Update cluster state management (#227)

* Fix function return types and dictionary key checks
* Update pre-commit Github actions
* Update cluster state management, i.e. present, absent, started, restarted, and stopped

Signed-off-by: Webster Mudge <[email protected]>

.github/workflows/pre-commit.yml

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-name: pre-commit
+name: Execute Precommit Linting and Checks
 
 on:
   pull_request:
@@ -23,6 +23,6 @@ jobs:
   pre-commit:
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v3
-    - uses: actions/setup-python@v3
+    - uses: actions/checkout@v4
+    - uses: actions/setup-python@v5
     - uses: pre-commit/[email protected]

plugins/modules/cluster.py

@@ -61,7 +61,6 @@
 }
 
 DOCUMENTATION = r"""
----
 module: cluster
 short_description: Manage the lifecycle and state of a cluster
 description:
@@ -77,7 +76,6 @@
 """
 
 EXAMPLES = r"""
----
 - name: Create an ECS cluster
   cloudera.cluster.cluster:
     host: example.cloudera.com
@@ -119,7 +117,6 @@
 """
 
 RETURN = r"""
----
 cloudera_manager:
     description: Details about Cloudera Manager Cluster
     type: dict
@@ -276,21 +273,18 @@ def process(self):
 
         elif self.state == "absent":
             # Delete cluster
-
             refresh = False
-
-            # TODO Check for status when deleting
-            # if existing and existing.entity_status == "":
-            #     self.wait_for_active_cmd(cluster_api, self.cluster_name)
-            # elif existing:
             if existing:
                 self.changed = True
                 if not self.module.check_mode:
-                    self.cluster_api.delete_cluster(cluster_name=self.name)
-                    self.wait_for_active_cmd(self.name)
+                    if existing.entity_status != "STOPPED":
+                        stop = self.cluster_api.stop_command(cluster_name=self.name)
+                        self.wait_command(stop, polling=self.timeout, delay=self.delay)
+
+                    delete = self.cluster_api.delete_cluster(cluster_name=self.name)
+                    self.wait_command(delete, polling=self.timeout, delay=self.delay)
 
         elif self.state == "started":
-            # TODO NONE seems to be fresh cluster, never run before
             # Already started
             if existing and existing.entity_status == "GOOD_HEALTH":
                 refresh = False
@@ -312,11 +306,11 @@ def process(self):
                     # If newly created or created by not yet initialize
                     if not existing or existing.entity_status == "NONE":
                         first_run = self.cluster_api.first_run(cluster_name=self.name)
-                        self.wait_for_composite_cmd(first_run.id)
+                        self.wait_command(first_run)
                     # Start the existing and previously initialized cluster
                     else:
                         start = self.cluster_api.start_command(cluster_name=self.name)
-                        self.wait_for_composite_cmd(start.id)
+                        self.wait_command(start, polling=self.timeout, delay=self.delay)
 
         if self.state == "stopped":
             # Already stopped
@@ -339,7 +333,7 @@ def process(self):
                     self.changed = True
                     if not self.module.check_mode:
                         stop = self.cluster_api.stop_command(cluster_name=self.name)
-                        self.wait_for_composite_cmd(stop.id)
+                        self.wait_command(stop, polling=self.timeout, delay=self.delay)
 
         if self.state == "restarted":
             # Start underway
@@ -357,7 +351,7 @@ def process(self):
                 self.changed = True
                 if not self.module.check_mode:
                     restart = self.cluster_api.restart_command(cluster_name=self.name)
-                    self.wait_for_composite_cmd(restart.id)
+                    self.wait_command(restart, polling=self.timeout, delay=self.delay)
 
         if refresh:
             # Retrieve the updated cluster details
@@ -547,7 +541,6 @@ def create_cluster_from_parameters(self):
                         timeout=self.timeout,
                     )
                     parcel.activate()
-
             # Apply host templates
             for ht, refs in template_map.items():
                 self.host_template_api.apply_host_template(
@@ -674,10 +667,10 @@ def create_cluster_from_parameters(self):
             if self.auto_assign:
                 self.cluster_api.auto_assign_roles(cluster_name=self.name)
 
-    def marshal_service(self, options: str) -> ApiService:
+    def marshal_service(self, options: dict) -> ApiService:
         service = ApiService(name=options["name"], type=options["type"])
 
-        if "display_name" in options:
+        if options["display_name"]:
             service.display_name = options["display_name"]
 
         # Service-wide configuration
@@ -741,9 +734,7 @@ def marshal_hostrefs(self, hosts: dict) -> list[ApiHostRef]:
             )
         return results
 
-    def find_base_role_group_name(
-        self, service_type: str, role_type: str
-    ) -> ApiRoleConfigGroup:
+    def find_base_role_group_name(self, service_type: str, role_type: str) -> str:
         rcgs = [
             rcg
             for s in self.service_api.read_services(cluster_name=self.name).items

tests/unit/plugins/modules/cluster/test_cluster.py → tests/unit/plugins/modules/cluster/test_base_cluster.py

@@ -365,97 +365,6 @@ def test_present_base_host_role_overrides(conn, module_args):
     LOG.info(str(e.value.cloudera_manager))
 
 
-def test_present_basic_cluster(conn, module_args):
-    args = """
-    name: Basic_Cluster
-    cluster_version: "7.1.9-1.cdh7.1.9.p0.44702451"
-    type: BASE_CLUSTER
-    state: present
-    services:
-      - name: core-settings-0
-        type: CORE_SETTINGS
-        display_name: CORE_SETTINGS_TEST
-      - name: zookeeper-0
-        type: ZOOKEEPER
-        display_name: ZK_TEST
-        config:
-          zookeeper_datadir_autocreate: yes
-      - name: hdfs-0
-        type: HDFS
-        display_name: HDFS_TEST
-        config:
-            zookeeper_service: zookeeper-0
-            core_connector: core-settings-0
-        role_groups:
-          - type: DATANODE
-            config:
-              dfs_data_dir_list: /dfs/dn
-          - type: NAMENODE
-            config:
-              dfs_name_dir_list: /dfs/nn
-          - type: SECONDARYNAMENODE
-            config:
-              fs_checkpoint_dir_list: /dfs/snn
-      - name: yarn-0
-        type: YARN
-        display_name: YARN_TEST
-        config:
-          hdfs_service: hdfs-0
-          zookeeper_service: zookeeper-0
-        role_groups:
-          - type: RESOURCEMANAGER
-            config:
-              yarn_scheduler_maximum_allocation_mb: 4096
-              yarn_scheduler_maximum_allocation_vcores: 4
-          - type: NODEMANAGER
-            config:
-              yarn_nodemanager_resource_memory_mb: 4096
-              yarn_nodemanager_resource_cpu_vcores: 4
-              yarn_nodemanager_local_dirs:  /tmp/nm
-              yarn_nodemanager_log_dirs: /var/log/nm
-          - type: GATEWAY
-            config:
-              mapred_submit_replication: 3
-              mapred_reduce_tasks: 6
-    host_templates:
-      - name: Master1
-        role_groups:
-          - service: HDFS
-            type: NAMENODE
-          - service: HDFS
-            type: SECONDARYNAMENODE
-          - service: YARN
-            type: RESOURCEMANAGER
-          - service: YARN
-            type: JOBHISTORY
-      - name: Worker
-        role_groups:
-          - service: HDFS
-            type: DATANODE
-          - service: YARN
-            type: NODEMANAGER
-          - service: ZOOKEEPER
-            type: SERVER
-    parcels:
-      CDH: "7.1.9-1.cdh7.1.9.p0.44702451"
-    hosts:
-      - name: test10-worker-free-01.cldr.internal
-        host_template: Master1
-      - name: test10-worker-free-02.cldr.internal
-        host_template: Worker
-      - name: test10-worker-free-03.cldr.internal
-        host_template: Worker
-    """
-    conn.update(yaml.safe_load(args))
-    module_args(conn)
-
-    with pytest.raises(AnsibleExitJson) as e:
-        cluster.main()
-
-    LOG.info(str(e.value.cloudera_manager))
-
-
-@pytest.mark.skip(reason="Not yet implemented")
 def test_started_base(conn, module_args):
     conn.update(
         name="PVC-Base",
@@ -470,7 +379,6 @@ def test_started_base(conn, module_args):
     LOG.info(str(e.value.cloudera_manager))
 
 
-@pytest.mark.skip(reason="Not yet implemented")
 def test_restarted_base(conn, module_args):
     conn.update(
         name="PVC-Base",
@@ -485,24 +393,20 @@ def test_restarted_base(conn, module_args):
     LOG.info(str(e.value.cloudera_manager))
 
 
-@pytest.mark.skip(reason="Not yet implemented")
 def test_stopped_base(conn, module_args):
     conn.update(
         name="PVC-Base",
         cluster_version="7.1.9",  # "1.5.1-b626.p0.42068229",
-        # type="COMPUTE_CLUSTER",
         state="stopped",
     )
     module_args(conn)
 
     with pytest.raises(AnsibleExitJson) as e:
         cluster.main()
 
-    # LOG.info(str(e.value))
     LOG.info(str(e.value.cloudera_manager))
 
 
-@pytest.mark.skip(reason="Not yet implemented")
 def test_absent_base(conn, module_args):
     conn.update(
         name="Example_Base",
@@ -516,78 +420,6 @@ def test_absent_base(conn, module_args):
     LOG.info(str(e.value.cloudera_manager))
 
 
-def test_present_compute_minimum(conn, module_args):
-    conn.update(
-        name="Example_Compute",
-        cluster_version="7.1.9",
-        contexts=["SDX"],
-        state="present",
-    )
-    module_args(conn)
-
-    with pytest.raises(AnsibleExitJson) as e:
-        cluster.main()
-
-    LOG.info(str(e.value.cloudera_manager))
-
-
-@pytest.mark.skip(reason="Not yet implemented")
-def test_started_compute_minimum(conn, module_args):
-    conn.update(
-        name="Example_Compute",
-        cluster_version="7.1.9",
-        contexts=["SDX"],
-        state="started",
-    )
-    module_args(conn)
-
-    with pytest.raises(AnsibleExitJson) as e:
-        cluster.main()
-
-    LOG.info(str(e.value.cloudera_manager))
-
-
-def test_absent_compute(conn, module_args):
-    conn.update(
-        name="Example_Compute",
-        state="absent",
-    )
-    module_args(conn)
-
-    with pytest.raises(AnsibleExitJson) as e:
-        cluster.main()
-
-    LOG.info(str(e.value.cloudera_manager))
-
-
-def test_present_experience_minimum(conn, module_args):
-    conn.update(
-        name="Example_Experience",
-        cluster_version="1.5.3",
-        type="EXPERIENCE_CLUSTER",
-        state="present",
-    )
-    module_args(conn)
-
-    with pytest.raises(AnsibleExitJson) as e:
-        cluster.main()
-
-    LOG.info(str(e.value.cloudera_manager))
-
-
-def test_absent_experience(conn, module_args):
-    conn.update(
-        name="Example_Experience",
-        state="absent",
-    )
-    module_args(conn)
-
-    with pytest.raises(AnsibleExitJson) as e:
-        cluster.main()
-
-    LOG.info(str(e.value.cloudera_manager))
-
-
 def test_pytest_cluster_with_template(module_args):
     module_args(
         {