sbcli-dev 3.8.49__zip → 3.8.51__zip

{sbcli_dev-3.8.49 → sbcli_dev-3.8.51}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 3.8.49
+Version: 3.8.51
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-3.8.49 → sbcli_dev-3.8.51}/env_var

@@ -1,5 +1,5 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
-SIMPLY_BLOCK_VERSION=3.8.49
+SIMPLY_BLOCK_VERSION=3.8.51
 
 
 SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:main

{sbcli_dev-3.8.49 → sbcli_dev-3.8.51}/sbcli_dev.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 3.8.49
+Version: 3.8.51
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-3.8.49 → sbcli_dev-3.8.51}/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-3.8.49 → sbcli_dev-3.8.51}/simplyblock_core/distr_controller.py

@@ -151,8 +151,11 @@ def parse_distr_cluster_map(map_string):
             }
             nd = db_controller.get_storage_node_by_id(node_id)
             if nd:
-                data["Actual Status"] = nd.status
-                if nd.status == status:
+                node_status = nd.status
+                if node_status == StorageNode.STATUS_SCHEDULABLE:
+                    node_status = StorageNode.STATUS_UNREACHABLE
+                data["Actual Status"] = node_status
+                if node_status == status:
                     data["Results"] = "ok"
                 else:
                     data["Results"] = "failed"

{sbcli_dev-3.8.49 → sbcli_dev-3.8.51}/simplyblock_core/models/storage_node.py

@@ -32,6 +32,8 @@ class StorageNode(BaseModel):
         STATUS_RESTARTING: 12,
 
         STATUS_UNREACHABLE: 20,
+
+        STATUS_SCHEDULABLE: 30,
     }
 
     attributes = {

{sbcli_dev-3.8.49 → sbcli_dev-3.8.51}/simplyblock_core/services/storage_node_monitor.py

@@ -130,52 +130,56 @@ while True:
             # 1- check node ping
             ping_check = health_controller._check_node_ping(snode.mgmt_ip)
             logger.info(f"Check: ping mgmt ip {snode.mgmt_ip} ... {ping_check}")
+            if not ping_check:
+                time.sleep(1)
+                ping_check = health_controller._check_node_ping(snode.mgmt_ip)
+                logger.info(f"Check 2: ping mgmt ip {snode.mgmt_ip} ... {ping_check}")
 
             # 2- check node API
-            node_api_check = health_controller._check_node_api(snode.mgmt_ip)
-            logger.info(f"Check: node API {snode.mgmt_ip}:5000 ... {node_api_check}")
-
-            # 3- check spdk_process
-            spdk_process = health_controller._check_spdk_process_up(snode.mgmt_ip)
-            logger.info(f"Check: spdk process {snode.mgmt_ip}:5000 ... {spdk_process}")
-
-            # node_rpc_check = True
-            # 3- check node RPC
-            node_rpc_check = health_controller._check_node_rpc(
-                snode.mgmt_ip, snode.rpc_port, snode.rpc_username, snode.rpc_password, timeout=5, retry=1)
-            logger.info(f"Check: node RPC {snode.mgmt_ip}:{snode.rpc_port} ... {node_rpc_check}")
-
-            # 4- docker API
-            node_docker_check = health_controller._check_node_docker_api(snode.mgmt_ip)
-            logger.info(f"Check: node docker API {snode.mgmt_ip}:2375 ... {node_docker_check}")
-
-            is_node_online = ping_check and node_api_check and node_rpc_check and node_docker_check and spdk_process
+            node_api_check = False
+            spdk_process = False
+            node_rpc_check = False
+            if ping_check:
+                node_api_check = health_controller._check_node_api(snode.mgmt_ip)
+                logger.info(f"Check: node API {snode.mgmt_ip}:5000 ... {node_api_check}")
+
+                # 3- check spdk_process
+                spdk_process = health_controller._check_spdk_process_up(snode.mgmt_ip)
+                logger.info(f"Check: spdk process {snode.mgmt_ip}:5000 ... {spdk_process}")
+
+                # node_rpc_check = True
+                # 3- check node RPC
+                node_rpc_check = health_controller._check_node_rpc(
+                    snode.mgmt_ip, snode.rpc_port, snode.rpc_username, snode.rpc_password, timeout=5, retry=1)
+                logger.info(f"Check: node RPC {snode.mgmt_ip}:{snode.rpc_port} ... {node_rpc_check}")
+
+                # check JM device
+                if snode.jm_device:
+                    if snode.jm_device.status in [JMDevice.STATUS_ONLINE, JMDevice.STATUS_UNAVAILABLE]:
+                        ret = health_controller.check_jm_device(snode.jm_device.get_id())
+                        if ret:
+                            logger.info(f"JM bdev is online: {snode.jm_device.get_id()}")
+                            if snode.jm_device.status != JMDevice.STATUS_ONLINE:
+                                device_controller.set_jm_device_state(snode.jm_device.get_id(), JMDevice.STATUS_ONLINE)
+                        else:
+                            logger.error(f"JM bdev is offline: {snode.jm_device.get_id()}")
+                            if snode.jm_device.status != JMDevice.STATUS_UNAVAILABLE:
+                                device_controller.set_jm_device_state(snode.jm_device.get_id(),
+                                                                      JMDevice.STATUS_UNAVAILABLE)
+
+            is_node_online = ping_check and node_api_check and node_rpc_check and spdk_process
             if is_node_online:
                 set_node_online(snode)
             else:
                 set_node_offline(snode)
-                if 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)
-
                 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)
-                continue
 
-            # check JM device
-            if snode.jm_device:
-                if snode.jm_device.status in [JMDevice.STATUS_ONLINE, JMDevice.STATUS_UNAVAILABLE]:
-                    ret = health_controller.check_jm_device(snode.jm_device.get_id())
-                    if ret:
-                        logger.info(f"JM bdev is online: {snode.jm_device.get_id()}")
-                        if snode.jm_device.status != JMDevice.STATUS_ONLINE:
-                            device_controller.set_jm_device_state(snode.jm_device.get_id(), JMDevice.STATUS_ONLINE)
-                    else:
-                        logger.error(f"JM bdev is offline: {snode.jm_device.get_id()}")
-                        if snode.jm_device.status != JMDevice.STATUS_UNAVAILABLE:
-                            device_controller.set_jm_device_state(snode.jm_device.get_id(), JMDevice.STATUS_UNAVAILABLE)
+                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)
 
         update_cluster_status(cluster_id)
 

{sbcli_dev-3.8.49 → sbcli_dev-3.8.51}/simplyblock_core/utils.py

@@ -261,7 +261,7 @@ def process_records(records, records_count):
 
 def ping_host(ip):
     logger.debug(f"Pinging ip ... {ip}")
-    response = os.system(f"ping -c 1 -W 3 {ip} > /dev/null")
+    response = os.system(f"ping -c 3 -W 3 {ip} > /dev/null")
     if response == 0:
         logger.debug(f"{ip} is UP")
         return True