sbcli-pre 25.5.4__tar.gz → 25.5.6__tar.gz

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-pre
-Version: 25.5.4
+Version: 25.5.6
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/sbcli_pre.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-pre
-Version: 25.5.4
+Version: 25.5.6
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/simplyblock_core/cluster_ops.py

@@ -683,6 +683,31 @@ def cluster_activate(cl_id, force=False, force_lvstore_create=False):
                 set_cluster_status(cl_id, ols_status)
                 return False
 
+    snodes = db_controller.get_storage_nodes_by_cluster_id(cl_id)
+    for snode in snodes:
+        if snode.status != StorageNode.STATUS_ONLINE:
+            continue
+
+        if not snode.is_secondary_node:
+            continue
+
+        logger.info(f"recreating secondary node {snode.get_id()}")
+        ret = storage_node_ops.recreate_lvstore_on_sec(snode)
+
+        snode = db_controller.get_storage_node_by_id(snode.get_id())
+        if ret:
+            snode.lvstore_status = "ready"
+            snode.write_to_db()
+
+        else:
+            snode.lvstore_status = "failed"
+            snode.write_to_db()
+            logger.error(f"Failed to restore lvstore on node {snode.get_id()}")
+            if not force:
+                logger.error("Failed to activate cluster")
+                set_cluster_status(cl_id, ols_status)
+                return False
+
 
     if not cluster.cluster_max_size:
         cluster = db_controller.get_cluster_by_id(cl_id)

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/simplyblock_core/env_var

@@ -1,6 +1,6 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-pre
-SIMPLY_BLOCK_VERSION=25.5.4
+SIMPLY_BLOCK_VERSION=25.5.6
 
 
-SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:main
+SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:R25.5-PRE
 SIMPLY_BLOCK_SPDK_ULTRA_IMAGE=simplyblock/spdk:main-latest

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/simplyblock_core/models/storage_node.py

@@ -146,7 +146,7 @@ class StorageNode(BaseNodeObject):
             #
             # raise
 
-    def create_hublvol(self, cluster_nqn):
+    def create_hublvol(self):
         """Create a hublvol for this node's lvstore
         """
         logger.info(f'Creating hublvol on {self.get_id()}')
@@ -159,7 +159,7 @@ class StorageNode(BaseNodeObject):
                 raise RPCException('Failed to create hublvol')
             self.hublvol = HubLVol({
                 'uuid': hublvol_uuid,
-                'nqn': f'{cluster_nqn}:lvol:{hublvol_uuid}',
+                'nqn': f'{self.host_nqn}:lvol:{hublvol_uuid}',
                 'bdev_name': f'{self.lvstore}/hublvol',
                 'model_number': str(uuid4()),
                 'nguid': utils.generate_hex_string(16),
@@ -212,6 +212,13 @@ class StorageNode(BaseNodeObject):
                 return True
             except RPCException:
                 pass
+        else:
+            try:
+                self.create_hublvol()
+                return True
+            except RPCException as e:
+                logger.error("Error establishing hublvol: %s", e.message)
+                # return False
 
         return self.hublvol
 

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/simplyblock_core/rpc_client.py

@@ -140,7 +140,7 @@ class RPCClient:
         params = {
             "trtype": trtype,
             "max_io_qpairs_per_ctrlr": 128,
-            "max_queue_depth": 64,
+            "max_queue_depth": 256,
             "abort_timeout_sec": 5,
             "ack_timeout": 2048,
             "zcopy": True,

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/simplyblock_core/services/health_check_service.py

@@ -197,7 +197,7 @@ while True:
                             connected_jms.append(remote_device.get_id())
 
                     for jm_id in snode.jm_ids:
-                        if jm_id not in connected_jms:
+                        if jm_id and jm_id not in connected_jms:
                             for nd in db.get_storage_nodes():
                                 if nd.jm_device and nd.jm_device.get_id() == jm_id:
                                     if nd.status == StorageNode.STATUS_ONLINE:

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/simplyblock_core/services/tasks_cluster_status.py

@@ -10,7 +10,7 @@ cluster_commands = [
     "sbcli-dev cluster show",
     "sbcli-dev cluster get-capacity",
     "sbcli-dev cluster get-io-stats",
-    "sbcli-dev cluster get-logs",
+    "sbcli-dev cluster get-logs --limit 0",
     "sbcli-dev lvol list --cluster-id",
 ]
 

{sbcli-pre-25.5.4 → sbcli-pre-25.5.6}/simplyblock_core/storage_node_ops.py

@@ -1761,6 +1761,15 @@ def restart_storage_node(
             if devs:
                 dev = db_controller.get_jm_device_by_id(devs[0])
                 snode.remote_jm_devices.append(dev)
+        if not snode.jm_ids:
+            snode.jm_ids = []
+            for rem_jm in snode.remote_jm_devices:
+                if rem_jm.get_id():
+                    snode.jm_ids.append(rem_jm.get_id())
+
+            snode.jm_ids = snode.jm_ids[:constants.HA_JM_COUNT-1]
+            snode.write_to_db(db_controller.kv_store)
+
         snode.remote_jm_devices = _connect_to_remote_jm_devs(snode)
     snode.health_check = True
     snode.lvstore_status = ""
@@ -2979,13 +2988,15 @@ def recreate_lvstore(snode):
             sec_node_api.firewall_set_port(snode.lvol_subsys_port, "tcp", "block", sec_node.rpc_port)
             tcp_ports_events.port_deny(sec_node, snode.lvol_subsys_port)
 
-            # time.sleep(1)
+            # time.sleep(0.2)
             ### 4- set leadership to false
             sec_rpc_client.bdev_lvol_set_leader(snode.lvstore, leader=False, bs_nonleadership=True)
             sec_rpc_client.bdev_distrib_force_to_non_leader(snode.jm_vuid)
             time.sleep(1)
 
     ### 5- examine
+    time.sleep(0.2)
+    rpc_client.bdev_distrib_force_to_non_leader(snode.jm_vuid)
     ret = rpc_client.bdev_examine(snode.raid)
     # time.sleep(1)
 
@@ -3088,30 +3099,37 @@ def add_lvol_thread(lvol, snode, lvol_ana_state="optimized"):
 def get_sorted_ha_jms(current_node):
     db_controller = DBController()
     jm_count = {}
-    mgmt_ips = []
+    jm_dev_to_mgmt_ip = {}
+
     for node in db_controller.get_storage_nodes_by_cluster_id(current_node.cluster_id):
-        if (node.get_id() == current_node.get_id() or node.status != StorageNode.STATUS_ONLINE):  # pass
-            continue
-        if node.mgmt_ip == current_node.mgmt_ip:
-            continue
-        if node.mgmt_ip in mgmt_ips:
+        if node.get_id() == current_node.get_id():  # pass
             continue
 
-        if node.jm_device and node.jm_device.status == JMDevice.STATUS_ONLINE:
-            jm_count[node.jm_device.get_id()] = 1
-            mgmt_ips.append(node.mgmt_ip)
+        if node.jm_device and node.jm_device.status == JMDevice.STATUS_ONLINE and node.jm_device.get_id():
+            jm_count[node.jm_device.get_id()] = 0
+            jm_dev_to_mgmt_ip[node.jm_device.get_id()] = node.mgmt_ip
 
     for node in db_controller.get_storage_nodes_by_cluster_id(current_node.cluster_id):
-        if (node.get_id() == current_node.get_id() or node.status != StorageNode.STATUS_ONLINE):  # pass
+        if node.get_id() == current_node.get_id():  # pass
             continue
-        if node.mgmt_ip == current_node.mgmt_ip or not node.jm_ids:
+        if not node.jm_ids:
             continue
         for rem_jm_id in node.jm_ids:
             if rem_jm_id in jm_count:
                 jm_count[rem_jm_id] += 1
 
+    mgmt_ips = []
     jm_count = dict(sorted(jm_count.items(), key=lambda x: x[1]))
-    return list(jm_count.keys())[:3]
+    out = []
+    for jm_id in jm_count.keys():
+        if jm_id:
+            if jm_dev_to_mgmt_ip[jm_id] in mgmt_ips:
+                continue
+            if jm_dev_to_mgmt_ip[jm_id] == current_node.mgmt_ip:
+                continue
+            mgmt_ips.append(jm_dev_to_mgmt_ip[jm_id])
+            out.append(jm_id)
+    return out[:constants.HA_JM_COUNT-1]
 
 
 def get_node_jm_names(current_node, remote_node=None):
@@ -3126,6 +3144,9 @@ def get_node_jm_names(current_node, remote_node=None):
 
     if current_node.enable_ha_jm:
         for jm_id in current_node.jm_ids:
+            if not jm_id:
+                continue
+
             if remote_node:
                 if remote_node.jm_device.get_id() == jm_id:
                     jm_list.append(remote_node.jm_device.jm_bdev)
@@ -3298,9 +3319,8 @@ def create_lvstore(snode, ndcs, npcs, distr_bs, distr_chunk_bs, page_size_in_blo
         sec_rpc_client.bdev_examine(snode.raid)
         sec_rpc_client.bdev_wait_for_examine()
 
-        cluster_nqn = db_controller.get_cluster_by_id(snode.cluster_id).nqn
         try:
-            snode.create_hublvol(cluster_nqn)
+            snode.create_hublvol()
 
         except RPCException as e:
             logger.error("Error establishing hublvol: %s", e.message)