fix: harden msteams monitor lifecycle edges (#24580) (thanks @chilu18)

CHANGELOG.md

@@ -47,6 +47,7 @@ Docs: https://docs.openclaw.ai
 
 ### Fixes
 
+- MSTeams/gateway lifecycle: keep `startAccount` pending until shutdown/abort to stop false auto-restart loops, fail fast on startup bind errors, and add lifecycle regressions for abort + startup failure behavior. (#24580) Thanks @chilu18.
 - Synology Chat/webhook compatibility: accept JSON and alias payload fields, allow token resolution from body/query/header sources, and ACK webhook requests with `204` to avoid persistent `Processing...` states in Synology Chat clients. (#26635) Thanks @memphislee09-source.
 - Synology Chat/webhook ingress hardening: enforce bounded body reads (size + timeout) via shared request-body guards to prevent unauthenticated slow-body hangs before token validation. (#25831) Thanks @bmendonca3.
 - Synology Chat/reply delivery: resolve webhook usernames to Chat API `user_id` values for outbound chatbot replies, avoiding mismatches between webhook user IDs and `method=chatbot` recipient IDs in multi-account setups. (#23709) Thanks @druide67.

extensions/msteams/src/monitor.lifecycle.test.ts

@@ -175,6 +175,24 @@ describe("monitorMSTeamsProvider lifecycle", () => {
     );
   });
 
+  it("returns a shutdown handle when abort signal is not provided", async () => {
+    const stores = createStores();
+    const result = await monitorMSTeamsProvider({
+      cfg: createConfig(0),
+      runtime: createRuntime(),
+      conversationStore: stores.conversationStore,
+      pollStore: stores.pollStore,
+    });
+
+    expect(result).toEqual(
+      expect.objectContaining({
+        shutdown: expect.any(Function),
+      }),
+    );
+
+    await expect(result.shutdown()).resolves.toBeUndefined();
+  });
+
   it("rejects startup when webhook port is already in use", async () => {
     expressControl.mode.value = "error";
     await expect(

extensions/msteams/src/monitor.ts

@@ -277,7 +277,6 @@ export async function monitorMSTeamsProvider(
   const httpServer = expressApp.listen(port);
   await new Promise<void>((resolve, reject) => {
     const onListening = () => {
-      httpServer.off("error", onError);
       log.info(`msteams provider started on port ${port}`);
       resolve();
     };
@@ -306,24 +305,30 @@ export async function monitorMSTeamsProvider(
     });
   };
 
-  // Handle abort signal
-  const onAbort = () => {
-    void shutdown();
-  };
-  if (opts.abortSignal) {
-    if (opts.abortSignal.aborted) {
-      onAbort();
-    } else {
-      opts.abortSignal.addEventListener("abort", onAbort, { once: true });
-    }
+  // Some direct callers may invoke monitor without lifecycle wiring.
+  // Return immediately so they can call shutdown explicitly.
+  if (!opts.abortSignal) {
+    return { app: expressApp, shutdown };
   }
 
-  // Keep this task alive until shutdown/close so gateway runtime does not treat startup as exit.
-  await new Promise<void>((resolve) => {
+  const closePromise = new Promise<void>((resolve) => {
     httpServer.once("close", () => {
       resolve();
     });
   });
+
+  // Handle abort signal
+  const onAbort = () => {
+    void shutdown();
+  };
+  if (opts.abortSignal.aborted) {
+    onAbort();
+  } else {
+    opts.abortSignal.addEventListener("abort", onAbort, { once: true });
+  }
+
+  // Keep this task alive until shutdown/close so gateway runtime does not treat startup as exit.
+  await closePromise;
   opts.abortSignal?.removeEventListener("abort", onAbort);
 
   return { app: expressApp, shutdown };