fix(e2e): bound kitchen sink RPC probes

CHANGELOG.md

@@ -78,6 +78,7 @@ Docs: https://docs.openclaw.ai
 - Telegram: treat `/command@TargetBot` bot-command entities as explicit mentions for the addressed bot so `requireMention` groups no longer drop targeted commands or captions. Fixes #84462. (#86553) Thanks @luoyanglang.
 - CI: bound Docker/Bash E2E tarball npm installs with `OPENCLAW_E2E_NPM_INSTALL_TIMEOUT` so package, onboarding, plugin, and upgrade lanes fail instead of hanging on a stuck npm install.
 - CI: fail Parallels npm-update smoke jobs after the guest command timeout and cleanup backstop instead of only logging a timeout line.
+- CI: bound kitchen-sink RPC HTTP probes so stalled gateway readiness or response bodies fail and retry instead of wedging the walker.
 - CI: keep `OPENCLAW_TESTBOX=1 pnpm check:changed` delegating to Blacksmith Testbox through Crabbox without forwarding local Testbox or worker env into the remote command.
 - CI: send KILL after the TERM grace period for manual checkout fetch timeouts so stuck Testbox and workflow checkout retries cannot hang behind a wedged `git fetch`.
 - CI: send KILL after the TERM grace period for Bun global install smoke command timeouts so trapped `openclaw` child processes cannot wedge the scheduled install smoke.

scripts/e2e/kitchen-sink-rpc-walk.mjs

@@ -27,6 +27,7 @@ const INSTALL_TIMEOUT_MS = readPositiveInt(
   Math.max(COMMAND_TIMEOUT_MS, 600000),
 );
 const RPC_TIMEOUT_MS = readPositiveInt(process.env.OPENCLAW_KITCHEN_SINK_RPC_CALL_MS, 60000);
+const FETCH_TIMEOUT_MS = readPositiveInt(process.env.OPENCLAW_KITCHEN_SINK_RPC_FETCH_MS, 10000);
 const MAX_RSS_MIB = readPositiveInt(process.env.OPENCLAW_KITCHEN_SINK_MAX_RSS_MIB, 2048);
 const GATEWAY_TEARDOWN_GRACE_MS = 10000;
 const GATEWAY_TEARDOWN_KILL_GRACE_MS = 2000;
@@ -439,11 +440,27 @@ function isRetryableTransientNetworkError(error, seen = new Set()) {
 
 export async function fetchJson(url, options = {}) {
   const attempts = Math.max(1, options.attempts ?? 3);
+  const timeoutMs = Math.max(1, options.timeoutMs ?? FETCH_TIMEOUT_MS);
   let lastError;
   for (let attempt = 1; attempt <= attempts; attempt += 1) {
+    const controller = new AbortController();
+    const timeoutError = Object.assign(new Error(`fetch ${url} timed out after ${timeoutMs}ms`), {
+      code: "ETIMEDOUT",
+    });
+    let timeout;
+    const timeoutPromise = new Promise((_, reject) => {
+      timeout = setTimeout(() => {
+        controller.abort(timeoutError);
+        reject(timeoutError);
+      }, timeoutMs);
+      timeout.unref?.();
+    });
     try {
-      const response = await (options.fetchImpl ?? fetch)(url);
-      const text = await response.text();
+      const response = await Promise.race([
+        (options.fetchImpl ?? fetch)(url, { signal: controller.signal }),
+        timeoutPromise,
+      ]);
+      const text = await Promise.race([response.text(), timeoutPromise]);
       let body = null;
       try {
         body = text ? JSON.parse(text) : null;
@@ -457,6 +474,10 @@ export async function fetchJson(url, options = {}) {
         throw error;
       }
       await delay(options.retryDelayMs ?? 250);
+    } finally {
+      if (timeout) {
+        clearTimeout(timeout);
+      }
     }
   }
   throw lastError ?? new Error(`fetch ${url} failed`);

test/scripts/kitchen-sink-rpc-walk.test.ts

@@ -3,7 +3,7 @@ import { existsSync, mkdirSync, mkdtempSync, readFileSync, rmSync, writeFileSync
 import { tmpdir } from "node:os";
 import path from "node:path";
 import { setTimeout as delay } from "node:timers/promises";
-import { describe, expect, it, vi } from "vitest";
+import { afterEach, describe, expect, it, vi } from "vitest";
 import {
   appendBoundedOutput,
   assertDiagnosticStabilityClean,
@@ -23,6 +23,10 @@ import {
 
 const posixIt = process.platform === "win32" ? it.skip : it;
 
+afterEach(() => {
+  vi.useRealTimers();
+});
+
 describe("kitchen-sink RPC isolated state", () => {
   it("cleans up the generated temporary home tree", async () => {
     const { root, env } = makeEnv();
@@ -459,6 +463,29 @@ describe("kitchen-sink RPC process sampling", () => {
     expect(fetchImpl).toHaveBeenCalledTimes(2);
   });
 
+  it("times out stalled HTTP probe response bodies", async () => {
+    vi.useFakeTimers();
+    const fetchImpl = vi.fn().mockResolvedValue({
+      ok: true,
+      status: 200,
+      text: () => new Promise(() => undefined),
+    });
+
+    const result = fetchJson("http://127.0.0.1:19680/readyz", {
+      attempts: 1,
+      fetchImpl,
+      timeoutMs: 100,
+    });
+    const rejection = expect(result).rejects.toMatchObject({
+      code: "ETIMEDOUT",
+      message: "fetch http://127.0.0.1:19680/readyz timed out after 100ms",
+    });
+
+    await vi.advanceTimersByTimeAsync(100);
+    await rejection;
+    expect(fetchImpl.mock.calls[0]?.[1]?.signal.aborted).toBe(true);
+  });
+
   it("fails when the sampled RSS exceeds the configured ceiling", () => {
     expect(() => assertResourceCeiling({ rssMiB: 2049 })).toThrow(
       "gateway RSS exceeded 2048 MiB: 2049 MiB",