docs(audit): clean 2 more stale circuit-breaker mentions found in /audit

- structured-logger.ts header bullet "External system integration via
  hooks with circuit breaker" -> updated to describe current
  fire-and-forget hook semantics + hookFailures metric counter.
- retry-utils.ts retryConfigs.rsocket comment referenced the
  prometheus-metrics -> circuit-breaker.ts circular import chain.
  circuit-breaker.ts is gone; trimmed the parenthetical to keep just
  the core reasoning (avoid ESM init-order issues via prometheus-metrics).

No code change. Build clean, 35/35 suites, 809/809 tests pass.

src/Defaults/index.ts

@@ -142,8 +142,6 @@ export const DEFAULT_CONNECTION_CONFIG: SocketConfig = {
 	getMessage: async () => undefined,
 	cachedGroupMetadata: async () => undefined,
 	makeSignalRepository: makeLibSignalRepository,
-	// Circuit breaker configuration
-	enableCircuitBreaker: true,
 	// Listener limits (memory leak prevention)
 	// WebSocket: 8 core events (open, close, error, message, ping, pong, upgrade, unexpected-response)
 	//          + 10 dynamic listeners (reconnect handlers, custom events)

src/Signal/libsignal.ts

@@ -6,7 +6,6 @@ import type { LIDMapping, SignalAuthState, SignalKeyStoreWithTransaction } from
 import type { BaileysEventEmitter } from '../Types/Events'
 import type { SignalRepositoryWithLIDStore } from '../Types/Signal'
 import { generateSignalPubKey } from '../Utils'
-import { CircuitBreaker } from '../Utils/circuit-breaker.js'
 import type { ILogger } from '../Utils/logger'
 import { metrics } from '../Utils/prometheus-metrics.js'
 import { isAnyLidUser, isAnyPnUser, jidDecode, transferDevice, WAJIDDomains } from '../WABinary'
@@ -67,8 +66,6 @@ export interface IdentitySaveResult {
 export interface LibSignalRepositoryOptions {
 	/** Event emitter for broadcasting identity changes */
 	ev?: BaileysEventEmitter
-	/** Circuit breaker for prekey operations (optional) */
-	preKeyCircuitBreaker?: CircuitBreaker
 }
 
 // ============================================
@@ -256,7 +253,7 @@ export function makeLibSignalRepository(
 	pnToLIDFunc?: (jids: string[]) => Promise<LIDMapping[] | undefined>,
 	options?: LibSignalRepositoryOptions
 ): SignalRepositoryWithLIDStore {
-	const { ev, preKeyCircuitBreaker } = options || {}
+	const { ev } = options || {}
 	const lidMapping = new LIDMappingStore(auth.keys as SignalKeyStoreWithTransaction, logger, pnToLIDFunc)
 
 	// Identity key cache to avoid repeated storage reads
@@ -277,7 +274,7 @@ export function makeLibSignalRepository(
 		cacheMetricsInterval.unref()
 	}
 
-	const storage = signalStorage(auth, lidMapping, identityKeyCache, ev, preKeyCircuitBreaker, logger)
+	const storage = signalStorage(auth, lidMapping, identityKeyCache, ev, logger)
 
 	const parsedKeys = auth.keys as SignalKeyStoreWithTransaction
 	const migratedSessionCache = new LRUCache<string, true>({
@@ -385,13 +382,9 @@ export function makeLibSignalRepository(
 									'Identity key changed - contact may have reinstalled WhatsApp, session will be re-established'
 								)
 
-								// Reset prekey circuit breaker since we identified the cause
-								// Reset regardless of state (could be open, half-open, or closed with accumulated failures)
-								// eslint-disable-next-line max-depth
-								if (preKeyCircuitBreaker) {
-									preKeyCircuitBreaker.reset()
-									logger.debug({ jid }, 'Reset prekey circuit breaker after identity key change detection')
-								}
+								// (No circuit breaker to reset — bounded-retry is stateless,
+								// each retry is independent so identity-change recovery
+								// happens automatically on the next operation.)
 							} else if (saveResult.isNew) {
 								logger.debug(
 									{ jid, addr: addrStr, fingerprint: saveResult.currentFingerprint },
@@ -769,7 +762,6 @@ function signalStorage(
 	lidMapping: LIDMappingStore,
 	identityKeyCache: LRUCache<string, Uint8Array>,
 	ev?: BaileysEventEmitter,
-	preKeyCircuitBreaker?: CircuitBreaker,
 	logger?: ILogger
 ): ExtendedSignalStorage {
 	// Shared function to resolve PN signal address to LID if mapping exists

src/Socket/chats.ts

@@ -54,7 +54,7 @@ import {
 	resolveLidToPn
 } from '../Utils'
 import { makeKeyedMutex, makeMutex } from '../Utils/make-mutex'
-import processMessage from '../Utils/process-message'
+import processMessage, { getChatId } from '../Utils/process-message'
 import { buildTcTokenFromJid } from '../Utils/tc-token-utils'
 import {
 	type BinaryNode,
@@ -130,6 +130,18 @@ export const makeChatsSocket = (config: SocketConfig) => {
 	/** this mutex ensures that notifications from the same chat are processed in order, while allowing parallel processing across chats */
 	const notificationMutex = makeKeyedMutex()
 
+	/**
+	 * Per-chat mutex dedicated to post-upsert work (history app-state sync +
+	 * processMessage side effects). Kept separate from `messageMutex` because
+	 * the inbound caller already holds `messageMutex(chatId)` while running
+	 * decrypt + upsertMessage; sharing the same mutex would let a concurrently-
+	 * arrived message N+1 enqueue *between* msg N's outer callback and msg N's
+	 * post-upsert task, so msg N+1's processMessage could run before msg N's
+	 * (breaking per-chat ordering of side effects). With a separate mutex,
+	 * post-upsert tasks enqueue strictly in upsertMessage call order.
+	 */
+	const postUpsertMutex = makeKeyedMutex()
+
 	// Timeout for AwaitingInitialSync state
 	let awaitingSyncTimeout: NodeJS.Timeout | undefined
 
@@ -1399,7 +1411,19 @@ export const makeChatsSocket = (config: SocketConfig) => {
 				blockedCollections.clear()
 
 				logger.info('Doing app state sync')
-				await resyncAppState(ALL_WA_PATCH_NAMES, true)
+				try {
+					await resyncAppState(ALL_WA_PATCH_NAMES, true)
+				} catch (err) {
+					// Failure recovery: without this, syncState would stay at Syncing
+					// and ev.flush() would never run, leaving the event buffer pinned
+					// until the buffer's own safety timeout expires. Force the state
+					// machine forward so live inbound events can flow even if the
+					// app-state resync failed (collections are already cleared, so
+					// blocked patches will be retried on the next creds.update tick).
+					syncState = SyncState.Online
+					ev.flush()
+					throw err
+				}
 
 				// Sync is complete, go online and flush everything
 				syncState = SyncState.Online
@@ -1411,29 +1435,131 @@ export const makeChatsSocket = (config: SocketConfig) => {
 			}
 		}
 
-		await Promise.all([
-			(async () => {
-				if (shouldProcessHistoryMsg) {
-					await doAppStateSync()
-				}
-			})(),
-			processMessage(msg, {
-				signalRepository,
-				shouldProcessHistoryMsg,
-				placeholderResendCache,
-				ev,
-				creds: authState.creds,
-				keyStore: authState.keys,
-				logger,
-				options: config.options,
-				getMessage
-			})
-		])
+		// Post-upsert work: history app-state sync + processMessage side effects.
+		// Awaiting here keeps `messages.upsert` pinned in the event buffer
+		// (createBufferedFunction only schedules flush after work() resolves), so
+		// the hot path detaches this work to release the emit on the next debounce
+		// tick.
+		//
+		// Use Promise.allSettled so the combined promise only settles after BOTH
+		// tasks finish. With a plain Promise.all, an early rejection from one task
+		// would release the keyed mutex while the other task is still mutating
+		// chat state — letting the next message of the same chat overtake it and
+		// break per-chat ordering.
+		//
+		// Returns the per-task settle status so the keyShare branch can know
+		// whether processMessage actually persisted the new app-state-sync key
+		// before triggering doAppStateSync (otherwise the sync would hit
+		// isMissingKeyError and park collections in blockedCollections).
+		const postUpsertTasks = async (): Promise<{ processMessageOk: boolean }> => {
+			const [historyResult, processResult] = await Promise.allSettled([
+				shouldProcessHistoryMsg ? doAppStateSync() : Promise.resolve(),
+				processMessage(msg, {
+					signalRepository,
+					shouldProcessHistoryMsg,
+					placeholderResendCache,
+					ev,
+					creds: authState.creds,
+					keyStore: authState.keys,
+					logger,
+					options: config.options,
+					getMessage
+				})
+			])
 
-		// If the app state key arrives and we are waiting to sync, trigger the sync now.
-		if (msg.message?.protocolMessage?.appStateSyncKeyShare && syncState === SyncState.Syncing) {
-			logger.info('App state sync key arrived, triggering app state sync')
-			await doAppStateSync()
+			if (historyResult.status === 'rejected') {
+				logger?.warn(
+					{ err: historyResult.reason, messageId: msg.key?.id, remoteJid: msg.key?.remoteJid },
+					'history doAppStateSync failed'
+				)
+			}
+
+			if (processResult.status === 'rejected') {
+				logger?.warn(
+					{ err: processResult.reason, messageId: msg.key?.id, remoteJid: msg.key?.remoteJid },
+					'processMessage failed'
+				)
+			}
+
+			return { processMessageOk: processResult.status === 'fulfilled' }
+		}
+
+		// Use getChatId + jidNormalizedUser so the mutex key matches the chat-id
+		// scheme processMessage uses for chat updates (broadcasts target the
+		// participant). When getChatId/jidNormalizedUser yields nothing usable
+		// (missing or malformed JID), prefer a message-derived fallback over a
+		// single global 'unknown' bucket — that bucket would head-of-line block
+		// every malformed message behind a shared queue. msg.key.id is unique
+		// per message so unrelated malformed inputs no longer serialize together;
+		// for valid messages we still hit the normalized chat-id path, so the
+		// per-chat ordering guarantee is unchanged where it matters.
+		const rawChatId = getChatId(msg.key)
+		const normalizedChatId = rawChatId ? jidNormalizedUser(rawChatId) : ''
+		const postUpsertChatId = normalizedChatId || msg.key?.id || 'unknown'
+
+		// Wrap in `postUpsertMutex(chatId)` (a SEPARATE keyed mutex from the outer
+		// `messageMutex` held by the inbound caller) so per-chat ordering of
+		// processMessage side effects (chat.unreadCount, LID/PN mapping,
+		// messages.update, history downloads) is preserved across messages of the
+		// same chat.
+		//
+		// Why a separate mutex: if we re-used messageMutex, a concurrently-arrived
+		// message N+1 could enqueue on the outer mutex BEFORE msg N's post-upsert
+		// task gets enqueued (because N's outer callback yields on `await decrypt()`
+		// before reaching the inner enqueue site). The queue would then be
+		// [OuterN+1, InnerN, ...], so InnerN+1 would beat InnerN to processMessage.
+		// With its own mutex, post-upsert tasks enqueue strictly in upsertMessage
+		// call order (which IS message arrival order because the outer
+		// messageMutex serializes the upserts per-chat).
+		const postUpsertWork = postUpsertMutex.mutex(postUpsertChatId, postUpsertTasks)
+
+		const isKeyShareDuringSync =
+			!!msg.message?.protocolMessage?.appStateSyncKeyShare && syncState === SyncState.Syncing
+
+		if (isKeyShareDuringSync) {
+			// appStateSyncKeyShare path: processMessage persists the new app-state-sync
+			// key in its APP_STATE_SYNC_KEY_SHARE handler (via keyStore.transaction).
+			// The follow-up doAppStateSync() needs that key to decrypt patches, so
+			// we MUST wait for processMessage to actually succeed before kicking off
+			// the sync — otherwise it would hit isMissingKeyError and park
+			// collections in blockedCollections, regressing the very issue this
+			// branch was added to fix.
+			//
+			// No deadlock with the inbound caller's messageMutex because
+			// postUpsertMutex is a different mutex instance.
+			logger.info('App state sync key arrived, awaiting persistence before triggering sync')
+			const { processMessageOk } = await postUpsertWork
+
+			if (!processMessageOk) {
+				logger?.warn(
+					{ messageId: msg.key?.id, remoteJid: msg.key?.remoteJid },
+					'processMessage failed during key-share — skipping doAppStateSync to avoid isMissingKeyError'
+				)
+			} else {
+				try {
+					await doAppStateSync()
+				} catch (err) {
+					logger?.warn(
+						{ err, messageId: msg.key?.id, remoteJid: msg.key?.remoteJid },
+						'doAppStateSync failed after key-share persistence'
+					)
+				}
+			}
+		} else {
+			// `postUpsertWork` is not expected to reject — `Promise.allSettled`
+			// inside `postUpsertTasks` never rejects, and per-task failures are
+			// already logged inline. The defensive catch routes any truly
+			// unexpected rejection (e.g. `postUpsertMutex` internal corruption,
+			// future synchronous throws inside processMessage) through
+			// `onUnexpectedError` instead of letting it surface as an
+			// UnhandledPromiseRejection — which on Node ≥15 can terminate the
+			// long-running socket process.
+			postUpsertWork.catch(err =>
+				onUnexpectedError(
+					err,
+					`processing post-upsert work for message ${msg.key?.id || 'unknown'} on ${msg.key?.remoteJid || 'unknown chat'}`
+				)
+			)
 		}
 	})
 

src/Socket/messages-recv.ts

@@ -2330,44 +2330,54 @@ export const makeMessagesRecvSocket = (config: SocketConfig) => {
 		)
 
 		const alt = msg.key.participantAlt || msg.key.remoteJidAlt
-		// Handle LID/PN mappings with hybrid approach:
-		// - Store mapping operation runs in background (non-critical for decrypt)
-		// - Session migration MUST complete before decrypt() to avoid "No session record" errors
-		// This addresses Codex/Copilot review concerns about race conditions with decrypt()
+		// Handle LID/PN mappings with optimized hot-path:
+		// - storeLIDPNMappings is fire-and-forget (background) — does NOT block decrypt
+		// - migrateSession is SYNC (await) — REQUIRED for decrypt to find session
+		//
+		// SAFETY: normalizeMessageJids has a fast-path that uses key.*Alt directly without
+		// hitting the store, so the just-arrived message normalizes correctly even before
+		// the background store completes. Subsequent messages in the same chat hit the
+		// store after the background write is done (ms later).
+		//
+		// Pre-check (getPNForLID/getLIDForPN) was removed — storeLIDPNMappings has internal
+		// LRU cache + dedup, the pre-check was a redundant store round-trip per inbound
+		// message that added latency under load.
+		//
+		// HISTORICAL: this restores the intent of d73cd28d39 (2026-02-03) which was
+		// partially reverted by c3fc792351 the same day due to a race-condition concern
+		// with migrateSession (kept sync here). storeLIDPNMappings was over-protected:
+		// it persists a mapping that downstream consumers can re-derive from key.*Alt,
+		// while migrateSession actually moves the Signal session record that decrypt()
+		// will load microseconds later — those two have very different criticality.
+		//
+		// DO NOT make migrateSession async — decrypt() depends on the session being at
+		// the correct identifier (LID vs PN) when it runs. Other code paths (USync
+		// device lookup in messages-send.ts) create LID/PN mappings without migrating
+		// the session, so we cannot skip migration even when the mapping already exists.
 		if (!!alt) {
 			const altServer = jidDecode(alt)?.server
 			const primaryJid = msg.key.participant || msg.key.remoteJid!
 
 			if (altServer === 'lid') {
-				// Check if mapping already exists to avoid unnecessary storage operations
-				const existingMapping = await signalRepository.lidMapping.getPNForLID(alt)
-				if (!existingMapping) {
-					// MUST await: normalizeMessageJids() runs after this and needs the mapping
-					// in the LIDMappingStore to resolve LID→PN for events delivered to consumers
-					await signalRepository.lidMapping
-						.storeLIDPNMappings([{ lid: alt, pn: primaryJid }])
-						.catch(error => logger.warn({ error, alt, primaryJid }, 'LID mapping storage failed'))
-				}
+				// Fire-and-forget: storeLIDPNMappings has internal cache+dedup,
+				// pre-check (getPNForLID) was redundant.
+				signalRepository.lidMapping
+					.storeLIDPNMappings([{ lid: alt, pn: primaryJid }])
+					.catch(error => logger.warn({ error, alt, primaryJid }, 'background LID mapping store failed'))
 
-				// CRITICAL: ALWAYS migrate session, even if mapping exists
-				// Other code paths (e.g., USync device lookup in messages-send.ts:310-319)
-				// may create mappings via storeLIDPNMappings() without calling migrateSession()
-				// This leaves sessions under PN format while decrypt() expects LID format
-				// Skipping migration based on mapping existence causes "No session record" errors
+				// CRITICAL: ALWAYS migrate session SYNC, even if mapping exists.
+				// Other code paths (e.g., USync device lookup in messages-send.ts) may create
+				// mappings via storeLIDPNMappings() without calling migrateSession(). This
+				// leaves sessions under PN format while decrypt() expects LID format.
+				// Skipping migration based on mapping existence causes "No session record" errors.
 				await signalRepository.migrateSession(primaryJid, alt)
 			} else {
-				// Check if reverse mapping exists
-				const existingMapping = await signalRepository.lidMapping.getLIDForPN(alt)
-				if (!existingMapping) {
-					// MUST await: normalizeMessageJids() runs after this and needs the mapping
-					// in the LIDMappingStore to resolve LID→PN for events delivered to consumers
-					await signalRepository.lidMapping
-						.storeLIDPNMappings([{ lid: primaryJid, pn: alt }])
-						.catch(error => logger.warn({ error, alt, primaryJid }, 'LID mapping storage failed'))
-				}
+				// Fire-and-forget: same rationale as above.
+				signalRepository.lidMapping
+					.storeLIDPNMappings([{ lid: primaryJid, pn: alt }])
+					.catch(error => logger.warn({ error, alt, primaryJid }, 'background LID mapping store failed'))
 
-				// CRITICAL: ALWAYS migrate session, even if mapping exists
-				// Same reasoning as above - mapping existence doesn't guarantee session migration
+				// CRITICAL: ALWAYS migrate session SYNC.
 				await signalRepository.migrateSession(alt, primaryJid)
 			}
 		}

src/Socket/socket.ts

@@ -39,12 +39,7 @@ import {
 	xmppSignedPreKey
 } from '../Utils'
 import { getPlatformId, isAndroidBrowser } from '../Utils/browser-utils'
-import {
-	CircuitBreaker,
-	CircuitOpenError,
-	createConnectionCircuitBreaker,
-	createPreKeyCircuitBreaker
-} from '../Utils/circuit-breaker'
+import { withBoundedRetry } from '../Utils/bounded-retry'
 import {
 	decrementActiveConnections,
 	incrementActiveConnections,
@@ -94,10 +89,6 @@ export const makeSocket = (config: SocketConfig) => {
 		transactionOpts,
 		qrTimeout,
 		makeSignalRepository,
-		enableCircuitBreaker = true,
-		queryCircuitBreaker: queryCircuitBreakerConfig,
-		connectionCircuitBreaker: connectionCircuitBreakerConfig,
-		preKeyCircuitBreaker: preKeyCircuitBreakerConfig,
 		// If enableUnifiedSession is explicitly set (true/false), use it
 		// Otherwise (undefined), check env var, then default to true
 		enableUnifiedSession: enableUnifiedSessionConfig
@@ -107,57 +98,6 @@ export const makeSocket = (config: SocketConfig) => {
 	const enableUnifiedSession =
 		enableUnifiedSessionConfig !== undefined ? enableUnifiedSessionConfig : shouldEnableUnifiedSession()
 
-	// Initialize circuit breakers if enabled
-	let queryCircuitBreaker: CircuitBreaker | undefined
-	let connectionCircuitBreaker: CircuitBreaker | undefined
-	let preKeyCircuitBreaker: CircuitBreaker | undefined
-
-	if (enableCircuitBreaker) {
-		// Circuit breaker for query operations (most critical)
-		queryCircuitBreaker = createConnectionCircuitBreaker({
-			name: 'socket-query',
-			failureThreshold: 5,
-			failureWindow: 60000,
-			resetTimeout: 30000,
-			successThreshold: 2,
-			timeout: defaultQueryTimeoutMs || 60000,
-			onStateChange: (from, to) => {
-				logger.info({ from, to }, 'Query circuit breaker state changed')
-			},
-			onOpen: () => {
-				logger.warn('Query circuit breaker OPENED - blocking requests')
-			},
-			onClose: () => {
-				logger.info('Query circuit breaker CLOSED - resuming normal operation')
-			},
-			...queryCircuitBreakerConfig
-		})
-
-		// Circuit breaker for connection operations
-		connectionCircuitBreaker = createConnectionCircuitBreaker({
-			name: 'socket-connection',
-			failureThreshold: 3,
-			failureWindow: 30000,
-			resetTimeout: 60000,
-			successThreshold: 1,
-			onStateChange: (from, to) => {
-				logger.info({ from, to }, 'Connection circuit breaker state changed')
-			},
-			...connectionCircuitBreakerConfig
-		})
-
-		// Circuit breaker for pre-key operations
-		preKeyCircuitBreaker = createPreKeyCircuitBreaker({
-			name: 'socket-prekey',
-			onStateChange: (from, to) => {
-				logger.info({ from, to }, 'PreKey circuit breaker state changed')
-			},
-			...preKeyCircuitBreakerConfig
-		})
-
-		logger.info('Circuit breakers initialized for socket operations')
-	}
-
 	// Unified Session Manager will be initialized after sendNode is defined
 	let unifiedSessionManager: UnifiedSessionManager | undefined
 
@@ -238,20 +178,17 @@ export const makeSocket = (config: SocketConfig) => {
 		})
 	}
 
-	/** send a raw buffer with circuit breaker protection */
+	/**
+	 * Send a raw buffer over the WebSocket.
+	 *
+	 * Replaces the previous connectionCircuitBreaker wrapping. WebSocket
+	 * lifecycle errors (Connection Closed / Lost / Replaced) were already
+	 * filtered out of the circuit's failure count, so the only failures it
+	 * tripped on were rare native send() errors — for which a state-machine
+	 * gate adds no value over a fast propagated error. Reconnection logic
+	 * in makeSocket already handles WS recovery independently.
+	 */
 	const sendRawMessage = async (data: Uint8Array | Buffer) => {
-		if (connectionCircuitBreaker) {
-			try {
-				return await connectionCircuitBreaker.execute(() => sendRawMessageInternal(data))
-			} catch (error) {
-				if (error instanceof CircuitOpenError) {
-					logger.warn({ circuitName: error.circuitName }, 'Send blocked by connection circuit breaker')
-				}
-
-				throw error
-			}
-		}
-
 		return sendRawMessageInternal(data)
 	}
 
@@ -266,7 +203,7 @@ export const makeSocket = (config: SocketConfig) => {
 	}
 
 	// Initialize Unified Session Manager now that sendNode is defined
-	// (single initialization to avoid duplicating circuit breakers and state)
+	// (single initialization to avoid duplicating manager state)
 	if (enableUnifiedSession) {
 		const sendNodeForSession = async (node: BinaryNode): Promise<void> => {
 			await sendNode(node)
@@ -275,7 +212,6 @@ export const makeSocket = (config: SocketConfig) => {
 		unifiedSessionManager = createUnifiedSessionManager({
 			enabled: true,
 			logger,
-			enableCircuitBreaker,
 			sendNode: sendNodeForSession
 		})
 		logger.info('Unified session manager initialized')
@@ -346,11 +282,8 @@ export const makeSocket = (config: SocketConfig) => {
 		// Register the response listener BEFORE sending — avoids a race where the server
 		// responds before we start listening.  waitForMessage already handles its own
 		// timeout (returns undefined) and connection-close errors (throws), so we do NOT
-		// wrap it in a second promiseTimeout.  The outer wrapper caused a race condition
-		// where both timers fired at ~the same deadline: the outer one threw
-		// Boom('Timed Out') while sendNode was still pending, producing a spurious error
-		// whose message contained "socket-query" → matched the circuit-breaker's
-		// "socket" pattern → incorrectly tripped the breaker after 5 timeouts.
+		// wrap it in a second promiseTimeout.  The outer wrapper would race the inner
+		// timeout near the deadline and throw a spurious Boom('Timed Out').
 		const responsePromise = waitForMessage<any>(msgId, timeoutMs)
 		// Prevent unhandled-rejection if sendNode throws before we reach
 		// `await responsePromise` below. The error from sendNode still propagates
@@ -372,23 +305,17 @@ export const makeSocket = (config: SocketConfig) => {
 		return result
 	}
 
-	/** send a query with circuit breaker protection */
+	/**
+	 * Send a query (iq stanza) and wait for its response.
+	 *
+	 * Replaces the previous queryCircuitBreaker wrapping (state-machine that
+	 * blocked ALL queries for 30s after 5 failures). The cascade behavior
+	 * was incompatible with WhatsApp Android's empirical retry pattern:
+	 * each operation retries independently with exponential backoff, no
+	 * global state. Callers that need retry semantics wrap query() with
+	 * withBoundedRetry() at their level (assertSessions, etc.).
+	 */
 	const query = async (node: BinaryNode, timeoutMs?: number) => {
-		// If circuit breaker is enabled, wrap the query
-		if (queryCircuitBreaker) {
-			try {
-				return await queryCircuitBreaker.execute(() => queryInternal(node, timeoutMs))
-			} catch (error) {
-				// If circuit is open, log and rethrow with context
-				if (error instanceof CircuitOpenError) {
-					logger.warn({ circuitName: error.circuitName, state: error.state }, 'Query blocked by circuit breaker')
-				}
-
-				throw error
-			}
-		}
-
-		// Fallback to direct query if circuit breaker is disabled
 		return queryInternal(node, timeoutMs)
 	}
 
@@ -685,20 +612,14 @@ export const makeSocket = (config: SocketConfig) => {
 	let lastUploadTime = 0
 
 	/** generates and uploads a set of pre-keys to the server */
-	const uploadPreKeys = async (count = MIN_PREKEY_COUNT, retryCount = 0) => {
-		// Check if pre-key circuit breaker is open
-		if (preKeyCircuitBreaker?.isOpen()) {
-			logger.warn('PreKey circuit breaker is open, skipping upload')
-			throw new CircuitOpenError('socket-prekey', 'open')
-		}
-
-		// Check minimum interval (except for retries)
-		if (retryCount === 0) {
-			const timeSinceLastUpload = Date.now() - lastUploadTime
-			if (timeSinceLastUpload < MIN_UPLOAD_INTERVAL) {
-				logger.debug(`Skipping upload, only ${timeSinceLastUpload}ms since last upload`)
-				return
-			}
+	const uploadPreKeys = async (count = MIN_PREKEY_COUNT) => {
+		// Check minimum interval. (Previously this was guarded by `retryCount === 0`
+		// because the function recursed on retry; with bounded-retry handling
+		// retries internally there is no recursion and the guard is implicit.)
+		const timeSinceLastUpload = Date.now() - lastUploadTime
+		if (timeSinceLastUpload < MIN_UPLOAD_INTERVAL) {
+			logger.debug(`Skipping upload, only ${timeSinceLastUpload}ms since last upload`)
+			return
 		}
 
 		// Prevent multiple concurrent uploads — if one is already running, wait for it and return:
@@ -709,8 +630,15 @@ export const makeSocket = (config: SocketConfig) => {
 			return
 		}
 
+		// Shared abort controller so the outer Promise.race can cancel the
+		// in-flight bounded-retry loop if its own timeout fires first.
+		// Without this, the outer race rejects but the bounded-retry loop
+		// keeps running in the background, mutating lastUploadTime / logs
+		// after the caller has given up (CodeRabbit + Copilot reviews).
+		const uploadAbort = new AbortController()
+
 		const uploadLogic = async () => {
-			logger.info({ count, retryCount }, 'uploading pre-keys')
+			logger.info({ count }, 'uploading pre-keys')
 
 			// Generate and save pre-keys atomically (prevents ID collisions on retry)
 			const node = await keys.transaction(async () => {
@@ -721,45 +649,69 @@ export const makeSocket = (config: SocketConfig) => {
 				return node // Only return node since update is already used
 			}, creds?.me?.id || 'upload-pre-keys')
 
-			// Upload to server with circuit breaker protection
-			const uploadToServer = async () => {
-				await query(node)
+			// Upload to server. Bounded-retry handles backoff (replaces both
+			// the previous circuit breaker AND the manual exponential backoff
+			// retry loop with maxRetries=3).
+			//
+			// CRITICAL: query() returns `undefined` on timeout (it does NOT
+			// throw — see waitForMessage). If we just `await query(node)`,
+			// a timed-out upload would resolve with undefined and bounded-retry
+			// would treat it as success. Validate the response shape so a
+			// timeout actually triggers a retry (CodeRabbit review on PR #393).
+			//
+			// bounded-retry's ttlMs MUST be < UPLOAD_TIMEOUT (the outer
+			// Promise.race below). With margin (28s vs 30s) bounded-retry
+			// always reaches its natural give-up first. We additionally pass
+			// an AbortController signal so that if the outer race ever fires
+			// first, the bounded-retry loop is aborted and does not keep
+			// running in the background (CodeRabbit + Copilot reviews).
+			const PER_ATTEMPT_TIMEOUT_MS = 8_000
+			const RETRY_TTL_MS = UPLOAD_TIMEOUT - 2_000
+
+			const uploadToServer = async (signal?: AbortSignal) => {
+				const result = await query(node, PER_ATTEMPT_TIMEOUT_MS)
+				if (signal?.aborted) {
+					throw new Error('aborted')
+				}
+
+				if (!result) {
+					// query() returned undefined → underlying iq response
+					// timed out. Throw so bounded-retry retries.
+					throw new Boom('Pre-key upload query timed out (no response)', { statusCode: 408 })
+				}
+
 				logger.info({ count }, 'uploaded pre-keys successfully')
 				lastUploadTime = Date.now()
 			}
 
 			try {
-				// Use circuit breaker if available
-				if (preKeyCircuitBreaker) {
-					await preKeyCircuitBreaker.execute(uploadToServer)
-				} else {
-					await uploadToServer()
-				}
+				await withBoundedRetry(uploadToServer, {
+					name: 'uploadPreKeys',
+					// 1s -> 2s -> 4s -> 8s (cap). Max 4 attempts fit comfortably
+					// within RETRY_TTL_MS (28s) given perAttemptTimeoutMs=8s.
+					delays: [1000, 2000, 4000, 8000],
+					ttlMs: RETRY_TTL_MS,
+					perAttemptTimeoutMs: PER_ATTEMPT_TIMEOUT_MS,
+					signal: uploadAbort.signal,
+					logger
+				})
 			} catch (uploadError) {
 				logger.error({ uploadError: (uploadError as Error).toString(), count }, 'Failed to upload pre-keys to server')
-
-				// Don't retry if circuit breaker is open
-				if (uploadError instanceof CircuitOpenError) {
-					throw uploadError
-				}
-
-				// Exponential backoff retry (max 3 retries)
-				if (retryCount < 3) {
-					const backoffDelay = Math.min(1000 * Math.pow(2, retryCount), 10000)
-					logger.info(`Retrying pre-key upload in ${backoffDelay}ms`)
-					await new Promise(resolve => setTimeout(resolve, backoffDelay))
-					return uploadPreKeys(count, retryCount + 1)
-				}
-
 				throw uploadError
 			}
 		}
 
-		// Add timeout protection
+		// Outer timeout protection. With ttlMs=28s vs UPLOAD_TIMEOUT=30s,
+		// bounded-retry should always win — but if anything ever blocks
+		// uploadLogic outside bounded-retry's reach, this race aborts the
+		// whole thing cleanly via uploadAbort.
 		uploadPreKeysPromise = Promise.race([
 			uploadLogic(),
 			new Promise<void>((_, reject) =>
-				setTimeout(() => reject(new Boom('Pre-key upload timeout', { statusCode: 408 })), UPLOAD_TIMEOUT)
+				setTimeout(() => {
+					uploadAbort.abort()
+					reject(new Boom('Pre-key upload timeout', { statusCode: 408 }))
+				}, UPLOAD_TIMEOUT)
 			)
 		])
 
@@ -1231,12 +1183,6 @@ export const makeSocket = (config: SocketConfig) => {
 		cleanupPreKeyAutoSync()
 		cleanupSessionTTL()
 
-		// CRITICAL: Destroy circuit breakers AFTER cleanup functions complete
-		// This ensures cleanup functions can still use circuit breakers if needed
-		queryCircuitBreaker?.destroy()
-		connectionCircuitBreaker?.destroy()
-		preKeyCircuitBreaker?.destroy()
-
 		// IMPORTANT: Do NOT use removeAllListeners('connection.update')
 		// It would remove consumer listeners, breaking their reconnection logic
 	}
@@ -1289,11 +1235,12 @@ export const makeSocket = (config: SocketConfig) => {
 				return // connection closing — do not reschedule
 			} else if (ws.isOpen) {
 				// Send keep-alive ping via sendNode() (fire-and-forget) instead of query().
-				// query() wraps the ping in the query circuit breaker — when that breaker is
-				// open or timing out, the ping is never sent, WA never responds, lastDateRecv
-				// goes stale, and the diff check above wrongly fires "Connection was lost".
-				// sendNode() bypasses the query circuit breaker entirely; WA's ping response
-				// still arrives as an incoming frame and updates lastDateRecv normally.
+				// We don't need the response correlator overhead — WA's pong arrives as an
+				// incoming frame and updates lastDateRecv via the standard receive path.
+				// (Historical context: this used to bypass the query circuit breaker,
+				// which would block pings during cascade failures. Circuit breaker has
+				// since been replaced by per-operation bounded-retry — but the
+				// fire-and-forget pattern stays correct on its own merits.)
 				sendNode({
 					tag: 'iq',
 					attrs: {
@@ -1809,25 +1756,6 @@ export const makeSocket = (config: SocketConfig) => {
 		sendWAMBuffer,
 		executeUSyncQuery,
 		onWhatsApp,
-		// Circuit breaker utilities
-		circuitBreakers: {
-			query: queryCircuitBreaker,
-			connection: connectionCircuitBreaker,
-			preKey: preKeyCircuitBreaker
-		},
-		/** Get circuit breaker statistics */
-		getCircuitBreakerStats: () => ({
-			query: queryCircuitBreaker?.getStats(),
-			connection: connectionCircuitBreaker?.getStats(),
-			preKey: preKeyCircuitBreaker?.getStats()
-		}),
-		/** Reset all circuit breakers to closed state */
-		resetCircuitBreakers: () => {
-			queryCircuitBreaker?.reset()
-			connectionCircuitBreaker?.reset()
-			preKeyCircuitBreaker?.reset()
-			logger.info('All circuit breakers reset to closed state')
-		},
 		// Unified Session Telemetry
 		/** Send unified_session telemetry manually */
 		sendUnifiedSession,

src/Types/Socket.ts

@@ -1,7 +1,6 @@
 import type { Agent } from 'https'
 import type { URL } from 'url'
 import { proto } from '../../WAProto/index.js'
-import type { CircuitBreakerOptions } from '../Utils/circuit-breaker'
 import type { ILogger } from '../Utils/logger'
 import type { AuthenticationState, LIDMapping, SignalAuthState, TransactionCapabilityOptions } from './Auth'
 import type { GroupMetadata } from './GroupMetadata'
@@ -203,23 +202,6 @@ export type SocketConfig = {
 		pnToLIDFunc?: (jids: string[]) => Promise<LIDMapping[] | undefined>
 	) => SignalRepositoryWithLIDStore
 
-	// === Circuit Breaker Configuration ===
-
-	/** Enable circuit breaker protection for socket operations (default: true) */
-	enableCircuitBreaker?: boolean
-
-	/** Circuit breaker configuration for query operations */
-	queryCircuitBreaker?: Partial<CircuitBreakerOptions>
-
-	/** Circuit breaker configuration for connection operations */
-	connectionCircuitBreaker?: Partial<CircuitBreakerOptions>
-
-	/** Circuit breaker configuration for pre-key operations */
-	preKeyCircuitBreaker?: Partial<CircuitBreakerOptions>
-
-	/** Circuit breaker configuration for message operations */
-	messageCircuitBreaker?: Partial<CircuitBreakerOptions>
-
 	// === Listener Limits (Memory Leak Prevention) ===
 
 	/**

src/Utils/auth-utils.ts

@@ -341,7 +341,18 @@ export const addTransactionCapability = (
 
 						return result
 					} catch (error) {
-						logger.error({ error }, 'transaction failed, rolling back')
+						// SessionError is part of the normal Bad MAC recovery flow
+						// (retry receipt → sender resends as pkmsg → new session within ~1.3s).
+						// Logging it as ERROR creates 2 noise lines per recoverable Bad MAC cycle.
+						// Downgrade to debug for SessionError; keep ERROR for everything else.
+						// The error is still re-thrown — recovery behavior is unchanged.
+						const errName = (error as { name?: string })?.name
+						if (errName === 'SessionError') {
+							logger.debug({ error }, 'transaction failed (SessionError — recoverable via retry receipt)')
+						} else {
+							logger.error({ error }, 'transaction failed, rolling back')
+						}
+
 						throw error
 					}
 				})

src/Utils/bounded-retry.ts

@@ -0,0 +1,393 @@
+/**
+ * Bounded retry — WhatsApp-aligned per-operation retry without global state.
+ *
+ * # Why this exists
+ *
+ * Replaces the circuit breaker pattern that was causing cascading failures
+ * in production: 5 timeouts in 60s would trip the global breaker, blocking
+ * EVERY socket query (typing indicators, profile pic fetches, contact
+ * validation, presence updates) for 30s — even queries to peers that were
+ * perfectly healthy.
+ *
+ * # Empirical justification
+ *
+ * Captured WhatsApp Android's actual retry behavior via Frida hooks
+ * (`hook-circuit-breaker-re-v2.js`, `hook-retry-bounds.js`) on a real
+ * device under controlled WiFi off/on cycles. Findings:
+ *
+ *   1. Delay sequence (per-operation): 3s -> 10s -> 60s -> ~64s -> 120s
+ *      (cap at 2 min). Last value reused for further attempts.
+ *   2. Memory profile during 5-min network outage: PSS dropped 53MB and
+ *      stabilised. FDs closed (305 -> 295). NO unbounded accumulation.
+ *   3. Recovery on reconnect: ~10s. No retry storm.
+ *   4. NO global state machine observed. Each operation has its own timer.
+ *      Failures of operation A do NOT block operation B.
+ *
+ * The default delays in WHATSAPP_BACKOFF_DELAYS below match the captured
+ * sequence directly. The default 10-min TTL is empirical 3-5 min stability
+ * + safety buffer.
+ *
+ * # Design properties
+ *
+ * - Per-operation isolation: each call to `withBoundedRetry` has its own
+ *   timer. No shared state. Operation A failing has zero effect on B.
+ * - Bounded by `ttlMs` (wall-clock budget). The TTL is enforced strictly
+ *   at three points: (a) before each attempt, (b) cap on per-attempt
+ *   timeout, (c) cap on retry delay.
+ * - Per-attempt timeout: each attempt has its own deadline (default 30s),
+ *   automatically capped to remaining TTL budget so total runtime never
+ *   exceeds ttlMs.
+ * - AbortSignal cancellation: external cancellation supported, both for
+ *   the sleep between retries AND (optionally) the in-flight operation
+ *   when the operation accepts a signal parameter.
+ * - Memory bound: at most one outstanding retry timer per call. Listeners
+ *   are explicitly removed when timers settle. Once the call resolves /
+ *   rejects / aborts, all state is freed.
+ *
+ * # Logging
+ *
+ * Pass `logger` in options to get structured logs at each retry attempt,
+ * give-up, and post-failure recovery. The module emits Prometheus metrics
+ * regardless of whether a logger is provided.
+ *
+ * # When to use this vs. plain query()
+ *
+ * - Use plain `query()` when you want fast-fail semantics (caller decides
+ *   what to do on failure). Most call sites in InfiniteAPI use this.
+ * - Use `withBoundedRetry(() => query(...), { name: 'X' })` when the
+ *   operation is "must-eventually-succeed" with no upstream retry — e.g.
+ *   `uploadPreKeys` or other write paths where the alternative is data loss.
+ *
+ * # Examples
+ *
+ * ```ts
+ * // Single attempt with 5-min TTL, give up after that:
+ * await withBoundedRetry(
+ *   () => assertSessions([jid], true),
+ *   { name: 'assertSessions', ttlMs: 5 * 60_000, logger }
+ * )
+ *
+ * // Tight deadline, fast give-up:
+ * await withBoundedRetry(
+ *   () => sendNode(node),
+ *   { name: 'send', ttlMs: 30_000, perAttemptTimeoutMs: 5_000 }
+ * )
+ *
+ * // Operation that respects abort signal (best practice):
+ * await withBoundedRetry(
+ *   (signal) => fetchSomething({ signal }),
+ *   { name: 'fetch', logger }
+ * )
+ * ```
+ *
+ * @module Utils/bounded-retry
+ */
+
+import type { ILogger } from './logger'
+import { metrics } from './prometheus-metrics.js'
+
+/**
+ * Default delay sequence (milliseconds) — matches WhatsApp Android empirical
+ * behavior. After exhausting the sequence, the last value is used (cap).
+ */
+export const WHATSAPP_BACKOFF_DELAYS = [3000, 10000, 60000, 60000, 120000] as const
+
+/**
+ * Default jitter (+/- 15%) to prevent thundering-herd retries.
+ */
+export const DEFAULT_JITTER_FACTOR = 0.15 as const
+
+/**
+ * Default time-to-live for retries: 10 minutes.
+ *
+ * Empirical justification: WhatsApp Android stabilises memory in ~3-5 min
+ * during a network outage; 10 min gives a generous safety buffer while
+ * preventing unbounded retry accumulation.
+ */
+export const DEFAULT_TTL_MS = 10 * 60 * 1000
+
+/**
+ * Default per-attempt timeout: 30 seconds.
+ *
+ * Most WhatsApp queries respond within seconds. A 30s timeout is generous
+ * enough for slow networks but prevents a single hang from blocking retries.
+ */
+export const DEFAULT_PER_ATTEMPT_TIMEOUT_MS = 30000
+
+export interface BoundedRetryOptions {
+	/** Operation name for logging/metrics */
+	name?: string
+	/** Sequence of delays (ms). Last value is used as cap. */
+	delays?: readonly number[]
+	/** Jitter factor 0..1 (default 0.15) */
+	jitter?: number
+	/** Total wall-clock budget — gives up after this. Default 10 min. */
+	ttlMs?: number
+	/** Per-attempt timeout (ms). Default 30s. Capped by remaining TTL. */
+	perAttemptTimeoutMs?: number
+	/** Predicate: should we retry on this error? Default: always */
+	shouldRetry?: (err: Error, attempt: number) => boolean
+	/** Hook fired before each retry */
+	onRetry?: (err: Error, attempt: number, delayMs: number) => void
+	/**
+	 * AbortSignal — when fired, the loop stops at the next observation point:
+	 *   - If the loop is sleeping between retries, the sleep rejects immediately.
+	 *   - If an attempt is in flight, the abort is forwarded to the operation
+	 *     via the per-attempt signal. The operation must ITSELF observe the
+	 *     signal (e.g. `(signal) => fetch({ signal })`) for cancellation to
+	 *     be truly immediate; otherwise it cancels at the next per-attempt
+	 *     timeout boundary.
+	 *   - Either way, on the next loop iteration BoundedRetryAbortedError is
+	 *     thrown, so the caller never sees more than one trailing attempt
+	 *     after the abort.
+	 */
+	signal?: AbortSignal
+	/** Optional logger for structured retry/give-up/recovery logs */
+	logger?: ILogger
+}
+
+export class BoundedRetryGiveUpError extends Error {
+	constructor(
+		public readonly opName: string,
+		public readonly attempts: number,
+		public readonly elapsedMs: number,
+		public readonly lastError: Error
+	) {
+		super(
+			`bounded-retry "${opName}" gave up after ${attempts} attempts ` +
+				`(${elapsedMs}ms elapsed). Last error: ${lastError.message}`
+		)
+		this.name = 'BoundedRetryGiveUpError'
+	}
+}
+
+export class BoundedRetryAbortedError extends Error {
+	constructor(public readonly opName: string) {
+		super(`bounded-retry "${opName}" aborted via signal`)
+		this.name = 'BoundedRetryAbortedError'
+	}
+}
+
+/**
+ * Apply jitter to a delay: returns delay * (1 +/- jitter)
+ */
+function withJitter(delayMs: number, jitter: number): number {
+	if (jitter <= 0) return delayMs
+	const factor = 1 + (Math.random() * 2 - 1) * jitter
+	return Math.max(0, Math.round(delayMs * factor))
+}
+
+/**
+ * Pick the delay for a given attempt index. Falls back to the last value
+ * (cap) once the sequence is exhausted.
+ */
+function pickDelay(attempt: number, delays: readonly number[]): number {
+	if (delays.length === 0) return 0
+	if (attempt < delays.length) return delays[attempt]!
+	return delays[delays.length - 1]!
+}
+
+/**
+ * Wrap a promise with a per-attempt timeout. Aborts the supplied controller
+ * when the timeout fires so the caller can cancel any in-flight work
+ * (operations that accept the signal). Always clears the timer on settlement.
+ */
+function withTimeout<T>(
+	promise: Promise<T>,
+	timeoutMs: number,
+	name: string,
+	abortOnTimeout: AbortController
+): Promise<T> {
+	return new Promise<T>((resolve, reject) => {
+		const timer = setTimeout(() => {
+			abortOnTimeout.abort()
+			reject(new Error(`bounded-retry "${name}" attempt timed out after ${timeoutMs}ms`))
+		}, timeoutMs)
+		promise
+			.then(value => {
+				clearTimeout(timer)
+				resolve(value)
+			})
+			.catch(err => {
+				clearTimeout(timer)
+				reject(err as Error)
+			})
+	})
+}
+
+/**
+ * Sleep with abort support. Always removes the abort listener on settlement
+ * so listeners do not accumulate on long-lived signals.
+ */
+function sleep(ms: number, signal?: AbortSignal): Promise<void> {
+	return new Promise((resolve, reject) => {
+		if (signal?.aborted) {
+			reject(new Error('aborted'))
+			return
+		}
+
+		let onAbort: (() => void) | undefined
+		const cleanup = () => {
+			if (onAbort && signal) {
+				signal.removeEventListener('abort', onAbort)
+			}
+		}
+
+		const timer = setTimeout(() => {
+			cleanup()
+			resolve()
+		}, ms)
+
+		if (signal) {
+			onAbort = () => {
+				clearTimeout(timer)
+				cleanup()
+				reject(new Error('aborted'))
+			}
+
+			signal.addEventListener('abort', onAbort, { once: true })
+		}
+	})
+}
+
+/**
+ * Run an async operation with bounded exponential-backoff retry.
+ *
+ * Independent per-call: no global state, no cross-operation interaction.
+ * Memory bound: at most one outstanding retry timer per call.
+ *
+ * The operation may optionally accept an `AbortSignal` parameter — when the
+ * per-attempt timeout fires (or the outer signal is aborted), the inner
+ * signal is aborted so the operation can stop in-flight work cleanly.
+ *
+ * @example
+ * ```ts
+ * const result = await withBoundedRetry(
+ *     (signal) => assertSessions([jid], true, { signal }),
+ *     { name: 'assertSessions', ttlMs: 5 * 60_000, logger }
+ * )
+ * ```
+ */
+export async function withBoundedRetry<T>(
+	operation: (signal?: AbortSignal) => Promise<T>,
+	options: BoundedRetryOptions = {}
+): Promise<T> {
+	const name = options.name ?? 'bounded-retry'
+	const delays = options.delays ?? WHATSAPP_BACKOFF_DELAYS
+	const jitter = options.jitter ?? DEFAULT_JITTER_FACTOR
+	const ttlMs = options.ttlMs ?? DEFAULT_TTL_MS
+	const perAttemptTimeoutMs = options.perAttemptTimeoutMs ?? DEFAULT_PER_ATTEMPT_TIMEOUT_MS
+	const shouldRetry = options.shouldRetry ?? (() => true)
+	const logger = options.logger
+
+	const start = Date.now()
+	let lastError: Error = new Error('unknown')
+	let attempt = 0
+
+	while (true) {
+		// Check abort + TTL BEFORE starting an attempt — strict wall-clock
+		// budget. Without this check the loop could begin a new attempt with
+		// 0ms remaining and then run for up to `perAttemptTimeoutMs`.
+		if (options.signal?.aborted) {
+			throw new BoundedRetryAbortedError(name)
+		}
+
+		const elapsedBeforeAttempt = Date.now() - start
+		const remainingBudget = ttlMs - elapsedBeforeAttempt
+		if (remainingBudget <= 0) {
+			metrics.errors?.inc({ category: 'bounded_retry', code: 'ttl_exceeded' })
+			logger?.warn?.(
+				{ op: name, attempts: attempt, elapsedMs: elapsedBeforeAttempt, ttlMs, lastError: lastError.message },
+				'bounded-retry: TTL exceeded before next attempt — giving up'
+			)
+			throw new BoundedRetryGiveUpError(name, attempt, elapsedBeforeAttempt, lastError)
+		}
+
+		// Cap the per-attempt timeout to the remaining TTL budget so a single
+		// attempt cannot run past the wall-clock deadline.
+		const attemptTimeoutMs = Math.min(perAttemptTimeoutMs, remainingBudget)
+		const attemptAbort = new AbortController()
+
+		// Forward outer abort to the per-attempt controller so the in-flight
+		// operation is cancelled when the user aborts. Cleaned up below.
+		let onOuterAbort: (() => void) | undefined
+		if (options.signal) {
+			onOuterAbort = () => attemptAbort.abort()
+			if (options.signal.aborted) {
+				attemptAbort.abort()
+			} else {
+				options.signal.addEventListener('abort', onOuterAbort, { once: true })
+			}
+		}
+
+		try {
+			const result = await withTimeout(operation(attemptAbort.signal), attemptTimeoutMs, name, attemptAbort)
+
+			if (attempt > 0) {
+				metrics.socketEvents?.inc({ event: 'bounded_retry_recovered' })
+				logger?.info?.(
+					{ op: name, attempts: attempt + 1, elapsedMs: Date.now() - start },
+					'bounded-retry: operation succeeded after retries'
+				)
+			}
+
+			return result
+		} catch (err) {
+			lastError = err as Error
+			attempt++
+
+			// (outer-abort listener detachment happens in the finally below —
+			// avoiding a double-remove that breaks listener-count assertions.)
+
+			// If the outer signal aborted us mid-attempt, surface that explicitly
+			// rather than as a generic operation failure.
+			if (options.signal?.aborted) {
+				throw new BoundedRetryAbortedError(name)
+			}
+
+			const elapsed = Date.now() - start
+
+			if (!shouldRetry(lastError, attempt)) {
+				metrics.errors?.inc({ category: 'bounded_retry', code: 'predicate_no_retry' })
+				logger?.warn?.(
+					{ op: name, attempts: attempt, elapsedMs: elapsed, error: lastError.message },
+					'bounded-retry: shouldRetry returned false — giving up'
+				)
+				throw lastError
+			}
+
+			// Re-check budget after the failure so we do not sleep past TTL.
+			const remainingAfterFailure = ttlMs - elapsed
+			if (remainingAfterFailure <= 0) {
+				metrics.errors?.inc({ category: 'bounded_retry', code: 'ttl_exceeded' })
+				logger?.warn?.(
+					{ op: name, attempts: attempt, elapsedMs: elapsed, ttlMs, lastError: lastError.message },
+					'bounded-retry: TTL exceeded after attempt — giving up'
+				)
+				throw new BoundedRetryGiveUpError(name, attempt, elapsed, lastError)
+			}
+
+			const baseDelay = pickDelay(attempt - 1, delays)
+			const delayMs = Math.min(withJitter(baseDelay, jitter), remainingAfterFailure)
+
+			options.onRetry?.(lastError, attempt, delayMs)
+			metrics.socketEvents?.inc({ event: 'bounded_retry_attempt' })
+			logger?.debug?.(
+				{ op: name, attempt, delayMs, elapsedMs: elapsed, error: lastError.message },
+				'bounded-retry: scheduling next attempt'
+			)
+
+			try {
+				await sleep(delayMs, options.signal)
+			} catch {
+				throw new BoundedRetryAbortedError(name)
+			}
+		} finally {
+			// Belt-and-suspenders: ensure the outer-abort listener is always
+			// removed even on early throws.
+			if (options.signal && onOuterAbort) {
+				options.signal.removeEventListener('abort', onOuterAbort)
+			}
+		}
+	}
+}

src/Utils/circuit-breaker.ts

@@ -1,766 +0,0 @@
-/**
- * Circuit Breaker Pattern Implementation
- *
- * Provides protection against cascading failures by monitoring operation outcomes
- * and temporarily blocking requests when failure thresholds are exceeded.
- *
- * States:
- * - CLOSED: Normal operation, requests pass through
- * - OPEN: Failures exceeded threshold, requests are blocked
- * - HALF_OPEN: Testing recovery, limited requests allowed
- *
- * @module Utils/circuit-breaker
- */
-
-import { EventEmitter } from 'events'
-import { metrics } from './prometheus-metrics.js'
-
-/**
- * Circuit breaker operational states
- */
-export type CircuitState = 'closed' | 'open' | 'half-open'
-
-/**
- * Failure record with timestamp for sliding window tracking
- */
-export interface FailureRecord {
-	timestamp: number
-	error: Error
-}
-
-/**
- * Circuit breaker configuration options
- */
-export interface CircuitBreakerOptions {
-	/** Unique identifier for this circuit breaker (used in metrics/logging) */
-	name: string
-	/** Number of failures within the window to trigger OPEN state (default: 5) */
-	failureThreshold?: number
-	/** Time window in ms for counting failures (default: 60000) */
-	failureWindow?: number
-	/** Number of successes required in HALF_OPEN to return to CLOSED (default: 2) */
-	successThreshold?: number
-	/** Time in ms to wait before transitioning from OPEN to HALF_OPEN (default: 30000) */
-	resetTimeout?: number
-	/** Timeout for individual operations in ms (default: 10000) */
-	timeout?: number
-	/** Minimum number of requests before circuit can trip (default: 5) */
-	volumeThreshold?: number
-	/** Predicate to determine if an error should count as a failure */
-	shouldCountError?: (error: Error) => boolean
-	/** Whether to collect Prometheus metrics (default: true) */
-	collectMetrics?: boolean
-	/** Fallback function when circuit is OPEN */
-	fallback?: <T>() => T | Promise<T>
-	/** Callback when state changes */
-	onStateChange?: (from: CircuitState, to: CircuitState) => void
-	/** Callback on failure */
-	onFailure?: (error: Error) => void
-	/** Callback on success */
-	onSuccess?: () => void
-	/** Callback when circuit opens */
-	onOpen?: () => void
-	/** Callback when circuit closes */
-	onClose?: () => void
-	/** Callback when circuit enters half-open */
-	onHalfOpen?: () => void
-}
-
-/**
- * Circuit breaker statistics
- */
-export interface CircuitBreakerStats {
-	state: CircuitState
-	failures: number
-	successes: number
-	consecutiveFailures: number
-	consecutiveSuccesses: number
-	totalCalls: number
-	totalFailures: number
-	totalSuccesses: number
-	totalRejected: number
-	failureRate: number
-	lastFailureTime?: number
-	lastSuccessTime?: number
-	lastStateChange?: number
-	isOpen: boolean
-	isClosed: boolean
-	isHalfOpen: boolean
-}
-
-/**
- * Error thrown when circuit is OPEN and request is rejected
- */
-export class CircuitOpenError extends Error {
-	constructor(
-		public readonly circuitName: string,
-		public readonly state: CircuitState
-	) {
-		super(`Circuit breaker "${circuitName}" is ${state}`)
-		this.name = 'CircuitOpenError'
-	}
-}
-
-/**
- * Error thrown when operation exceeds timeout
- */
-export class CircuitTimeoutError extends Error {
-	constructor(
-		public readonly circuitName: string,
-		public readonly timeoutMs: number
-	) {
-		super(`Circuit breaker "${circuitName}" operation timed out after ${timeoutMs}ms`)
-		this.name = 'CircuitTimeoutError'
-	}
-}
-
-/**
- * Circuit Breaker implementation with sliding window failure tracking
- *
- * @example
- * ```typescript
- * const breaker = new CircuitBreaker({
- *   name: 'whatsapp-api',
- *   failureThreshold: 5,
- *   resetTimeout: 30000
- * })
- *
- * try {
- *   const result = await breaker.execute(() => sendMessage(msg))
- * } catch (error) {
- *   if (error instanceof CircuitOpenError) {
- *     // Circuit is open, use fallback
- *   }
- * }
- * ```
- */
-export class CircuitBreaker extends EventEmitter {
-	private state: CircuitState = 'closed'
-	private failureRecords: FailureRecord[] = []
-	private consecutiveFailures = 0
-	private consecutiveSuccesses = 0
-	private totalCalls = 0
-	private totalFailures = 0
-	private totalSuccesses = 0
-	private totalRejected = 0
-	private lastFailureTime?: number
-	private lastSuccessTime?: number
-	private lastStateChange: number
-	private resetTimer?: ReturnType<typeof setTimeout>
-	private readonly options: Required<CircuitBreakerOptions>
-
-	constructor(options: CircuitBreakerOptions) {
-		super()
-
-		this.options = {
-			name: options.name,
-			failureThreshold: options.failureThreshold ?? 5,
-			failureWindow: options.failureWindow ?? 60000,
-			successThreshold: options.successThreshold ?? 2,
-			resetTimeout: options.resetTimeout ?? 30000,
-			timeout: options.timeout ?? 10000,
-			volumeThreshold: options.volumeThreshold ?? 5,
-			shouldCountError: options.shouldCountError ?? (() => true),
-			collectMetrics: options.collectMetrics ?? true,
-			fallback:
-				options.fallback ??
-				(() => {
-					throw new CircuitOpenError(this.options.name, this.state)
-				}),
-			onStateChange: options.onStateChange ?? (() => {}),
-			onFailure: options.onFailure ?? (() => {}),
-			onSuccess: options.onSuccess ?? (() => {}),
-			onOpen: options.onOpen ?? (() => {}),
-			onClose: options.onClose ?? (() => {}),
-			onHalfOpen: options.onHalfOpen ?? (() => {})
-		}
-
-		this.lastStateChange = Date.now()
-	}
-
-	/**
-	 * Check if the circuit allows execution
-	 */
-	canExecute(): boolean {
-		if (this.state === 'closed') {
-			return true
-		}
-
-		if (this.state === 'open') {
-			return false
-		}
-
-		// HALF_OPEN: allow limited requests for testing
-		return true
-	}
-
-	/**
-	 * Execute an async operation with circuit breaker protection
-	 */
-	async execute<T>(operation: () => T | Promise<T>): Promise<T> {
-		this.totalCalls++
-
-		// Check if circuit allows execution
-		if (this.state === 'open') {
-			this.totalRejected++
-
-			if (this.options.collectMetrics) {
-				metrics.errors.inc({ category: 'circuit_breaker', code: 'rejected' })
-			}
-
-			return this.options.fallback() as T
-		}
-
-		// Execute with timeout protection
-		try {
-			const result = await this.executeWithTimeout(operation)
-			this.recordSuccess()
-			return result
-		} catch (error) {
-			this.recordFailure(error as Error)
-			throw error
-		}
-	}
-
-	/**
-	 * Execute a synchronous operation with circuit breaker protection
-	 */
-	executeSync<T>(operation: () => T): T {
-		this.totalCalls++
-
-		if (this.state === 'open') {
-			this.totalRejected++
-
-			if (this.options.collectMetrics) {
-				metrics.errors.inc({ category: 'circuit_breaker', code: 'rejected' })
-			}
-
-			return this.options.fallback() as T
-		}
-
-		try {
-			const result = operation()
-			this.recordSuccess()
-			return result
-		} catch (error) {
-			this.recordFailure(error as Error)
-			throw error
-		}
-	}
-
-	/**
-	 * Execute operation with timeout wrapper
-	 */
-	private async executeWithTimeout<T>(operation: () => T | Promise<T>): Promise<T> {
-		return new Promise<T>((resolve, reject) => {
-			const timer = setTimeout(() => {
-				reject(new CircuitTimeoutError(this.options.name, this.options.timeout))
-			}, this.options.timeout)
-
-			Promise.resolve(operation())
-				.then(result => {
-					clearTimeout(timer)
-					resolve(result)
-				})
-				.catch(error => {
-					clearTimeout(timer)
-					reject(error)
-				})
-		})
-	}
-
-	/**
-	 * Record a successful operation
-	 */
-	private recordSuccess(): void {
-		this.totalSuccesses++
-		this.lastSuccessTime = Date.now()
-		this.consecutiveSuccesses++
-		this.consecutiveFailures = 0
-
-		if (this.options.collectMetrics) {
-			metrics.socketEvents.inc({ event: 'circuit_success' })
-		}
-
-		this.options.onSuccess()
-		this.emit('success')
-
-		// In HALF_OPEN state, check if we can close the circuit
-		if (this.state === 'half-open') {
-			if (this.consecutiveSuccesses >= this.options.successThreshold) {
-				this.transitionTo('closed')
-			}
-		}
-	}
-
-	/**
-	 * Record a failed operation
-	 */
-	private recordFailure(error: Error): void {
-		// Check if this error should count as a failure
-		if (!this.options.shouldCountError(error)) {
-			return
-		}
-
-		const now = Date.now()
-		this.totalFailures++
-		this.lastFailureTime = now
-		this.consecutiveFailures++
-		this.consecutiveSuccesses = 0
-
-		// Add to failure records for sliding window
-		this.failureRecords.push({ timestamp: now, error })
-
-		// Clean old failures outside the window
-		this.cleanOldFailures()
-
-		if (this.options.collectMetrics) {
-			metrics.errors.inc({ category: 'circuit_breaker', code: 'failure' })
-		}
-
-		this.options.onFailure(error)
-		this.emit('failure', error)
-
-		// State transition logic
-		if (this.state === 'half-open') {
-			// Any failure in HALF_OPEN immediately reopens the circuit
-			this.transitionTo('open')
-		} else if (this.state === 'closed') {
-			// Check if we should trip the circuit
-			const recentFailures = this.failureRecords.length
-
-			if (this.totalCalls >= this.options.volumeThreshold && recentFailures >= this.options.failureThreshold) {
-				this.transitionTo('open')
-			}
-		}
-	}
-
-	/**
-	 * Remove failure records outside the sliding window
-	 */
-	private cleanOldFailures(): void {
-		const cutoff = Date.now() - this.options.failureWindow
-		this.failureRecords = this.failureRecords.filter(record => record.timestamp > cutoff)
-	}
-
-	/**
-	 * Transition to a new state
-	 */
-	private transitionTo(newState: CircuitState): void {
-		const oldState = this.state
-
-		if (oldState === newState) {
-			return
-		}
-
-		this.state = newState
-		this.lastStateChange = Date.now()
-
-		// Clear existing reset timer
-		if (this.resetTimer) {
-			clearTimeout(this.resetTimer)
-			this.resetTimer = undefined
-		}
-
-		// State-specific actions
-		switch (newState) {
-			case 'closed':
-				this.consecutiveFailures = 0
-				this.consecutiveSuccesses = 0
-				this.failureRecords = []
-				this.options.onClose()
-				this.emit('close')
-				break
-
-			case 'open':
-				this.consecutiveSuccesses = 0
-				this.options.onOpen()
-				this.emit('open')
-
-				// Record circuit breaker trip metric
-				if (this.options.collectMetrics) {
-					metrics.circuitBreakerTrips?.inc({ name: this.options.name })
-				}
-
-				// Schedule transition to HALF_OPEN
-				this.resetTimer = setTimeout(() => {
-					this.transitionTo('half-open')
-				}, this.options.resetTimeout)
-				break
-
-			case 'half-open':
-				this.consecutiveSuccesses = 0
-				this.consecutiveFailures = 0
-				this.options.onHalfOpen()
-				this.emit('half-open')
-				break
-		}
-
-		this.options.onStateChange(oldState, newState)
-		this.emit('state-change', { from: oldState, to: newState })
-	}
-
-	/**
-	 * Manually trip the circuit to OPEN state
-	 */
-	trip(): void {
-		this.transitionTo('open')
-	}
-
-	/**
-	 * Manually reset the circuit to CLOSED state
-	 */
-	reset(): void {
-		this.transitionTo('closed')
-	}
-
-	/**
-	 * Get current circuit state
-	 */
-	getState(): CircuitState {
-		return this.state
-	}
-
-	/**
-	 * Check if circuit is OPEN
-	 */
-	isOpen(): boolean {
-		return this.state === 'open'
-	}
-
-	/**
-	 * Check if circuit is CLOSED
-	 */
-	isClosed(): boolean {
-		return this.state === 'closed'
-	}
-
-	/**
-	 * Check if circuit is HALF_OPEN
-	 */
-	isHalfOpen(): boolean {
-		return this.state === 'half-open'
-	}
-
-	/**
-	 * Get circuit breaker statistics
-	 */
-	getStats(): CircuitBreakerStats {
-		this.cleanOldFailures()
-
-		const failureRate = this.totalCalls > 0 ? (this.totalFailures / this.totalCalls) * 100 : 0
-
-		return {
-			state: this.state,
-			failures: this.failureRecords.length,
-			successes: this.consecutiveSuccesses,
-			consecutiveFailures: this.consecutiveFailures,
-			consecutiveSuccesses: this.consecutiveSuccesses,
-			totalCalls: this.totalCalls,
-			totalFailures: this.totalFailures,
-			totalSuccesses: this.totalSuccesses,
-			totalRejected: this.totalRejected,
-			failureRate,
-			lastFailureTime: this.lastFailureTime,
-			lastSuccessTime: this.lastSuccessTime,
-			lastStateChange: this.lastStateChange,
-			isOpen: this.isOpen(),
-			isClosed: this.isClosed(),
-			isHalfOpen: this.isHalfOpen()
-		}
-	}
-
-	/**
-	 * Get circuit breaker name
-	 */
-	getName(): string {
-		return this.options.name
-	}
-
-	/**
-	 * Destroy circuit breaker and clean up resources
-	 */
-	destroy(): void {
-		if (this.resetTimer) {
-			clearTimeout(this.resetTimer)
-		}
-
-		this.failureRecords = []
-		this.removeAllListeners()
-	}
-}
-
-/**
- * Factory function to create a circuit breaker
- */
-export function createCircuitBreaker(options: CircuitBreakerOptions): CircuitBreaker {
-	return new CircuitBreaker(options)
-}
-
-/**
- * Registry for managing multiple circuit breakers
- */
-export class CircuitBreakerRegistry {
-	private breakers: Map<string, CircuitBreaker> = new Map()
-
-	/**
-	 * Get or create a circuit breaker by name
-	 */
-	get(name: string, options?: Omit<CircuitBreakerOptions, 'name'>): CircuitBreaker {
-		if (!this.breakers.has(name)) {
-			const breaker = new CircuitBreaker({ ...options, name })
-			this.breakers.set(name, breaker)
-		}
-
-		return this.breakers.get(name)!
-	}
-
-	/**
-	 * Check if a circuit breaker exists
-	 */
-	has(name: string): boolean {
-		return this.breakers.has(name)
-	}
-
-	/**
-	 * Remove a circuit breaker
-	 */
-	remove(name: string): boolean {
-		const breaker = this.breakers.get(name)
-		if (breaker) {
-			breaker.destroy()
-			return this.breakers.delete(name)
-		}
-
-		return false
-	}
-
-	/**
-	 * Get all circuit breakers
-	 */
-	getAll(): Map<string, CircuitBreaker> {
-		return new Map(this.breakers)
-	}
-
-	/**
-	 * Get statistics for all circuit breakers
-	 */
-	getAllStats(): Record<string, CircuitBreakerStats> {
-		const stats: Record<string, CircuitBreakerStats> = {}
-		for (const [name, breaker] of this.breakers) {
-			stats[name] = breaker.getStats()
-		}
-
-		return stats
-	}
-
-	/**
-	 * Reset all circuit breakers to CLOSED state
-	 */
-	resetAll(): void {
-		for (const breaker of this.breakers.values()) {
-			breaker.reset()
-		}
-	}
-
-	/**
-	 * Destroy all circuit breakers
-	 */
-	destroyAll(): void {
-		for (const breaker of this.breakers.values()) {
-			breaker.destroy()
-		}
-
-		this.breakers.clear()
-	}
-}
-
-/**
- * Global circuit breaker registry instance
- */
-export const globalCircuitRegistry = new CircuitBreakerRegistry()
-
-/**
- * Decorator to protect a method with circuit breaker
- *
- * @example
- * ```typescript
- * class MyService {
- *   @circuitBreaker({ failureThreshold: 3 })
- *   async fetchData() {
- *     return await api.getData()
- *   }
- * }
- * ```
- */
-export function circuitBreaker(options: Omit<CircuitBreakerOptions, 'name'> & { name?: string } = {}) {
-	return function (
-		_target: unknown,
-		propertyKey: string,
-		descriptor: TypedPropertyDescriptor<(...args: unknown[]) => unknown>
-	) {
-		const originalMethod = descriptor.value
-		if (!originalMethod) return descriptor
-
-		const name = options.name || propertyKey
-		const breaker = globalCircuitRegistry.get(name, options)
-
-		descriptor.value = async function (...args: unknown[]): Promise<unknown> {
-			return breaker.execute(() => originalMethod.apply(this, args))
-		}
-
-		return descriptor
-	}
-}
-
-/**
- * Wrap a function with circuit breaker protection
- *
- * @example
- * ```typescript
- * const protectedFetch = withCircuitBreaker(
- *   fetchData,
- *   { name: 'api-fetch', failureThreshold: 5 }
- * )
- * ```
- */
-// eslint-disable-next-line space-before-function-paren
-export function withCircuitBreaker<T extends (...args: unknown[]) => unknown>(
-	fn: T,
-	options: CircuitBreakerOptions
-): T {
-	const breaker = new CircuitBreaker(options)
-
-	return (async (...args: Parameters<T>): Promise<ReturnType<T>> => {
-		return breaker.execute(() => fn(...args)) as Promise<ReturnType<T>>
-	}) as unknown as T
-}
-
-/**
- * Get health status of all circuit breakers
- */
-export function getCircuitHealth(): {
-	healthy: boolean
-	openCircuits: string[]
-	stats: Record<string, CircuitBreakerStats>
-} {
-	const stats = globalCircuitRegistry.getAllStats()
-	const openCircuits: string[] = []
-
-	for (const [name, stat] of Object.entries(stats)) {
-		if (stat.isOpen) {
-			openCircuits.push(name)
-		}
-	}
-
-	return {
-		healthy: openCircuits.length === 0,
-		openCircuits,
-		stats
-	}
-}
-
-/**
- * Create a pre-configured circuit breaker for WhatsApp PreKey operations
- *
- * This circuit breaker is optimized for handling encryption/session errors
- * that commonly occur with WhatsApp's Signal protocol implementation.
- *
- * @example
- * ```typescript
- * const preKeyBreaker = createPreKeyCircuitBreaker()
- *
- * async function sendEncryptedMessage(msg) {
- *   return preKeyBreaker.execute(async () => {
- *     return await encryptAndSend(msg)
- *   })
- * }
- * ```
- */
-export function createPreKeyCircuitBreaker(customOptions?: Partial<CircuitBreakerOptions>): CircuitBreaker {
-	const preKeyErrorPatterns = ['prekey', 'pre-key', 'session', 'signal', 'encrypt', 'decrypt', 'cipher', 'key']
-
-	return new CircuitBreaker({
-		name: 'prekey-operations',
-		failureThreshold: 5,
-		failureWindow: 60000,
-		resetTimeout: 30000,
-		successThreshold: 2,
-		shouldCountError: (error: Error) => {
-			const message = error.message.toLowerCase()
-			return preKeyErrorPatterns.some(pattern => message.includes(pattern))
-		},
-		...customOptions
-	})
-}
-
-/**
- * Create a circuit breaker for WebSocket connection operations
- */
-export function createConnectionCircuitBreaker(customOptions?: Partial<CircuitBreakerOptions>): CircuitBreaker {
-	const connectionErrorPatterns = [
-		'econnrefused',
-		'econnreset',
-		'etimedout',
-		'enotfound',
-		'socket',
-		'websocket',
-		'connection',
-		'network'
-	]
-
-	// Normal WS lifecycle status codes that must NEVER trip the query circuit breaker.
-	// These are transient events that happen on every reconnect and carry no information
-	// about persistent server-side failures.  The reconnection logic in makeSocket handles
-	// them independently.
-	const WS_LIFECYCLE_STATUS_CODES = new Set([
-		428, // connectionClosed
-		408, // connectionLost / timedOut
-		440, // connectionReplaced
-	])
-
-	return new CircuitBreaker({
-		name: 'connection-operations',
-		failureThreshold: 3,
-		failureWindow: 30000,
-		resetTimeout: 60000,
-		successThreshold: 1,
-		shouldCountError: (error: Error) => {
-			// CircuitTimeoutError messages embed the circuit name (e.g. "socket-query"), which
-			// accidentally matches the "socket" pattern below and causes a self-reinforcing loop
-			// where the breaker's own timeouts keep tripping the breaker. Exclude them explicitly.
-			if (error instanceof CircuitTimeoutError) return false
-
-			// Exclude normal WS reconnect events (Connection Closed, Connection Lost, Timed Out, etc.).
-			// These are not server-side failures — they happen on every restart/redeploy and should
-			// never cause the circuit to open.  If we counted them, the 3 concurrent parallel
-			// post-login queries (sendPassiveIq, uploadPreKeysToServerIfRequired, digestKeyBundle) could all
-			// fail simultaneously when the WS drops, instantly opening the circuit and blocking
-			// profile-picture fetches and message delivery for the next 30 s.
-			const statusCode = (error as { output?: { statusCode?: number } })?.output?.statusCode
-			if (statusCode !== undefined && WS_LIFECYCLE_STATUS_CODES.has(statusCode)) return false
-
-			const message = error.message.toLowerCase()
-			const code = (error as NodeJS.ErrnoException).code?.toLowerCase() || ''
-			return connectionErrorPatterns.some(pattern => message.includes(pattern) || code.includes(pattern))
-		},
-		...customOptions
-	})
-}
-
-/**
- * Create a circuit breaker for message sending operations
- */
-export function createMessageCircuitBreaker(customOptions?: Partial<CircuitBreakerOptions>): CircuitBreaker {
-	return new CircuitBreaker({
-		name: 'message-operations',
-		failureThreshold: 5,
-		failureWindow: 60000,
-		resetTimeout: 15000,
-		successThreshold: 2,
-		timeout: 30000,
-		...customOptions
-	})
-}
-
-export default CircuitBreaker

src/Utils/decode-wa-message.ts

@@ -58,7 +58,11 @@ export const BAD_MAC_ERROR_TEXT = 'Bad MAC'
 export const DECRYPTION_RETRY_CONFIG = {
 	maxRetries: 3,
 	baseDelayMs: 100,
-	sessionRecordErrors: ['No session record', 'SessionError: No session record'],
+	// 'No matching sessions found' is the libsignal error when decryptWithSessions exhausts
+	// all stored sessions for a JID. Same recovery flow (retry receipt → pkmsg → new session)
+	// — categorise it as session-record so the caller logs DEBUG on retry, ERROR only when
+	// retries are exhausted (instead of dumping the full stack as an unknown error).
+	sessionRecordErrors: ['No session record', 'SessionError: No session record', 'No matching sessions found'],
 	corruptedSessionErrors: ['Bad MAC', 'MessageCounterError', MISSING_KEYS_ERROR_TEXT]
 }
 
@@ -421,9 +425,26 @@ export const decryptMessageNode = (
 						const isCorrupted = isCorruptedSessionError(originalError)
 						const isSessionRecord = isSessionRecordError(originalError)
 
+						// Slim error projection — keep name/message/type for diagnosis,
+						// drop `stack` which adds 4-5 lines of node_modules paths per log
+						// for known-recoverable libsignal errors.
+						//
+						// CRITICAL: only slim for KNOWN-RECOVERABLE categories (corrupted /
+						// session-record). The unknown-error branch keeps the full Error so
+						// protobuf/parsing/runtime bugs still emit a stack trace where it
+						// matters most. Catches Copilot/Codex P2 review on PR #391.
+						const slimErr = originalError
+							? {
+									name: (originalError as { name?: string }).name,
+									message: (originalError as { message?: string }).message,
+									type: (originalError as { type?: string }).type
+								}
+							: undefined
+						const isRecoverableCategory = isCorrupted || isSessionRecord
+
 						const errorContext = {
 							key: fullMessage.key,
-							err: originalError,
+							err: isRecoverableCategory ? slimErr : originalError,
 							messageType: tag === 'plaintext' ? 'plaintext' : attrs.type,
 							sender,
 							author,

src/Utils/health-status.ts

@@ -6,11 +6,12 @@
  * @module Utils/health-status
  */
 
-import { globalCircuitRegistry } from './circuit-breaker.js'
 import { getVersionCacheStatus } from './version-cache.js'
 
 /**
- * Circuit breaker health information
+ * Circuit breaker health information (kept for k8s probe schema stability —
+ * always reports an empty list now that circuit breakers were removed in
+ * favour of bounded-retry).
  */
 export interface CircuitBreakerHealth {
 	name: string
@@ -105,44 +106,10 @@ export function getHealthStatus(): HealthStatus {
 		})
 	}
 
-	// 2. Check circuit breakers
+	// 2. Circuit breakers were removed in favour of bounded-retry.
+	// Keep the field for backward-compat with k8s probe schema, always empty.
 	const circuitBreakers: CircuitBreakerHealth[] = []
-	let openCircuits = 0
-
-	for (const [name, breaker] of globalCircuitRegistry.getAll()) {
-		const stats = breaker.getStats()
-		const state = breaker.getState()
-
-		circuitBreakers.push({
-			name,
-			state,
-			failures: stats.totalFailures,
-			successes: stats.totalSuccesses,
-			totalCalls: stats.totalCalls
-		})
-
-		if (state === 'open') {
-			openCircuits++
-		}
-	}
-
-	if (openCircuits > 0) {
-		checks.push({
-			name: 'circuit_breakers',
-			status: openCircuits > 2 ? 'fail' : 'warn',
-			message: `${openCircuits} circuit breaker(s) are open`
-		})
-		if (openCircuits > 2) {
-			overallStatus = 'unhealthy'
-		} else if (overallStatus === 'healthy') {
-			overallStatus = 'degraded'
-		}
-	} else {
-		checks.push({
-			name: 'circuit_breakers',
-			status: 'pass'
-		})
-	}
+	checks.push({ name: 'circuit_breakers', status: 'pass' })
 
 	// 3. Check memory usage (warn if > 90%)
 	const memUsage = process.memoryUsage()

src/Utils/index.ts

@@ -33,8 +33,8 @@ export * from './trace-context'
 export * from './prometheus-metrics'
 
 // Resilience and performance
+export * from './bounded-retry'
 export * from './cache-utils'
-export * from './circuit-breaker'
 export * from './retry-utils'
 
 // Telemetry and detection mitigation

src/Utils/process-message.ts

@@ -82,6 +82,47 @@ const REAL_MSG_REQ_ME_STUB_TYPES = new Set([WAMessageStubType.GROUP_PARTICIPANT_
  * (TC_TOKEN_INDEX_KEY) via buildMergedTcTokenIndexWrite, so the 24h prune sweep in
  * messages-recv picks them up across sessions.
  */
+/**
+ * Single-concurrency queue for `storeTcTokensFromHistorySync` calls.
+ *
+ * Why: the function does read-then-write merges (`keyStore.get('tctoken', ...)` →
+ * compute → `keyStore.set(...)`) which are NOT atomic at the store level. If two
+ * history-sync chunks invoke this concurrently (common during reconnect / QR
+ * scan), an older chunk that started first can `keyStore.set` AFTER a newer
+ * chunk, overwriting the newer entry — and worse, the merged `__index` write
+ * can drop JIDs the other chunk just added. Result: stale tcTokens / repeat 463
+ * sends until the next opportunistic refetch.
+ *
+ * Serialising via a chained Promise keeps the runs ordered while still freeing
+ * the calling `processMessage` to emit `messaging-history.set` immediately
+ * (the chain is fire-and-forget at the call site). Errors don't break the chain
+ * — each `catch` resets it to `Promise.resolve()` so a single failure can't
+ * stall future runs.
+ *
+ * The chain is module-scoped (one per Node process). Multiple Baileys instances
+ * sharing this module will serialise across instances too, but their writes
+ * target different keyStores so there's no correctness gain — only a tiny loss
+ * of inter-instance parallelism for tcToken syncs, which is acceptable given
+ * how rarely this runs vs. how rare cross-instance contention is.
+ */
+let historyTcTokenChain: Promise<void> = Promise.resolve()
+
+function scheduleHistoryTcTokenSync(
+	chats: Chat[],
+	signalRepository: SignalRepositoryWithLIDStore,
+	keyStore: SignalKeyStoreWithTransaction,
+	logger?: ILogger
+): void {
+	historyTcTokenChain = historyTcTokenChain
+		.catch(() => {
+			/* swallow prior error so chain stays alive */
+		})
+		.then(() => storeTcTokensFromHistorySync(chats, signalRepository, keyStore, logger))
+		.catch(err => {
+			logger?.warn({ err }, 'background tctoken history-sync persistence failed')
+		})
+}
+
 async function storeTcTokensFromHistorySync(
 	chats: Chat[],
 	signalRepository: SignalRepositoryWithLIDStore,
@@ -565,11 +606,28 @@ const processMessage = async (
 						}
 					}
 
-					// Persist tctokens carried by history-sync chats BEFORE emitting messaging-history.set
-					// — listeners may immediately fire outbound sends that need the tctoken, and the store
-					// has to be populated first to avoid an error 463 on the first multi-device send.
-					// Runs AFTER storeLIDPNMappings (see comment above) so LID resolution works.
-					await storeTcTokensFromHistorySync(data.chats, signalRepository, keyStore, logger)
+					// Persist tctokens carried by history-sync chats in BACKGROUND, serialised.
+					//
+					// Originally awaited (PR #386) to avoid 463 on first multi-device send, but in
+					// production this drained the event buffer per-chunk and added visible delivery
+					// latency (especially after restart / QR scan when many chunks arrived at once).
+					//
+					// `scheduleHistoryTcTokenSync` enqueues onto a single-concurrency promise chain
+					// (see definition above) — chunks persist sequentially in the order they were
+					// emitted, preserving timestamp monotonicity AND keeping the `__index` write
+					// safe from concurrent merge clobbers. The call returns immediately so the
+					// `messaging-history.set` emit is not blocked.
+					//
+					// TRADE-OFF: a listener that fires an outbound send IMMEDIATELY after the emit
+					// may race the still-pending persistence and get a 463 on that specific send.
+					// The existing 463 handler in messages-recv.ts triggers a getPrivacyTokens()
+					// refetch that auto-recovers within seconds. Net result is much better UX than
+					// per-chunk stalls.
+					//
+					// DO NOT add `await` back here without re-evaluating production latency, AND
+					// DO NOT call storeTcTokensFromHistorySync directly — it must go through the
+					// chain to preserve write ordering across overlapping chunks.
+					scheduleHistoryTcTokenSync(data.chats, signalRepository, keyStore, logger)
 
 					ev.emit('messaging-history.set', {
 						...data,

src/Utils/prometheus-metrics.ts

@@ -1570,26 +1570,6 @@ export const metrics = {
 		new Counter('socket_reconnects_total', 'Total socket reconnection attempts', ['reason'])
 	),
 
-	// ========== Circuit Breaker Metrics ==========
-	circuitBreakerState: baileysMetrics.register(
-		new Gauge('circuit_breaker_state', 'Circuit breaker state (0=closed, 1=open, 2=half-open)', ['name'])
-	),
-	circuitBreakerTrips: baileysMetrics.register(
-		new Counter('circuit_breaker_trips_total', 'Total circuit breaker trips', ['name'])
-	),
-	circuitBreakerRecoveries: baileysMetrics.register(
-		new Counter('circuit_breaker_recoveries_total', 'Total circuit breaker recoveries', ['name'])
-	),
-	circuitBreakerRejections: baileysMetrics.register(
-		new Counter('circuit_breaker_rejections_total', 'Total requests rejected by circuit breaker', ['name'])
-	),
-	circuitBreakerSuccesses: baileysMetrics.register(
-		new Counter('circuit_breaker_successes_total', 'Total successful requests through circuit breaker', ['name'])
-	),
-	circuitBreakerFailures: baileysMetrics.register(
-		new Counter('circuit_breaker_failures_total', 'Total failed requests through circuit breaker', ['name'])
-	),
-
 	// ========== Encryption Metrics ==========
 	encryptionOperations: baileysMetrics.register(
 		new Counter('encryption_operations_total', 'Total encryption operations', ['operation'])
@@ -2167,9 +2147,6 @@ function initializeMetricsWithLabels(): void {
 		metrics.messageFailures?.inc({ type: 'text', reason: 'max_retries' }, 0)
 		metrics.messageFailures?.inc({ type: 'other', reason: 'max_retries' }, 0)
 
-		// Circuit breaker metric
-		metrics.circuitBreakerTrips?.inc({ name: 'main' }, 0)
-
 		console.log('[Prometheus] Initialized metrics with labels')
 	} catch (error) {
 		console.error('[Prometheus] Error initializing metrics with labels:', error)

src/Utils/retry-utils.ts

@@ -6,15 +6,21 @@
  * - Jitter to avoid thundering herd
  * - Configurable max attempts
  * - Customizable retry predicates
- * - Circuit breaker integration
- * - Event hooks
- * - Cancellation support
+ * - Cancellation support (AbortSignal)
+ *
+ * Used by `decode-wa-message.ts` for Bad MAC retry recovery on the decrypt
+ * path. For socket-operation retries (uploadPreKeys etc.) prefer
+ * `withBoundedRetry` from `./bounded-retry.ts`, which is empirically
+ * aligned with WhatsApp Android's per-operation backoff.
+ *
+ * NOTE: the previous `circuitBreaker` integration option, the `withRetry`
+ * decorator, the `retryable` wrapper and the `RetryManager` class were
+ * removed when the socket-level circuit breaker was retired — see PR #393
+ * for the rationale.
  *
  * @module Utils/retry-utils
  */
 
-import { EventEmitter } from 'events'
-import type { CircuitBreaker } from './circuit-breaker.js'
 import { metrics } from './prometheus-metrics.js'
 
 /**
@@ -59,8 +65,6 @@ export interface RetryOptions {
 	operationName?: string
 	/** Collect metrics */
 	collectMetrics?: boolean
-	/** Circuit breaker for integration */
-	circuitBreaker?: CircuitBreaker
 	/** Callback before each retry */
 	onRetry?: (error: Error, attempt: number, delay: number) => void | Promise<void>
 	/** Callback on success */
@@ -283,7 +287,6 @@ export async function retry<T>(
 		timeout: options.timeout,
 		operationName: options.operationName ?? 'operation',
 		collectMetrics: options.collectMetrics ?? true,
-		circuitBreaker: options.circuitBreaker,
 		onRetry: options.onRetry ?? (() => {}),
 		onSuccess: options.onSuccess ?? (() => {}),
 		onFailure: options.onFailure ?? (() => {}),
@@ -313,11 +316,6 @@ export async function retry<T>(
 			throw new RetryAbortedError(attempt)
 		}
 
-		// Check circuit breaker
-		if (config.circuitBreaker?.isOpen()) {
-			throw new Error(`Circuit breaker "${config.circuitBreaker.getName()}" is open`)
-		}
-
 		try {
 			// Execute operation
 			let result: T
@@ -425,138 +423,6 @@ export function createRetrier(defaultOptions: RetryOptions = {}) {
 	}
 }
 
-/**
- * Decorator to add retry to method
- */
-export function withRetry(options: RetryOptions = {}) {
-	return function (
-		_target: unknown,
-		propertyKey: string,
-		descriptor: TypedPropertyDescriptor<(...args: unknown[]) => unknown>
-	) {
-		const originalMethod = descriptor.value
-		if (!originalMethod) return descriptor
-
-		descriptor.value = async function (...args: unknown[]): Promise<unknown> {
-			return retry(() => originalMethod.apply(this, args), {
-				...options,
-				operationName: options.operationName || propertyKey
-			})
-		}
-
-		return descriptor
-	}
-}
-
-/**
- * Wrapper for function with retry
- */
-// eslint-disable-next-line space-before-function-paren
-export function retryable<T extends (...args: unknown[]) => unknown>(
-	fn: T,
-	options: RetryOptions = {}
-): (...args: Parameters<T>) => Promise<ReturnType<T>> {
-	return async (...args: Parameters<T>): Promise<ReturnType<T>> => {
-		return retry(() => fn(...args), options) as Promise<ReturnType<T>>
-	}
-}
-
-/**
- * Class to manage retries with state
- */
-export class RetryManager extends EventEmitter {
-	private activeRetries: Map<string, { cancel: () => void; context: RetryContext }> = new Map()
-	private defaultOptions: RetryOptions
-
-	constructor(defaultOptions: RetryOptions = {}) {
-		super()
-		this.defaultOptions = defaultOptions
-	}
-
-	/**
-	 * Execute operation with retry
-	 */
-	async execute<T>(
-		id: string,
-		operation: (context: RetryContext) => T | Promise<T>,
-		options?: RetryOptions
-	): Promise<T> {
-		// Cancel previous retry with same ID
-		this.cancel(id)
-
-		const abortController = new AbortController()
-		const mergedOptions = { ...this.defaultOptions, ...options, abortSignal: abortController.signal }
-
-		const retryPromise = retry(context => {
-			this.activeRetries.set(id, {
-				cancel: () => abortController.abort(),
-				context
-			})
-			this.emit('attempt', { id, attempt: context.attempt })
-			return operation(context)
-		}, mergedOptions)
-
-		try {
-			const result = await retryPromise
-			this.emit('success', { id })
-			return result
-		} catch (error) {
-			this.emit('failure', { id, error })
-			throw error
-		} finally {
-			this.activeRetries.delete(id)
-		}
-	}
-
-	/**
-	 * Cancel in-progress retry
-	 */
-	cancel(id: string): boolean {
-		const active = this.activeRetries.get(id)
-		if (active) {
-			active.cancel()
-			this.activeRetries.delete(id)
-			this.emit('cancelled', { id })
-			return true
-		}
-
-		return false
-	}
-
-	/**
-	 * Cancel all retries
-	 */
-	cancelAll(): void {
-		for (const [id, active] of this.activeRetries) {
-			active.cancel()
-			this.emit('cancelled', { id })
-		}
-
-		this.activeRetries.clear()
-	}
-
-	/**
-	 * Check if there is an active retry
-	 */
-	isActive(id: string): boolean {
-		return this.activeRetries.has(id)
-	}
-
-	/**
-	 * Return active retry context
-	 */
-	getContext(id: string): RetryContext | undefined {
-		return this.activeRetries.get(id)?.context
-	}
-
-	/**
-	 * Return active retry IDs
-	 */
-	getActiveIds(): string[] {
-		return Array.from(this.activeRetries.keys())
-	}
-}
-
 /**
  * Common predicates for shouldRetry
  */
@@ -660,9 +526,8 @@ export const retryConfigs = {
 	 * Uses fixed delay array: 1s, 2s, 5s, 10s, 20s (with ±15% jitter)
 	 *
 	 * NOTE: Values are hardcoded instead of referencing RETRY_BACKOFF_DELAYS/RETRY_JITTER_FACTOR
-	 * to prevent "Cannot access before initialization" errors in ESM environments.
-	 * This occurs when modules are loaded in specific orders due to indirect circular imports
-	 * (e.g., via prometheus-metrics.ts -> circuit-breaker.ts chain).
+	 * to prevent "Cannot access before initialization" errors in ESM environments
+	 * caused by indirect circular imports through prometheus-metrics.ts.
 	 * Keep these values in sync with the constants above (lines 25, 31).
 	 */
 	rsocket: {

src/Utils/structured-logger.ts

@@ -6,7 +6,9 @@
  * - Configurable log levels (trace, debug, info, warn, error, fatal)
  * - JSON formatting for log analysis
  * - Hierarchical context with child loggers
- * - External system integration via hooks with circuit breaker
+ * - External system integration via fire-and-forget hooks (failures
+ *   counted in hookFailures metric; no per-hook circuit breaker — chronic
+ *   breakage is visible to operators via the metric)
  * - Logging metrics with Prometheus integration
  * - Sensitive data sanitization
  * - Log buffering for batch writes
@@ -73,10 +75,6 @@ export interface StructuredLoggerConfig {
 	maxLogsPerSecond?: number
 	/** Enable async logging queue (default: false) */
 	enableAsyncQueue?: boolean
-	/** Circuit breaker failure threshold for external hooks (default: 5) */
-	circuitBreakerThreshold?: number
-	/** Circuit breaker reset timeout in ms (default: 30000) */
-	circuitBreakerResetMs?: number
 	/** Enable Prometheus metrics integration (default: false) */
 	enableMetrics?: boolean
 }
@@ -155,8 +153,6 @@ export interface LoggerMetrics {
 	bufferFlushes: number
 	/** External hook failures */
 	hookFailures: number
-	/** Circuit breaker trips */
-	circuitBreakerTrips: number
 	/** Average log processing time in ms */
 	avgProcessingTimeMs: number
 }
@@ -169,8 +165,6 @@ export interface LoggerStatistics extends LoggerMetrics {
 	bufferSize: number
 	/** Rate limiter tokens available */
 	rateLimiterTokens: number
-	/** Circuit breaker state */
-	circuitBreakerState: 'closed' | 'open' | 'half-open'
 	/** Queue size (if async enabled) */
 	queueSize: number
 	/** Created timestamp */
@@ -242,69 +236,6 @@ class RateLimiter {
 	}
 }
 
-// ============================================================================
-// CIRCUIT BREAKER
-// ============================================================================
-
-/**
- * Circuit breaker for external hook protection
- */
-class CircuitBreaker {
-	private failures = 0
-	private lastFailure = 0
-	private state: 'closed' | 'open' | 'half-open' = 'closed'
-
-	constructor(
-		private readonly threshold: number,
-		private readonly resetTimeoutMs: number
-	) {}
-
-	async execute<T>(operation: () => Promise<T>): Promise<T | null> {
-		if (this.state === 'open') {
-			if (Date.now() - this.lastFailure > this.resetTimeoutMs) {
-				this.state = 'half-open'
-			} else {
-				return null // Circuit is open, skip
-			}
-		}
-
-		try {
-			const result = await operation()
-			this.onSuccess()
-			return result
-		} catch (error) {
-			this.onFailure()
-			throw error
-		}
-	}
-
-	private onSuccess(): void {
-		this.failures = 0
-		this.state = 'closed'
-	}
-
-	private onFailure(): void {
-		this.failures++
-		this.lastFailure = Date.now()
-		if (this.failures >= this.threshold) {
-			this.state = 'open'
-		}
-	}
-
-	getState(): 'closed' | 'open' | 'half-open' {
-		// Check if should transition from open to half-open
-		if (this.state === 'open' && Date.now() - this.lastFailure > this.resetTimeoutMs) {
-			this.state = 'half-open'
-		}
-
-		return this.state
-	}
-
-	getFailures(): number {
-		return this.failures
-	}
-}
-
 // ============================================================================
 // ASYNC LOG QUEUE
 // ============================================================================
@@ -403,8 +334,6 @@ export class StructuredLogger implements ILogger {
 	private rateLimiter: RateLimiter | null = null
 
 	// Circuit breaker for external hook
-	private circuitBreaker: CircuitBreaker | null = null
-
 	// Async queue
 	private asyncQueue: AsyncLogQueue | null = null
 
@@ -435,8 +364,6 @@ export class StructuredLogger implements ILogger {
 			enableRateLimiting: config.enableRateLimiting ?? envConfig.enableRateLimiting ?? false,
 			maxLogsPerSecond: config.maxLogsPerSecond ?? envConfig.maxLogsPerSecond ?? 1000,
 			enableAsyncQueue: config.enableAsyncQueue ?? envConfig.enableAsyncQueue ?? false,
-			circuitBreakerThreshold: config.circuitBreakerThreshold ?? 5,
-			circuitBreakerResetMs: config.circuitBreakerResetMs ?? 30000,
 			enableMetrics: config.enableMetrics ?? envConfig.enableMetrics ?? false
 		}
 
@@ -455,7 +382,6 @@ export class StructuredLogger implements ILogger {
 			droppedLogs: 0,
 			bufferFlushes: 0,
 			hookFailures: 0,
-			circuitBreakerTrips: 0,
 			avgProcessingTimeMs: 0
 		}
 
@@ -464,11 +390,6 @@ export class StructuredLogger implements ILogger {
 			this.rateLimiter = new RateLimiter(this.config.maxLogsPerSecond)
 		}
 
-		// Initialize circuit breaker for external hook
-		if (this.config.externalHook) {
-			this.circuitBreaker = new CircuitBreaker(this.config.circuitBreakerThreshold, this.config.circuitBreakerResetMs)
-		}
-
 		// Initialize async queue
 		if (this.config.enableAsyncQueue) {
 			this.asyncQueue = new AsyncLogQueue()
@@ -569,19 +490,27 @@ export class StructuredLogger implements ILogger {
 				this.output(entry)
 			}
 
-			// External hook with circuit breaker
+			// External hook — fire-and-forget. Failures are counted in
+			// hookFailures metric so chronic breakage is visible to operators
+			// who can fix their hook. We do not gate further calls — that is
+			// the consumer's responsibility (consistent with the rest of the
+			// codebase after the circuit-breaker removal).
 			const externalHook = this.config.externalHook
-			if (externalHook && this.circuitBreaker) {
-				this.circuitBreaker
-					.execute(async () => {
-						await Promise.resolve(externalHook(entry))
-					})
-					.catch(() => {
-						this.metrics.hookFailures++
-						if (this.circuitBreaker?.getState() === 'open') {
-							this.metrics.circuitBreakerTrips++
-						}
-					})
+			if (externalHook) {
+				try {
+					const hookResult = externalHook(entry) as unknown
+					// Use Promise.resolve().catch() so any thenable (including
+					// Promise-likes that only implement .then) is handled
+					// safely — calling .catch directly would throw on such
+					// objects (Copilot review on PR #393).
+					if (hookResult && typeof (hookResult as PromiseLike<unknown>).then === 'function') {
+						void Promise.resolve(hookResult as PromiseLike<unknown>).catch(() => {
+							this.metrics.hookFailures++
+						})
+					}
+				} catch {
+					this.metrics.hookFailures++
+				}
 			}
 
 			// Track processing time
@@ -834,7 +763,6 @@ export class StructuredLogger implements ILogger {
 			...this.metrics,
 			bufferSize: this.buffer.length,
 			rateLimiterTokens: this.rateLimiter?.getTokens() ?? 0,
-			circuitBreakerState: this.circuitBreaker?.getState() ?? 'closed',
 			queueSize: this.asyncQueue?.getSize() ?? 0,
 			createdAt: this.createdAt,
 			uptimeMs: Date.now() - this.createdAt
@@ -860,7 +788,6 @@ export class StructuredLogger implements ILogger {
 			droppedLogs: 0,
 			bufferFlushes: 0,
 			hookFailures: 0,
-			circuitBreakerTrips: 0,
 			avgProcessingTimeMs: 0
 		}
 		this.totalProcessingTime = 0
@@ -893,7 +820,6 @@ export class StructuredLogger implements ILogger {
 
 		// Clear references
 		this.rateLimiter = null
-		this.circuitBreaker = null
 		this.metricsModule = null
 	}
 }

src/Utils/unified-session.ts

@@ -15,7 +15,6 @@
  */
 
 import type { BinaryNode } from '../WABinary/types.js'
-import { CircuitBreaker, CircuitOpenError, createConnectionCircuitBreaker } from './circuit-breaker.js'
 import type { ILogger } from './logger.js'
 import { metrics } from './prometheus-metrics.js'
 
@@ -58,8 +57,6 @@ export interface UnifiedSessionOptions {
 	enabled?: boolean
 	/** Logger instance for debugging */
 	logger?: ILogger
-	/** Enable circuit breaker protection for send operations */
-	enableCircuitBreaker?: boolean
 	/** Function to send binary nodes to WhatsApp */
 	sendNode?: (node: BinaryNode) => Promise<void>
 }
@@ -93,7 +90,9 @@ export type UnifiedSessionTrigger = 'login' | 'pairing' | 'presence' | 'manual'
  * Manages the unified_session telemetry feature with:
  * - Server time synchronization
  * - Rate limiting (prevents spam)
- * - Circuit breaker protection
+ * - Best-effort send semantics (failures are swallowed and counted —
+ *   telemetry is non-critical so a single attempt with the underlying
+ *   sendNode timeout is enough; no retry, no circuit breaker)
  * - Prometheus metrics integration
  * - Structured logging
  *
@@ -121,7 +120,6 @@ export class UnifiedSessionManager {
 	}
 
 	private readonly options: Required<UnifiedSessionOptions>
-	private circuitBreaker: CircuitBreaker | null = null
 
 	/** Minimum interval between unified_session sends (1 minute) */
 	private static readonly MIN_SEND_INTERVAL_MS = TimeMs.Minute
@@ -130,7 +128,6 @@ export class UnifiedSessionManager {
 		this.options = {
 			enabled: options.enabled ?? true,
 			logger: options.logger ?? (console as unknown as ILogger),
-			enableCircuitBreaker: options.enableCircuitBreaker ?? true,
 			sendNode:
 				options.sendNode ??
 				(async () => {
@@ -138,25 +135,6 @@ export class UnifiedSessionManager {
 				})
 		}
 
-		// Initialize circuit breaker if enabled
-		if (this.options.enableCircuitBreaker) {
-			this.circuitBreaker = createConnectionCircuitBreaker({
-				name: 'unified-session',
-				failureThreshold: 3,
-				failureWindow: 60000,
-				resetTimeout: 30000,
-				successThreshold: 1,
-				timeout: 10000,
-				onStateChange: (from, to) => {
-					this.options.logger.debug?.({ from, to }, 'Unified session circuit breaker state changed')
-				},
-				onOpen: () => {
-					this.options.logger.warn?.('Unified session circuit breaker OPENED')
-					metrics.circuitBreakerTrips?.inc({ name: 'unified-session' })
-				}
-			})
-		}
-
 		this.state.isInitialized = true
 		this.options.logger.debug?.('UnifiedSessionManager initialized')
 	}
@@ -292,20 +270,12 @@ export class UnifiedSessionManager {
 		}
 
 		try {
-			// Execute with circuit breaker if available
-			if (this.circuitBreaker) {
-				await this.circuitBreaker.execute(sendOperation)
-			} else {
-				await sendOperation()
-			}
+			// Telemetry is non-critical: a single send attempt with the
+			// underlying sendNode timeout is enough. No retry/circuit needed.
+			await sendOperation()
 		} catch (error) {
 			const errorMessage = error instanceof Error ? error.message : String(error)
-
-			if (error instanceof CircuitOpenError) {
-				this.options.logger.warn?.({ trigger, circuitState: error.state }, 'Unified session blocked by circuit breaker')
-			} else {
-				this.options.logger.warn?.({ trigger, error: errorMessage }, 'Failed to send unified session telemetry')
-			}
+			this.options.logger.warn?.({ trigger, error: errorMessage }, 'Failed to send unified session telemetry')
 
 			// Record failure metric
 			metrics.errors?.inc({ category: 'unified_session', code: 'send_failed' })
@@ -331,7 +301,6 @@ export class UnifiedSessionManager {
 			sendCount: 0,
 			isInitialized: true
 		}
-		this.circuitBreaker?.reset()
 		this.options.logger.debug?.('UnifiedSessionManager reset')
 	}
 
@@ -339,7 +308,6 @@ export class UnifiedSessionManager {
 	 * Destroy the manager and clean up resources.
 	 */
 	destroy(): void {
-		this.circuitBreaker?.destroy()
 		this.state.isInitialized = false
 		this.options.logger.debug?.('UnifiedSessionManager destroyed')
 	}

src/__tests__/Utils/bounded-retry.test.ts

@@ -0,0 +1,391 @@
+import { jest } from '@jest/globals'
+import {
+	withBoundedRetry,
+	BoundedRetryGiveUpError,
+	BoundedRetryAbortedError,
+	WHATSAPP_BACKOFF_DELAYS,
+	DEFAULT_TTL_MS
+} from '../../Utils/bounded-retry'
+
+// All tests use very short real delays (1-50ms) so they run fast.
+// This sidesteps the fragility of jest fake timers + async chains.
+
+describe('bounded-retry — WhatsApp-aligned per-operation retry', () => {
+	test('default delay sequence matches WhatsApp Android empirical capture', () => {
+		// Frida trace: 3000 -> 10000 -> 60000 -> ~64000 -> 120000 ms
+		expect(WHATSAPP_BACKOFF_DELAYS).toEqual([3000, 10000, 60000, 60000, 120000])
+	})
+
+	test('default TTL is 10 minutes (matches empirical stabilisation + safety buffer)', () => {
+		expect(DEFAULT_TTL_MS).toBe(600_000)
+	})
+
+	test('first-attempt success returns immediately, no retries', async () => {
+		const op = jest.fn<() => Promise<string>>().mockResolvedValue('ok')
+		const result = await withBoundedRetry(op, { name: 'fast-success' })
+		expect(result).toBe('ok')
+		expect(op).toHaveBeenCalledTimes(1)
+	})
+
+	test('retries with the configured delay sequence after failures', async () => {
+		const op = jest
+			.fn<() => Promise<number>>()
+			.mockRejectedValueOnce(new Error('try1'))
+			.mockRejectedValueOnce(new Error('try2'))
+			.mockResolvedValueOnce(42)
+
+		const onRetry = jest.fn<(err: Error, attempt: number, delayMs: number) => void>()
+		const result = await withBoundedRetry(op, {
+			name: 'three-tries',
+			delays: [10, 20, 40],
+			jitter: 0,
+			onRetry
+		})
+
+		expect(result).toBe(42)
+		expect(op).toHaveBeenCalledTimes(3)
+		expect(onRetry).toHaveBeenCalledTimes(2)
+		expect(onRetry.mock.calls[0]![2]).toBe(10)
+		expect(onRetry.mock.calls[1]![2]).toBe(20)
+	})
+
+	test('TTL exhaustion throws BoundedRetryGiveUpError (no infinite retries)', async () => {
+		const op = jest.fn<() => Promise<string>>().mockRejectedValue(new Error('always-fails'))
+
+		await expect(
+			withBoundedRetry(op, {
+				name: 'ttl-test',
+				delays: [10],
+				jitter: 0,
+				ttlMs: 50
+			})
+		).rejects.toBeInstanceOf(BoundedRetryGiveUpError)
+
+		// Should have tried multiple times before giving up
+		expect(op.mock.calls.length).toBeGreaterThanOrEqual(2)
+	})
+
+	test('per-attempt timeout fires for slow operations', async () => {
+		const slowOp = jest.fn<() => Promise<string>>(() => new Promise(() => {})) // hangs
+
+		await expect(
+			withBoundedRetry(slowOp, {
+				name: 'slow',
+				delays: [5],
+				jitter: 0,
+				ttlMs: 200,
+				perAttemptTimeoutMs: 30
+			})
+		).rejects.toBeInstanceOf(BoundedRetryGiveUpError)
+
+		// Multiple attempts should have been tried (each timed out at 30ms)
+		expect(slowOp.mock.calls.length).toBeGreaterThan(1)
+	})
+
+	test('shouldRetry predicate can short-circuit retries', async () => {
+		const op = jest
+			.fn<() => Promise<string>>()
+			.mockRejectedValueOnce(new Error('transient'))
+			.mockRejectedValueOnce(new Error('FATAL: do not retry'))
+
+		await expect(
+			withBoundedRetry(op, {
+				name: 'predicate-test',
+				delays: [1],
+				jitter: 0,
+				shouldRetry: err => !err.message.startsWith('FATAL')
+			})
+		).rejects.toThrow('FATAL: do not retry')
+
+		expect(op).toHaveBeenCalledTimes(2)
+	})
+
+	test('AbortSignal cancels in-flight retry sleep', async () => {
+		const op = jest.fn<() => Promise<string>>().mockRejectedValue(new Error('fail'))
+		const ctrl = new AbortController()
+
+		const promise = withBoundedRetry(op, {
+			name: 'abort-test',
+			delays: [10_000], // long sleep so abort can happen during it
+			jitter: 0,
+			signal: ctrl.signal
+		})
+
+		// First attempt fires immediately; sleep starts
+		await new Promise(resolve => setTimeout(resolve, 10))
+		ctrl.abort()
+
+		await expect(promise).rejects.toBeInstanceOf(BoundedRetryAbortedError)
+	})
+
+	test('two concurrent retry chains do NOT interfere with each other (per-op isolation)', async () => {
+		// Key advantage over circuit breaker: failures in one chain do not
+		// block another chain. With circuit breaker, 5 failures from chain A
+		// would block chain B too. With bounded-retry, B is independent.
+
+		const flakyOp = jest
+			.fn<() => Promise<string>>()
+			.mockRejectedValueOnce(new Error('a1'))
+			.mockResolvedValueOnce('a-ok')
+
+		const fastOp = jest.fn<() => Promise<string>>().mockResolvedValue('b-ok')
+
+		// Start a (will retry once after 30ms) and b (immediate)
+		const aPromise = withBoundedRetry(flakyOp, { name: 'a', delays: [30], jitter: 0 })
+		const bPromise = withBoundedRetry(fastOp, { name: 'b', delays: [30], jitter: 0 })
+
+		// b should resolve right away
+		const bResult = await bPromise
+		expect(bResult).toBe('b-ok')
+		expect(fastOp).toHaveBeenCalledTimes(1)
+
+		// a still running — wait for it
+		const aResult = await aPromise
+		expect(aResult).toBe('a-ok')
+		expect(flakyOp).toHaveBeenCalledTimes(2)
+	})
+
+	test('delay sequence caps at last value (no unbounded growth)', async () => {
+		const op = jest.fn<() => Promise<string>>().mockRejectedValue(new Error('fail'))
+		const onRetry = jest.fn<(err: Error, attempt: number, delayMs: number) => void>()
+
+		await expect(
+			withBoundedRetry(op, {
+				name: 'cap-test',
+				delays: [5, 10, 15], // 3 distinct steps; 4th+ retry should reuse 15
+				jitter: 0,
+				ttlMs: 300,
+				onRetry
+			})
+		).rejects.toBeInstanceOf(BoundedRetryGiveUpError)
+
+		// Find delays >= 4th retry — all should be capped at 15
+		const lateRetries = onRetry.mock.calls.filter(c => c[1] >= 4)
+		expect(lateRetries.length).toBeGreaterThan(0)
+		for (const call of lateRetries) {
+			expect(call[2]).toBeLessThanOrEqual(15)
+		}
+	})
+
+	test('jitter randomises delay within +/- factor', async () => {
+		const op = jest.fn<() => Promise<string>>().mockRejectedValue(new Error('fail'))
+		const observed: number[] = []
+
+		// Single run, capture multiple onRetry calls (each with different jitter)
+		await withBoundedRetry(op, {
+			name: 'jitter-test',
+			delays: [50], // base delay
+			jitter: 0.3, // +/- 30% -> [35, 65]
+			ttlMs: 1000, // ttl high enough for ~10 retries
+			perAttemptTimeoutMs: 5,
+			onRetry: (_, __, d) => observed.push(d)
+		}).catch(() => {})
+
+		expect(observed.length).toBeGreaterThanOrEqual(3)
+		const unique = new Set(observed)
+		expect(unique.size).toBeGreaterThan(1)
+		// Every observed delay must be within jitter window — but the LAST one
+		// may be capped by remaining ttl budget, so check all but the last
+		const inWindow = observed.slice(0, -1)
+		for (const d of inWindow) {
+			expect(d).toBeGreaterThanOrEqual(35)
+			expect(d).toBeLessThanOrEqual(65)
+		}
+	})
+
+	test('delay does not exceed remaining TTL budget', async () => {
+		const op = jest.fn<() => Promise<string>>().mockRejectedValue(new Error('fail'))
+		const onRetry = jest.fn<(err: Error, attempt: number, delayMs: number) => void>()
+
+		await withBoundedRetry(op, {
+			name: 'budget-cap',
+			delays: [200], // would exceed ttl
+			jitter: 0,
+			ttlMs: 50, // very short
+			onRetry
+		}).catch(() => {})
+
+		// All scheduled delays should be capped to remaining budget
+		for (const call of onRetry.mock.calls) {
+			expect(call[2]).toBeLessThanOrEqual(50)
+		}
+	})
+
+	// ─── Tests for fixes from PR #393 review ──────────────────────────────
+
+	test('TTL is enforced BEFORE the next attempt (not just after)', async () => {
+		// Regression: previously TTL was only checked after a failure, so a
+		// new attempt could start with 0ms remaining and run for the full
+		// per-attempt timeout, overshooting wall-clock budget by that much.
+		const slowOp = jest.fn<() => Promise<string>>(() => new Promise(resolve => setTimeout(() => resolve('late'), 200)))
+
+		const start = Date.now()
+		await expect(
+			withBoundedRetry(slowOp, {
+				name: 'ttl-strict',
+				delays: [10],
+				jitter: 0,
+				ttlMs: 50,
+				perAttemptTimeoutMs: 1000 // huge, must be capped by TTL
+			})
+		).rejects.toBeInstanceOf(BoundedRetryGiveUpError)
+
+		const elapsed = Date.now() - start
+		// With strict TTL enforcement, total runtime should be very close to ttlMs.
+		// Allow up to 100ms slack for Node timer + microtask scheduling.
+		expect(elapsed).toBeLessThan(150)
+	})
+
+	test('per-attempt timeout is capped by remaining TTL budget', async () => {
+		// If user passes perAttemptTimeoutMs > ttlMs, the cap should apply.
+		const slowOp = jest.fn<() => Promise<string>>(() => new Promise(() => {})) // hangs forever
+
+		const start = Date.now()
+		await expect(
+			withBoundedRetry(slowOp, {
+				name: 'attempt-cap',
+				delays: [1],
+				jitter: 0,
+				ttlMs: 30,
+				perAttemptTimeoutMs: 5000 // way bigger than ttl
+			})
+		).rejects.toBeInstanceOf(BoundedRetryGiveUpError)
+
+		const elapsed = Date.now() - start
+		expect(elapsed).toBeLessThan(150) // not 5000ms!
+	})
+
+	test('operation receives an AbortSignal that is aborted on per-attempt timeout', async () => {
+		// New API: operation can opt into cancellation via the signal arg.
+		const aborts: boolean[] = []
+		const opThatRespectsSignal = jest.fn(
+			(signal?: AbortSignal) =>
+				new Promise<string>((resolve, reject) => {
+					const timer = setTimeout(() => resolve('late'), 100)
+					signal?.addEventListener('abort', () => {
+						clearTimeout(timer)
+						aborts.push(true)
+						reject(new Error('aborted by signal'))
+					})
+				})
+		)
+
+		await withBoundedRetry(opThatRespectsSignal, {
+			name: 'signal-aware',
+			delays: [1],
+			jitter: 0,
+			ttlMs: 80,
+			perAttemptTimeoutMs: 20 // smaller than 100ms op duration
+		}).catch(() => {})
+
+		expect(aborts.length).toBeGreaterThan(0) // at least one attempt was aborted
+	})
+
+	test('sleep listener is removed when timeout completes normally (no leak)', async () => {
+		// Regression: sleep used signal.addEventListener with { once: true } but
+		// never removed the listener on timer-resolve. With many retries on a
+		// shared signal, listeners would accumulate.
+		//
+		// Make the failure mode observable by spying on add/removeEventListener
+		// directly so the assertion fails if cleanup is removed.
+		const ctrl = new AbortController()
+		const realAdd = ctrl.signal.addEventListener.bind(ctrl.signal)
+		const realRemove = ctrl.signal.removeEventListener.bind(ctrl.signal)
+		let addCount = 0
+		let removeCount = 0
+		ctrl.signal.addEventListener = ((type: string, listener: never, opts?: never) => {
+			if (type === 'abort') addCount++
+			return realAdd(type, listener, opts)
+		}) as typeof ctrl.signal.addEventListener
+		ctrl.signal.removeEventListener = ((type: string, listener: never, opts?: never) => {
+			if (type === 'abort') removeCount++
+			return realRemove(type, listener, opts)
+		}) as typeof ctrl.signal.removeEventListener
+
+		const op = jest
+			.fn<() => Promise<string>>()
+			.mockRejectedValueOnce(new Error('1'))
+			.mockRejectedValueOnce(new Error('2'))
+			.mockRejectedValueOnce(new Error('3'))
+			.mockResolvedValueOnce('ok')
+
+		const result = await withBoundedRetry(op, {
+			name: 'no-leak',
+			delays: [5],
+			jitter: 0,
+			ttlMs: 500,
+			signal: ctrl.signal
+		})
+
+		expect(result).toBe('ok')
+		expect(op).toHaveBeenCalledTimes(4)
+		// If sleep() leaks listeners, addCount > removeCount. Each retry adds
+		// listeners (1 sleep + 1 outer-abort forwarding) and must remove the
+		// same number on settlement.
+		expect(addCount).toBeGreaterThan(0)
+		expect(addCount).toBe(removeCount)
+	})
+
+	test('logger receives structured logs for retry, give-up, and recovery', async () => {
+		const debugCalls: unknown[][] = []
+		const infoCalls: unknown[][] = []
+		const warnCalls: unknown[][] = []
+		const fakeLogger = {
+			debug: (...args: unknown[]) => debugCalls.push(args),
+			info: (...args: unknown[]) => infoCalls.push(args),
+			warn: (...args: unknown[]) => warnCalls.push(args),
+			error: () => {},
+			fatal: () => {},
+			trace: () => {},
+			child: () => fakeLogger,
+			level: 'debug'
+		}
+
+		const op = jest
+			.fn<() => Promise<string>>()
+			.mockRejectedValueOnce(new Error('boom'))
+			.mockResolvedValueOnce('ok')
+
+		const result = await withBoundedRetry(op, {
+			name: 'logged-op',
+			delays: [5],
+			jitter: 0,
+			ttlMs: 200,
+			logger: fakeLogger as never
+		})
+
+		expect(result).toBe('ok')
+		// Should have logged the scheduling at debug level
+		expect(debugCalls.length).toBeGreaterThanOrEqual(1)
+		// Should have logged the recovery at info level
+		expect(infoCalls.length).toBeGreaterThanOrEqual(1)
+	})
+
+	test('logger reports give-up via warn when TTL exceeded', async () => {
+		const warnCalls: unknown[][] = []
+		const fakeLogger = {
+			debug: () => {},
+			info: () => {},
+			warn: (...args: unknown[]) => warnCalls.push(args),
+			error: () => {},
+			fatal: () => {},
+			trace: () => {},
+			child: () => fakeLogger,
+			level: 'debug'
+		}
+
+		const op = jest.fn<() => Promise<string>>().mockRejectedValue(new Error('always-fails'))
+
+		await expect(
+			withBoundedRetry(op, {
+				name: 'logged-fail',
+				delays: [5],
+				jitter: 0,
+				ttlMs: 30,
+				logger: fakeLogger as never
+			})
+		).rejects.toBeInstanceOf(BoundedRetryGiveUpError)
+
+		expect(warnCalls.length).toBeGreaterThanOrEqual(1)
+	})
+})

src/__tests__/Utils/circuit-breaker.test.ts

@@ -1,371 +0,0 @@
-/**
- * Testes unitários para circuit-breaker.ts
- */
-
-import { afterEach, beforeEach, describe, expect, it } from '@jest/globals'
-import {
-	CircuitBreaker,
-	CircuitBreakerRegistry,
-	CircuitOpenError,
-	type CircuitState,
-	CircuitTimeoutError,
-	createCircuitBreaker,
-	getCircuitHealth,
-	globalCircuitRegistry,
-	withCircuitBreaker
-} from '../../Utils/circuit-breaker.js'
-
-describe('CircuitBreaker', () => {
-	let breaker: CircuitBreaker
-
-	beforeEach(() => {
-		breaker = createCircuitBreaker({
-			name: 'test',
-			failureThreshold: 3,
-			successThreshold: 2,
-			resetTimeout: 100,
-			timeout: 1000,
-			volumeThreshold: 3,
-			collectMetrics: false
-		})
-	})
-
-	afterEach(() => {
-		breaker.destroy()
-	})
-
-	describe('initial state', () => {
-		it('should start in closed state', () => {
-			expect(breaker.getState()).toBe('closed')
-			expect(breaker.isClosed()).toBe(true)
-			expect(breaker.isOpen()).toBe(false)
-			expect(breaker.isHalfOpen()).toBe(false)
-		})
-	})
-
-	describe('successful operations', () => {
-		it('should execute successful operations', async () => {
-			const result = await breaker.execute(() => 'success')
-			expect(result).toBe('success')
-		})
-
-		it('should execute async operations', async () => {
-			const result = await breaker.execute(async () => {
-				await new Promise(resolve => setTimeout(resolve, 10))
-				return 'async success'
-			})
-			expect(result).toBe('async success')
-		})
-
-		it('should track successful calls in stats', async () => {
-			await breaker.execute(() => 'success')
-			await breaker.execute(() => 'success')
-
-			const stats = breaker.getStats()
-			expect(stats.totalCalls).toBe(2)
-			expect(stats.totalSuccesses).toBe(2)
-			expect(stats.totalFailures).toBe(0)
-		})
-	})
-
-	describe('failure handling', () => {
-		it('should open after reaching failure threshold', async () => {
-			const failingOp = () => {
-				throw new Error('Failure')
-			}
-
-			// Reach failure threshold
-			for (let i = 0; i < 3; i++) {
-				await expect(breaker.execute(failingOp)).rejects.toThrow('Failure')
-			}
-
-			expect(breaker.isOpen()).toBe(true)
-		})
-
-		it('should not open before reaching threshold', async () => {
-			const failingOp = () => {
-				throw new Error('Failure')
-			}
-
-			// Below threshold
-			for (let i = 0; i < 2; i++) {
-				await expect(breaker.execute(failingOp)).rejects.toThrow('Failure')
-			}
-
-			expect(breaker.isClosed()).toBe(true)
-		})
-
-		it('should reset failure count on success', async () => {
-			const failingOp = () => {
-				throw new Error('Failure')
-			}
-
-			await expect(breaker.execute(failingOp)).rejects.toThrow()
-			await expect(breaker.execute(failingOp)).rejects.toThrow()
-
-			// Success resets failures
-			await breaker.execute(() => 'success')
-
-			// Only 1 more failure needed (already have 2 in window, need 3 total)
-			await expect(breaker.execute(failingOp)).rejects.toThrow()
-
-			// Should open now (3 failures in window)
-			expect(breaker.isOpen()).toBe(true)
-		})
-	})
-
-	describe('open state', () => {
-		beforeEach(async () => {
-			// Force open
-			breaker.trip()
-		})
-
-		it('should reject operations when open', async () => {
-			await expect(breaker.execute(() => 'success')).rejects.toThrow(CircuitOpenError)
-		})
-
-		it('should transition to half-open after reset timeout', async () => {
-			await new Promise(resolve => setTimeout(resolve, 150))
-			expect(breaker.isHalfOpen()).toBe(true)
-		})
-	})
-
-	describe('half-open state', () => {
-		beforeEach(async () => {
-			breaker.trip()
-			await new Promise(resolve => setTimeout(resolve, 150))
-		})
-
-		it('should close after success threshold', async () => {
-			expect(breaker.isHalfOpen()).toBe(true)
-
-			// Success threshold is 2
-			await breaker.execute(() => 'success')
-			await breaker.execute(() => 'success')
-
-			expect(breaker.isClosed()).toBe(true)
-		})
-
-		it('should reopen on failure', async () => {
-			expect(breaker.isHalfOpen()).toBe(true)
-
-			await expect(
-				breaker.execute(() => {
-					throw new Error('Failure')
-				})
-			).rejects.toThrow()
-
-			expect(breaker.isOpen()).toBe(true)
-		})
-	})
-
-	describe('manual controls', () => {
-		it('should force open with trip()', () => {
-			breaker.trip()
-			expect(breaker.isOpen()).toBe(true)
-		})
-
-		it('should force close with reset()', async () => {
-			breaker.trip()
-			breaker.reset()
-			expect(breaker.isClosed()).toBe(true)
-		})
-	})
-
-	describe('timeout', () => {
-		it('should timeout slow operations', async () => {
-			const slowBreaker = createCircuitBreaker({
-				name: 'slow',
-				timeout: 50,
-				collectMetrics: false
-			})
-
-			await expect(slowBreaker.execute(() => new Promise(resolve => setTimeout(resolve, 200)))).rejects.toThrow(
-				CircuitTimeoutError
-			)
-
-			slowBreaker.destroy()
-		})
-	})
-
-	describe('events', () => {
-		it('should emit state-change event', async () => {
-			const stateChanges: Array<{ from: CircuitState; to: CircuitState }> = []
-
-			breaker.on('state-change', change => {
-				stateChanges.push(change)
-			})
-
-			breaker.trip()
-			breaker.reset()
-
-			expect(stateChanges).toHaveLength(2)
-			expect(stateChanges[0]).toEqual({ from: 'closed', to: 'open' })
-			expect(stateChanges[1]).toEqual({ from: 'open', to: 'closed' })
-		})
-
-		it('should emit success and failure events', async () => {
-			let successCount = 0
-			let failureCount = 0
-
-			breaker.on('success', () => successCount++)
-			breaker.on('failure', () => failureCount++)
-
-			await breaker.execute(() => 'success')
-			await expect(
-				breaker.execute(() => {
-					throw new Error()
-				})
-			).rejects.toThrow()
-
-			expect(successCount).toBe(1)
-			expect(failureCount).toBe(1)
-		})
-	})
-
-	describe('isFailure predicate', () => {
-		it('should use custom isFailure predicate', async () => {
-			const customBreaker = createCircuitBreaker({
-				name: 'custom',
-				failureThreshold: 1,
-				volumeThreshold: 1,
-				shouldCountError: (error: any) => error.message !== 'Ignored',
-				collectMetrics: false
-			})
-
-			// This error should be ignored
-			await expect(
-				customBreaker.execute(() => {
-					throw new Error('Ignored')
-				})
-			).rejects.toThrow()
-
-			expect(customBreaker.isClosed()).toBe(true)
-
-			// This should trip the breaker
-			await expect(
-				customBreaker.execute(() => {
-					throw new Error('Real failure')
-				})
-			).rejects.toThrow()
-
-			expect(customBreaker.isOpen()).toBe(true)
-
-			customBreaker.destroy()
-		})
-	})
-
-	describe('sync operations', () => {
-		it('should execute sync operations', () => {
-			const result = breaker.executeSync(() => 'sync result')
-			expect(result).toBe('sync result')
-		})
-
-		it('should handle sync failures', () => {
-			expect(() =>
-				breaker.executeSync(() => {
-					throw new Error('Sync failure')
-				})
-			).toThrow('Sync failure')
-		})
-	})
-})
-
-describe('CircuitBreakerRegistry', () => {
-	let registry: CircuitBreakerRegistry
-
-	beforeEach(() => {
-		registry = new CircuitBreakerRegistry()
-	})
-
-	afterEach(() => {
-		registry.destroyAll()
-	})
-
-	it('should create and retrieve circuit breakers', () => {
-		const breaker1 = registry.get('test1')
-		const breaker2 = registry.get('test1')
-
-		expect(breaker1).toBe(breaker2)
-	})
-
-	it('should check if breaker exists', () => {
-		registry.get('exists')
-
-		expect(registry.has('exists')).toBe(true)
-		expect(registry.has('notexists')).toBe(false)
-	})
-
-	it('should remove breaker', () => {
-		registry.get('toRemove')
-		expect(registry.remove('toRemove')).toBe(true)
-		expect(registry.has('toRemove')).toBe(false)
-	})
-
-	it('should get all stats', () => {
-		registry.get('breaker1')
-		registry.get('breaker2')
-
-		const stats = registry.getAllStats()
-
-		expect(stats).toHaveProperty('breaker1')
-		expect(stats).toHaveProperty('breaker2')
-	})
-
-	it('should reset all breakers', async () => {
-		const breaker = registry.get('test')
-		breaker.trip()
-
-		registry.resetAll()
-
-		expect(breaker.isClosed()).toBe(true)
-	})
-})
-
-describe('withCircuitBreaker', () => {
-	it('should wrap function with circuit breaker', async () => {
-		let callCount = 0
-
-		const protectedFn = withCircuitBreaker(
-			async () => {
-				callCount++
-				return 'result'
-			},
-			{
-				name: 'wrapped-fn',
-				collectMetrics: false
-			}
-		)
-
-		const result = await protectedFn()
-
-		expect(result).toBe('result')
-		expect(callCount).toBe(1)
-	})
-})
-
-describe('getCircuitHealth', () => {
-	beforeEach(() => {
-		globalCircuitRegistry.destroyAll()
-	})
-
-	it('should report healthy when all circuits closed', () => {
-		globalCircuitRegistry.get('healthy1')
-		globalCircuitRegistry.get('healthy2')
-
-		const health = getCircuitHealth()
-
-		expect(health.healthy).toBe(true)
-		expect(health.openCircuits).toHaveLength(0)
-	})
-
-	it('should report unhealthy when circuit is open', () => {
-		const breaker = globalCircuitRegistry.get('unhealthy')
-		breaker.trip()
-
-		const health = getCircuitHealth()
-
-		expect(health.healthy).toBe(false)
-		expect(health.openCircuits).toContain('unhealthy')
-	})
-})

src/__tests__/Utils/retry-utils.test.ts

@@ -2,7 +2,7 @@
  * Testes unitários para retry-utils.ts
  */
 
-import { beforeEach, describe, expect, it, jest } from '@jest/globals'
+import { describe, expect, it, jest } from '@jest/globals'
 import {
 	calculateDelay,
 	createRetrier,
@@ -11,12 +11,10 @@ import {
 	retry,
 	RETRY_BACKOFF_DELAYS,
 	RETRY_JITTER_FACTOR,
-	retryable,
 	RetryAbortedError,
 	retryConfigs,
 	type RetryContext,
 	RetryExhaustedError,
-	RetryManager,
 	retryPredicates,
 	retryWithResult
 } from '../../Utils/retry-utils.js'
@@ -315,103 +313,6 @@ describe('createRetrier', () => {
 	})
 })
 
-describe('retryable', () => {
-	it('should wrap function with retry', async () => {
-		let attempts = 0
-
-		const fn = retryable(
-			() => {
-				attempts++
-				if (attempts < 2) throw new Error('Failing')
-				return 'wrapped result'
-			},
-			{
-				maxAttempts: 5,
-				baseDelay: 10,
-				collectMetrics: false
-			}
-		)
-
-		const result = await fn()
-
-		expect(result).toBe('wrapped result')
-		expect(attempts).toBe(2)
-	})
-})
-
-describe('RetryManager', () => {
-	let manager: RetryManager
-
-	beforeEach(() => {
-		manager = new RetryManager({ baseDelay: 200, collectMetrics: false })
-	})
-
-	it('should execute operation with id', async () => {
-		const result = await manager.execute('op1', () => 'result')
-		expect(result).toBe('result')
-	})
-
-	it('should cancel active retry', async () => {
-		let attempt = 0
-
-		const promise = manager.execute('cancelable', async () => {
-			attempt++
-			if (attempt === 1) {
-				throw new Error('Temporary failure')
-			}
-
-			await new Promise(resolve => setTimeout(resolve, 1000))
-			return 'result'
-		})
-
-		// Cancel during retry delay
-		setTimeout(() => manager.cancel('cancelable'), 100)
-
-		await expect(promise).rejects.toThrow()
-	})
-
-	it('should check if operation is active', async () => {
-		let resolve: () => void
-		const promise = manager.execute(
-			'active',
-			() =>
-				new Promise<string>(r => {
-					resolve = () => r('done')
-				})
-		)
-
-		expect(manager.isActive('active')).toBe(true)
-
-		resolve!()
-		await promise
-
-		expect(manager.isActive('active')).toBe(false)
-	})
-
-	it('should cancel all operations', async () => {
-		const promises = [
-			manager.execute('op1', () => new Promise((_, reject) => setTimeout(() => reject(new Error()), 1000))),
-			manager.execute('op2', () => new Promise((_, reject) => setTimeout(() => reject(new Error()), 1000)))
-		]
-
-		setTimeout(() => manager.cancelAll(), 20)
-
-		await expect(Promise.all(promises)).rejects.toThrow()
-	})
-
-	it('should emit events', async () => {
-		const events: string[] = []
-
-		manager.on('attempt', () => events.push('attempt'))
-		manager.on('success', () => events.push('success'))
-
-		await manager.execute('test', () => 'result')
-
-		expect(events).toContain('attempt')
-		expect(events).toContain('success')
-	})
-})
-
 describe('retryPredicates', () => {
 	describe('always', () => {
 		it('should always return true', () => {

src/__tests__/Utils/unified-session.test.ts

@@ -29,8 +29,7 @@ const TimeMs = {
 jest.mock('../../Utils/prometheus-metrics.js', () => ({
 	metrics: {
 		socketEvents: { inc: jest.fn() },
-		errors: { inc: jest.fn() },
-		circuitBreakerTrips: { inc: jest.fn() }
+		errors: { inc: jest.fn() }
 	}
 }))
 
@@ -59,7 +58,6 @@ describe('UnifiedSessionManager', () => {
 		manager = createUnifiedSessionManager({
 			enabled: true,
 			logger: mockLogger as any,
-			enableCircuitBreaker: false, // Disable for simpler testing
 			sendNode: mockSendNode
 		})
 	})
@@ -356,7 +354,11 @@ describe('shouldEnableUnifiedSession', () => {
 	})
 })
 
-describe('UnifiedSessionManager with CircuitBreaker', () => {
+describe('UnifiedSessionManager — non-critical send semantics', () => {
+	// The previous "with CircuitBreaker" suite is gone — telemetry no longer
+	// uses a circuit breaker. Verify equivalent behavior still holds:
+	// failures are swallowed (telemetry is non-critical) and don't propagate.
+
 	let manager: UnifiedSessionManager
 	let mockSendNode: MockSendNode
 
@@ -366,7 +368,6 @@ describe('UnifiedSessionManager with CircuitBreaker', () => {
 		manager = createUnifiedSessionManager({
 			enabled: true,
 			logger: createMockLogger() as any,
-			enableCircuitBreaker: true,
 			sendNode: mockSendNode
 		})
 	})
@@ -375,24 +376,13 @@ describe('UnifiedSessionManager with CircuitBreaker', () => {
 		manager.destroy()
 	})
 
-	it('should work with circuit breaker enabled', async () => {
+	it('should send telemetry once per call', async () => {
 		await manager.send('login')
 		expect(mockSendNode).toHaveBeenCalledTimes(1)
 	})
 
-	it('should handle circuit breaker failures gracefully', async () => {
-		// Make send fail multiple times to trigger circuit breaker
-		let callCount = 0
-		mockSendNode.mockImplementation(async () => {
-			callCount++
-			if (callCount <= 3) {
-				throw new Error(`Fail ${callCount}`)
-			}
-		})
-
-		// These should not throw even with failures
-		await expect(manager.send('login')).resolves.toBeUndefined()
-		await expect(manager.send('login')).resolves.toBeUndefined()
+	it('should swallow send failures gracefully (telemetry is non-critical)', async () => {
+		mockSendNode.mockRejectedValueOnce(new Error('boom'))
 		await expect(manager.send('login')).resolves.toBeUndefined()
 	})
 })

src/index.ts

@@ -28,7 +28,16 @@ console.info = function (...args: unknown[]) {
 
 // Track errors by type + JID to avoid duplicates (using Map for better performance)
 const _errorTimestamps = new Map<string, number>()
-const DEDUP_WINDOW_MS = 150
+// Dedup window for repeated decrypt-error console lines (Bad MAC / Counter / etc).
+// Was 150ms, but retry attempts of the SAME message are typically ~300-1000ms apart,
+// so the second attempt fell outside the window and double-printed.
+//
+// TRADE-OFF: dedup key is `errorType + JID` (no message-id). With 5s, a burst of
+// errors for the SAME JID — even of slightly different categories or different
+// messages — collapses to one log line every 5s. This is intentional for a noisy
+// production stream; if you need per-message visibility, set BAILEYS_LOG_LEVEL=debug
+// to bypass this console-side dedup and see the structured pino logs in full.
+const DEDUP_WINDOW_MS = 5000
 
 console.error = function (...args: unknown[]) {
 	if (args.length > 0 && typeof args[0] === 'string') {
@@ -70,7 +79,7 @@ console.error = function (...args: unknown[]) {
 				const lastTime = _errorTimestamps.get(dedupeKey)
 
 				if (lastTime && now - lastTime < DEDUP_WINDOW_MS) {
-					return // Skip duplicate within 150ms window
+					return // Skip duplicate within DEDUP_WINDOW_MS window
 				}
 
 				_errorTimestamps.set(dedupeKey, now)