fix(qa-lab): handle gateway child spawn errors

vincentkoc · vincentkoc · commit 8e375242bed2 · 2026-06-20T07:18:07.000+02:00
diff --git a/extensions/qa-lab/src/gateway-child.test.ts b/extensions/qa-lab/src/gateway-child.test.ts
@@ -134,6 +134,32 @@ describe("buildQaRuntimeEnv", () => {
     await expect(readdir(tempParent)).resolves.toStrictEqual([]);
   });
 
+  it("reports command spawn errors instead of leaking unhandled child errors", async () => {
+    const tempParent = await mkdtemp(path.join(os.tmpdir(), "qa-gateway-spawn-fail-"));
+    cleanups.push(async () => {
+      await rm(tempParent, { recursive: true, force: true });
+    });
+    qaTempPathState.preferredTmpDir = tempParent;
+    const missingExecutable = path.join(tempParent, "missing-openclaw-node");
+
+    await expect(
+      startQaGatewayChild({
+        repoRoot: process.cwd(),
+        command: {
+          executablePath: missingExecutable,
+          usePackagedPlugins: true,
+        },
+        transport: {
+          requiredPluginIds: [],
+          createGatewayConfig: () => ({}),
+        },
+        transportBaseUrl: "http://127.0.0.1:43123",
+      }),
+    ).rejects.toThrow(/gateway failed to spawn: .*ENOENT/u);
+
+    await expect(readdir(tempParent)).resolves.toStrictEqual([]);
+  });
+
   it("keeps the slow-reply QA opt-out enabled under fast mode", () => {
     const env = buildQaRuntimeEnv({
       ...createParams(),
diff --git a/extensions/qa-lab/src/gateway-child.ts b/extensions/qa-lab/src/gateway-child.ts
@@ -279,6 +279,20 @@ function createQaGatewayChildLogCollector() {
   };
 }
 
+function monitorQaGatewayChildSpawnError(
+  child: ChildProcess,
+  output: { push(chunk: Buffer): void },
+) {
+  let spawnError: unknown = null;
+  child.once("error", (error) => {
+    spawnError = error;
+    output.push(
+      Buffer.from(`[qa-lab] gateway child process error: ${formatErrorMessage(error)}\n`),
+    );
+  });
+  return () => spawnError;
+}
+
 async function fetchLocalGatewayHealth(params: {
   baseUrl: string;
   healthPath: "/readyz" | "/healthz";
@@ -487,10 +501,19 @@ async function waitForGatewayReady(params: {
     exitCode: number | null;
     signalCode: NodeJS.Signals | null;
   };
+  getSpawnError?: () => unknown;
   timeoutMs?: number;
 }) {
   const startedAt = Date.now();
   while (Date.now() - startedAt < (params.timeoutMs ?? 60_000)) {
+    const spawnError = params.getSpawnError?.();
+    if (spawnError) {
+      throw new QaSuiteInfraError(
+        "gateway_startup_unhealthy",
+        `gateway failed to spawn: ${formatErrorMessage(spawnError)}\n${params.logs()}`,
+        { cause: spawnError },
+      );
+    }
     if (params.child.exitCode !== null || params.child.signalCode !== null) {
       throw new QaSuiteInfraError(
         "gateway_startup_unhealthy",
@@ -768,12 +791,14 @@ export async function startQaGatewayChild(params: {
         stderrLog.write(buffer);
       });
       child = attemptChild;
+      const getAttemptSpawnError = monitorQaGatewayChildSpawnError(attemptChild, output);
 
       try {
         await waitForGatewayReady({
           baseUrl,
           logs,
           child: attemptChild,
+          getSpawnError: getAttemptSpawnError,
           timeoutMs: 120_000,
         });
         const attemptRpcClient = await startQaGatewayRpcClient({
@@ -805,6 +830,7 @@ export async function startQaGatewayChild(params: {
                 baseUrl,
                 logs,
                 child: attemptChild,
+                getSpawnError: getAttemptSpawnError,
                 timeoutMs: QA_GATEWAY_CHILD_RPC_RETRY_HEALTH_TIMEOUT_MS,
               });
             }
@@ -871,12 +897,14 @@ export async function startQaGatewayChild(params: {
         output.push(buffer);
         stderrLog.write(buffer);
       });
+      const getNextSpawnError = monitorQaGatewayChildSpawnError(nextChild, output);
 
       try {
         await waitForGatewayReady({
           baseUrl,
           logs,
           child: nextChild,
+          getSpawnError: getNextSpawnError,
           timeoutMs: 120_000,
         });
         const nextRpcClient = await startQaGatewayRpcClient({
@@ -908,6 +936,7 @@ export async function startQaGatewayChild(params: {
                 baseUrl,
                 logs,
                 child: nextChild,
+                getSpawnError: getNextSpawnError,
                 timeoutMs: 15_000,
               });
             }
