From 0e99a6a346b956c4bad5780bee4c2e7036b8b51b Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Mon, 1 Jun 2026 11:46:29 +0200
Subject: [PATCH 01/14] [FIXUP] hardcode to 29 to test if boot works on
 different hosts

---
 src/drunc/connectivity_service/client.py      | 10 ++++-----
 src/drunc/process_manager/configuration.py    |  2 ++
 src/drunc/process_manager/oks_parser.py       |  5 +++++
 .../process_manager/process_manager_driver.py | 21 ++++++++++++++++---
 4 files changed, 30 insertions(+), 8 deletions(-)

diff --git a/src/drunc/connectivity_service/client.py b/src/drunc/connectivity_service/client.py
index 5e7eb124f..87fa1406f 100644
--- a/src/drunc/connectivity_service/client.py
+++ b/src/drunc/connectivity_service/client.py
@@ -41,16 +41,16 @@ def is_ready(self, timeout: int = 10):
             attempt += 1
             elapsed = time.time() - start
 
-            self.log.debug(f"Health check attempt {attempt} at {elapsed:.2f}s elapsed")
-            self.log.debug(f"Polling address: {self.address}")
+            self.log.info(f"Health check attempt {attempt} at {elapsed:.2f}s elapsed")
+            self.log.info(f"Polling address: {self.address}")
             try:
                 r = get(self.address)
             except Exception as e:
-                self.log.debug(f"Polling failed with exception: {e}")
+                self.log.info(f"Polling failed with exception: {e}")
                 r = None
             # Request failed - service is NOT ready.
             if r is None or not r.ok:
-                self.log.debug(
+                self.log.info(
                     f"Connectivity service not ready, retrying in {delay:.2f}s"
                 )
                 time.sleep(delay)
@@ -58,7 +58,7 @@ def is_ready(self, timeout: int = 10):
                 delay = min(delay * 2, max_delay)
             # Request succeeded - service is ready.
             else:
-                self.log.debug(
+                self.log.info(
                     f"Connectivity service ready after {attempt} attempts ({elapsed:.2f}s)"
                 )
                 return True
diff --git a/src/drunc/process_manager/configuration.py b/src/drunc/process_manager/configuration.py
index da9198722..21be97eb2 100644
--- a/src/drunc/process_manager/configuration.py
+++ b/src/drunc/process_manager/configuration.py
@@ -149,6 +149,8 @@ def get_commandline_parameters(
         Ordered list of command-line arguments for process launch.
     """
     runs_on = obj.runs_on.runs_on.id
+    runs_on = "np04-srv-029.cern.ch"
+
     control_service_port = -1
     control_service_protocol = ""
     for svc in obj.exposes_service:
diff --git a/src/drunc/process_manager/oks_parser.py b/src/drunc/process_manager/oks_parser.py
index 5feb6ba50..dea72e64e 100644
--- a/src/drunc/process_manager/oks_parser.py
+++ b/src/drunc/process_manager/oks_parser.py
@@ -183,6 +183,7 @@ def collect_apps(
     collect_variables(controller.application_environment, rc_env)
     rc_env["DUNEDAQ_APPLICATION_NAME"] = controller.id
     host = controller.runs_on.runs_on.id
+    host = "np04-srv-029.cern.ch"
 
     tree_id_str = ".".join(map(str, tree_prefix))
     apps.append(
@@ -260,6 +261,10 @@ def collect_apps(
         )
         log.debug(f"Collecting app {app.id} with args {args}")
 
+        host = "np04-srv-029.cern.ch"
+
+        log.info(f"the host is {host}")
+
         data_path = get_writer_directory_path(app, log)
         if not data_path:
             log.debug(f"No data path found for app {app.id}")
diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index ea239f4d6..9b31305a3 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -141,15 +141,18 @@ def boot(
         # Step 3 - check for port conflicts and update configuration/DAL as needed
         db, session_dal = self.check_port_conflicts(db, session_dal)
 
+        self.log.info("step 4")
         # Step 4 - connect to the connection service
         csc, connection_server, connection_port = self._connect_to_service(
             session_dal, session_name
         )
 
+        self.log.info("step 5")
         # Step 5 - track boot timings per host
         last_boot_on_host_at = {}
         previous_host = None
 
+        self.log.info("step 6")
         # Step 6: iterate over boot requests
         for request in self._convert_oks_to_boot_request(
             oks_conf=conf_file,
@@ -159,6 +162,8 @@ def boot(
             override_logs=override_logs,
             **kwargs,
         ):
+            self.log.warning(request)
+
             if not request:
                 self.log.error("[red]No boot request was generated, ending boot.[/red]")
                 return None
@@ -169,10 +174,11 @@ def boot(
                     f"Skipping connectivity service readiness check for application {request.process_description.metadata.name}"
                 )
             else:
-                self.log.debug(
+                self.log.info(
                     f"Checking connectivity service readiness before booting application {request.process_description.metadata.name}"
                 )
                 if csc and not csc.is_ready(timeout=10):
+                    self.log.info("whoops its not ready?")
                     raise DruncSetupException(
                         "Connectivity service did not respond within timeout."
                     )
@@ -205,6 +211,7 @@ def boot(
                     )
                 handle_grpc_error(e)
 
+        self.log.info("step 7")
         # Step 7: discover segment root controller
         self._discover_controller(
             session_dal, session_name, csc, connection_server, connection_port
@@ -239,7 +246,7 @@ def _collect_all_apps(
 
         apps = infra_apps + apps
 
-        self.log.debug(f"{json.dumps(apps, indent=4)}")
+        self.log.info(f"{json.dumps(apps, indent=4)}")
 
         return apps
 
@@ -542,19 +549,27 @@ def _connect_to_service(
         self, session_dal: "conffwk.dal.Session", session_name: str
     ) -> ConnectivityServiceClient | None:
         if session_dal.connectivity_service:
+            self.log.info("we have connectivity service")
+
             connection_server = session_dal.connectivity_service.host
             connection_port = session_dal.connectivity_service.service.port
 
             if connection_server == "localhost":
+                self.log.info("Apparently its localhost?")
                 resolved_server = resolve_localhost_to_hostname(connection_server)
-                self.log.debug(
+                self.log.info(
                     f"Resolved connection server 'localhost' to '{resolved_server}' to avoid K8s hairpinning."
                 )
                 connection_server = resolved_server
 
+            connection_server = "np04-srv-029.cern.ch"
             client = ConnectivityServiceClient(
                 session_name, f"{connection_server}:{connection_port}"
             )
+
+            self.log.info(
+                f"returning {client=}, {connection_server=}, {connection_port=}"
+            )
             return client, connection_server, connection_port
         return None, None, None
 

From 29f01727b10ab70914df03a1bb3e18d1d8d3f370 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Mon, 1 Jun 2026 12:27:29 +0200
Subject: [PATCH 02/14] minimum fix needed to fix the LCS issue

---
 src/drunc/process_manager/configuration.py       |  2 +-
 src/drunc/process_manager/oks_parser.py          |  4 ++--
 .../process_manager/process_manager_driver.py    | 16 ++++++++++++++--
 3 files changed, 17 insertions(+), 5 deletions(-)

diff --git a/src/drunc/process_manager/configuration.py b/src/drunc/process_manager/configuration.py
index 21be97eb2..fa6e821cd 100644
--- a/src/drunc/process_manager/configuration.py
+++ b/src/drunc/process_manager/configuration.py
@@ -149,7 +149,7 @@ def get_commandline_parameters(
         Ordered list of command-line arguments for process launch.
     """
     runs_on = obj.runs_on.runs_on.id
-    runs_on = "np04-srv-029.cern.ch"
+    # runs_on = "np04-srv-029.cern.ch"
 
     control_service_port = -1
     control_service_protocol = ""
diff --git a/src/drunc/process_manager/oks_parser.py b/src/drunc/process_manager/oks_parser.py
index dea72e64e..e73a4cae4 100644
--- a/src/drunc/process_manager/oks_parser.py
+++ b/src/drunc/process_manager/oks_parser.py
@@ -183,7 +183,7 @@ def collect_apps(
     collect_variables(controller.application_environment, rc_env)
     rc_env["DUNEDAQ_APPLICATION_NAME"] = controller.id
     host = controller.runs_on.runs_on.id
-    host = "np04-srv-029.cern.ch"
+    # host = "np04-srv-029.cern.ch"
 
     tree_id_str = ".".join(map(str, tree_prefix))
     apps.append(
@@ -261,7 +261,7 @@ def collect_apps(
         )
         log.debug(f"Collecting app {app.id} with args {args}")
 
-        host = "np04-srv-029.cern.ch"
+        # host = "np04-srv-029.cern.ch"
 
         log.info(f"the host is {host}")
 
diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index 9b31305a3..5a29eadd4 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -7,6 +7,7 @@
 from collections.abc import Iterator
 from time import sleep
 from typing import Dict, List
+from urllib.parse import urlparse
 
 import conffwk
 import grpc
@@ -129,6 +130,9 @@ def boot(
     ) -> Iterator[ProcessInstanceList] | None:
         self.log.info(f"Booting session [green]{session_name}[/green]")
 
+        self.log.warning(
+            f"parameters: {conf_file=}, {conf_id=}, {user=}, {self.address=}"
+        )
         # Assume oksconflibs if no framework is defined
         conf_file = f"oksconflibs:{conf_file}" if ":" not in conf_file else conf_file
 
@@ -556,13 +560,21 @@ def _connect_to_service(
 
             if connection_server == "localhost":
                 self.log.info("Apparently its localhost?")
-                resolved_server = resolve_localhost_to_hostname(connection_server)
+                # resolved_server = resolve_localhost_to_hostname(connection_server)
+
+                test_address = self.address
+                if "://" not in test_address:
+                    test_address = "grpc://" + test_address
+
+                resolved_server = urlparse(test_address).hostname
                 self.log.info(
                     f"Resolved connection server 'localhost' to '{resolved_server}' to avoid K8s hairpinning."
                 )
                 connection_server = resolved_server
 
-            connection_server = "np04-srv-029.cern.ch"
+            # Funnily enough this is the minimum amount of change needed...s
+            # so basically instead of trying to resolve the localhost hostname, resolve to whatever we are connecting to?
+            # connection_server = "np04-srv-028"
             client = ConnectivityServiceClient(
                 session_name, f"{connection_server}:{connection_port}"
             )

From 77280e1e035dd0b721901fa12d06d1c2397d9820 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Mon, 1 Jun 2026 12:47:33 +0200
Subject: [PATCH 03/14] Remove the confirmation?

---
 src/drunc/process_manager/interface/commands.py | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/src/drunc/process_manager/interface/commands.py b/src/drunc/process_manager/interface/commands.py
index cbfde6250..05deb10de 100644
--- a/src/drunc/process_manager/interface/commands.py
+++ b/src/drunc/process_manager/interface/commands.py
@@ -48,11 +48,15 @@ def boot(
     processes = obj.get_driver("process_manager").ps(ProcessQuery(user=user))
 
     if len(processes.values) > 0:
-        click.confirm(
-            f"You already have {len(processes.values)} processes running, are you sure you want to boot a session?",
-            abort=True,
+        log.info(
+            f"Note that there are already {len(processes.values)} processes running."
         )
 
+        # click.confirm(
+        #     f"You already have {len(processes.values)} processes running, are you sure you want to boot a session?",
+        #     abort=True,
+        # )
+
     log.debug(
         f"Booting session {session_name} with boot configuration file {configuration_file} and id {configuration_id}, requested by user {user}"
     )

From 446f7747d8b313d2eb648e0b868456cdf47c761f Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Mon, 1 Jun 2026 15:04:49 +0200
Subject: [PATCH 04/14] update controller advertising output

---
 src/drunc/process_manager/interface/commands.py |  3 ++-
 src/drunc/utils/utils.py                        | 12 ++++++++++++
 tests/utils/test_utils.py                       |  7 +++++++
 3 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/src/drunc/process_manager/interface/commands.py b/src/drunc/process_manager/interface/commands.py
index 05deb10de..eefbbb1d5 100644
--- a/src/drunc/process_manager/interface/commands.py
+++ b/src/drunc/process_manager/interface/commands.py
@@ -13,7 +13,7 @@
 from drunc.process_manager.interface.context import ProcessManagerContext
 from drunc.process_manager.utils import tabulate_process_instance_list
 from drunc.utils.shell_utils import InterruptedCommand, log_pm_cmd
-from drunc.utils.utils import get_logger
+from drunc.utils.utils import get_logger, resolve_context_peer
 
 
 @click.command("boot")
@@ -82,6 +82,7 @@ def boot(
         raise e
 
     controller_address = obj.get_driver("process_manager").controller_address
+    controller_address = resolve_context_peer(controller_address)
     if controller_address:
         obj.print(
             Panel(
diff --git a/src/drunc/utils/utils.py b/src/drunc/utils/utils.py
index 0109c00c4..2dc9456e3 100644
--- a/src/drunc/utils/utils.py
+++ b/src/drunc/utils/utils.py
@@ -524,6 +524,18 @@ def resolve_context_peer(peer: str) -> str:
         str: The original peer string, or a resolved ``host:port`` representation.
     """
 
+    if not peer:
+        return peer
+
+    # Some callers pass a plain host:port string without a transport prefix.
+    # Handle those directly instead of assuming the first token is always a transport.
+    if peer.startswith("[") or peer.count(":") == 1:
+        parsed = _parse_host_port(peer)
+        if parsed is not None:
+            host, port = parsed
+            resolved_host = _resolve_host(host)
+            return f"{resolved_host}:{port}"
+
     match = re.match(r"^(?P<transport>[^:]+):(?P<address>.+)$", peer)
     if not match:
         return peer
diff --git a/tests/utils/test_utils.py b/tests/utils/test_utils.py
index 94b3573d6..5094824b0 100644
--- a/tests/utils/test_utils.py
+++ b/tests/utils/test_utils.py
@@ -20,6 +20,7 @@
     now_str,
     parent_death_pact,
     regex_match,
+    resolve_context_peer,
     resolve_localhost_and_127_ip_to_network_ip,
     resolve_localhost_to_hostname,
     validate_command_facility,
@@ -157,6 +158,12 @@ def test_resolve_localhost_and_127_ip_to_network_ip():
     assert resolved == generate_address(this_ip)
 
 
+def test_resolve_context_peer():
+    assert resolve_context_peer("grpc:np04-srv-028:50000") == "np04-srv-028:50000"
+    assert resolve_context_peer("np04-srv-028:50000") == "np04-srv-028:50000"
+    assert resolve_context_peer("") == ""
+
+
 def test_host_is_local():
     this_ip = socket.gethostbyname(socket.gethostname())
     hostname = socket.gethostname()

From 053c4db41080edccceb257d24e28f258da90a93b Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Mon, 1 Jun 2026 15:36:59 +0200
Subject: [PATCH 05/14] [HACK] remoce shared context to enable controller shell
 to work

---
 src/drunc/apps/controller_shell.py            |  1 +
 src/drunc/controller/controller_driver.py     |  1 +
 src/drunc/controller/interface/shell.py       |  6 ++++--
 src/drunc/controller/interface/shell_utils.py | 16 +++++++++++++---
 4 files changed, 19 insertions(+), 5 deletions(-)

diff --git a/src/drunc/apps/controller_shell.py b/src/drunc/apps/controller_shell.py
index 5ff6ca7d0..79068740e 100644
--- a/src/drunc/apps/controller_shell.py
+++ b/src/drunc/apps/controller_shell.py
@@ -14,6 +14,7 @@ def main() -> None:
         log = get_logger("controller_shell", rich_handler=True)
         log.error("[red bold]:fire::fire: Exception thrown :fire::fire:")
         log.exception(e)
+        # raise e
         exit(1)
 
 
diff --git a/src/drunc/controller/controller_driver.py b/src/drunc/controller/controller_driver.py
index 376530660..6522058fb 100644
--- a/src/drunc/controller/controller_driver.py
+++ b/src/drunc/controller/controller_driver.py
@@ -126,6 +126,7 @@ def describe(
         execute_on_all_subsequent_children_in_path: bool = True,
         timeout: int | float = 60,
     ) -> DescribeResponse:
+        # self.log.("in describe")
         request = DescribeRequest(
             target=target,
             execute_along_path=execute_along_path,
diff --git a/src/drunc/controller/interface/shell.py b/src/drunc/controller/interface/shell.py
index 5c71bde1c..401dd9744 100644
--- a/src/drunc/controller/interface/shell.py
+++ b/src/drunc/controller/interface/shell.py
@@ -53,16 +53,18 @@ def controller_shell(ctx, controller_address: str, log_level: str) -> None:
     get_root_logger(log_level)
     controller_shell_log = get_logger("controller.iface.shell", rich_handler=True)
 
-    controller_shell_log.debug("Resetting the context instance address")
+    controller_shell_log.info("Resetting the context instance address")
     ctx.obj.reset(address=controller_address)
     ctx.call_on_close(controller_cleanup_wrapper(ctx.obj))
 
     desc = None
-    controller_shell_log.debug(
+    controller_shell_log.info(
         f"[green]{getpass.getuser()}[/green] connecting to the [green]controller[/green] through a [green]controller-shell[/green] via address [green]{controller_address}[/green]"
     )
     try:
+        controller_shell_log.info("test 1")
         desc = controller_setup(ctx.obj, controller_address)
+        controller_shell_log.info("test 2")
     except ServerUnreachable as e:
         controller_shell_log.critical("Could not connect to the controller")
         controller_shell_log.exception(
diff --git a/src/drunc/controller/interface/shell_utils.py b/src/drunc/controller/interface/shell_utils.py
index 3ffba7455..42e8a1572 100644
--- a/src/drunc/controller/interface/shell_utils.py
+++ b/src/drunc/controller/interface/shell_utils.py
@@ -46,7 +46,7 @@
     pack_to_any,
     unpack_any,
 )
-from drunc.utils.utils import format_name_for_cli, get_logger, get_shared_rich_console
+from drunc.utils.utils import format_name_for_cli, get_logger
 
 log = get_logger("controller.iface.shell_utils")
 
@@ -188,7 +188,8 @@ def __init__(self, ctx, refresh_per_second=2, *args, **kwargs) -> None:
         # Get the instance of the console that the logger is using with the rich handler
         # so that the progress bar can be rendered in the same console, and not mess up
         # the logs
-        shared_console = get_shared_rich_console(self.ctx.log)
+        # shared_console = get_shared_rich_console(self.ctx.log)
+        shared_console = None
         if shared_console:
             kwargs["console"] = shared_console
 
@@ -197,7 +198,7 @@ def __init__(self, ctx, refresh_per_second=2, *args, **kwargs) -> None:
     def update_table(self):
         # The following debug log line will be used in an integration test to validate
         # that issue 817 does not appear again (rich table overriding the log entries)
-        self.ctx.log.debug("Updating the status table...")
+        # self.ctx.log.debug("Updating the status table...")
         statuses = self.ctx.get_driver("controller").status()
         descriptions = self.ctx.get_driver("controller").describe()
         self.table = get_status_table(statuses, descriptions)
@@ -251,6 +252,8 @@ def controller_setup(ctx, controller_address):
 
     timeout = 60
 
+    log.warning("into progrss")
+
     with Progress(
         SpinnerColumn(),
         TextColumn("[progress.description]{task.description}"),
@@ -268,22 +271,29 @@ def controller_setup(ctx, controller_address):
         start_time = time.time()
         while time.time() - start_time < timeout:
             progress.update(waiting, completed=time.time() - start_time)
+            log.warning("trying this thingy")
 
             try:
+                log.warning("getting desc")
                 desc = ctx.get_driver("controller").describe().description
+                log.warning("desc done")
                 stored_exception = None
                 break
             except ServerUnreachable as e:
+                log.warning("yee hawe")
                 stored_exception = e
                 time.sleep(1)
 
             except Exception as e:
+                log.warning("weewoo")
                 ctx.critical("Could not get the controller's status")
                 ctx.critical(e)
                 ctx.critical("Exiting.")
                 ctx.terminate()
                 raise e
+            log.warning(start_time)
 
+    log.info("survived progress")
     if stored_exception is not None:
         raise stored_exception
 

From 5ce8e03352fa556abff3d5505cf6a24773fee039 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Tue, 2 Jun 2026 14:39:35 +0200
Subject: [PATCH 06/14] add some notes and logging

---
 src/drunc/connectivity_service/client.py      |   2 +-
 src/drunc/process_manager/oks_parser.py       |   7 +-
 .../process_manager/process_manager_driver.py | 102 +++++++++++++++++-
 .../process_manager/ssh_process_manager.py    |   2 +-
 4 files changed, 109 insertions(+), 4 deletions(-)

diff --git a/src/drunc/connectivity_service/client.py b/src/drunc/connectivity_service/client.py
index 87fa1406f..111b349c5 100644
--- a/src/drunc/connectivity_service/client.py
+++ b/src/drunc/connectivity_service/client.py
@@ -17,7 +17,7 @@ def __init__(self, session: str, address: str):
             # assume the simplest case here
             self.address = f"http://{address}"
 
-        self.log.debug(
+        self.log.error(
             f"Connectivity service address: {self.address}, session: {self.session}"
         )
 
diff --git a/src/drunc/process_manager/oks_parser.py b/src/drunc/process_manager/oks_parser.py
index e73a4cae4..bdb4d99d2 100644
--- a/src/drunc/process_manager/oks_parser.py
+++ b/src/drunc/process_manager/oks_parser.py
@@ -179,11 +179,16 @@ def collect_apps(
 
     # Add controller for this segment to list of apps
     controller = segment_obj.controller
+
     rc_env = defenv.copy()
     collect_variables(controller.application_environment, rc_env)
     rc_env["DUNEDAQ_APPLICATION_NAME"] = controller.id
+
     host = controller.runs_on.runs_on.id
+    # okay fine if we change this to 29 that works..
+    # but we will need this to be clean
     # host = "np04-srv-029.cern.ch"
+    log.info(f"main host is {host}")
 
     tree_id_str = ".".join(map(str, tree_prefix))
     apps.append(
@@ -263,7 +268,7 @@ def collect_apps(
 
         # host = "np04-srv-029.cern.ch"
 
-        log.info(f"the host is {host}")
+        log.info(f"app host is {host}")
 
         data_path = get_writer_directory_path(app, log)
         if not data_path:
diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index 5a29eadd4..3f2eed278 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -56,6 +56,88 @@
 class ProcessManagerDriver:
     controller_address = ""
 
+
+    def _dump_oks_value(self, value):
+        if value is None or isinstance(value, (str, int, float, bool)):
+            return value
+
+        if isinstance(value, (list, tuple, set)):
+            return [self._dump_oks_value(item) for item in value]
+
+        if hasattr(value, "className"):
+            class_name = value.className()
+            if class_name == "VariableSet":
+                return {
+                    "class": class_name,
+                    "contains": self._dump_oks_value(getattr(value, "contains", [])),
+                }
+
+            if class_name == "Variable":
+                return {
+                    "class": class_name,
+                    "name": getattr(value, "name", None),
+                    "value": getattr(value, "value", None),
+                }
+
+            snapshot = {
+                "class": class_name,
+                "id": getattr(value, "id", None),
+            }
+
+            field_map = {
+                "Session": [
+                    "log_path",
+                    "rte_script",
+                    "environment",
+                    "segment",
+                    "connectivity_service",
+                    "infrastructure_applications",
+                    "disabled",
+                ],
+                "Segment": ["controller", "segments", "applications", "disabled"],
+                "Application": [
+                    "application_name",
+                    "application_environment",
+                    "runs_on",
+                    "log_path",
+                    "data_path",
+                ],
+                "Controller": [
+                    "application_name",
+                    "application_environment",
+                    "runs_on",
+                    "exposes_service",
+                    "fsm",
+                    "log_path",
+                ],
+                "ConnectivityService": ["host", "service"],
+                "Service": ["host", "port"],
+                "Resource": ["id"],
+            }
+
+            for field_name in field_map.get(class_name, ["host", "service", "port", "name", "value"]):
+                try:
+                    field_value = getattr(value, field_name)
+                except Exception:
+                    continue
+                snapshot[field_name] = self._dump_oks_value(field_value)
+
+            return snapshot
+
+        if isinstance(value, dict):
+            return {key: self._dump_oks_value(item) for key, item in value.items()}
+
+        return repr(value)
+
+    def _dump_database(self, db: conffwk.Configuration) -> str:
+        configuration_file = db.active_database
+        full_db_path = get_full_db_path(configuration_file)
+        with open(full_db_path, encoding="utf-8") as config_file:
+            return config_file.read()
+
+    def _dump_session_dal(self, session_dal: "conffwk.dal.Session") -> str:
+        return json.dumps(self._dump_oks_value(session_dal), indent=2, sort_keys=False)
+
     def __init__(self, address: str, token: Token):
         self.log = get_logger("process_manager_driver", rich_handler=True)
         self.address = address
@@ -145,13 +227,22 @@ def boot(
         # Step 3 - check for port conflicts and update configuration/DAL as needed
         db, session_dal = self.check_port_conflicts(db, session_dal)
 
+        # Step 3.5 - modify the host mapping (explicitly!!)
+        # self.log.warning("prechange session_dal:\n%s", self._dump_session_dal(session_dal))
+
+        # session_dal = self.update_dal_host(session_dal)
+
+
+        # self.log.warning("db dump:\n%s", self._dump_database(db))
+        # self.log.warning("postchange session_dal:\n%s", self._dump_session_dal(session_dal))
+
         self.log.info("step 4")
         # Step 4 - connect to the connection service
         csc, connection_server, connection_port = self._connect_to_service(
             session_dal, session_name
         )
 
-        self.log.info("step 5")
+        self.log.info(f"step 5 + {csc.address}")
         # Step 5 - track boot timings per host
         last_boot_on_host_at = {}
         previous_host = None
@@ -250,6 +341,7 @@ def _collect_all_apps(
 
         apps = infra_apps + apps
 
+        self.log.warning("json jumps")
         self.log.info(f"{json.dumps(apps, indent=4)}")
 
         return apps
@@ -428,6 +520,13 @@ def update_connectivity_port_dal(
                     if item.name == "CONNECTION_PORT":
                         item.value = new_port
 
+
+    
+    def update_dal_host(self, session_dal):
+        session_dal.connectivity_service.host = "np04-srv-029"
+        return session_dal
+
+
     def check_port_conflicts(
         self, db: conffwk.Configuration, session_dal: "conffwk.dal.Session"
     ) -> tuple[conffwk.Configuration, "conffwk.dal.Session"]:
@@ -572,6 +671,7 @@ def _connect_to_service(
                 )
                 connection_server = resolved_server
 
+            self.log.info(f"connectivity service: {connection_server=}")
             # Funnily enough this is the minimum amount of change needed...s
             # so basically instead of trying to resolve the localhost hostname, resolve to whatever we are connecting to?
             # connection_server = "np04-srv-028"
diff --git a/src/drunc/process_manager/ssh_process_manager.py b/src/drunc/process_manager/ssh_process_manager.py
index 09a2f55cd..f4eef6679 100644
--- a/src/drunc/process_manager/ssh_process_manager.py
+++ b/src/drunc/process_manager/ssh_process_manager.py
@@ -394,7 +394,7 @@ def __boot(self, boot_request: BootRequest, uuid: str) -> ProcessInstance:
         self.log.info(
             f"Booted '{boot_request.process_description.metadata.name}' "
             f"from session '{boot_request.process_description.metadata.session}' "
-            f"with UUID {uuid}"
+            f"with UUID {uuid} in hosts '{boot_request.process_restriction.allowed_hosts}"
         )
 
         # Query current process status

From 33e0ae628de791dc694462b70c5775fcb58d2510 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Tue, 2 Jun 2026 14:40:06 +0200
Subject: [PATCH 07/14] new, add minimal working for it to run anywhere?

---
 .../process_manager/process_manager_driver.py   | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index 3f2eed278..478e400bb 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -56,7 +56,6 @@
 class ProcessManagerDriver:
     controller_address = ""
 
-
     def _dump_oks_value(self, value):
         if value is None or isinstance(value, (str, int, float, bool)):
             return value
@@ -115,7 +114,9 @@ def _dump_oks_value(self, value):
                 "Resource": ["id"],
             }
 
-            for field_name in field_map.get(class_name, ["host", "service", "port", "name", "value"]):
+            for field_name in field_map.get(
+                class_name, ["host", "service", "port", "name", "value"]
+            ):
                 try:
                     field_value = getattr(value, field_name)
                 except Exception:
@@ -232,7 +233,6 @@ def boot(
 
         # session_dal = self.update_dal_host(session_dal)
 
-
         # self.log.warning("db dump:\n%s", self._dump_database(db))
         # self.log.warning("postchange session_dal:\n%s", self._dump_session_dal(session_dal))
 
@@ -390,7 +390,9 @@ def _build_boot_request(
         override_logs: bool,
         pwd: str,
     ) -> BootRequest:
-        host = format_hostname(app["restriction"])
+        # host = format_hostname(app["restriction"])
+        host = "np04-srv-029"  #! GOD BLESS THIS WORKSS
+        # this is one of the two minimal changes needed to get this working in general?
         name = app["name"]
         exe = app["type"]
         args = app["args"]
@@ -520,13 +522,10 @@ def update_connectivity_port_dal(
                     if item.name == "CONNECTION_PORT":
                         item.value = new_port
 
-
-    
     def update_dal_host(self, session_dal):
         session_dal.connectivity_service.host = "np04-srv-029"
         return session_dal
 
-
     def check_port_conflicts(
         self, db: conffwk.Configuration, session_dal: "conffwk.dal.Session"
     ) -> tuple[conffwk.Configuration, "conffwk.dal.Session"]:
@@ -661,7 +660,9 @@ def _connect_to_service(
                 self.log.info("Apparently its localhost?")
                 # resolved_server = resolve_localhost_to_hostname(connection_server)
 
-                test_address = self.address
+                # test_address = self.address
+                test_address = "np04-srv-029"
+                # this is one of the two minimal changes needed to get this working in general?
                 if "://" not in test_address:
                     test_address = "grpc://" + test_address
 

From 0107676f7d9f55dce1b1c2572dae9f9f8775a187 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Tue, 2 Jun 2026 14:58:47 +0200
Subject: [PATCH 08/14] finally the multi host stuff works

---
 .../process_manager/process_manager_driver.py   | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index 478e400bb..4c6c33294 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -390,8 +390,14 @@ def _build_boot_request(
         override_logs: bool,
         pwd: str,
     ) -> BootRequest:
-        # host = format_hostname(app["restriction"])
-        host = "np04-srv-029"  #! GOD BLESS THIS WORKSS
+        host_test = format_hostname(app["restriction"])
+        resolved_server_test = resolve_localhost_to_hostname(host_test)
+        self.log.info(f"boot resolve {resolved_server_test}")
+
+        host = resolved_server_test
+
+        # host = "np04-srv-029"  #! GOD BLESS THIS WORKSS
+
         # this is one of the two minimal changes needed to get this working in general?
         name = app["name"]
         exe = app["type"]
@@ -658,10 +664,13 @@ def _connect_to_service(
 
             if connection_server == "localhost":
                 self.log.info("Apparently its localhost?")
-                # resolved_server = resolve_localhost_to_hostname(connection_server)
+                resolved_server_test = resolve_localhost_to_hostname(connection_server)
+                self.log.info(resolved_server_test)
 
                 # test_address = self.address
-                test_address = "np04-srv-029"
+                # test_address = "np04-srv-029"
+                test_address = resolved_server_test
+
                 # this is one of the two minimal changes needed to get this working in general?
                 if "://" not in test_address:
                     test_address = "grpc://" + test_address

From 59944672a5f5c1b7620678d5a33db81e7d25de49 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Thu, 4 Jun 2026 11:39:37 +0200
Subject: [PATCH 09/14] pawel push

---
 src/drunc/process_manager/process_manager_driver.py | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index 4c6c33294..4c3fb4ee6 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -599,6 +599,10 @@ def check_port_conflicts(
         if session_dal.infrastructure_applications:  # Check if the own application needs to be spawned, or if an externally managed one is in use (e.g. if using ehn1 connectivity service or integration tests.)
             connectivity_service_host: str = session_dal.connectivity_service.host
             connectivity_service_port = session_dal.connectivity_service.service.port
+            self.log.critical(
+                f"{connectivity_service_host=}, {connectivity_service_port=}"
+            )
+
             if not is_port_available(
                 connectivity_service_host, connectivity_service_port
             ):

From df9c2a097701c97994f2d5550a9d3f8e5e833818 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Fri, 5 Jun 2026 17:49:56 +0200
Subject: [PATCH 10/14] clean up comments

---
 src/drunc/apps/controller_shell.py            |  1 -
 src/drunc/connectivity_service/client.py      | 12 ++---
 src/drunc/controller/controller_driver.py     |  1 -
 src/drunc/controller/interface/shell.py       |  6 +--
 src/drunc/controller/interface/shell_utils.py | 16 ++-----
 src/drunc/process_manager/configuration.py    |  2 -
 src/drunc/process_manager/oks_parser.py       | 10 ----
 .../process_manager/process_manager_driver.py | 47 ++-----------------
 8 files changed, 14 insertions(+), 81 deletions(-)

diff --git a/src/drunc/apps/controller_shell.py b/src/drunc/apps/controller_shell.py
index 79068740e..5ff6ca7d0 100644
--- a/src/drunc/apps/controller_shell.py
+++ b/src/drunc/apps/controller_shell.py
@@ -14,7 +14,6 @@ def main() -> None:
         log = get_logger("controller_shell", rich_handler=True)
         log.error("[red bold]:fire::fire: Exception thrown :fire::fire:")
         log.exception(e)
-        # raise e
         exit(1)
 
 
diff --git a/src/drunc/connectivity_service/client.py b/src/drunc/connectivity_service/client.py
index 111b349c5..5e7eb124f 100644
--- a/src/drunc/connectivity_service/client.py
+++ b/src/drunc/connectivity_service/client.py
@@ -17,7 +17,7 @@ def __init__(self, session: str, address: str):
             # assume the simplest case here
             self.address = f"http://{address}"
 
-        self.log.error(
+        self.log.debug(
             f"Connectivity service address: {self.address}, session: {self.session}"
         )
 
@@ -41,16 +41,16 @@ def is_ready(self, timeout: int = 10):
             attempt += 1
             elapsed = time.time() - start
 
-            self.log.info(f"Health check attempt {attempt} at {elapsed:.2f}s elapsed")
-            self.log.info(f"Polling address: {self.address}")
+            self.log.debug(f"Health check attempt {attempt} at {elapsed:.2f}s elapsed")
+            self.log.debug(f"Polling address: {self.address}")
             try:
                 r = get(self.address)
             except Exception as e:
-                self.log.info(f"Polling failed with exception: {e}")
+                self.log.debug(f"Polling failed with exception: {e}")
                 r = None
             # Request failed - service is NOT ready.
             if r is None or not r.ok:
-                self.log.info(
+                self.log.debug(
                     f"Connectivity service not ready, retrying in {delay:.2f}s"
                 )
                 time.sleep(delay)
@@ -58,7 +58,7 @@ def is_ready(self, timeout: int = 10):
                 delay = min(delay * 2, max_delay)
             # Request succeeded - service is ready.
             else:
-                self.log.info(
+                self.log.debug(
                     f"Connectivity service ready after {attempt} attempts ({elapsed:.2f}s)"
                 )
                 return True
diff --git a/src/drunc/controller/controller_driver.py b/src/drunc/controller/controller_driver.py
index 6522058fb..376530660 100644
--- a/src/drunc/controller/controller_driver.py
+++ b/src/drunc/controller/controller_driver.py
@@ -126,7 +126,6 @@ def describe(
         execute_on_all_subsequent_children_in_path: bool = True,
         timeout: int | float = 60,
     ) -> DescribeResponse:
-        # self.log.("in describe")
         request = DescribeRequest(
             target=target,
             execute_along_path=execute_along_path,
diff --git a/src/drunc/controller/interface/shell.py b/src/drunc/controller/interface/shell.py
index 401dd9744..5c71bde1c 100644
--- a/src/drunc/controller/interface/shell.py
+++ b/src/drunc/controller/interface/shell.py
@@ -53,18 +53,16 @@ def controller_shell(ctx, controller_address: str, log_level: str) -> None:
     get_root_logger(log_level)
     controller_shell_log = get_logger("controller.iface.shell", rich_handler=True)
 
-    controller_shell_log.info("Resetting the context instance address")
+    controller_shell_log.debug("Resetting the context instance address")
     ctx.obj.reset(address=controller_address)
     ctx.call_on_close(controller_cleanup_wrapper(ctx.obj))
 
     desc = None
-    controller_shell_log.info(
+    controller_shell_log.debug(
         f"[green]{getpass.getuser()}[/green] connecting to the [green]controller[/green] through a [green]controller-shell[/green] via address [green]{controller_address}[/green]"
     )
     try:
-        controller_shell_log.info("test 1")
         desc = controller_setup(ctx.obj, controller_address)
-        controller_shell_log.info("test 2")
     except ServerUnreachable as e:
         controller_shell_log.critical("Could not connect to the controller")
         controller_shell_log.exception(
diff --git a/src/drunc/controller/interface/shell_utils.py b/src/drunc/controller/interface/shell_utils.py
index 42e8a1572..3ffba7455 100644
--- a/src/drunc/controller/interface/shell_utils.py
+++ b/src/drunc/controller/interface/shell_utils.py
@@ -46,7 +46,7 @@
     pack_to_any,
     unpack_any,
 )
-from drunc.utils.utils import format_name_for_cli, get_logger
+from drunc.utils.utils import format_name_for_cli, get_logger, get_shared_rich_console
 
 log = get_logger("controller.iface.shell_utils")
 
@@ -188,8 +188,7 @@ def __init__(self, ctx, refresh_per_second=2, *args, **kwargs) -> None:
         # Get the instance of the console that the logger is using with the rich handler
         # so that the progress bar can be rendered in the same console, and not mess up
         # the logs
-        # shared_console = get_shared_rich_console(self.ctx.log)
-        shared_console = None
+        shared_console = get_shared_rich_console(self.ctx.log)
         if shared_console:
             kwargs["console"] = shared_console
 
@@ -198,7 +197,7 @@ def __init__(self, ctx, refresh_per_second=2, *args, **kwargs) -> None:
     def update_table(self):
         # The following debug log line will be used in an integration test to validate
         # that issue 817 does not appear again (rich table overriding the log entries)
-        # self.ctx.log.debug("Updating the status table...")
+        self.ctx.log.debug("Updating the status table...")
         statuses = self.ctx.get_driver("controller").status()
         descriptions = self.ctx.get_driver("controller").describe()
         self.table = get_status_table(statuses, descriptions)
@@ -252,8 +251,6 @@ def controller_setup(ctx, controller_address):
 
     timeout = 60
 
-    log.warning("into progrss")
-
     with Progress(
         SpinnerColumn(),
         TextColumn("[progress.description]{task.description}"),
@@ -271,29 +268,22 @@ def controller_setup(ctx, controller_address):
         start_time = time.time()
         while time.time() - start_time < timeout:
             progress.update(waiting, completed=time.time() - start_time)
-            log.warning("trying this thingy")
 
             try:
-                log.warning("getting desc")
                 desc = ctx.get_driver("controller").describe().description
-                log.warning("desc done")
                 stored_exception = None
                 break
             except ServerUnreachable as e:
-                log.warning("yee hawe")
                 stored_exception = e
                 time.sleep(1)
 
             except Exception as e:
-                log.warning("weewoo")
                 ctx.critical("Could not get the controller's status")
                 ctx.critical(e)
                 ctx.critical("Exiting.")
                 ctx.terminate()
                 raise e
-            log.warning(start_time)
 
-    log.info("survived progress")
     if stored_exception is not None:
         raise stored_exception
 
diff --git a/src/drunc/process_manager/configuration.py b/src/drunc/process_manager/configuration.py
index fa6e821cd..da9198722 100644
--- a/src/drunc/process_manager/configuration.py
+++ b/src/drunc/process_manager/configuration.py
@@ -149,8 +149,6 @@ def get_commandline_parameters(
         Ordered list of command-line arguments for process launch.
     """
     runs_on = obj.runs_on.runs_on.id
-    # runs_on = "np04-srv-029.cern.ch"
-
     control_service_port = -1
     control_service_protocol = ""
     for svc in obj.exposes_service:
diff --git a/src/drunc/process_manager/oks_parser.py b/src/drunc/process_manager/oks_parser.py
index bdb4d99d2..5feb6ba50 100644
--- a/src/drunc/process_manager/oks_parser.py
+++ b/src/drunc/process_manager/oks_parser.py
@@ -179,16 +179,10 @@ def collect_apps(
 
     # Add controller for this segment to list of apps
     controller = segment_obj.controller
-
     rc_env = defenv.copy()
     collect_variables(controller.application_environment, rc_env)
     rc_env["DUNEDAQ_APPLICATION_NAME"] = controller.id
-
     host = controller.runs_on.runs_on.id
-    # okay fine if we change this to 29 that works..
-    # but we will need this to be clean
-    # host = "np04-srv-029.cern.ch"
-    log.info(f"main host is {host}")
 
     tree_id_str = ".".join(map(str, tree_prefix))
     apps.append(
@@ -266,10 +260,6 @@ def collect_apps(
         )
         log.debug(f"Collecting app {app.id} with args {args}")
 
-        # host = "np04-srv-029.cern.ch"
-
-        log.info(f"app host is {host}")
-
         data_path = get_writer_directory_path(app, log)
         if not data_path:
             log.debug(f"No data path found for app {app.id}")
diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index 4c3fb4ee6..8446fb218 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -213,9 +213,6 @@ def boot(
     ) -> Iterator[ProcessInstanceList] | None:
         self.log.info(f"Booting session [green]{session_name}[/green]")
 
-        self.log.warning(
-            f"parameters: {conf_file=}, {conf_id=}, {user=}, {self.address=}"
-        )
         # Assume oksconflibs if no framework is defined
         conf_file = f"oksconflibs:{conf_file}" if ":" not in conf_file else conf_file
 
@@ -228,26 +225,15 @@ def boot(
         # Step 3 - check for port conflicts and update configuration/DAL as needed
         db, session_dal = self.check_port_conflicts(db, session_dal)
 
-        # Step 3.5 - modify the host mapping (explicitly!!)
-        # self.log.warning("prechange session_dal:\n%s", self._dump_session_dal(session_dal))
-
-        # session_dal = self.update_dal_host(session_dal)
-
-        # self.log.warning("db dump:\n%s", self._dump_database(db))
-        # self.log.warning("postchange session_dal:\n%s", self._dump_session_dal(session_dal))
-
-        self.log.info("step 4")
         # Step 4 - connect to the connection service
         csc, connection_server, connection_port = self._connect_to_service(
             session_dal, session_name
         )
 
-        self.log.info(f"step 5 + {csc.address}")
         # Step 5 - track boot timings per host
         last_boot_on_host_at = {}
         previous_host = None
 
-        self.log.info("step 6")
         # Step 6: iterate over boot requests
         for request in self._convert_oks_to_boot_request(
             oks_conf=conf_file,
@@ -257,8 +243,6 @@ def boot(
             override_logs=override_logs,
             **kwargs,
         ):
-            self.log.warning(request)
-
             if not request:
                 self.log.error("[red]No boot request was generated, ending boot.[/red]")
                 return None
@@ -269,11 +253,10 @@ def boot(
                     f"Skipping connectivity service readiness check for application {request.process_description.metadata.name}"
                 )
             else:
-                self.log.info(
+                self.log.debug(
                     f"Checking connectivity service readiness before booting application {request.process_description.metadata.name}"
                 )
                 if csc and not csc.is_ready(timeout=10):
-                    self.log.info("whoops its not ready?")
                     raise DruncSetupException(
                         "Connectivity service did not respond within timeout."
                     )
@@ -306,7 +289,6 @@ def boot(
                     )
                 handle_grpc_error(e)
 
-        self.log.info("step 7")
         # Step 7: discover segment root controller
         self._discover_controller(
             session_dal, session_name, csc, connection_server, connection_port
@@ -341,8 +323,7 @@ def _collect_all_apps(
 
         apps = infra_apps + apps
 
-        self.log.warning("json jumps")
-        self.log.info(f"{json.dumps(apps, indent=4)}")
+        self.log.debug(f"{json.dumps(apps, indent=4)}")
 
         return apps
 
@@ -393,11 +374,7 @@ def _build_boot_request(
         host_test = format_hostname(app["restriction"])
         resolved_server_test = resolve_localhost_to_hostname(host_test)
         self.log.info(f"boot resolve {resolved_server_test}")
-
         host = resolved_server_test
-
-        # host = "np04-srv-029"  #! GOD BLESS THIS WORKSS
-
         # this is one of the two minimal changes needed to get this working in general?
         name = app["name"]
         exe = app["type"]
@@ -599,10 +576,6 @@ def check_port_conflicts(
         if session_dal.infrastructure_applications:  # Check if the own application needs to be spawned, or if an externally managed one is in use (e.g. if using ehn1 connectivity service or integration tests.)
             connectivity_service_host: str = session_dal.connectivity_service.host
             connectivity_service_port = session_dal.connectivity_service.service.port
-            self.log.critical(
-                f"{connectivity_service_host=}, {connectivity_service_port=}"
-            )
-
             if not is_port_available(
                 connectivity_service_host, connectivity_service_port
             ):
@@ -661,18 +634,12 @@ def _connect_to_service(
         self, session_dal: "conffwk.dal.Session", session_name: str
     ) -> ConnectivityServiceClient | None:
         if session_dal.connectivity_service:
-            self.log.info("we have connectivity service")
-
             connection_server = session_dal.connectivity_service.host
             connection_port = session_dal.connectivity_service.service.port
 
             if connection_server == "localhost":
-                self.log.info("Apparently its localhost?")
                 resolved_server_test = resolve_localhost_to_hostname(connection_server)
-                self.log.info(resolved_server_test)
 
-                # test_address = self.address
-                # test_address = "np04-srv-029"
                 test_address = resolved_server_test
 
                 # this is one of the two minimal changes needed to get this working in general?
@@ -680,22 +647,14 @@ def _connect_to_service(
                     test_address = "grpc://" + test_address
 
                 resolved_server = urlparse(test_address).hostname
-                self.log.info(
+                self.log.debug(
                     f"Resolved connection server 'localhost' to '{resolved_server}' to avoid K8s hairpinning."
                 )
                 connection_server = resolved_server
 
-            self.log.info(f"connectivity service: {connection_server=}")
-            # Funnily enough this is the minimum amount of change needed...s
-            # so basically instead of trying to resolve the localhost hostname, resolve to whatever we are connecting to?
-            # connection_server = "np04-srv-028"
             client = ConnectivityServiceClient(
                 session_name, f"{connection_server}:{connection_port}"
             )
-
-            self.log.info(
-                f"returning {client=}, {connection_server=}, {connection_port=}"
-            )
             return client, connection_server, connection_port
         return None, None, None
 

From 926814e83fce1cf9b7745e3d2fb859a375838f2f Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Fri, 5 Jun 2026 17:52:47 +0200
Subject: [PATCH 11/14] Clean up even more

---
 .../process_manager/process_manager_driver.py | 83 -------------------
 1 file changed, 83 deletions(-)

diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index 8446fb218..155798263 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -56,89 +56,6 @@
 class ProcessManagerDriver:
     controller_address = ""
 
-    def _dump_oks_value(self, value):
-        if value is None or isinstance(value, (str, int, float, bool)):
-            return value
-
-        if isinstance(value, (list, tuple, set)):
-            return [self._dump_oks_value(item) for item in value]
-
-        if hasattr(value, "className"):
-            class_name = value.className()
-            if class_name == "VariableSet":
-                return {
-                    "class": class_name,
-                    "contains": self._dump_oks_value(getattr(value, "contains", [])),
-                }
-
-            if class_name == "Variable":
-                return {
-                    "class": class_name,
-                    "name": getattr(value, "name", None),
-                    "value": getattr(value, "value", None),
-                }
-
-            snapshot = {
-                "class": class_name,
-                "id": getattr(value, "id", None),
-            }
-
-            field_map = {
-                "Session": [
-                    "log_path",
-                    "rte_script",
-                    "environment",
-                    "segment",
-                    "connectivity_service",
-                    "infrastructure_applications",
-                    "disabled",
-                ],
-                "Segment": ["controller", "segments", "applications", "disabled"],
-                "Application": [
-                    "application_name",
-                    "application_environment",
-                    "runs_on",
-                    "log_path",
-                    "data_path",
-                ],
-                "Controller": [
-                    "application_name",
-                    "application_environment",
-                    "runs_on",
-                    "exposes_service",
-                    "fsm",
-                    "log_path",
-                ],
-                "ConnectivityService": ["host", "service"],
-                "Service": ["host", "port"],
-                "Resource": ["id"],
-            }
-
-            for field_name in field_map.get(
-                class_name, ["host", "service", "port", "name", "value"]
-            ):
-                try:
-                    field_value = getattr(value, field_name)
-                except Exception:
-                    continue
-                snapshot[field_name] = self._dump_oks_value(field_value)
-
-            return snapshot
-
-        if isinstance(value, dict):
-            return {key: self._dump_oks_value(item) for key, item in value.items()}
-
-        return repr(value)
-
-    def _dump_database(self, db: conffwk.Configuration) -> str:
-        configuration_file = db.active_database
-        full_db_path = get_full_db_path(configuration_file)
-        with open(full_db_path, encoding="utf-8") as config_file:
-            return config_file.read()
-
-    def _dump_session_dal(self, session_dal: "conffwk.dal.Session") -> str:
-        return json.dumps(self._dump_oks_value(session_dal), indent=2, sort_keys=False)
-
     def __init__(self, address: str, token: Token):
         self.log = get_logger("process_manager_driver", rich_handler=True)
         self.address = address

From b54a79c6faea552d87fb7b7a3dfa3b47305d3c01 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Mon, 8 Jun 2026 11:02:04 +0200
Subject: [PATCH 12/14] block boot on existing sessions

---
 src/drunc/process_manager/interface/commands.py | 14 ++++++--------
 src/drunc/unified_shell/commands.py             | 12 ++++++------
 2 files changed, 12 insertions(+), 14 deletions(-)

diff --git a/src/drunc/process_manager/interface/commands.py b/src/drunc/process_manager/interface/commands.py
index eefbbb1d5..b2644b288 100644
--- a/src/drunc/process_manager/interface/commands.py
+++ b/src/drunc/process_manager/interface/commands.py
@@ -45,18 +45,16 @@ def boot(
 ) -> None:
     log = get_logger("process_manager.shell")
     log_pm_cmd(obj)
-    processes = obj.get_driver("process_manager").ps(ProcessQuery(user=user))
+    processes = obj.get_driver("process_manager").ps(
+        ProcessQuery(user=user, session=session_name)
+    )
 
     if len(processes.values) > 0:
-        log.info(
-            f"Note that there are already {len(processes.values)} processes running."
+        click.confirm(
+            f"You already have {len(processes.values)} processes running for {session_name}, are you sure you want to boot a session?",
+            abort=True,
         )
 
-        # click.confirm(
-        #     f"You already have {len(processes.values)} processes running, are you sure you want to boot a session?",
-        #     abort=True,
-        # )
-
     log.debug(
         f"Booting session {session_name} with boot configuration file {configuration_file} and id {configuration_id}, requested by user {user}"
     )
diff --git a/src/drunc/unified_shell/commands.py b/src/drunc/unified_shell/commands.py
index be093c134..7dc1bc950 100644
--- a/src/drunc/unified_shell/commands.py
+++ b/src/drunc/unified_shell/commands.py
@@ -58,12 +58,12 @@ def boot(
         override_logs_boot = obj.override_logs
     else:
         override_logs_boot = override_logs
-    # if len(processes.values) > 0:
-    #     log.error(
-    #         f"Cannot boot: session {session_name} already has {len(processes.values)} processes running. "
-    #         "Please terminate the existing session first."
-    #     )
-    #     return
+    if len(processes.values) > 0:
+        log.error(
+            f"Cannot boot: session {session_name} already has {len(processes.values)} processes running. "
+            "Please terminate the existing session first."
+        )
+        return
 
     try:
         results = obj.get_driver("process_manager").boot(

From a26a9b2858c4016363908d3943d5a154473020aa Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Mon, 8 Jun 2026 12:50:42 +0200
Subject: [PATCH 13/14] Cleanup

---
 .../process_manager/process_manager_driver.py | 25 +++++++------------
 1 file changed, 9 insertions(+), 16 deletions(-)

diff --git a/src/drunc/process_manager/process_manager_driver.py b/src/drunc/process_manager/process_manager_driver.py
index 155798263..aa38f314a 100644
--- a/src/drunc/process_manager/process_manager_driver.py
+++ b/src/drunc/process_manager/process_manager_driver.py
@@ -288,10 +288,10 @@ def _build_boot_request(
         override_logs: bool,
         pwd: str,
     ) -> BootRequest:
-        host_test = format_hostname(app["restriction"])
-        resolved_server_test = resolve_localhost_to_hostname(host_test)
-        self.log.info(f"boot resolve {resolved_server_test}")
-        host = resolved_server_test
+        # Run mapping to physical hostname to enable multi host usage
+        host = resolve_localhost_to_hostname(format_hostname(app["restriction"]))
+        self.log.info(f"boot resolve {host}")  # keep this until big PR gets merged
+
         # this is one of the two minimal changes needed to get this working in general?
         name = app["name"]
         exe = app["type"]
@@ -422,10 +422,6 @@ def update_connectivity_port_dal(
                     if item.name == "CONNECTION_PORT":
                         item.value = new_port
 
-    def update_dal_host(self, session_dal):
-        session_dal.connectivity_service.host = "np04-srv-029"
-        return session_dal
-
     def check_port_conflicts(
         self, db: conffwk.Configuration, session_dal: "conffwk.dal.Session"
     ) -> tuple[conffwk.Configuration, "conffwk.dal.Session"]:
@@ -555,15 +551,12 @@ def _connect_to_service(
             connection_port = session_dal.connectivity_service.service.port
 
             if connection_server == "localhost":
-                resolved_server_test = resolve_localhost_to_hostname(connection_server)
-
-                test_address = resolved_server_test
-
-                # this is one of the two minimal changes needed to get this working in general?
-                if "://" not in test_address:
-                    test_address = "grpc://" + test_address
+                # resolve localhost to hostname to enable multi host
+                resolved_address = resolve_localhost_to_hostname(connection_server)
+                if "://" not in resolved_address:
+                    resolved_address = "grpc://" + resolved_address
 
-                resolved_server = urlparse(test_address).hostname
+                resolved_server = urlparse(resolved_address).hostname
                 self.log.debug(
                     f"Resolved connection server 'localhost' to '{resolved_server}' to avoid K8s hairpinning."
                 )

From e26bc6d98b427f53dfc781fd025429758f0a5763 Mon Sep 17 00:00:00 2001
From: Emir Muhammad <emir.muhammad@cern.ch>
Date: Mon, 8 Jun 2026 15:06:39 +0200
Subject: [PATCH 14/14] Fix pytest

---
 src/drunc/process_manager/interface/commands.py  | 2 +-
 tests/process_manager/conftest.py                | 6 ++++--
 tests/process_manager/interface/test_commands.py | 5 +++--
 3 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/src/drunc/process_manager/interface/commands.py b/src/drunc/process_manager/interface/commands.py
index b2644b288..b5a479086 100644
--- a/src/drunc/process_manager/interface/commands.py
+++ b/src/drunc/process_manager/interface/commands.py
@@ -289,7 +289,7 @@ def logs_impl(
     log = get_logger("process_manager.shell")
     # TODO: MOVE BACK TO DEBUG BEFORE MERGE
     # THIS IS USEFUL FOR TESTING THOUGH
-    log.error(f"Running logs with query {query}")
+    log.debug(f"Running logs with query {query}")
     log_req = LogRequest(
         how_far=how_far,
         query=query,
diff --git a/tests/process_manager/conftest.py b/tests/process_manager/conftest.py
index 6a2ba4bfd..78aed2525 100644
--- a/tests/process_manager/conftest.py
+++ b/tests/process_manager/conftest.py
@@ -7,6 +7,8 @@
 to be back in line with druncschema definitions.
 """
 
+import socket
+
 import google.protobuf.any_pb2
 import pytest
 from druncschema.description_pb2 import Description
@@ -36,7 +38,7 @@ def app_data():
     Provides a mock application dictionary with required keys.
     """
     return {
-        "restriction": "localhost",
+        "restriction": socket.gethostname(),
         "name": "TestApp",
         "type": "binary",
         "args": ["--arg1"],
@@ -55,7 +57,7 @@ def bootrequest(app_data):
                 user="test_user",
                 session="session1",
                 name=app_data["name"],
-                hostname="localhost",
+                hostname=socket.gethostname(),
                 tree_id=app_data["tree_id"],
             ),
             executable_and_arguments=[{"exec": "binary", "args": ["--arg1"]}],
diff --git a/tests/process_manager/interface/test_commands.py b/tests/process_manager/interface/test_commands.py
index 6ba1e10ad..f1a5c5e4b 100644
--- a/tests/process_manager/interface/test_commands.py
+++ b/tests/process_manager/interface/test_commands.py
@@ -237,8 +237,9 @@ def test_boot_exiting_processes_abort(boot_arguments):
     # check that 'boot' was never called
     mock_driver.boot.assert_not_called()
 
+    # conf-id-123 from session name in this file
     assert (
-        "You already have 2 processes running, are you sure you want to boot a session?"
+        "You already have 2 processes running for conf-id-123, are you sure you want to boot a session?"
         in result.output
     )
 
@@ -267,7 +268,7 @@ def test_boot_exiting_processes_user_confirm(boot_arguments):
     mock_driver.boot.assert_called()
 
     assert (
-        "You already have 2 processes running, are you sure you want to boot a session?"
+        "You already have 2 processes running for conf-id-123, are you sure you want to boot a session?"
         in result.output
     )