[healthd] Use unix_socket_path instead of loopback ip (#14843)

- Why I did it

interfaces-config service restarts networking service, which in-turn results in loopback interface address is being removed and reassigned back

If the system-health happens to start during that instance expections and logs like this are seen:

Apr 15 18:14:49.357869 r-panther-20 ERR healthd: update system status exception:Unable to connect to redis: Cannot assign requested address
Apr 15 18:14:49.429778 r-panther-20 ERR healthd: subscribe_statedb exited- Unable to connect to redis: Cannot assign requested address
Apr 15 18:14:52.218594 r-panther-20 ERR healthd: system_service_Map_base::at
Apr 15 18:14:52.219714 r-panther-20 ERR healthd: system_service_Map_base::at
Apr 15 18:14:55.218636 r-panther-20 ERR healthd: system_service_Map_base::at
Apr 15 18:14:55.218722 r-panther-20 ERR healthd: system_service_Map_base::at

- How I did it
use unix socket path

Signed-off-by: Vivek Reddy Karri <vkarri@nvidia.com>

src/system-health/health_checker/hardware_checker.py

@@ -15,7 +15,7 @@ class HardwareChecker(HealthChecker):
 
     def __init__(self):
         HealthChecker.__init__(self)
-        self._db = SonicV2Connector(host="127.0.0.1")
+        self._db = SonicV2Connector(use_unix_socket_path=True)
         self._db.connect(self._db.STATE_DB)
 
     def get_category(self):

src/system-health/health_checker/service_checker.py

@@ -269,7 +269,7 @@ class ServiceChecker(HealthChecker):
             config (config.Config): Health checker configuration.
         """
         if not self.config_db:
-            self.config_db = swsscommon.ConfigDBConnector()
+            self.config_db = swsscommon.ConfigDBConnector(use_unix_socket_path=True)
             self.config_db.connect()
         feature_table = self.config_db.get_table("FEATURE")
         expected_running_containers, self.container_feature_dict = self.get_expected_running_containers(feature_table)

src/system-health/health_checker/sysmonitor.py

@@ -31,7 +31,7 @@ class MonitorStateDbTask(ProcessTaskBase):
         self.task_queue = myQ
 
     def subscribe_statedb(self):
-        state_db = swsscommon.DBConnector("STATE_DB", REDIS_TIMEOUT_MS, True)
+        state_db = swsscommon.DBConnector("STATE_DB", REDIS_TIMEOUT_MS, False)
         sel = swsscommon.Select()
         cst = swsscommon.SubscriberStateTable(state_db, "FEATURE")
         sel.addSelectable(cst)
@@ -122,7 +122,7 @@ class Sysmonitor(ProcessTaskBase):
     def post_system_status(self, state):
         try:
             if not self.state_db:
-                self.state_db = swsscommon.SonicV2Connector(host='127.0.0.1')
+                self.state_db = swsscommon.SonicV2Connector(use_unix_socket_path=True)
                 self.state_db.connect(self.state_db.STATE_DB)
 
             self.state_db.set(self.state_db.STATE_DB, "SYSTEM_READY|SYSTEM_STATE", "Status", state)
@@ -135,7 +135,7 @@ class Sysmonitor(ProcessTaskBase):
     def get_all_service_list(self):
 
         if not self.config_db:
-            self.config_db = swsscommon.ConfigDBConnector()
+            self.config_db = swsscommon.ConfigDBConnector(use_unix_socket_path=True)
             self.config_db.connect()
 
         dir_list = []
@@ -197,10 +197,10 @@ class Sysmonitor(ProcessTaskBase):
     #else, just return Up
     def get_app_ready_status(self, service):
         if not self.state_db:
-            self.state_db = swsscommon.SonicV2Connector(host='127.0.0.1')
+            self.state_db = swsscommon.SonicV2Connector(use_unix_socket_path=True)
             self.state_db.connect(self.state_db.STATE_DB)
         if not self.config_db:
-            self.config_db = swsscommon.ConfigDBConnector()
+            self.config_db = swsscommon.ConfigDBConnector(use_unix_socket_path=True)
             self.config_db.connect()
 
         fail_reason = ""
@@ -248,7 +248,7 @@ class Sysmonitor(ProcessTaskBase):
     #Sets the service status to state db
     def post_unit_status(self, srv_name, srv_status, app_status, fail_reason, update_time):
         if not self.state_db:
-            self.state_db = swsscommon.SonicV2Connector(host='127.0.0.1')
+            self.state_db = swsscommon.SonicV2Connector(use_unix_socket_path=True)
             self.state_db.connect(self.state_db.STATE_DB)
 
         key = 'ALL_SERVICE_STATUS|{}'.format(srv_name)
@@ -378,7 +378,7 @@ class Sysmonitor(ProcessTaskBase):
     def check_unit_status(self, event):
         #global dnsrvs_name
         if not self.state_db:
-            self.state_db = swsscommon.SonicV2Connector(host='127.0.0.1')
+            self.state_db = swsscommon.SonicV2Connector(use_unix_socket_path=True)
             self.state_db.connect(self.state_db.STATE_DB)
         astate = "DOWN"
 
@@ -419,7 +419,7 @@ class Sysmonitor(ProcessTaskBase):
 
     def system_service(self):
         if not self.state_db:
-            self.state_db = swsscommon.SonicV2Connector(host='127.0.0.1')
+            self.state_db = swsscommon.SonicV2Connector(use_unix_socket_path=True)
             self.state_db.connect(self.state_db.STATE_DB)
 
         mpmgr = multiprocessing.Manager()

src/system-health/scripts/healthd

@@ -31,7 +31,7 @@ class HealthDaemon(DaemonBase):
         Constructor of HealthDaemon.
         """
         DaemonBase.__init__(self, SYSLOG_IDENTIFIER)
-        self._db = SonicV2Connector(host="127.0.0.1")
+        self._db = SonicV2Connector(use_unix_socket_path=True)
         self._db.connect(self._db.STATE_DB)
         self.stop_event = threading.Event()
 

src/system-health/tests/mock_connector.py

@@ -2,7 +2,7 @@ class MockConnector(object):
     STATE_DB = None
     data = {}
 
-    def __init__(self, host):
+    def __init__(self, use_unix_socket_path):
         pass
 
     def connect(self, db_id):