sbcli-dev 4.0.65__zip → 4.0.66__zip

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 4.0.65
+Version: 4.0.66
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/env_var

@@ -1,5 +1,5 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
-SIMPLY_BLOCK_VERSION=4.0.65
+SIMPLY_BLOCK_VERSION=4.0.66
 
 
 SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:main

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/sbcli_dev.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 4.0.65
+Version: 4.0.66
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/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("--namespace", help='k8s namespace to deploy on',)
 
 
         # delete storage node
@@ -293,7 +294,7 @@ class CLIWrapper:
         sub_command.add_argument("--ha-type", help='LVol HA type (single, ha), default is cluster HA type',
                                  dest='ha_type', choices=["single", "ha", "default"], default='single')
         sub_command.add_argument("--enable-node-affinity", help='Enable node affinity for storage nodes', action='store_true')
-        sub_command.add_argument("--qpair-count", help='tcp transport qpair count', type=int, dest='qpair_count', default=256)
+
         # add cluster
         sub_command = self.add_sub_command(subparser, 'add', 'Add new cluster')
         sub_command.add_argument("--blk_size", help='The block size in bytes', type=int, choices=[512, 4096], default=512)
@@ -315,7 +316,6 @@ class CLIWrapper:
         sub_command.add_argument("--ha-type", help='LVol HA type (single, ha), default is cluster HA type',
                                  dest='ha_type', choices=["single", "ha", "default"], default='default')
         sub_command.add_argument("--enable-node-affinity", help='Enable node affinity for storage nodes', action='store_true')
-        sub_command.add_argument("--qpair-count", help='tcp transport qpair count', type=int, dest='qpair_count', default=256)
 
         # Activate cluster
         sub_command = self.add_sub_command(subparser, 'activate', 'Create distribs and raid0 bdevs on all the storage node and move the cluster to active state')
@@ -694,7 +694,7 @@ class CLIWrapper:
         if args.debug:
             self.logger.setLevel(logging.DEBUG)
         else:
-            self.logger.setLevel(constants.LOG_LEVEL)
+            self.logger.setLevel(logging.INFO)
         logging.getLogger("urllib3.connectionpool").setLevel(logging.WARNING)
 
         args_dict = args.__dict__
@@ -740,6 +740,7 @@ class CLIWrapper:
                 enable_test_device = args.enable_test_device
                 enable_ha_jm = args.enable_ha_jm
                 number_of_distribs = args.number_of_distribs
+                namespace = args.namespace
 
                 out = storage_ops.add_node(
                     cluster_id=cluster_id,
@@ -758,7 +759,7 @@ class CLIWrapper:
                     jm_percent=jm_percent,
                     number_of_devices=number_of_devices,
                     enable_test_device=enable_test_device,
-                    namespace=None,
+                    namespace=namespace,
                     number_of_distribs=number_of_distribs,
                     enable_ha_jm=enable_ha_jm
                 )
@@ -1249,11 +1250,10 @@ class CLIWrapper:
         distr_chunk_bs = args.distr_chunk_bs
         ha_type = args.ha_type
         enable_node_affinity = args.enable_node_affinity
-        qpair_count = args.qpair_count
 
         return cluster_ops.add_cluster(
             blk_size, page_size_in_blocks, cap_warn, cap_crit, prov_cap_warn, prov_cap_crit,
-            distr_ndcs, distr_npcs, distr_bs, distr_chunk_bs, ha_type, enable_node_affinity, qpair_count)
+            distr_ndcs, distr_npcs, distr_bs, distr_chunk_bs, ha_type, enable_node_affinity)
 
     def cluster_create(self, args):
         page_size_in_blocks = args.page_size
@@ -1274,13 +1274,12 @@ class CLIWrapper:
         contact_point = args.contact_point
         grafana_endpoint = args.grafana_endpoint
         enable_node_affinity = args.enable_node_affinity
-        qpair_count = args.qpair_count
 
         return cluster_ops.create_cluster(
             blk_size, page_size_in_blocks,
             CLI_PASS, cap_warn, cap_crit, prov_cap_warn, prov_cap_crit,
             ifname, log_del_interval, metrics_retention_period, contact_point, grafana_endpoint,
-            distr_ndcs, distr_npcs, distr_bs, distr_chunk_bs, ha_type, enable_node_affinity, qpair_count)
+            distr_ndcs, distr_npcs, distr_bs, distr_chunk_bs, ha_type, enable_node_affinity)
 
     def query_yes_no(self, question, default="yes"):
         """Ask a yes/no question via raw_input() and return their answer.

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/cluster_ops.py

@@ -102,8 +102,7 @@ def _add_graylog_input(cluster_ip, password):
 
 def create_cluster(blk_size, page_size_in_blocks, cli_pass,
                    cap_warn, cap_crit, prov_cap_warn, prov_cap_crit, ifname, log_del_interval, metrics_retention_period,
-                   contact_point, grafana_endpoint, distr_ndcs, distr_npcs, distr_bs, distr_chunk_bs, ha_type,
-                   enable_node_affinity, qpair_count):
+                   contact_point, grafana_endpoint, distr_ndcs, distr_npcs, distr_bs, distr_chunk_bs, ha_type, enable_node_affinity):
     logger.info("Installing dependencies...")
     ret = scripts.install_deps()
     logger.info("Installing dependencies > Done")
@@ -170,8 +169,7 @@ def create_cluster(blk_size, page_size_in_blocks, cli_pass,
     else:
         c.grafana_endpoint = f"http://{DEV_IP}/grafana"
     c.enable_node_affinity = enable_node_affinity
-    c.qpair_count = qpair_count or 256
-
+    
     alerts_template_folder = os.path.join(TOP_DIR, "simplyblock_core/scripts/alerting/")
     alert_resources_file = "alert_resources.yaml"
 
@@ -289,7 +287,7 @@ def deploy_spdk(node_docker, spdk_cpu_mask, spdk_mem):
 
 
 def add_cluster(blk_size, page_size_in_blocks, cap_warn, cap_crit, prov_cap_warn, prov_cap_crit,
-                distr_ndcs, distr_npcs, distr_bs, distr_chunk_bs, ha_type, enable_node_affinity, qpair_count):
+                distr_ndcs, distr_npcs, distr_bs, distr_chunk_bs, ha_type, enable_node_affinity):
     db_controller = DBController()
     clusters = db_controller.get_clusters()
     if not clusters:
@@ -321,7 +319,6 @@ def add_cluster(blk_size, page_size_in_blocks, cap_warn, cap_crit, prov_cap_warn
     cluster.distr_chunk_bs = distr_chunk_bs
     cluster.ha_type = ha_type
     cluster.enable_node_affinity = enable_node_affinity
-    cluster.qpair_count = qpair_count or 256
     if cap_warn and cap_warn > 0:
         cluster.cap_warn = cap_warn
     if cap_crit and cap_crit > 0:

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/constants.py

@@ -6,8 +6,8 @@ KVD_DB_FILE_PATH = '/etc/foundationdb/fdb.cluster'
 KVD_DB_TIMEOUT_MS = 10000
 SPK_DIR = '/home/ec2-user/spdk'
 RPC_HTTP_PROXY_PORT = 8080
-LOG_LEVEL = logging.INFO
-LOG_WEB_DEBUG = False
+LOG_LEVEL = logging.DEBUG
+LOG_WEB_DEBUG = True
 
 INSTALL_DIR = os.path.dirname(os.path.realpath(__file__))
 

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/controllers/device_controller.py

@@ -818,7 +818,7 @@ def restart_jm_device(device_id, force=False, format_alceml=False):
             bdevs_names = [d['name'] for d in rpc_client.get_bdevs()]
             jm_nvme_bdevs = []
             for dev in snode.nvme_devices:
-                if dev.status != NVMeDevice.STATUS_ONLINE:
+                if dev.status not in [NVMeDevice.STATUS_ONLINE, NVMeDevice.STATUS_NEW]:
                     continue
                 dev_part = f"{dev.nvme_bdev[:-2]}p1"
                 if dev_part in bdevs_names:

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/controllers/health_controller.py

@@ -40,15 +40,15 @@ def check_cluster(cluster_id):
                 "Status": "ok" if ret else "failed"
             })
 
-        for lvol in db_controller.get_lvols(cluster_id):
-            ret = check_lvol(lvol.get_id())
-            result &= ret
-            print("*" * 100)
-            data.append({
-                "Kind": "LVol",
-                "UUID": lvol.get_id(),
-                "Status": "ok" if ret else "failed"
-            })
+    for lvol in db_controller.get_lvols(cluster_id):
+        ret = check_lvol(lvol.get_id())
+        result &= ret
+        print("*" * 100)
+        data.append({
+            "Kind": "LVol",
+            "UUID": lvol.get_id(),
+            "Status": "ok" if ret else "failed"
+        })
     print(utils.print_table(data))
     return result
 
@@ -163,6 +163,8 @@ def check_node(node_id, with_devices=True):
             if dev.status in [NVMeDevice.STATUS_ONLINE, NVMeDevice.STATUS_UNAVAILABLE]:
                 ret = check_device(dev.get_id())
                 node_devices_check &= ret
+            else:
+                logger.info(f"Device skipped: {dev.get_id()} status: {dev.status}")
             print("*" * 100)
 
         logger.info(f"Node remote device: {len(snode.remote_devices)}")

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/controllers/lvol_controller.py

@@ -635,7 +635,6 @@ def add_lvol_on_node(lvol, snode, ha_comm_addrs=None, ha_inode_self=None):
     ret = rpc_client.subsystem_create(lvol.nqn, 'sbcli-cn', lvol.uuid)
     logger.debug(ret)
 
-    cluster = db_controller.get_cluster_by_id(snode.cluster_id)
     # add listeners
     logger.info("adding listeners")
     for iface in snode.data_nics:
@@ -648,7 +647,7 @@ def add_lvol_on_node(lvol, snode, ha_comm_addrs=None, ha_inode_self=None):
                     if ty['trtype'] == tr_type:
                         found = True
             if found is False:
-                ret = rpc_client.transport_create(tr_type, cluster.qpair_count)
+                ret = rpc_client.transport_create(tr_type)
             logger.info("adding listener for %s on IP %s" % (lvol.nqn, iface.ip4_address))
             ret = rpc_client.listeners_create(lvol.nqn, tr_type, iface.ip4_address, "4420")
             is_optimized = False
@@ -696,7 +695,6 @@ def recreate_lvol_on_node(lvol, snode, ha_comm_addrs=None, ha_inode_self=None):
     ret = rpc_client.subsystem_create(lvol.nqn, 'sbcli-cn', lvol.uuid)
     logger.debug(ret)
 
-    cluster = db_controller.get_cluster_by_id(snode.cluster_id)
     # add listeners
     logger.info("adding listeners")
     for iface in snode.data_nics:
@@ -709,7 +707,7 @@ def recreate_lvol_on_node(lvol, snode, ha_comm_addrs=None, ha_inode_self=None):
                     if ty['trtype'] == tr_type:
                         found = True
             if found is False:
-                ret = rpc_client.transport_create(tr_type, cluster.qpair_count)
+                ret = rpc_client.transport_create(tr_type)
             logger.info("adding listener for %s on IP %s" % (lvol.nqn, iface.ip4_address))
             ret = rpc_client.listeners_create(lvol.nqn, tr_type, iface.ip4_address, "4420")
             is_optimized = False

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/controllers/snapshot_controller.py

@@ -307,13 +307,12 @@ def clone(snapshot_id, clone_name, new_size=0):
     logger.info("creating subsystem %s", subsystem_nqn)
     ret = rpc_client.subsystem_create(subsystem_nqn, 'sbcli-cn', lvol.uuid)
 
-    cluster = db_controller.get_cluster_by_id(snode.cluster_id)
     # add listeners
     logger.info("adding listeners")
     for iface in snode.data_nics:
         if iface.ip4_address:
             tr_type = iface.get_transport_type()
-            ret = rpc_client.transport_create(tr_type, cluster.qpair_count)
+            ret = rpc_client.transport_create(tr_type)
             logger.info("adding listener for %s on IP %s" % (subsystem_nqn, iface.ip4_address))
             ret = rpc_client.listeners_create(subsystem_nqn, tr_type, iface.ip4_address, "4420")
 

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/distr_controller.py

@@ -35,6 +35,8 @@ def send_node_status_event(node, node_status, target_node=None):
 
 
 def send_dev_status_event(device, dev_status, target_node=None):
+    if dev_status == NVMeDevice.STATUS_NEW:
+        return
     db_controller = DBController()
     storage_ID = device.cluster_device_order
     node_status_event = {

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/models/cluster.py

@@ -50,8 +50,6 @@ class Cluster(BaseModel):
         "prov_cap_warn": {"type": int, "default": 180},
         "prov_cap_crit": {"type": int, "default": 190},
 
-        "qpair_count": {"type": int, "default": 256},
-
         "secret": {"type": str, "default": ""},
         "status": {"type": str, "default": ""},
         "updated_at": {"type": str, "default": ""},

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/models/job_schedule.py

@@ -9,7 +9,7 @@ class JobSchedule(BaseModel):
 
     STATUS_NEW = 'new'
     STATUS_RUNNING = 'running'
-    # STATUS_RETRYING = 'retrying'
+    STATUS_SUSPENDED = 'suspended'
     STATUS_DONE = 'done'
 
     FN_DEV_RESTART = "device_restart"

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/models/storage_node.py

@@ -79,6 +79,7 @@ class StorageNode(BaseModel):
         "lvstore": {"type": str, 'default': ""},
         "raid": {"type": str, 'default': ""},
         "lvstore_stack": {"type": List[dict], 'default': []},
+        "jm_vuid": {"type": int, "default": 0},
 
         # spdk params
         "spdk_cpu_mask": {"type": str, "default": ""},

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/rpc_client.py

@@ -119,7 +119,7 @@ class RPCClient:
             params = {"trtype": trtype}
         return self._request("nvmf_get_transports", params)
 
-    def transport_create(self, trtype, qpair_count=256):
+    def transport_create(self, trtype):
         """
             [{'trtype': 'TCP', 'max_queue_depth': 128,
                'max_io_qpairs_per_ctrlr': 127, 'in_capsule_data_size': 4096,
@@ -134,7 +134,7 @@ class RPCClient:
         """
         params = {
             "trtype": trtype,
-            "max_io_qpairs_per_ctrlr": qpair_count,
+            "max_io_qpairs_per_ctrlr": 256,
             "max_queue_depth": 512,
             "abort_timeout_sec": 5,
             "ack_timeout": 512,
@@ -837,3 +837,9 @@ class RPCClient:
             "file": file,
         }
         return self._request("bdev_lvs_dump", params)
+
+    def jc_explicit_synchronization(self, jm_vuid):
+        params = {
+            "jm_vuid": jm_vuid
+        }
+        return self._request("jc_explicit_synchronization", params)

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/services/tasks_runner_failed_migration.py

@@ -32,14 +32,14 @@ def task_runner(task):
         task.write_to_db(db_controller.kv_store)
         return True
 
-    if task.status == JobSchedule.STATUS_NEW:
+    if task.status in [JobSchedule.STATUS_NEW ,JobSchedule.STATUS_SUSPENDED]:
         task.status = JobSchedule.STATUS_RUNNING
         task.write_to_db(db_controller.kv_store)
         tasks_events.task_updated(task)
 
     if snode.status != StorageNode.STATUS_ONLINE:
         task.function_result = "node is not online, retrying"
-        task.status = JobSchedule.STATUS_NEW
+        task.status = JobSchedule.STATUS_SUSPENDED
         task.retry += 1
         task.write_to_db(db_controller.kv_store)
         return False
@@ -62,6 +62,7 @@ def task_runner(task):
             logger.error(f"Failed to start device migration task, storage_ID: {device.cluster_device_order}")
             task.function_result = "Failed to start device migration task"
             task.retry += 1
+            task.status = JobSchedule.STATUS_SUSPENDED
             task.write_to_db(db_controller.kv_store)
             return False
 
@@ -83,7 +84,7 @@ def task_runner(task):
                     task.status = JobSchedule.STATUS_DONE
                 else:
                     task.function_result = "Failed to complete migration, retrying"
-                    task.status = JobSchedule.STATUS_NEW
+                    task.status = JobSchedule.STATUS_SUSPENDED
                     task.retry += 1
                     del task.function_params["migration"]
                 task.write_to_db(db_controller.kv_store)
@@ -91,7 +92,7 @@ def task_runner(task):
 
             elif migration_status == "failed":
                 task.function_result = "Failed to complete migration, retrying"
-                task.status = JobSchedule.STATUS_NEW
+                task.status = JobSchedule.STATUS_SUSPENDED
                 task.retry += 1
                 del task.function_params["migration"]
                 task.write_to_db(db_controller.kv_store)
@@ -131,7 +132,7 @@ while True:
             for task in tasks:
                 delay_seconds = 5
                 if task.function_name == JobSchedule.FN_FAILED_DEV_MIG:
-                    if task.status == JobSchedule.STATUS_NEW:
+                    if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
                         active_task = tasks_controller.get_active_node_mig_task(task.cluster_id, task.node_id)
                         if active_task:
                             logger.info("task found on same node, retry")

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/services/tasks_runner_migration.py

@@ -29,7 +29,7 @@ def task_runner(task):
         task.write_to_db(db_controller.kv_store)
         return True
 
-    if task.status == JobSchedule.STATUS_NEW:
+    if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
         task.status = JobSchedule.STATUS_RUNNING
         task.write_to_db(db_controller.kv_store)
         tasks_events.task_updated(task)
@@ -37,7 +37,7 @@ def task_runner(task):
 
     if snode.status != StorageNode.STATUS_ONLINE:
         task.function_result = "node is not online, retrying"
-        task.status = JobSchedule.STATUS_NEW
+        task.status = JobSchedule.STATUS_SUSPENDED
         task.retry += 1
         task.write_to_db(db_controller.kv_store)
         return False
@@ -56,7 +56,7 @@ def task_runner(task):
 
         if not all_devs_online:
             task.function_result = "Some devs are offline, retrying"
-            task.status = JobSchedule.STATUS_NEW
+            task.status = JobSchedule.STATUS_SUSPENDED
             task.retry += 1
             task.write_to_db(db_controller.kv_store)
             return False
@@ -90,7 +90,7 @@ def task_runner(task):
                 if st['error'] == 1:
                     task.function_result = "mig completed with errors, retrying"
                     task.retry += 1
-                    task.status = JobSchedule.STATUS_NEW
+                    task.status = JobSchedule.STATUS_SUSPENDED
                     del task.function_params['migration']
                 else:
                     task.status = JobSchedule.STATUS_DONE
@@ -122,7 +122,7 @@ while True:
             for task in tasks:
                 delay_seconds = 5
                 if task.function_name == JobSchedule.FN_DEV_MIG:
-                    if task.status == JobSchedule.STATUS_NEW:
+                    if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
                         active_task = tasks_controller.get_active_node_mig_task(task.cluster_id, task.node_id)
                         if active_task:
                             logger.info("task found on same node, retry")

{sbcli_dev-4.0.65 → sbcli_dev-4.0.66}/simplyblock_core/services/tasks_runner_new_dev_migration.py

@@ -32,7 +32,7 @@ def task_runner(task):
         task.write_to_db(db_controller.kv_store)
         return True
 
-    if task.status == JobSchedule.STATUS_NEW:
+    if task.status in [JobSchedule.STATUS_NEW ,JobSchedule.STATUS_SUSPENDED]:
         task.status = JobSchedule.STATUS_RUNNING
         task.write_to_db(db_controller.kv_store)
         tasks_events.task_updated(task)
@@ -40,7 +40,7 @@ def task_runner(task):
     if snode.status != StorageNode.STATUS_ONLINE:
         task.function_result = "node is not online, retrying"
         task.retry += 1
-        task.status = JobSchedule.STATUS_NEW
+        task.status = JobSchedule.STATUS_SUSPENDED
         task.write_to_db(db_controller.kv_store)
         return False
 
@@ -56,7 +56,7 @@ def task_runner(task):
         if not all_devs_online:
             task.function_result = "Some devs are offline, retrying"
             task.retry += 1
-            task.status = JobSchedule.STATUS_NEW
+            task.status = JobSchedule.STATUS_SUSPENDED
             task.write_to_db(db_controller.kv_store)
             return False
 
@@ -94,7 +94,7 @@ def task_runner(task):
                 if res_data['error'] == 1:
                     task.function_result = "mig completed with errors, retrying"
                     task.retry += 1
-                    task.status = JobSchedule.STATUS_NEW
+                    task.status = JobSchedule.STATUS_SUSPENDED
                     del task.function_params['migration']
                 else:
                     task.function_result = "Done"
@@ -143,7 +143,7 @@ while True:
             for task in tasks:
                 delay_seconds = 5
                 if task.function_name == JobSchedule.FN_NEW_DEV_MIG:
-                    if task.status == JobSchedule.STATUS_NEW:
+                    if task.status in [JobSchedule.STATUS_NEW, JobSchedule.STATUS_SUSPENDED]:
                         active_task = tasks_controller.get_active_node_mig_task(task.cluster_id, task.node_id)
                         if active_task:
                             logger.info("task found on same node, retry")