gitea_admin/sientia-dataops-opc-ingestor
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

SIENTIAPDE-1083: add prometheus metrics to ingestor_manager.

Browse Source
This commit is contained in:
Bruno Domingues
2025-05-29 17:47:53 -03:00
parent c49f04ded6
commit 0fe2343891
2 changed files with 39 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
ingestor/managers/ingestor_manager.py
View File

@@ -7,6 +7,7 @@ from sientia_do.notifications.models import NotificationLevel
from ingestor.managers.data_manager import DataManager
from ingestor.managers.opc_manager import OpcManager
from ingestor.managers.resource_manager import ResourceManager
import ingestor.metrics as metrics
class IngestorManager():
@@ -162,6 +163,8 @@ class IngestorManager():
self.opc_managers[server].disconnect()
self.opc_managers.pop(server, None)
metrics.OPC_MANAGERS_ACTIVE.labels(pod_id=self.pod_id).set(len(self.opc_managers))
def check_opc_servers_integrity(self):
"""
Checks the integrity of the OPC servers and updates the OPC servers if necessary.
@@ -179,6 +182,8 @@ class IngestorManager():
for slot, _config in self.managed_tags.items():
self.managed_tags[slot].pop(server, None)
metrics.OPC_MANAGERS_ACTIVE.labels(pod_id=self.pod_id).set(len(self.opc_managers))
def declare_active(self):
"""
Declares the ingestor as active by sending a heartbeat signal to the resource manager.
@@ -211,6 +216,7 @@ class IngestorManager():
leases = self.resource_manager.get_all_leases()
self.number_of_slots = len(leases) if leases else 0
metrics.LEASES_TOTAL.set(self.number_of_slots)
return self.number_of_slots
def get_number_of_slots(self) -> int:
@@ -224,6 +230,7 @@ class IngestorManager():
slots = self.resource_manager.get_all_slots()
self.number_of_slots = len(slots) if slots else 0
metrics.SLOTS_TOTAL.set(self.number_of_slots)
return self.number_of_slots
def get_slot_leases(self, max_slots: int = 1) -> Dict:
@@ -255,9 +262,11 @@ class IngestorManager():
if slots is None:
continue
acquired[str(i)] = slots
metrics.SLOTS_ACQUIRED.labels(pod_id=self.pod_id).inc()
if len(acquired) >= max_slots:
self.managed_tags.update(acquired)
metrics.SLOTS_MANAGED.labels(pod_id=self.pod_id).set(len(self.managed_tags))
return acquired
self.logger.warning(
@@ -265,6 +274,7 @@ class IngestorManager():
f"Only {acquired} slots were leased."
)
self.managed_tags.update(acquired)
metrics.SLOTS_MANAGED.labels(pod_id=self.pod_id).set(len(self.managed_tags))
return acquired
def unsubscribe_slot(self, slot: str):
@@ -332,6 +342,7 @@ class IngestorManager():
for lease_id in ids:
self.resource_manager.drop_tag_lease(lease_id)
metrics.SLOTS_RELEASED.labels(pod_id=self.pod_id).inc()
def manage_server(self, slot: str, server: str, server_config: dict, tags: dict) -> int:
"""
@@ -390,6 +401,7 @@ class IngestorManager():
tags_to_sub
)
except Exception as e:
metrics.OPC_SUBSCRIPTION_ERRORS.labels(pod_id=self.pod_id, server=server, slot=slot).inc()
trace = traceback.format_exc()
self.notification_handler.build_and_send_notification(
notification_id=f'OPC_SUBSCRIPTION_ERROR_{slot}:{server}',