sbcli-dev 3.8.48__zip → 3.8.50__zip

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 3.8.48
+Version: 3.8.50
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/env_var

@@ -1,5 +1,5 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
-SIMPLY_BLOCK_VERSION=3.8.48
+SIMPLY_BLOCK_VERSION=3.8.50
 
 
 SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:main

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/sbcli_dev.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 3.8.48
+Version: 3.8.50
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/simplyblock_cli/cli.py

@@ -96,6 +96,8 @@ class CLIWrapper:
         sub_command.add_argument("--iobuf_small_bufsize", help='bdev_set_options param', dest='small_bufsize',  type=int, default=0)
         sub_command.add_argument("--iobuf_large_bufsize", help='bdev_set_options param', dest='large_bufsize',  type=int, default=0)
 
+        sub_command.add_argument("--force", help='Force restart', required=False, action='store_true')
+
         # sub_command.add_argument("-t", '--test', help='Run smart test on the NVMe devices', action='store_true')
 
         # Shutdown storage node
@@ -773,7 +775,7 @@ class CLIWrapper:
                 ret = storage_ops.restart_storage_node(
                     node_id, max_lvol, max_snap, max_prov,
                     spdk_image, spdk_debug,
-                    small_bufsize, large_bufsize, number_of_devices, node_ip=args.node_ip)
+                    small_bufsize, large_bufsize, number_of_devices, node_ip=args.node_ip, force=args.force)
 
             elif sub_command == "list-devices":
                 ret = self.storage_node_list_devices(args)

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/simplyblock_core/controllers/health_controller.py

@@ -82,7 +82,7 @@ def _check_node_rpc(rpc_ip, rpc_port, rpc_username, rpc_password, timeout=60, re
 
 def _check_node_api(ip):
     try:
-        snode_api = SNodeClient(f"{ip}:5000", timeout=3, retry=1)
+        snode_api = SNodeClient(f"{ip}:5000", timeout=10, retry=2)
         logger.debug(f"Node API={ip}:5000")
         info, _ = snode_api.info()
         if info:
@@ -95,7 +95,7 @@ def _check_node_api(ip):
 
 def _check_spdk_process_up(ip):
     try:
-        snode_api = SNodeClient(f"{ip}:5000", timeout=3, retry=1)
+        snode_api = SNodeClient(f"{ip}:5000", timeout=10, retry=2)
         logger.debug(f"Node API={ip}:5000")
         is_up, _ = snode_api.spdk_process_is_up()
         logger.debug(f"SPDK is {is_up}")

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/simplyblock_core/distr_controller.py

@@ -108,8 +108,11 @@ def get_distr_cluster_map(snodes, target_node):
             dev_w_map.append({
                 "weight": dev_w,
                 "id": dev.cluster_device_order})
+        node_status = snode.status
+        if node_status == StorageNode.STATUS_SCHEDULABLE:
+            node_status = StorageNode.STATUS_UNREACHABLE
         map_cluster[snode.get_id()] = {
-            "status": snode.status,
+            "status": node_status,
             "devices": dev_map}
         map_prob.append({
             "weight": node_w,

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/simplyblock_core/models/storage_node.py

@@ -19,6 +19,7 @@ class StorageNode(BaseModel):
 
     STATUS_IN_CREATION = 'in_creation'
     STATUS_UNREACHABLE = 'unreachable'
+    STATUS_SCHEDULABLE = 'schedulable'
 
     STATUS_CODE_MAP = {
         STATUS_ONLINE: 0,
@@ -31,6 +32,8 @@ class StorageNode(BaseModel):
         STATUS_RESTARTING: 12,
 
         STATUS_UNREACHABLE: 20,
+
+        STATUS_SCHEDULABLE: 30,
     }
 
     attributes = {

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/simplyblock_core/services/storage_node_monitor.py

@@ -154,11 +154,17 @@ while True:
                 set_node_online(snode)
             else:
                 set_node_offline(snode)
-                if ping_check and node_api_check and not spdk_process:
+                if not ping_check and not node_api_check and not spdk_process:
+                    # restart on new node
+                    storage_node_ops.set_node_status(snode.get_id(), StorageNode.STATUS_SCHEDULABLE)
+
+                elif ping_check and node_api_check and not spdk_process:
                     # add node to auto restart
                     if cluster.status != Cluster.STATUS_UNREADY:
                         tasks_controller.add_node_to_auto_restart(snode)
 
+                continue
+
             # check JM device
             if snode.jm_device:
                 if snode.jm_device.status in [JMDevice.STATUS_ONLINE, JMDevice.STATUS_UNAVAILABLE]:

{sbcli_dev-3.8.48 → sbcli_dev-3.8.50}/simplyblock_core/storage_node_ops.py

@@ -1372,7 +1372,8 @@ def restart_storage_node(
 
     if snode.status == StorageNode.STATUS_RESTARTING:
         logger.error(f"Node is in restart: {node_id}")
-        return False
+        if force is False:
+            return False
 
     task_id = tasks_controller.get_active_node_restart_task(snode.cluster_id, snode.get_id())
     if task_id: