Implement network resilience features for improved upload stability during network changes

- Enable network events by default in configuration - Integrate network resilience manager into upload handling - Add support for automatic upload pause/resume during WLAN to 5G transitions - Enhance documentation with network resilience settings and testing procedures - Create a test script for validating network resilience functionality
2025-08-24 13:32:44 +00:00
parent 3887feb12c
commit 91128f2861
9 changed files with 792 additions and 22 deletions
--- a/cmd/server/config_simplified.go
+++ b/cmd/server/config_simplified.go
@ -33,6 +33,7 @@ func DefaultConfig() *Config {
 			EnableDynamicWorkers:  true,
 			WorkerScaleUpThresh:   40,  // Optimized from previous session
 			WorkerScaleDownThresh: 10,
+			NetworkEvents:         true,  // Enable network resilience by default
 		},
 		Uploads: UploadsConfig{
 			AllowedExtensions:       []string{".zip", ".rar", ".7z", ".tar.gz", ".tgz", ".gpg", ".enc", ".pgp", ".txt", ".pdf", ".png", ".jpg", ".jpeg"},
@ -104,6 +105,14 @@ func DefaultConfig() *Config {
 			NumWorkers:      4,
 			UploadQueueSize: 100, // Optimized from previous session
 		},
+		NetworkResilience: NetworkResilienceConfig{
+			FastDetection:        true,   // Enable fast 1-second detection
+			QualityMonitoring:    true,   // Monitor connection quality
+			PredictiveSwitching:  true,   // Switch before complete failure
+			MobileOptimizations:  true,   // Mobile-friendly thresholds
+			DetectionInterval:    "1s",   // Fast detection
+			QualityCheckInterval: "5s",   // Regular quality checks
+		},
 		File: FileConfig{},
 		Build: BuildConfig{
 			Version: "3.2",
@ -254,13 +263,26 @@ worker_scale_up_thresh = 40
 worker_scale_down_thresh = 10

 [uploads]
-allowed_extensions = [".zip", ".rar", ".7z", ".tar.gz", ".tgz", ".gpg", ".enc", ".pgp"]
+allowed_extensions = [".zip", ".rar", ".7z", ".tar.gz", ".tgz", ".gpg", ".enc", ".pgp", ".txt", ".pdf", ".jpg", ".jpeg", ".png", ".gif", ".webp", ".mp4", ".mov", ".ogg", ".mp3", ".doc", ".docx"]
 chunked_uploads_enabled = true
 chunk_size = "10MB"
 resumable_uploads_enabled = true
 max_resumable_age = "48h"
 sessiontimeout = "60m"
 maxretries = 3
+networkevents = false                      # Enable network event monitoring for resilience
+
+# Upload resilience and session management
+session_persistence = true                 # Persist sessions across restarts
+session_recovery_timeout = "300s"          # Session recovery timeout after network changes
+client_reconnect_window = "120s"           # Time window for client reconnection
+upload_slot_ttl = "3600s"                  # Upload slot validity time
+retry_failed_uploads = true                # Auto-retry failed uploads
+max_upload_retries = 3                     # Maximum retry attempts
+allow_session_resume = true                # Allow resume from different IPs
+session_persistence_duration = "24h"       # How long to keep session data
+detect_duplicate_uploads = true            # Detect same upload from different IPs
+merge_duplicate_sessions = true            # Merge sessions from same client

 [downloads]
 allowed_extensions = [".txt", ".pdf", ".png", ".jpg", ".jpeg", ".gif", ".bmp", ".tiff", ".svg", ".webp"]
@ -322,6 +344,45 @@ redishealthcheckinterval = "120s"
 numworkers = 4
 uploadqueuesize = 100

+# Network Resilience Configuration (v3.2+)
+[network_resilience]
+enabled = true                              # Enable network resilience system
+fast_detection = true                       # Enable 1-second network change detection
+quality_monitoring = true                   # Monitor RTT and packet loss per interface
+predictive_switching = true                 # Switch proactively before network failure
+mobile_optimizations = true                # Use mobile-friendly thresholds for cellular networks
+upload_resilience = true                    # Resume uploads across network changes
+detection_interval = "1s"                  # Network change detection interval
+quality_check_interval = "5s"              # Connection quality monitoring interval
+max_detection_interval = "10s"             # Maximum detection interval during stable periods
+network_change_threshold = 3               # Switches required to trigger network change
+interface_stability_time = "30s"           # Time to wait before marking interface stable
+upload_pause_timeout = "5m"               # Maximum time to pause uploads during network changes
+upload_retry_timeout = "10m"              # Maximum time to retry uploads after network changes
+rtt_warning_threshold = "200ms"            # RTT threshold for warning
+rtt_critical_threshold = "1000ms"          # RTT threshold for critical
+packet_loss_warning_threshold = 2.0        # Packet loss % for warning
+packet_loss_critical_threshold = 10.0      # Packet loss % for critical
+
+# Multi-Interface Support (v3.2+)
+multi_interface_enabled = false            # Enable multi-interface management
+interface_priority = ["eth0", "wlan0", "wwan0", "ppp0"]  # Interface priority order
+auto_switch_enabled = true                 # Enable automatic interface switching
+switch_threshold_latency = "500ms"         # Latency threshold for switching
+switch_threshold_packet_loss = 5.0         # Packet loss threshold for switching
+quality_degradation_threshold = 0.5        # Quality degradation threshold
+max_switch_attempts = 3                    # Maximum switch attempts per detection
+switch_detection_interval = "10s"          # Switch detection interval
+
+# Client Network Support (v3.2+)
+[client_network_support]
+session_based_tracking = false             # Track sessions by ID instead of IP
+allow_ip_changes = true                    # Allow session continuation from different IPs
+session_migration_timeout = "5m"           # Time to wait for client reconnection
+max_ip_changes_per_session = 10           # Maximum IP changes per session
+client_connection_detection = false        # Detect client network type
+adapt_to_client_network = false           # Optimize parameters based on client connection
+
 [build]
 version = "3.2"
 `
--- a/cmd/server/main.go
+++ b/cmd/server/main.go
@ -748,10 +748,11 @@ func main() {
 	ctx, cancel := context.WithCancel(context.Background())
 	defer cancel()

-	if conf.Server.NetworkEvents { // Corrected field name
-		go monitorNetwork(ctx)      // Assuming monitorNetwork is defined in helpers.go or elsewhere
-		go handleNetworkEvents(ctx) // Assuming handleNetworkEvents is defined in helpers.go or elsewhere
-	}
+	// Legacy network monitoring disabled - now handled by NetworkResilienceManager
+	// if conf.Server.NetworkEvents { // Corrected field name
+	//	go monitorNetwork(ctx)      // OLD: Basic network monitoring (replaced by NetworkResilienceManager)
+	//	go handleNetworkEvents(ctx) // OLD: Basic event logging (replaced by NetworkResilienceManager)
+	// }
 	go updateSystemMetrics(ctx)

 	if conf.ClamAV.ClamAVEnabled {
@ -1640,8 +1641,21 @@ func handleUpload(w http.ResponseWriter, r *http.Request) {
 	}
 	defer dst.Close()

-	// Copy file content
-	written, err := io.Copy(dst, file)
+	// Register upload with network resilience manager for WLAN/5G switching support
+	var uploadCtx *UploadContext
+	var sessionID string
+	if networkManager != nil {
+		sessionID = r.Header.Get("X-Upload-Session-ID")
+		if sessionID == "" {
+			sessionID = fmt.Sprintf("upload_%s_%d", getClientIP(r), time.Now().UnixNano())
+		}
+		uploadCtx = networkManager.RegisterUpload(sessionID)
+		defer networkManager.UnregisterUpload(sessionID)
+		log.Debugf("Registered upload with network resilience: %s", sessionID)
+	}
+
+	// Copy file content with network resilience support
+	written, err := copyWithNetworkResilience(dst, file, uploadCtx)
 	if err != nil {
 		http.Error(w, fmt.Sprintf("Error saving file: %v", err), http.StatusInternalServerError)
 		uploadErrorsTotal.Inc()
--- a/cmd/server/network_resilience.go
+++ b/cmd/server/network_resilience.go
@ -4,11 +4,13 @@ package main

 import (
 	"context"
+	"fmt"
+	"io"
 	"net"
 	"net/http"
+	"os/exec"
 	"sync"
 	"time"
-	"os/exec"
 )

 // NetworkResilienceManager handles network change detection and upload pausing
@ -841,3 +843,56 @@ func InitializeNetworkResilience() {
 	ConfigureEnhancedTimeouts()
 	log.Info("Network resilience system initialized")
 }
+
+// copyWithNetworkResilience performs io.Copy with network resilience support
+func copyWithNetworkResilience(dst io.Writer, src io.Reader, uploadCtx *UploadContext) (int64, error) {
+	if uploadCtx == nil {
+		// Fallback to regular copy if no network resilience
+		return io.Copy(dst, src)
+	}
+	
+	const bufferSize = 32 * 1024 // 32KB buffer
+	buf := make([]byte, bufferSize)
+	var written int64
+	
+	for {
+		// Check for network resilience signals before each read
+		select {
+		case <-uploadCtx.PauseChan:
+			log.Debug("Upload paused due to network change, waiting for resume...")
+			uploadCtx.IsPaused = true
+			// Wait for resume signal
+			<-uploadCtx.ResumeChan
+			uploadCtx.IsPaused = false
+			log.Debug("Upload resumed after network stabilization")
+		case <-uploadCtx.CancelChan:
+			return written, fmt.Errorf("upload cancelled due to network issues")
+		default:
+			// Continue with upload
+		}
+		
+		// Read data
+		nr, readErr := src.Read(buf)
+		if nr > 0 {
+			// Write data
+			nw, writeErr := dst.Write(buf[:nr])
+			if nw > 0 {
+				written += int64(nw)
+			}
+			if writeErr != nil {
+				return written, writeErr
+			}
+			if nr != nw {
+				return written, io.ErrShortWrite
+			}
+		}
+		if readErr != nil {
+			if readErr != io.EOF {
+				return written, readErr
+			}
+			break
+		}
+	}
+	
+	return written, nil
+}