Improve BLE mesh reliability for large transfers (#964)

* Improve BLE mesh reliability for large transfers

Major reliability improvements for fragment-based transfers (photos, files):

**Notification Queue Fixes**
- Fix silent packet loss when notification queue is full - now queues for retry
- Fix retry queue bug where remaining items were lost when one retry failed
- Add periodic drain mechanism as backup (every 5 seconds)

**Write Queue Fixes**
- Fix drainPendingWrites to use atomic take-send-requeue pattern
- Add logging when peripheral is ready for more writes
- Add periodic drain for pending writes as backup

**Fragment Pacing**
- Increase fragment spacing from 4-5ms to 25-30ms to prevent buffer overflow
- Conservative pacing prevents packet loss on congested BLE connections

**Thread Safety**
- Fix race conditions in stopServices() and emergencyDisconnectAll()
- Synchronize access to peripherals/centrals dictionaries during cleanup
- Clear pending message queues in emergencyDisconnectAll()

**Error Recovery**
- Add handlers for all BLE state transitions (poweredOff, unauthorized, etc.)
- Clear Noise session and re-initiate handshake on decryption failure
- Queue ACKs/receipts for delivery after handshake instead of dropping
- Re-queue failed pending messages for retry

**Other Improvements**
- Add route freshness validation in MeshTopologyTracker (60s threshold)
- Add TTL (24h) and size limits (100 per peer) for MessageRouter outbox
- Fix connection timeout to check peripheral.state before canceling
- Add NoiseEncryptionService.clearSession(for:) method

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

* Fix race condition and increase test timeouts

- Fix race in sendNoisePayload: use sync barrier instead of async
  to ensure payload is queued before initiating handshake
  (addresses Codex review feedback)

- Increase BLEServiceTests sleep from 0.5s to 1.0s for CI reliability

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

---------

Co-authored-by: jack <jackjackbits@users.noreply.github.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
jack
2026-01-17 07:39:31 -10:00
committed by GitHub
co-authored by jack Claude Opus 4.5
parent 806c420313
commit 9964710de2
6 changed files with 366 additions and 94 deletions
+50 -9
View File
@@ -5,7 +5,20 @@ import Foundation
@MainActor
final class MessageRouter {
private let transports: [Transport]
private var outbox: [PeerID: [(content: String, nickname: String, messageID: String)]] = [:] // peerID -> queued messages
// Outbox entry with timestamp for TTL-based eviction
private struct QueuedMessage {
let content: String
let nickname: String
let messageID: String
let timestamp: Date
}
private var outbox: [PeerID: [QueuedMessage]] = [:]
// Outbox limits to prevent unbounded memory growth
private static let maxMessagesPerPeer = 100
private static let messageTTLSeconds: TimeInterval = 24 * 60 * 60 // 24 hours
init(transports: [Transport]) {
self.transports = transports
@@ -51,10 +64,19 @@ final class MessageRouter {
SecureLogger.debug("Routing PM via \(type(of: transport)) to \(peerID.id.prefix(8))… id=\(messageID.prefix(8))", category: .session)
transport.sendPrivateMessage(content, to: peerID, recipientNickname: recipientNickname, messageID: messageID)
} else {
// Queue for later
// Queue for later with timestamp for TTL tracking
if outbox[peerID] == nil { outbox[peerID] = [] }
outbox[peerID]?.append((content, recipientNickname, messageID))
SecureLogger.debug("Queued PM for \(peerID.id.prefix(8))… (no reachable transport) id=\(messageID.prefix(8))", category: .session)
let message = QueuedMessage(content: content, nickname: recipientNickname, messageID: messageID, timestamp: Date())
outbox[peerID]?.append(message)
// Enforce per-peer size limit with FIFO eviction
if let count = outbox[peerID]?.count, count > Self.maxMessagesPerPeer {
let evicted = outbox[peerID]?.removeFirst()
SecureLogger.warning("📤 Outbox overflow for \(peerID.id.prefix(8))… - evicted oldest message: \(evicted?.messageID.prefix(8) ?? "?")", category: .session)
}
SecureLogger.debug("Queued PM for \(peerID.id.prefix(8))… (no reachable transport) id=\(messageID.prefix(8))… queue=\(outbox[peerID]?.count ?? 0)", category: .session)
}
}
@@ -87,14 +109,22 @@ final class MessageRouter {
func flushOutbox(for peerID: PeerID) {
guard let queued = outbox[peerID], !queued.isEmpty else { return }
SecureLogger.debug("Flushing outbox for \(peerID.id.prefix(8))… count=\(queued.count)", category: .session)
var remaining: [(content: String, nickname: String, messageID: String)] = []
for (content, nickname, messageID) in queued {
let now = Date()
var remaining: [QueuedMessage] = []
for message in queued {
// Skip expired messages (TTL exceeded)
if now.timeIntervalSince(message.timestamp) > Self.messageTTLSeconds {
SecureLogger.debug("⏰ Expired queued message for \(peerID.id.prefix(8))… id=\(message.messageID.prefix(8))… (age: \(Int(now.timeIntervalSince(message.timestamp)))s)", category: .session)
continue
}
if let transport = reachableTransport(for: peerID) {
SecureLogger.debug("Outbox -> \(type(of: transport)) for \(peerID.id.prefix(8))… id=\(messageID.prefix(8))", category: .session)
transport.sendPrivateMessage(content, to: peerID, recipientNickname: nickname, messageID: messageID)
SecureLogger.debug("Outbox -> \(type(of: transport)) for \(peerID.id.prefix(8))… id=\(message.messageID.prefix(8))", category: .session)
transport.sendPrivateMessage(message.content, to: peerID, recipientNickname: message.nickname, messageID: message.messageID)
} else {
remaining.append((content, nickname, messageID))
remaining.append(message)
}
}
@@ -108,4 +138,15 @@ final class MessageRouter {
func flushAllOutbox() {
for key in Array(outbox.keys) { flushOutbox(for: key) }
}
/// Periodically clean up expired messages from all outboxes
func cleanupExpiredMessages() {
let now = Date()
for peerID in Array(outbox.keys) {
outbox[peerID]?.removeAll { now.timeIntervalSince($0.timestamp) > Self.messageTTLSeconds }
if outbox[peerID]?.isEmpty == true {
outbox.removeValue(forKey: peerID)
}
}
}
}