X-Git-Url: https://gerrit.opnfv.org/gerrit/gitweb?a=blobdiff_plain;f=doctor_tests%2Fmain.py;h=3dea89d13ae6f079a48720d4735b36882198d4af;hb=d8eb12f4200c21f569df5bc01d378a846b4c0db0;hp=2a8abda765788f819ee68d1743607cf71016560f;hpb=b54cbc5dd2d32fcb27238680b4657ed384d021c5;p=doctor.git

diff --git a/doctor_tests/main.py b/doctor_tests/main.py
index 2a8abda7..3dea89d1 100644
--- a/doctor_tests/main.py
+++ b/doctor_tests/main.py
@@ -1,5 +1,5 @@
 ##############################################################################
-# Copyright (c) 2017 ZTE Corporation and others.
+# Copyright (c) 2019 ZTE Corporation and others.
 #
 # All rights reserved. This program and the accompanying materials
 # are made available under the terms of the Apache License, Version 2.0
@@ -43,7 +43,6 @@ class DoctorTest(object):
     def setup(self):
         # prepare the cloud env
         self.installer.setup()
-
         # preparing VM image...
         self.image.create()
 
@@ -51,37 +50,51 @@ class DoctorTest(object):
         self.user.create()
 
     def test_fault_management(self):
-        try:
-            LOG.info('doctor fault management test starting.......')
-
-            self.fault_management = \
-                FaultManagement(self.conf, self.installer, self.user, LOG)
-
-            # prepare test env
-            self.fault_management.setup()
-
-            # wait for aodh alarms are updated in caches for event evaluator,
-            # sleep time should be larger than event_alarm_cache_ttl
-            # (default 60)
-            # (tojuvone) Fraser currently needs 120
-            time.sleep(120)
-
-            # injecting host failure...
-            # NOTE (umar) add INTERFACE_NAME logic to host injection
-            self.fault_management.start()
-            time.sleep(30)
-
-            # verify the test results
-            # NOTE (umar) copy remote monitor.log file when monitor=collectd
-            self.fault_management.check_host_status('down')
-            self.fault_management.check_notification_time()
-
-        except Exception as e:
-            LOG.error('doctor fault management test failed, '
-                      'Exception=%s' % e)
-            sys.exit(1)
-        finally:
-            self.fault_management.cleanup()
+        retry = 2
+        # Retry once if notified_time is None
+        while retry > 0:
+            try:
+                self.fault_management = None
+                LOG.info('doctor fault management test starting.......')
+                transport_url = self.installer.get_transport_url()
+                self.fault_management = \
+                    FaultManagement(self.conf, self.installer, self.user, LOG,
+                                    transport_url)
+
+                # prepare test env
+                self.fault_management.setup()
+
+                # wait for aodh alarms are updated in caches for event
+                # evaluator,
+                # sleep time should be larger than event_alarm_cache_ttl
+                # (default 60)
+                # (tojuvone) Fraser currently needs 120
+                time.sleep(120)
+
+                # injecting host failure...
+                # NOTE (umar) add INTERFACE_NAME logic to host injection
+                self.fault_management.start()
+                time.sleep(30)
+
+                # verify the test results
+                # NOTE (umar) copy remote monitor.log file when
+                # monitor=collectd
+                self.fault_management.check_host_status('down')
+                self.fault_management.check_notification_time()
+                retry = 0
+
+            except Exception as e:
+                LOG.error('doctor fault management test failed, '
+                          'Exception=%s' % e)
+                if 'notified_time=None' in str(e):
+                    retry -= 1
+                    LOG.info('doctor fault management retry')
+                    continue
+                LOG.error(format_exc())
+                sys.exit(1)
+            finally:
+                if self.fault_management is not None:
+                    self.fault_management.cleanup()
 
     def _amount_compute_nodes(self):
         services = self.nova.services.list(binary='nova-compute')
@@ -94,24 +107,35 @@ class DoctorTest(object):
             LOG.info('not enough compute nodes, skipping doctor '
                      'maintenance test')
             return
-        elif self.conf.installer.type != 'apex':
+        elif self.conf.installer.type not in ['apex', 'fuel', 'devstack']:
             LOG.info('not supported installer, skipping doctor '
                      'maintenance test')
             return
         try:
+            maintenance = None
             LOG.info('doctor maintenance test starting.......')
-
-            maintenance = Maintenance(self.conf, LOG)
+            trasport_url = self.installer.get_transport_url()
+            maintenance = Maintenance(trasport_url, self.conf, LOG)
             maintenance.setup_maintenance(self.user)
 
-            # TODO (tojuvone) actual test
+            # wait for aodh alarms are updated in caches for event evaluator,
+            # sleep time should be larger than event_alarm_cache_ttl
+            # (default 60)
+            LOG.info('wait aodh for 120s.......')
+            time.sleep(120)
+
+            session_id = maintenance.start_maintenance()
+            maintenance.wait_maintenance_complete(session_id)
+
+            LOG.info('doctor maintenance complete.......')
 
         except Exception as e:
             LOG.error('doctor maintenance test failed, Exception=%s' % e)
             LOG.error(format_exc())
             sys.exit(1)
         finally:
-            maintenance.cleanup_maintenance()
+            if maintenance is not None:
+                maintenance.cleanup_maintenance()
 
     def run(self):
         """run doctor tests"""
@@ -134,6 +158,7 @@ class DoctorTest(object):
                                     % function)
         except Exception as e:
             LOG.error('doctor test failed, Exception=%s' % e)
+            LOG.error(format_exc())
             sys.exit(1)
         finally:
             self.cleanup()