Update playbook.yml

checker.py

@@ -138,7 +138,10 @@ def format_number(number_str):
     try:
         number = int(number_str)
         if number >= 1000:
-            return f"{number//1000}k"
+            value_in_k = number / 1000.0
+            # Format to 3 decimal places if needed, remove trailing zeros and potentially the dot
+            formatted_num = f"{value_in_k:.3f}".rstrip('0').rstrip('.')
+            return f"{formatted_num}k"
         return str(number)
     except (ValueError, TypeError):
         return "NaN" # Or some other indicator of invalid input
@@ -202,7 +205,7 @@ def check_logs(logger, initial_sync_count, previous_status):
             formatted_id = format_number(last_checking_info["last_sub_id"])
             if last_checking_info["num_subs_to_sync"] > 0:
                 current_status_type = "Sync"
-                status_message = f"Sync: {formatted_id} ({current_sync_count})" # Use current_sync_count
+                status_message = f"Sync: {formatted_id}" # Use current_sync_count
                 logger.info(f"Node is syncing. Last sub ID: {last_checking_info['last_sub_id']}, Num subs to sync: {last_checking_info['num_subs_to_sync']}")
             else:
                 current_status_type = "OK"
@@ -210,14 +213,14 @@ def check_logs(logger, initial_sync_count, previous_status):
                 if previous_status == "Sync":
                     current_sync_count += 1 # Increment local count
                     logger.info(f"Sync completed. Sync count incremented to {current_sync_count}.")
-                status_message = f"OK: {formatted_id} ({current_sync_count})" # Use current_sync_count
+                status_message = f"OK: {formatted_id}" # Use current_sync_count
                 logger.info(f"Node is OK. Last sub ID: {last_checking_info['last_sub_id']}")
         
         elif last_ignored_id:
             # Fallback to "Ignored" logs if "Checking" is missing
             formatted_id = format_number(last_ignored_id)
             current_status_type = "Sync" # Assume sync if we only see ignored creations recently
-            status_message = f"Sync: {formatted_id} ({current_sync_count})" # Use current_sync_count
+            status_message = f"Sync: {formatted_id}" # Use current_sync_count
             logger.info(f"Node possibly syncing (based on ignored logs). Last ignored ID: {last_ignored_id}")
 
         elif last_head_sub_id:
@@ -225,7 +228,7 @@ def check_logs(logger, initial_sync_count, previous_status):
             formatted_id = format_number(last_head_sub_id)
             current_status_type = "OK" # Assume OK if this is the latest relevant info
             # Don't increment sync count here, only on Sync -> OK transition based on "Checking" logs
-            status_message = f"OK: {formatted_id} ({current_sync_count})" # Use current_sync_count
+            status_message = f"OK: {formatted_id}" # Use current_sync_count
             logger.info(f"Node status based on head sub id. Head sub ID: {last_head_sub_id}")
             
         else:
@@ -284,8 +287,31 @@ if __name__ == "__main__":
     current_vm = grist.find_record(name=GRIST_ROW_NAME, table=NODES_TABLE)[0]
     def grist_callback(msg): grist.update(current_vm.id, msg, NODES_TABLE)
 
-    # Get initial state from Grist
-    initial_sync_count = int(current_vm.Syncs or 0)
+    # Initialize updates dictionary
+    initial_updates = {}
+    # Check and prepare update for Syncs if it's None or empty
+    if not current_vm.Syncs: # Handles None, empty string, potentially 0 if that's how Grist stores it
+        initial_updates["Syncs"] = 0
+    # Check and prepare update for Reboots if it's None or empty
+    if not current_vm.Reboots: # Handles None, empty string, potentially 0
+        initial_updates["Reboots"] = 0
+
+    # If there are updates, send them to Grist
+    if initial_updates:
+        try:
+            logger.info(f"Found empty initial values, updating Grist: {initial_updates}")
+            grist.update(current_vm.id, initial_updates, NODES_TABLE)
+            # Re-fetch the record to ensure subsequent logic uses the updated values
+            current_vm = grist.find_record(name=GRIST_ROW_NAME, table=NODES_TABLE)[0]
+            logger.info("Grist updated successfully with initial zeros.")
+        except Exception as e:
+            logger.error(f"Failed to update Grist with initial zeros: {e}")
+            # Decide how to proceed: maybe exit, maybe continue with potentially incorrect defaults
+            # For now, we'll log the error and continue using the potentially incorrect defaults from the first fetch
+
+    # Get initial state from Grist (now potentially updated)
+    initial_sync_count = int(current_vm.Syncs or 0) # 'or 0' still useful as fallback
+    reboot_count = int(current_vm.Reboots or 0) # 'or 0' still useful as fallback
     # Determine previous status type based on Health string (simplified)
     previous_health_status = current_vm.Health or "Idle"
     previous_status_type = "Idle" # Default
@@ -296,10 +322,19 @@ if __name__ == "__main__":
     elif previous_health_status.startswith("Error"):
         previous_status_type = "Error" # Consider error state
 
-    logger.info(f"Initial state from Grist - Syncs: {initial_sync_count}, Health: {previous_health_status} (interpreted as: {previous_status_type})")
+    logger.info(f"Initial state from Grist - Syncs: {initial_sync_count}, Health: {previous_health_status}, Reboots: {reboot_count}")
 
     for attempt in range(3):
         try:
+            vm_ip = os.popen("ip -4 addr show eth0 | grep -oP '(?<=inet )[^/]+'").read()
+            vm_ip = vm_ip.strip()
+            if vm_ip == "":
+                logger.error("Failed to get VM IP address")
+            else:
+                logger.info(f"VM IP address: {vm_ip}")
+                grist_callback({"IP": f"{vm_ip}"})
+
+
             # Pass initial state to check_logs
             result = check_logs(logger, initial_sync_count, previous_status_type)
 
@@ -318,13 +353,13 @@ if __name__ == "__main__":
             if result["current_status_type"] == "Idle": # Check type, not message
                 uptime_seconds = os.popen("cat /proc/uptime | cut -d'.' -f1").read()
                 uptime_seconds = int(uptime_seconds)
-                if uptime_seconds > 60*60*1:
+                if uptime_seconds > 60*60*4:
                     reboot_count = int(current_vm.Reboots or 0)
                     reboot_count += 1
                     # Include reboot count in the final Grist update before rebooting
                     grist_updates = { "Health": "Rebooting", "Reboots": reboot_count }
                     grist_callback(grist_updates)
-                    logger.info(f"Idle detected for >1 hour (uptime: {uptime_seconds}s). Rebooting. Reboot count: {reboot_count}")
+                    logger.info(f"Idle detected for >4 hours (uptime: {uptime_seconds}s). Rebooting. Reboot count: {reboot_count}")
                     os.system("reboot")
             break # Exit loop on success
         except Exception as e:

playbook.yml

@@ -199,8 +199,8 @@
         rm -rf {{ ansible_env.HOME }}/node/projects/hello-world/contracts/lib/infernet-sdk
         cd {{ ansible_env.HOME }}/foundry && source {{ ansible_env.HOME }}/.bashrc && foundryup
         cd {{ ansible_env.HOME }}/node/projects/hello-world/contracts
-        forge install --no-commit foundry-rs/forge-std
-        forge install --no-commit ritual-net/infernet-sdk
+        forge install foundry-rs/forge-std
+        forge install ritual-net/infernet-sdk
       args:
         executable: /bin/bash