Merge pull request #5134 from wazuh/fix/5101-system-tests

Fix enrollment cluster system test

CHANGELOG.md

@@ -49,6 +49,7 @@ All notable changes to this project will be documented in this file.
 
 ### Fixed
 
+- Fix enrollment cluster system tests ([#5134](https://github.com/wazuh/wazuh-qa/pull/5134/)) \- (Tests)
 - Fix `test_synchronization` system test ([#5089](https://github.com/wazuh/wazuh-qa/pull/5089)) \- (Framework + Tests)
 - Fix number of files and their size for `test_zip_size_limit` ([#5133](https://github.com/wazuh/wazuh-qa/pull/5133)) \- (Tests)
 - Fix test_shutdown_message system test ([#5087](https://github.com/wazuh/wazuh-qa/pull/5087)) \- (Tests)

tests/system/test_cluster/test_agent_enrollment/test_agent_enrollment.py

@@ -26,11 +26,13 @@
 # Remove the agent once the test has finished
 @pytest.fixture(scope='module')
 def clean_environment():
+
     yield
+
+    host_manager.control_service(host='wazuh-agent1', service='wazuh', state="stopped")
     agent_id = host_manager.run_command('wazuh-master', f'cut -c 1-3 {WAZUH_PATH}/etc/client.keys')
     host_manager.get_host('wazuh-master').ansible("command", f'{WAZUH_PATH}/bin/manage_agents -r {agent_id}',
                                                   check=False)
-    host_manager.control_service(host='wazuh-agent1', service='wazuh', state="stopped")
     host_manager.clear_file(host='wazuh-agent1', file_path=os.path.join(WAZUH_PATH, 'etc', 'client.keys'))
     host_manager.clear_file(host='wazuh-agent1', file_path=os.path.join(WAZUH_LOGS_PATH, 'ossec.log'))
 
@@ -47,7 +49,7 @@ def test_agent_enrollment(clean_environment):
     # Start the agent enrollment process by restarting the wazuh-agent
     host_manager.control_service(host='wazuh-master', service='wazuh', state="restarted")
     host_manager.control_service(host='wazuh-worker1', service='wazuh', state="restarted")
-    host_manager.get_host('wazuh-agent1').ansible('command', f'service wazuh-agent restart', check=False)
+    host_manager.control_service(host='wazuh-agent1', service='wazuh', state="restarted")
 
     # Run the callback checks for the ossec.log and the cluster.log
     HostMonitor(inventory_path=inventory_path,

tests/system/test_cluster/test_agent_groups/data/guess_group_messages_master.yaml

@@ -1,10 +1,10 @@
 wazuh-master:
   - regex: .*Agent 'AGENT_ID' with file 'merged.mg' MD5 .*
     path: var/ossec/logs/ossec.log
-    timeout: 30
+    timeout: 60
   - regex: ".*Group assigned: 'GROUP_ID'"
     path: /var/ossec/logs/ossec.log
-    timeout: 10
+    timeout: 60
   - regex: .*Agent 'AGENT_ID' group is 'GROUP_ID'
     path: /var/ossec/logs/ossec.log
-    timeout: 10
+    timeout: 60

tests/system/test_cluster/test_agent_groups/data/guess_group_messages_worker.yaml

@@ -2,17 +2,17 @@ wazuh-worker1:
   - regex: ".*Sending message to master node: '{\"daemon_name\":\"remoted\",\"message\":{\"command\":\"assigngroup\",\
             \"parameters\":{\"agent\":\"AGENT_ID\",\"md5\":.*"
     path: /var/ossec/logs/ossec.log
-    timeout: 30
+    timeout: 60
   - regex: ".*Message received from master node: '{\"error\":0,\"message\":\"ok\",\"data\":{\"group\":\"GROUP_ID\"}}'"
     path: /var/ossec/logs/ossec.log
-    timeout: 10
+    timeout: 60
   - regex: .*Agent 'AGENT_ID' group is 'GROUP_ID'
     path: /var/ossec/logs/ossec.log
-    timeout: 10
+    timeout: 60
 wazuh-master:
   - regex: .*Agent 'AGENT_ID' with file 'merged.mg' MD5 .*
     path: /var/ossec/logs/ossec.log
-    timeout: 30
+    timeout: 60
   - regex: ".*Group assigned: 'GROUP_ID'"
     path: /var/ossec/logs/ossec.log
-    timeout: 10
+    timeout: 60

tests/system/test_cluster/test_agent_groups/test_assign_agent_to_a_group.py

@@ -66,7 +66,7 @@
 
 # Variables
 test_group = 'group_test'
-timeout = 25
+timeout = 40
 
 
 # Tests
@@ -107,12 +107,15 @@ def test_assign_agent_to_a_group(agent_target, initial_status, clean_environment
     restart_cluster(test_infra_agents, host_manager)
 
     time.sleep(timeout)
+
     # Check that agent status is active in cluster
     check_agent_status(agent_id, agent_name, agent_ip, AGENT_STATUS_ACTIVE, host_manager, test_infra_managers)
 
     if (initial_status == AGENT_STATUS_DISCONNECTED):
         host_manager.control_service(host='wazuh-agent1', service=WAZUH_SERVICE_PREFIX, state=WAZUH_SERVICES_STOPPED)
+
         time.sleep(timeout)
+
         check_agent_status(agent_id, agent_name, agent_ip, AGENT_STATUS_DISCONNECTED, host_manager, test_infra_managers)
 
     try:
@@ -123,6 +126,7 @@ def test_assign_agent_to_a_group(agent_target, initial_status, clean_environment
         assign_agent_to_new_group('wazuh-master', test_group, agent_id, host_manager)
 
         time.sleep(timeout)
+
         # Check that agent has group set to group_test on Managers
         check_agent_groups(agent_id, test_group, test_infra_managers, host_manager)
 

tests/system/test_cluster/test_agent_groups/test_assign_agent_to_a_group_api.py

@@ -63,7 +63,7 @@
 tmp_path = os.path.join(local_path, 'tmp')
 
 # Variables
-timeout = 10
+timeout = 30
 test_group = 'group_test'
 
 

tests/system/test_cluster/test_agent_groups/test_assign_groups_guess.py

@@ -276,6 +276,7 @@ def test_guess_multigroups(n_agents, target_node, status_guess_agent_group, clea
         # Run the callback checks for the ossec.log
         messages_path = master_messages_path if target_node == 'wazuh-master' else worker_messages_path
         replace_regex_in_file(['AGENT_ID', 'GROUP_ID'], [agent1_id, expected_group], messages_path)
+
         HostMonitor(inventory_path=inventory_path,
                     messages_path=messages_path,
                     tmp_path=tmp_path).run(update_position=True)

tests/system/test_cluster/test_agent_groups/test_group_hash.py

@@ -107,13 +107,13 @@ def test_group_hash(target_node, group, n_agents, configure_groups, clean_enviro
 
     # Restart agent
     restart_cluster(test_infra_agents, host_manager)
-    time.sleep(fw.T_10)
+    time.sleep(fw.T_20)
 
     # Assing group for multigroups case
     for agent in range(n_agents):
         if group != 'default':
             assign_agent_to_new_group(test_infra_managers[0], group, agents_data[agent][1], host_manager)
-    time.sleep(fw.T_10)
+    time.sleep(fw.T_20)
 
     # Calculate global hash
     expected_global_hash = calculate_global_hash(test_infra_managers[0], host_manager)
@@ -129,7 +129,7 @@ def test_group_hash(target_node, group, n_agents, configure_groups, clean_enviro
         # Unassign one agent from group
 
         unassign_agent_from_group(test_infra_managers[0], group, agents_data[0][1], host_manager)
-        time.sleep(fw.T_10)
+        time.sleep(fw.T_20)
 
         # Calculate global hash
         expected_global_hash = calculate_global_hash(test_infra_managers[0], host_manager)

tests/system/test_cluster/test_agent_groups/test_group_sync_status.py

@@ -59,7 +59,7 @@
                                '..', '..', 'provisioning', 'enrollment_cluster', 'roles', 'agent-role',
                                'files', 'ossec.conf')
 t1_configuration_parameters, t1_configuration_metadata, t1_case_ids = get_test_cases_data(test_cases_yaml)
-
+TIMEOUT_SECOND_CHECK = 10
 
 @pytest.fixture()
 def group_creation_and_assignation(metadata, target_node):
@@ -141,34 +141,40 @@ def test_group_sync_status(metadata, target_node, clean_environment, group_creat
         cluster recreates groups without syncreq status.
     '''
     # Delete group folder
+
     delete_agent_group(metadata['delete_target'], metadata['group_folder_deleted'], host_manager, 'folder')
 
     # Set values
     first_time_check = 'synced'
     second_time_check = ''
 
-    # Check each 0.25 seconds/10 seconds sync_status
+    # Check each 0.10 seconds/10 seconds sync_status
     for _ in range(T_10):
-        time.sleep(T_025)
-        agent1_status = json.loads(execute_wdb_query(query, test_infra_hosts[0], host_manager))[1]['group_sync_status']
-        agent2_status = json.loads(execute_wdb_query(query, test_infra_hosts[0], host_manager))[2]['group_sync_status']
+        status_info = json.loads(execute_wdb_query(query, test_infra_hosts[0], host_manager))[1:3]
+        agent1_status = status_info[0]['group_sync_status']
+        agent2_status = status_info[1]['group_sync_status']
 
         if metadata['agent_in_group'] == 'agent1':
-            if 'syncreq' == agent1_status and 'synced' == agent2_status:
+            if agent1_status == 'syncreq' and agent2_status == 'synced':
                 first_time_check = "syncreq"
+                break
 
         elif metadata['agent_in_group'] == 'agent2':
-            if 'synced' == agent1_status and 'syncreq' == agent2_status:
+            if agent1_status == 'synced' and agent2_status == 'syncreq':
                 first_time_check = "syncreq"
+                break
 
         else:
             if agent1_status == 'syncreq' and agent2_status == 'syncreq':
                 first_time_check = 'syncreq'
+                break
 
-    time.sleep(T_5)
+        time.sleep(0.10)
 
     assert metadata['expected_first_check'] == first_time_check
 
+    time.sleep(TIMEOUT_SECOND_CHECK)
+
     # Check after 5 seconds, sync_status
     if 'syncreq' in execute_wdb_query(query, test_infra_hosts[0], host_manager):
         second_time_check = 'syncreq'

tests/system/test_cluster/test_agent_groups/test_remove_group.py

@@ -29,7 +29,7 @@
 # Variables
 t1_configuration_parameters, t1_configuration_metadata, t1_case_ids = get_test_cases_data(t1_cases_path)
 queries = ['sql select `group` from agent;', 'sql select name from `group`;', 'sql select id_group from belongs;']
-
+TIMEOUT_GET_GROUPS_ID = 3
 
 # Fixtures
 @pytest.fixture()
@@ -112,6 +112,8 @@ def test_remove_group(metadata, group, target_node, pre_configured_groups, clean
                     messages_path=messages_path,
                     tmp_path=tmp_path).run(update_position=True)
 
+        sleep(TIMEOUT_GET_GROUPS_ID)
+
         for manager in test_infra_managers:
             group_ids[manager] = str(get_group_id(group, manager, host_manager))