sbcli-dev 10.9.3__tar.gz → 10.9.4__tar.gz

{sbcli_dev-10.9.3 → sbcli_dev-10.9.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: sbcli-dev
-Version: 10.9.3
+Version: 10.9.4
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-10.9.3 → sbcli_dev-10.9.4}/sbcli_dev.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: sbcli-dev
-Version: 10.9.3
+Version: 10.9.4
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-10.9.3 → sbcli_dev-10.9.4}/simplyblock_cli/cli.py

@@ -57,6 +57,7 @@ class CLIWrapper:
         sub_command.add_argument("--iobuf_large_bufsize", help='bdev_set_options param', dest='large_bufsize',  type=int, default=0)
         sub_command.add_argument("--enable-test-device", help='Enable creation of test device', action='store_true')
         sub_command.add_argument("--disable-ha-jm", help='Disable HA JM for distrib creation', action='store_false', dest='enable_ha_jm', default=True)
+        sub_command.add_argument("--ha-jm-count", help='HA JM count', dest='ha_jm_count', type=int, default=constants.HA_JM_COUNT)
         sub_command.add_argument("--is-secondary-node", help='add as secondary node', action='store_true', dest='is_secondary_node', default=False)
         sub_command.add_argument("--namespace", help='k8s namespace to deploy on',)
         sub_command.add_argument("--id-device-by-nqn", help='Use device nqn to identify it instead of serial number', action='store_true', dest='id_device_by_nqn', default=False)
@@ -781,6 +782,7 @@ class CLIWrapper:
                 enable_ha_jm = args.enable_ha_jm
                 number_of_distribs = args.number_of_distribs
                 namespace = args.namespace
+                ha_jm_count = args.ha_jm_count
 
                 out = storage_ops.add_node(
                     cluster_id=cluster_id,
@@ -805,6 +807,7 @@ class CLIWrapper:
                     is_secondary_node=args.is_secondary_node,
                     id_device_by_nqn=args.id_device_by_nqn,
                     partition_size=args.partition_size,
+                    ha_jm_count=ha_jm_count,
                 )
 
                 return out

{sbcli_dev-10.9.3 → sbcli_dev-10.9.4}/simplyblock_core/constants.py

@@ -129,4 +129,4 @@ QPAIR_COUNT=32
 NVME_TIMEOUT_US=20000000
 
 NVMF_MAX_SUBSYSTEMS=50000
-HA_JM_COUNT=3
+HA_JM_COUNT=3

{sbcli_dev-10.9.3 → sbcli_dev-10.9.4}/simplyblock_core/env_var

@@ -1,5 +1,5 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
-SIMPLY_BLOCK_VERSION=10.9.3
+SIMPLY_BLOCK_VERSION=10.9.4
 
 SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:main
 DOCKER_USER=hamdysimplyblock

{sbcli_dev-10.9.3 → sbcli_dev-10.9.4}/simplyblock_core/models/storage_node.py

@@ -28,6 +28,7 @@ class StorageNode(BaseNodeObject):
     distrib_cpu_index: int = 0
     distrib_cpu_mask: str = ""
     enable_ha_jm: bool = False
+    ha_jm_count: int = 3
     enable_test_device: bool = False
     health_check: bool = True
     host_nqn: str = ""

{sbcli_dev-10.9.3 → sbcli_dev-10.9.4}/simplyblock_core/services/tasks_runner_migration.py

@@ -29,24 +29,6 @@ def task_runner(task):
         task.write_to_db(db_controller.kv_store)
         return True
 
-    if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
-        if task.status == JobSchedule.STATUS_NEW:
-            for node in db_controller.get_storage_nodes_by_cluster_id(task.cluster_id):
-                if node.online_since:
-                    try:
-                        diff = datetime.now() - datetime.fromisoformat(node.online_since)
-                        if diff.total_seconds() < 60:
-                            task.function_result = "node is online < 1 min, retrying"
-                            task.status = JobSchedule.STATUS_SUSPENDED
-                            task.retry += 1
-                            task.write_to_db(db_controller.kv_store)
-                            return False
-                    except Exception as e:
-                        logger.error(f"Failed to get online since: {e}")
-
-        task.status = JobSchedule.STATUS_RUNNING
-        task.write_to_db(db_controller.kv_store)
-
     if snode.status != StorageNode.STATUS_ONLINE:
         task.function_result = "node is not online, retrying"
         task.status = JobSchedule.STATUS_SUSPENDED
@@ -54,9 +36,8 @@ def task_runner(task):
         task.write_to_db(db_controller.kv_store)
         return False
 
-    rpc_client = RPCClient(snode.mgmt_ip, snode.rpc_port, snode.rpc_username, snode.rpc_password,
-                           timeout=5, retry=2)
-    if "migration" not in task.function_params:
+
+    if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
         all_devs_online = True
         for node in db_controller.get_storage_nodes_by_cluster_id(task.cluster_id):
             for dev in node.nvme_devices:
@@ -73,6 +54,27 @@ def task_runner(task):
             task.write_to_db(db_controller.kv_store)
             return False
 
+        if node.online_since:
+            try:
+                diff = datetime.now() - datetime.fromisoformat(node.online_since)
+                if diff.total_seconds() < 60:
+                    task.function_result = "node is online < 1 min, retrying"
+                    task.status = JobSchedule.STATUS_SUSPENDED
+                    task.retry += 1
+                    task.write_to_db(db_controller.kv_store)
+                    return False
+            except Exception as e:
+                logger.error(f"Failed to get online since: {e}")
+
+        task.status = JobSchedule.STATUS_RUNNING
+        task.function_result = ""
+        task.write_to_db(db_controller.kv_store)
+
+
+    rpc_client = RPCClient(snode.mgmt_ip, snode.rpc_port, snode.rpc_username, snode.rpc_password,
+                           timeout=5, retry=2)
+    if "migration" not in task.function_params:
+
         device = db_controller.get_storage_device_by_id(task.device_id)
         distr_name = task.function_params["distr_name"]
 
@@ -95,7 +97,7 @@ def task_runner(task):
         task.function_params['migration'] = {
             "name": distr_name}
         task.write_to_db(db_controller.kv_store)
-        time.sleep(3)
+        time.sleep(1)
 
     if "migration" in task.function_params:
         mig_info = task.function_params["migration"]
@@ -112,7 +114,6 @@ db_controller = db_controller.DBController()
 
 logger.info("Starting Tasks runner...")
 while True:
-    time.sleep(3)
     clusters = db_controller.get_clusters()
     if not clusters:
         logger.error("No clusters found!")
@@ -135,4 +136,4 @@ while True:
                             tasks_events.task_updated(task)
                             continue
 
-                        time.sleep(delay_seconds)
+    time.sleep(3)

{sbcli_dev-10.9.3 → sbcli_dev-10.9.4}/simplyblock_core/storage_node_ops.py

@@ -936,7 +936,7 @@ def add_node(cluster_id, node_ip, iface_name, data_nics_list,
              small_bufsize=0, large_bufsize=0, spdk_cpu_mask=None,
              num_partitions_per_dev=0, jm_percent=0, number_of_devices=0, enable_test_device=False,
              namespace=None, number_of_distribs=2, enable_ha_jm=False, is_secondary_node=False, id_device_by_nqn=False,
-             partition_size=""):
+             partition_size="", ha_jm_count=3):
 
     db_controller = DBController()
     kv_store = db_controller.kv_store
@@ -1175,6 +1175,7 @@ def add_node(cluster_id, node_ip, iface_name, data_nics_list,
     snode.number_of_distribs = number_of_distribs
     snode.enable_ha_jm = enable_ha_jm
     snode.is_secondary_node = is_secondary_node
+    snode.ha_jm_count = ha_jm_count
 
     if 'cpu_count' in node_info:
         snode.cpu = node_info['cpu_count']
@@ -2785,7 +2786,7 @@ def recreate_lvstore_on_sec(snode):
             remote_rpc_client.bdev_distrib_force_to_non_leader(node.jm_vuid)
 
         ret, err = _create_bdev_stack(snode, node.lvstore_stack, primary_node=node)
-        # ret = rpc_client.bdev_examine(node.raid)
+        ret = rpc_client.bdev_examine(node.raid)
         ret = rpc_client.bdev_wait_for_examine()
         ret = rpc_client.bdev_lvol_set_lvs_groupid(node.lvstore, node.jm_vuid)
 
@@ -2866,7 +2867,7 @@ def recreate_lvstore(snode):
             for did in devs:
                 dev = db_controller.get_jm_device_by_id(did)
                 online_devs.append(dev)
-                if len(online_devs) > constants.HA_JM_COUNT - 1:
+                if len(online_devs) > snode.ha_jm_count - 1:
                     break
 
         snode.remote_jm_devices = online_devs
@@ -2957,7 +2958,7 @@ def get_node_jm_names(current_node):
         jm_list.append("JM_LOCAL")
 
     if current_node.enable_ha_jm:
-        for jm_dev in current_node.remote_jm_devices[:constants.HA_JM_COUNT-1]:
+        for jm_dev in current_node.remote_jm_devices[:current_node.ha_jm_count-1]:
             jm_list.append(jm_dev.remote_bdev)
     return jm_list
 
@@ -3079,7 +3080,7 @@ def create_lvstore(snode, ndcs, npcs, distr_bs, distr_chunk_bs, page_size_in_blo
                 sec_node_1.mgmt_ip, sec_node_1.rpc_port,
                 sec_node_1.rpc_username, sec_node_1.rpc_password)
 
-        # ret = temp_rpc_client.bdev_examine(snode.raid)
+        ret = temp_rpc_client.bdev_examine(snode.raid)
         ret = temp_rpc_client.bdev_wait_for_examine()
         ret = temp_rpc_client.bdev_lvol_set_lvs_groupid(snode.lvstore, snode.jm_vuid)
 
@@ -3120,7 +3121,7 @@ def _create_bdev_stack(snode, lvstore_stack=None, primary_node=None):
                     jm_list.append(bdev_name)
                 else:
                     jm_list.append("JM_LOCAL")
-                for jm_dev in primary_node.remote_jm_devices[:constants.HA_JM_COUNT-1]:
+                for jm_dev in primary_node.remote_jm_devices[:primary_node.ha_jm_count-1]:
                     jm_list.append(jm_dev.remote_bdev)
                 params['jm_names'] = jm_list
             else: