+    if [ -n "$data" ]; then
+        response=$(curl -s -w "\n%{http_code}" -X "$method" \
+            -H "Content-Type: application/json" \
+            -d "$data" \
+            "$endpoint")
+    else
+        response=$(curl -s -w "\n%{http_code}" -X "$method" "$endpoint")
+    fi
+    
+    http_code=$(echo "$response" | tail -n1)
+    body=$(echo "$response" | sed '$d')
+    
+    if [ "$http_code" -ge 200 ] && [ "$http_code" -lt 300 ]; then
+        echo -e "  ${GREEN}✓ Success (HTTP $http_code)${NC}"
+        echo "$body" | jq '.' 2>/dev/null || echo "$body"
+    elif [ "$http_code" -ge 400 ] && [ "$http_code" -lt 500 ]; then
+        echo -e "  ${YELLOW}⚠ Client Error (HTTP $http_code)${NC}"
+        echo "$body" | jq '.' 2>/dev/null || echo "$body"
+    else
+        echo -e "  ${RED}✗ Error (HTTP $http_code)${NC}"
+        echo "$body"
+    fi
+    echo ""
+}
+
+# Check if server is running
+echo "Checking if server is running..."
+if ! curl -s "$BASE_URL/healthz" > /dev/null; then
+    echo -e "${RED}Error: Server is not running at $BASE_URL${NC}"
+    echo "Start the server with: go run ./cmd/atlas-api/main.go"
+    exit 1
+fi
+echo -e "${GREEN}Server is running!${NC}"
+echo ""
+
+# Health & Metrics
+echo "=========================================="
+echo "1. Health & Metrics"
+echo "=========================================="
+test_endpoint "GET" "$BASE_URL/healthz" "" "Health check"
+test_endpoint "GET" "$BASE_URL/metrics" "" "Prometheus metrics"
+echo ""
+
+# Disk Discovery
+echo "=========================================="
+echo "2. Disk Discovery"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/disks" "" "List available disks"
+echo ""
+
+# ZFS Pools
+echo "=========================================="
+echo "3. ZFS Pool Management"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/pools" "" "List ZFS pools"
+test_endpoint "GET" "$API_URL/pools/testpool" "" "Get pool details (if exists)"
+echo ""
+
+# Datasets
+echo "=========================================="
+echo "4. Dataset Management"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/datasets" "" "List all datasets"
+test_endpoint "GET" "$API_URL/datasets?pool=testpool" "" "List datasets in pool"
+echo ""
+
+# ZVOLs
+echo "=========================================="
+echo "5. ZVOL Management"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/zvols" "" "List all ZVOLs"
+test_endpoint "GET" "$API_URL/zvols?pool=testpool" "" "List ZVOLs in pool"
+echo ""
+
+# Snapshots
+echo "=========================================="
+echo "6. Snapshot Management"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/snapshots" "" "List all snapshots"
+test_endpoint "GET" "$API_URL/snapshots?dataset=testpool/test" "" "List snapshots for dataset"
+echo ""
+
+# Snapshot Policies
+echo "=========================================="
+echo "7. Snapshot Policies"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/snapshot-policies" "" "List snapshot policies"
+
+# Create a test policy
+test_endpoint "POST" "$API_URL/snapshot-policies" \
+    '{"dataset":"testpool/test","frequent":4,"hourly":24,"daily":7,"weekly":4,"monthly":12,"yearly":2,"autosnap":true,"autoprune":true}' \
+    "Create snapshot policy"
+
+test_endpoint "GET" "$API_URL/snapshot-policies/testpool/test" "" "Get snapshot policy"
+echo ""
+
+# Storage Services
+echo "=========================================="
+echo "8. Storage Services"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/shares/smb" "" "List SMB shares"
+test_endpoint "GET" "$API_URL/exports/nfs" "" "List NFS exports"
+test_endpoint "GET" "$API_URL/iscsi/targets" "" "List iSCSI targets"
+echo ""
+
+# Jobs
+echo "=========================================="
+echo "9. Job Management"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/jobs" "" "List jobs"
+echo ""
+
+# Users & Auth
+echo "=========================================="
+echo "10. Authentication & Users"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/users" "" "List users"
+echo ""
+
+# Audit Logs
+echo "=========================================="
+echo "11. Audit Logs"
+echo "=========================================="
+test_endpoint "GET" "$API_URL/audit" "" "List audit logs"
+echo ""
+
+echo "=========================================="
+echo "Test Suite Complete!"
+echo "=========================================="