[TO REVERT] test on controller ers

Author: emmuhamm

src/drunc/controller/configuration.py

@@ -41,6 +41,8 @@ class cler:
         self.controller.broadcaster = id_able()
         self.controller.fsm = id_able()
 
+        # Maybe we can have something here? but what uses tihs
+
 
 class ControllerConfHandler(ConfHandler):
     @staticmethod
@@ -74,6 +76,8 @@ def _post_process_oks(self, *args, **kwargs):
         if self.this_host in ["localhost"] or self.this_host.startswith("127."):
             self.this_host = socket.gethostname()
 
+        print(self.session.opmon_uri)
+        print(self.session)
         self.opmon_publisher = None
         self.opmon_conf = parse_opmon_conf(
             log=self.log,

src/drunc/controller/controller.py

@@ -226,7 +226,9 @@ def __init__(self, configuration, name: str, session: str, token: Token):
         self.broadcast_service = None
         self.monitoring_metrics = ControllerMonitoringMetrics()
 
-        self.log = get_logger("controller", stream_handlers=True)
+        self.log = get_logger("controller", rich_handler=True)
+        test_loggess = get_logger("controller.heyo")
+        test_loggess.error("Hello, world")
         log_init = get_logger("controller.__init__")
         log_init.info(f"Initialising controller '{name}' with session '{session}'")
 
@@ -250,6 +252,8 @@ def __init__(self, configuration, name: str, session: str, token: Token):
             data=self.configuration.data.controller.broadcaster,
         )
 
+        # And then its gonna go here im sure
+
         self.broadcast_service = BroadcastSender(
             name=name,
             session=session,
@@ -974,9 +978,11 @@ def execute_fsm_command(
 
         command = request.command
         command_name = command.command_name
-        self.log.debug(f"FSM command: {command_name}")
+        newlog = get_logger("mytestlogger", rich_handler=True)
+        newlog.warning("Hello there")
+        self.log.warning(f"FSM command: {command_name}")
         transition = self.stateful_node.get_fsm_transition(command_name)
-        self.log.debug(f"FSM transition: {transition}")
+        self.log.warning(f"FSM transition: {transition}")
 
         # Check controller readiness.
         if not self.stateful_node.get_ready_state():

src/drunc/controller/controller_driver.py

@@ -79,6 +79,11 @@ def __init__(self, address: str, token: Token):
             ("grpc.keepalive_time_ms", 60000)  # pings the server every 60 seconds
         ]
 
+        self.log.error(f"address: {address}, token: {token}")
+
+        #! SO HERE!!! is where we initialise the add ers protobuf handler. The  question is now how do we get the correct config?
+        #! But once yoh ave this it should be fine
+
         try:
             resolved_address = self._resolve_address_to_ipv4(address)
         except ValueError as e:
@@ -189,6 +194,18 @@ def execute_fsm_command(
         execute_on_all_subsequent_children_in_path: bool = True,
         timeout: int | float = 60,
     ) -> ExecuteFSMCommandResponse:
+        test_log = get_logger("my_test_logger", rich_handler=True)
+
+        test_log.warning(
+            f"Command: {command}, target: {target}, command_name: {command.command_name}"
+        )
+
+        if command.command_name == "disable_triggers":
+            test_log.error("there is an error here!")
+            raise ValueError("Putting an error in here to stop the transition")
+
+        #! this is not where we want the value error, this is where we want the logger to be!!
+
         request = ExecuteFSMCommandRequest(
             target=target,
             execute_along_path=execute_along_path,
@@ -199,7 +216,9 @@ def execute_fsm_command(
 
         try:
             response = self.stub.execute_fsm_command(request, timeout=timeout)
+            #! Here is where we do a fake 'failure'
         except grpc.RpcError as e:
+            #! And then handle the logging
             handle_grpc_error(e)
 
         return response

src/drunc/controller/interface/shell_utils.py

@@ -480,6 +480,9 @@ def run_one_fsm_command(
         ArgumentException: If there is an issue with the arguments
         ServerTimeout: If the server times out
     """
+    log = get_logger("controller.shell_utils")
+
+    #! Is this where we want the protobuf handler?
     log.info(
         f"Running transition '{transition_name}' on controller '{controller_name}', targeting: '{target if target else controller_name}'"
     )
@@ -543,7 +546,10 @@ class DummyCommand:
         time_start = time.time()
         result = None
 
+        log.info("About to execute the command")
+
         with ThreadPoolExecutor() as executor:
+            # so the shell utils hands this over to the actual controller, which will then execute its own fsm command. Wow
             future = executor.submit(
                 obj.get_driver("controller").execute_fsm_command,
                 command=data,
@@ -581,6 +587,10 @@ class DummyCommand:
             "Alternatively, if you are patient, you can try to wait a bit longer and send [yellow]'status'[/yellow] to check if the command ends up being executed (you may want to check the logs of the controller and application with the [yellow]'logs'[/yellow] command)."
         )
         return
+    except ValueError as e:
+        log.error(e)
+        log.error("Lets see what we have here")
+        # log.error("This should go to ERS")
 
     if not result:
         return
@@ -633,6 +643,8 @@ def add_to_table(table, response, prefix=""):
 
 
 def generate_fsm_command(ctx, transition: FSMCommandDescription, controller_name: str):
+    log = get_logger("controller.shell_utils")
+    # log.debug("inside generate_fsm_command")
     cmd = partial(run_one_fsm_command, controller_name, transition.name)
     cmd = click.pass_obj(cmd)
     cmd = click.option(

src/drunc/process_manager/interface/commands.py

@@ -13,7 +13,6 @@
 from drunc.process_manager.utils import tabulate_process_instance_list
 from drunc.utils.shell_utils import InterruptedCommand
 from drunc.utils.utils import get_logger
-from erskafka.ERSKafkaLogHandler import ERSKafkaLogHandler
 
 
 @click.command("boot")

src/drunc/process_manager/process_manager.py

@@ -3,6 +3,7 @@
 import threading
 import time
 
+from daqpytools.logging.handlers import add_ers_protobuf_handler
 from druncschema.authoriser_pb2 import ActionType, SystemType
 from druncschema.broadcast_pb2 import BroadcastType
 from druncschema.description_pb2 import CommandDescription, Description
@@ -23,6 +24,9 @@
     Request,
     ResponseFlag,
 )
+
+# import logging
+from erskafka.ERSKafkaLogHandler import ERSKafkaLogHandler
 from google.rpc import code_pb2
 from grpc import ServicerContext
 
@@ -40,11 +44,6 @@
 from drunc.utils.configuration import ConfTypes
 from drunc.utils.utils import get_logger, pid_info_str
 
-from daqpytools.logging.handlers import add_file_handler, add_ers_protobuf_handler
-
-# import logging
-from erskafka.ERSKafkaLogHandler import ERSKafkaLogHandler
-
 
 class BadQuery(DruncCommandException):
     def __init__(self, txt):
@@ -231,9 +230,9 @@ def publish(self, q: ProcessQuery, interval_s: float = 10.0):
             )
             #! Obviously hacky but proof of principle
             # Checks if n_dead changed compared to the previous iteration
-            self.log.warning(
-                f"Processes: {n_running}, {n_dead}, {n_session}, {n_dead_keep}"
-            )
+            # self.log.warning(
+            #     f"Processes: {n_running}, {n_dead}, {n_session}, {n_dead_keep}"
+            # )
             if n_dead_keep != n_dead:
                 self.log.error("We have a real dead one!", extra={"use_ers": True})
             n_dead_keep = n_dead