conductor(checkpoint): Checkpoint end of Phase 2

scripts/cluster_status/litefs_client.py

@@ -0,0 +1,51 @@
+import requests
+
+def get_node_status(node_address, port=20202):
+    """
+    Queries the LiteFS HTTP API on a specific node for its status.
+    Tries /status first, then falls back to /debug/vars.
+    """
+    # Try /status first
+    url = f"http://{node_address}:{port}/status"
+    try:
+        response = requests.get(url, timeout=3)
+        if response.status_code == 200:
+            data = response.json()
+            status = {
+                "is_primary": data.get("primary", False),
+                "uptime": data.get("uptime", 0),
+                "advertise_url": data.get("advertiseURL", "")
+            }
+            if "replicationLag" in data:
+                status["replication_lag"] = data["replicationLag"]
+            if "primaryURL" in data:
+                status["primary_url"] = data["primaryURL"]
+            return status
+    except Exception:
+        pass
+
+    # Fallback to /debug/vars
+    url = f"http://{node_address}:{port}/debug/vars"
+    try:
+        response = requests.get(url, timeout=3)
+        response.raise_for_status()
+        data = response.json()
+        store = data.get("store", {})
+        
+        status = {
+            "is_primary": store.get("isPrimary", False),
+            "uptime": "N/A", # Not available in /debug/vars
+            "advertise_url": f"http://{node_address}:{port}" # Best guess
+        }
+        
+        # Look for lag in dbs or store if it exists in other versions
+        if "replicationLag" in store:
+            status["replication_lag"] = store["replicationLag"]
+            
+        return status
+    except Exception as e:
+        return {
+            "error": str(e),
+            "is_primary": False,
+            "uptime": "N/A"
+        }