removed timeout waiting for NDJSON, change workflow timeout to 30mins

Signed-off-by: Lior Sventitzky <liorsve@amazon.com>

Author: liorsve

.github/workflows/benchmark.yml

@@ -65,7 +65,7 @@ env:
 jobs:
   benchmark:
     runs-on: [self-hosted, Linux, x86, ephemeral, metal]
-    timeout-minutes: 120
+    timeout-minutes: 30
 
     steps:
       - uses: actions/checkout@v4

.github/workflows/benchmark_orchestrator.py

@@ -349,25 +349,10 @@ def __init__(self, metrics_path: Path):
         self.watcher_thread: Optional[threading.Thread] = None
         self.phase_records: dict = {}
 
-    def start(self, timeout_seconds: int = 600,
-              benchmark_proc: Optional[subprocess.Popen] = None):
+    def start(self, benchmark_proc: Optional[subprocess.Popen] = None):
         print(f"Waiting for metrics file: {self.metrics_path}")
-        start_time = time.time()
+        last_status_time = time.time()
         while not self.metrics_path.exists():
-            if time.time() - start_time > timeout_seconds:
-                # Try to capture benchmark output if process provided
-                if benchmark_proc:
-                    benchmark_proc.terminate()
-                    try:
-                        stdout, stderr = benchmark_proc.communicate(timeout=5)
-                        print("=== Benchmark stdout ===")
-                        print(stdout.decode() if stdout else "(empty)")
-                        print("=== Benchmark stderr ===")
-                        print(stderr.decode() if stderr else "(empty)")
-                    except Exception as e:
-                        print(f"Could not capture benchmark output: {e}")
-                raise RuntimeError(
-                    f"Timeout waiting for metrics file: {self.metrics_path}")
             # Check if benchmark process died
             if benchmark_proc and benchmark_proc.poll() is not None:
                 stdout, stderr = benchmark_proc.communicate()
@@ -378,6 +363,11 @@ def start(self, timeout_seconds: int = 600,
                 print(stderr.decode() if stderr else "(empty)")
                 raise RuntimeError(
                     f"Benchmark process died with exit code {benchmark_proc.returncode}")
+            # Print status every 30 seconds
+            if time.time() - last_status_time > 30:
+                elapsed = int(time.time() - last_status_time)
+                print(f"Still waiting for metrics file... (benchmark process running)")
+                last_status_time = time.time()
             time.sleep(0.1)
 
         self.tail_process = subprocess.Popen(