fix: giảm spam 429 từ Claude OAuth usage endpoint

- claudeAutoPing: cache resetAt in-mem, bỏ qua poll usage cho tới gần reset
- ProviderLimits: throttle auto-refresh Claude 3 phút, nút bấm tay vẫn refresh ngay
- claude.js: 429 ở OAuth usage → cooldown 3 phút, fallback legacy (không ảnh hưởng chat)

Co-authored-by: Cursor <cursoragent@cursor.com>

open-sse/services/usage/claude.js

@@ -14,11 +14,19 @@ const CLAUDE_CONFIG = {
   apiVersion: ANTHROPIC_API_VERSION,
 };
 
-/**
- * Claude Usage - Primary: OAuth endpoint, Fallback: legacy settings/org endpoint
- */
+// OAuth usage endpoint rate-limits (429); cool down per-token to stop hammering it.
+// Only the quota endpoint is affected — chat with the same token still works.
+const OAUTH_429_COOLDOWN_MS = 180000;
+const oauthCooldown = new Map();
+
 export async function getClaudeUsage(accessToken, proxyOptions = null) {
   try {
+    // Skip OAuth usage call while this token is cooling down from a recent 429
+    const cooldownUntil = oauthCooldown.get(accessToken);
+    if (cooldownUntil && Date.now() < cooldownUntil) {
+      return await getClaudeUsageLegacy(accessToken, proxyOptions);
+    }
+
     // Primary: OAuth usage endpoint (Claude Code consumer OAuth tokens)
     const oauthResponse = await proxyAwareFetch(CLAUDE_CONFIG.oauthUsageUrl, {
       method: "GET",
@@ -73,6 +81,11 @@ export async function getClaudeUsage(accessToken, proxyOptions = null) {
       };
     }
 
+    // Cool down OAuth usage polling after a 429 (quota endpoint only)
+    if (oauthResponse.status === 429) {
+      oauthCooldown.set(accessToken, Date.now() + OAUTH_429_COOLDOWN_MS);
+    }
+
     // Fallback: legacy settings + org usage endpoint
     console.warn(`[Claude Usage] OAuth endpoint returned ${oauthResponse.status}, falling back to legacy`);
     return await getClaudeUsageLegacy(accessToken, proxyOptions);

src/app/(dashboard)/dashboard/usage/components/ProviderLimits/index.js

@@ -26,6 +26,7 @@ import {
   setQuotaCache,
   QUOTA_CACHE_KEY,
   REFRESH_INTERVAL_MS,
+  CLAUDE_REFRESH_INTERVAL_MS,
   DEPLETED_QUOTA_THRESHOLD,
   AUTO_REFRESH_STORAGE_KEY,
   CONNECTIONS_PAGE_SIZE,
@@ -131,6 +132,7 @@ export default function ProviderLimits() {
 
   const intervalRef = useRef(null);
   const countdownRef = useRef(null);
+  const tickCountRef = useRef(0);
 
   const fetchConnections = useCallback(
     async (targetPage = page) => {
@@ -401,12 +403,18 @@ export default function ProviderLimits() {
     };
   }, []);
 
-  const refreshAll = useCallback(async () => {
+  const refreshAll = useCallback(async (force = false) => {
     if (refreshingAll) return;
 
     setRefreshingAll(true);
     setCountdown(60);
 
+    // Throttle Claude: poll its quota every Nth auto-tick (manual force bypasses)
+    const tick = (tickCountRef.current += 1);
+    const claudeEvery = Math.round(CLAUDE_REFRESH_INTERVAL_MS / REFRESH_INTERVAL_MS);
+    const shouldFetch = (conn) =>
+      force || conn.provider !== "claude" || tick % claudeEvery === 0;
+
     try {
       const visibleConnections = await fetchConnections(page);
 
@@ -419,7 +427,9 @@ export default function ProviderLimits() {
       );
 
       await Promise.all(
-        visibleConnections.map((conn) => fetchQuota(conn.id, conn.provider)),
+        visibleConnections
+          .filter(shouldFetch)
+          .map((conn) => fetchQuota(conn.id, conn.provider)),
       );
 
       setLastUpdated(new Date());
@@ -539,7 +549,7 @@ export default function ProviderLimits() {
         }
       } else if (autoRefresh && hasHydratedAutoRefresh) {
         // Resume auto-refresh when tab becomes visible
-        intervalRef.current = setInterval(refreshAll, REFRESH_INTERVAL_MS);
+        intervalRef.current = setInterval(() => refreshAll(), REFRESH_INTERVAL_MS);
         countdownRef.current = setInterval(() => {
           setCountdown((prev) => (prev <= 1 ? 60 : prev - 1));
         }, 1000);
@@ -866,7 +876,7 @@ export default function ProviderLimits() {
           {/* Refresh all button */}
           <button
             type="button"
-            onClick={refreshAll}
+            onClick={() => refreshAll(true)}
             disabled={refreshingAll}
             className="flex h-8 shrink-0 items-center gap-1 rounded-lg border border-black/10 px-2 text-xs text-text-primary transition-colors hover:bg-black/5 dark:border-white/10 dark:hover:bg-white/5 disabled:opacity-50"
             title="Refresh all"

src/app/(dashboard)/dashboard/usage/components/ProviderLimits/utils.js

@@ -3,6 +3,8 @@ import { getModelsByProviderId } from "open-sse/config/providerModels.js";
 // ─── Constants ───────────────────────────────────────────────────────────────
 export const QUOTA_CACHE_KEY = "quotaCacheData";
 export const REFRESH_INTERVAL_MS = 60000;
+// Claude usage/quota endpoint rate-limits; poll it less often than other providers
+export const CLAUDE_REFRESH_INTERVAL_MS = 180000;
 export const DEPLETED_QUOTA_THRESHOLD = 5;
 export const AUTO_REFRESH_STORAGE_KEY = "quotaAutoRefresh";
 export const CONNECTIONS_PAGE_SIZE = 20;

src/shared/services/claudeAutoPing.js

@@ -12,7 +12,7 @@ import { CLAUDE_AUTOPING_CONFIG } from "@/shared/constants/config";
 const C = CLAUDE_AUTOPING_CONFIG;
 const PING_URL = "https://api.anthropic.com/v1/messages?beta=true";
 
-const g = (global.__claudeAutoPing ??= { interval: null, running: false });
+const g = (global.__claudeAutoPing ??= { interval: null, running: false, resetCache: {} });
 
 function buildProxyOptions(cfg) {
   return {
@@ -43,6 +43,10 @@ async function sendPing(accessToken, proxyOptions) {
 }
 
 async function pingConnection(conn) {
+  // Cached resetAt is stable for the whole 5h window; skip usage poll until near reset
+  const cachedReset = g.resetCache[conn.id];
+  if (cachedReset && Date.now() < new Date(cachedReset).getTime() - C.refreshAheadMs) return;
+
   const proxyCfg = await resolveConnectionProxyConfig(conn.providerSpecificData);
   const proxyOptions = buildProxyOptions(proxyCfg);
 
@@ -60,6 +64,9 @@ async function pingConnection(conn) {
   const resetAt = usage?.quotas?.[C.fiveHourKey]?.resetAt;
   if (!resetAt) return;
 
+  // Cache resetAt to gate future ticks
+  g.resetCache[conn.id] = resetAt;
+
   const resetMs = new Date(resetAt).getTime();
   const now = Date.now();