Merge pull request #4143 from wazuh/3953-group_sync_statusv3

Syncronization testing after group deletingv3

CHANGELOG.md

@@ -7,6 +7,9 @@ All notable changes to this project will be documented in this file.
 Wazuh commit: TBD \
 Release report: TBD
 
+### Added
+- Agent syncronization testing after group deleting ([#3953](https://github.com/wazuh/wazuh-qa/pull/4143)) \- (Tests)
+
 ### Changed
 
 - Enable Ubuntu Vulnerability Detector E2E. ([#4252](https://github.com/wazuh/wazuh-qa/pull/4252)) \- (Tests)

deps/wazuh_testing/wazuh_testing/__init__.py

@@ -71,7 +71,11 @@
 ANALYSISD_STATE = os.path.join(WAZUH_PATH, 'var', 'run', 'wazuh-analysisd.state')
 
 # Timeouts
+
+T_025 = 0.25
+T_1 = 1
 T_2 = 2
+T_3 = 3
 T_5 = 5
 T_10 = 10
 T_20 = 20

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

@@ -0,0 +1,89 @@
+- name: delete_folder_master_agent_all
+  description: all agents in all groups, deleting folder group_master from wazuh-master, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-master
+    agent_in_group: all
+    group_folder_deleted: group_master
+    expected_first_check: syncreq
+    expected_second_check: synced
+
+- name: delete_folder_worker1_agent_all
+  description: all agents in all groups, deleting folder group_worker1 from wazuh-worker1, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-worker1
+    agent_in_group: all
+    group_folder_deleted: group_worker1
+    expected_first_check: synced
+    expected_second_check: synced
+
+- name: delete_folder_worker2_agent_all
+  description: all agents in all groups, deleting folder group_worker2 from wazuh-worker2, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-worker2
+    agent_in_group: all
+    group_folder_deleted: group_worker2
+    expected_first_check: synced
+    expected_second_check: synced
+
+- name: delete_folder_master_agent_agent1
+  description: agent1 in all groups, deleting folder group_master from wazuh-master, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-master
+    agent_in_group: agent1
+    group_folder_deleted: group_master
+    expected_first_check: syncreq
+    expected_second_check: synced
+
+- name: delete_folder_worker1_agent1
+  description: agent1 in all groups, deleting folder group_worker1 from wazuh-worker1, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-worker1
+    agent_in_group: agent1
+    group_folder_deleted: group_worker1
+    expected_first_check: synced
+    expected_second_check: synced
+
+- name: delete_folder_worker2_agent1
+  description: agent1 in all groups, deleting folder group_worker2 from wazuh-worker2, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-worker2
+    agent_in_group: agent1
+    group_folder_deleted: group_worker2
+    expected_first_check: synced
+    expected_second_check: synced
+
+- name: delete_folder_master_agent_agent2
+  description: agent2 in all groups, deleting folder group_master from wazuh-master, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-master
+    agent_in_group: agent2
+    group_folder_deleted: group_master
+    expected_first_check: syncreq
+    expected_second_check: synced
+
+- name: delete_folder_worker1_agent2
+  description: agent2 in all groups, deleting folder group_worker1 from wazuh-worker1, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-worker1
+    agent_in_group: agent2
+    group_folder_deleted: group_worker1
+    expected_first_check: synced
+    expected_second_check: synced
+
+- name: delete_folder_worker2_agent2
+  description: agent2 in all groups, deleting folder group_worker2 from wazuh-worker2, check sync_status
+  configuration_parameters: null
+  metadata:
+    delete_target: wazuh-worker2
+    agent_in_group: agent2
+    group_folder_deleted: group_worker2
+    expected_first_check: synced
+    expected_second_check: synced

tests/system/test_cluster/test_agent_groups/test_group_sync_status.py

@@ -0,0 +1,178 @@
+'''
+copyright: Copyright (C) 2015-2023, Wazuh Inc.
+            Created by Wazuh, Inc. <info@wazuh.com>.
+            This program is free software; you can redistribute it and/or modify it under the terms of GPLv2
+type: system
+brief: Wazuh manager handles agent groups.
+        If a group is deleted from a master cluster, there will be an instance where the agents require a
+        resynchronization (syncreq).
+        If the group is deleted from a worker cluster, the cluster master will take care of reestablishing the
+        group structure without the need for resynchronization.
+        This test suite tests the correct functioning of the mentioned use case.
+tier: 0
+modules:
+    - enrollment
+components:
+    - manager
+    - agent
+daemons:
+    - wazuh-authd
+    - wazuh-agentd
+os_platform:
+    - linux
+os_version:
+    - Debian Buster
+references:
+    - https://documentation.wazuh.com/current/user-manual/registering/agent-enrollment.html
+'''
+
+import json
+import os
+import pytest
+import time
+from time import time as current_time
+from wazuh_testing import T_025, T_1, T_5, T_10
+from wazuh_testing.tools.system import HostManager
+from system import (assign_agent_to_new_group, create_new_agent_group, delete_agent_group, execute_wdb_query,
+                    restart_cluster)
+from wazuh_testing.tools.configuration import get_test_cases_data
+from system.test_cluster.test_agent_groups.common import register_agent
+
+pytestmark = [pytest.mark.cluster, pytest.mark.enrollment_cluster_env]
+
+test_infra_hosts = ['wazuh-master', 'wazuh-worker1', 'wazuh-worker2', 'wazuh-agent1', 'wazuh-agent2']
+test_infra_managers = ['wazuh-master', 'wazuh-worker1', 'wazuh-worker2']
+test_infra_agents = ['wazuh-agent1', 'wazuh-agent2']
+groups = ['group_master', 'group_worker1', 'group_worker2']
+workers = ['wazuh-worker1', 'wazuh-worker2']
+groups_created = []
+query = "global 'sql select name, group_sync_status from agent;'"
+
+inventory_path = os.path.join(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))),
+                              'provisioning', 'enrollment_cluster', 'inventory.yml')
+host_manager = HostManager(inventory_path)
+data_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'data')
+local_path = os.path.dirname(os.path.abspath(__file__))
+test_cases_yaml = os.path.join(data_path, 'cases_group_sync.yaml')
+wdb_query = os.path.join(os.path.dirname(os.path.realpath(__file__)), 'script/wdb-query.py')
+agent_conf_file = os.path.join(os.path.dirname(os.path.realpath(__file__)),
+                               '..', '..', '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)
+
+
+@pytest.fixture()
+def group_creation_and_assignation(metadata, target_node):
+
+    agent_ids = []
+    for agent in test_infra_agents:
+        agent_ip, agent_id, agent_name, manager_ip = register_agent(agent, test_infra_hosts[0], host_manager)
+        agent_ids.append(agent_id)
+
+    restart_cluster(test_infra_agents, host_manager)
+
+    time.sleep(T_10)
+    for group in groups:
+        create_new_agent_group(target_node, group, host_manager)
+
+    if metadata['agent_in_group'] == 'agent1':
+        for group in groups:
+            assign_agent_to_new_group(target_node, group, agent_ids[0], host_manager)
+
+    elif metadata['agent_in_group'] == 'agent2':
+        for group in groups:
+            assign_agent_to_new_group(target_node, group, agent_ids[1], host_manager)
+
+    else:
+        for group in groups:
+            for agent_id in agent_ids:
+                assign_agent_to_new_group(target_node, group, agent_id, host_manager)
+
+    yield
+
+    for group in groups:
+        delete_agent_group(test_infra_hosts[0], group, host_manager, 'api')
+
+
+@pytest.fixture()
+def wait_end_initial_syncreq():
+    timeout = current_time() + T_10
+    result = execute_wdb_query(query, test_infra_hosts[0], host_manager)
+
+    while 'syncreq' in result:
+        time.sleep(T_1)
+        if current_time() >= timeout:
+            pytest.fail('Test failure due to unstable environment, syncreq does not disappear after group management')
+        result = execute_wdb_query(query, test_infra_hosts[0], host_manager)
+
+
+@pytest.mark.parametrize('target_node', ['wazuh-master', 'wazuh-worker1', 'wazuh-worker2'])
+@pytest.mark.parametrize('metadata', t1_configuration_metadata, ids=t1_case_ids)
+def test_group_sync_status(metadata, target_node, clean_environment, group_creation_and_assignation,
+                           wait_end_initial_syncreq):
+    '''
+    description: Delete a group folder in wazuh server cluster and check group_sync status in 2 times.
+    wazuh_min_version: 4.4.0
+    metadata:
+        - metadata:
+            type: list
+            brief: List of tests to be performed.
+        - target_node:
+            type: list
+            brief: List of nodes from the groups will be managed.
+        - clean_environment:
+            type: fixture
+            brief: Cleaning logs and resetting environment before testing.
+        - group_creation_and_assignation:
+            type: fixture
+            brief: Delete and create from zero all the groups that are going to be used for testing.
+                    It includes group cleaning after tests.
+        - wait_end_initial_syncreq:
+            type: fixture
+            brief: Wait until syncreqs related with the test-environment setting get neutralized
+    assertions:
+        - Verify that group_sync status changes according the trigger.
+        - Verify same conditions creating and assigning groups from all wazuh-manager clusters (Master and Workers)
+    input_description: Different use cases are found in the test module and include parameters.
+    expected_output:
+        - If the group-folder is deleted from master cluster, it is expected to find a
+        syncreq group_sync status until it gets synced.
+        - If the group-folder is deletef rom a worker cluster, it is expected that master
+        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
+    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']
+
+        if metadata['agent_in_group'] == 'agent1':
+            if 'syncreq' == agent1_status and 'synced' == agent2_status:
+                first_time_check = "syncreq"
+
+        elif metadata['agent_in_group'] == 'agent2':
+            if 'synced' == agent1_status and 'syncreq' == agent2_status:
+                first_time_check = "syncreq"
+
+        else:
+            if agent1_status == 'syncreq' and agent2_status == 'syncreq':
+                first_time_check = 'syncreq'
+
+    time.sleep(T_5)
+
+    assert metadata['expected_first_check'] == first_time_check
+
+    # Check after 5 seconds, sync_status
+    if 'syncreq' in execute_wdb_query(query, test_infra_hosts[0], host_manager):
+        second_time_check = 'syncreq'
+    else:
+        second_time_check = 'synced'
+
+    assert metadata['expected_second_check'] == second_time_check