fix(v2): make EC2 bootstrap readiness deterministic

backend/src/services/ssh.ts

@@ -63,22 +63,32 @@ function tryConnect(host: string, privateKey: string, timeoutMs: number): Promis
       username: "root",
       privateKey,
       readyTimeout: timeoutMs,
+      keepaliveInterval: 10000,
+      keepaliveCountMax: 3,
       algorithms: {
-        serverHostKey: ["ssh-rsa", "ecdsa-sha2-nistp256", "ecdsa-sha2-nistp384", "ecdsa-sha2-nistp521"],
+        serverHostKey: ["ssh-rsa", "ecdsa-sha2-nistp256", "ecdsa-sha2-nistp384", "ecdsa-sha2-nistp521", "ssh-ed25519"],
       },
     });
   });
 }
 
+const EXEC_TIMEOUT_MS = 30_000;
+
 function execOnConn(conn: Client, command: string): Promise<{ stdout: string; stderr: string; code: number }> {
   return new Promise((resolve, reject) => {
+    const timer = setTimeout(() => reject(new Error(`SSH exec timed out: ${command.slice(0, 60)}`)), EXEC_TIMEOUT_MS);
+
     conn.exec(command, (err, stream) => {
-      if (err) return reject(err);
+      if (err) {
+        clearTimeout(timer);
+        return reject(err);
+      }
       let stdout = "";
       let stderr = "";
       stream.on("data", (d: Buffer) => { stdout += d.toString(); });
       stream.stderr.on("data", (d: Buffer) => { stderr += d.toString(); });
       stream.on("close", (code: number) => {
+        clearTimeout(timer);
         resolve({ stdout, stderr, code: code ?? 0 });
       });
     });
@@ -88,8 +98,12 @@ function execOnConn(conn: Client, command: string): Promise<{ stdout: string; st
 export async function waitForBootstrap(session: SshSession, maxWaitMs = 600_000): Promise<void> {
   const start = Date.now();
   while (Date.now() - start < maxWaitMs) {
-    const res = await session.exec("test -f /var/lib/pp-bootstrap-done && echo done || echo waiting");
-    if (res.stdout.trim() === "done") return;
+    try {
+      const res = await session.exec("test -f /var/lib/pp-bootstrap-done && echo done || echo waiting");
+      if (res.stdout.trim() === "done") return;
+    } catch {
+      // transient SSH exec error (e.g. timeout, channel reset) — retry
+    }
     await sleep(10_000);
   }
   throw new Error("EC2 bootstrap did not complete within timeout");