Hotfix 2 main

simplyblock_cli/cli-reference.yaml

@@ -38,13 +38,9 @@ commands:
             type: int
             required: true
           - name: "--max-size"
-            help: "Maximum amount of GB to be utilized on this storage node"
+            help: "Maximum amount of Huge Pages to be set on the node"
             description: >
-              Maximum amount of GB to be utilized on this storage node. This cannot be larger than the total effective
-              cluster capacity. A safe value is `1/n * 2.0` of effective cluster capacity. Meaning, if you have three
-              storage nodes, each with 100 TiB of raw capacity and a cluster with erasure coding scheme 1+1
-              (two replicas), the effective cluster capacity is _100 TiB * 3 / 2 = 150 TiB_. Setting this parameter to
-              _150 TiB / 3 * 2 = 100TiB_ would be a safe choice.
+              Maximum amount of Huge Pages to be set on the node, it will override the auto-calculated value.
             dest: max_prov
             type: str
             required: false

simplyblock_cli/cli.py

@@ -93,7 +93,7 @@ def init_storage_node__deploy(self, subparser):
     def init_storage_node__configure(self, subparser):
         subcommand = self.add_sub_command(subparser, 'configure', 'Prepare a configuration file to be used when adding the storage node')
         argument = subcommand.add_argument('--max-lvol', help='Max logical volume per storage node', type=int, dest='max_lvol', required=True)
-        argument = subcommand.add_argument('--max-size', help='Maximum amount of GB to be utilized on this storage node', type=str, dest='max_prov', required=False)
+        argument = subcommand.add_argument('--max-size', help='Maximum amount of Huge Pages to be set on the node', type=str, dest='max_prov', required=False)
         argument = subcommand.add_argument('--nodes-per-socket', help='number of each node to be added per each socket.', type=int, default=1, dest='nodes_per_socket')
         argument = subcommand.add_argument('--sockets-to-use', help='The system socket to use when adding the storage nodes', type=str, default='0', dest='sockets_to_use')
         argument = subcommand.add_argument('--cores-percentage', help='The percentage of cores to be used for spdk (0-99)', type=range_type(0, 99), default=0, dest='cores_percentage')

simplyblock_core/cluster_ops.py

@@ -1352,8 +1352,12 @@ def update_cluster(cluster_id, mgmt_only=False, restart=False, spdk_image=None,
 
 
 def cluster_grace_startup(cl_id, clear_data=False, spdk_image=None) -> None:
-    db_controller.get_cluster_by_id(cl_id)  # ensure exists
+    get_cluster = db_controller.get_cluster_by_id(cl_id)  # ensure exists
 
+    st = db_controller.get_storage_nodes_by_cluster_id(cl_id)
+    for node in st:
+        logger.info(f"Shutting down node: {node.get_id()}")
+        storage_node_ops.shutdown_storage_node(node.get_id(), force=True)
     st = db_controller.get_storage_nodes_by_cluster_id(cl_id)
     for node in st:
         logger.info(f"Restarting node: {node.get_id()}")
@@ -1362,18 +1366,30 @@ def cluster_grace_startup(cl_id, clear_data=False, spdk_image=None) -> None:
         get_node = db_controller.get_storage_node_by_id(node.get_id())
         if get_node.status != StorageNode.STATUS_ONLINE:
             raise ValueError("failed to restart node")
+    if get_cluster.status == Cluster.STATUS_UNREADY:
+        logger.info("Cluster is not activated yet, please manually activate it")
+
+    else:
+        while True:
+            get_cluster = db_controller.get_cluster_by_id(cl_id)
+            if get_cluster.status != Cluster.STATUS_ACTIVE:
+                logger.info(f"wait for cluster to be active, current status is: {get_cluster.status}")
+                time.sleep(5)
+            else:
+                break
+    logger.info("Cluster gracefully started")
+
 
 
 def cluster_grace_shutdown(cl_id) -> None:
     db_controller.get_cluster_by_id(cl_id)  # ensure exists
 
     st = db_controller.get_storage_nodes_by_cluster_id(cl_id)
     for node in st:
-        if node.status == StorageNode.STATUS_ONLINE:
-            logger.info(f"Suspending node: {node.get_id()}")
-            storage_node_ops.suspend_storage_node(node.get_id())
-            logger.info(f"Shutting down node: {node.get_id()}")
-            storage_node_ops.shutdown_storage_node(node.get_id())
+        logger.info(f"Suspending node: {node.get_id()}")
+        storage_node_ops.suspend_storage_node(node.get_id(), force=True)
+        logger.info(f"Shutting down node: {node.get_id()}")
+        storage_node_ops.shutdown_storage_node(node.get_id(), force=True)
 
 
 def delete_cluster(cl_id) -> None:

simplyblock_core/storage_node_ops.py

@@ -1650,11 +1650,12 @@ def restart_storage_node(
             snode.mgmt_ip = node_ip.split(":")[0]
             data_nics = []
             for nic in snode.data_nics:
-                device = node_info['network_interface'][nic.if_name]
+                if_name = nic["if_name"]
+                device = node_info['network_interface'][if_name]
                 data_nics.append(
                     IFace({
                         'uuid': str(uuid.uuid4()),
-                        'if_name': nic,
+                        'if_name': if_name,
                         'ip4_address': device['ip'],
                         'status': device['status'],
                         'net_type': device['net_type']}))