sbcli-dev 17.1.8__tar.gz → 17.2.0__tar.gz

{sbcli_dev-17.1.8 → sbcli_dev-17.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sbcli-dev
-Version: 17.1.8
+Version: 17.2.0
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-17.1.8 → sbcli_dev-17.2.0}/sbcli_dev.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sbcli-dev
-Version: 17.1.8
+Version: 17.2.0
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-17.1.8 → sbcli_dev-17.2.0}/simplyblock_core/env_var

@@ -1,5 +1,5 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
-SIMPLY_BLOCK_VERSION=17.1.8
+SIMPLY_BLOCK_VERSION=17.2.0
 
 SIMPLY_BLOCK_DOCKER_IMAGE=public.ecr.aws/simply-block/simplyblock:main
 SIMPLY_BLOCK_SPDK_ULTRA_IMAGE=public.ecr.aws/simply-block/ultra:main-latest

{sbcli_dev-17.1.8 → sbcli_dev-17.2.0}/simplyblock_core/services/health_check_service.py

@@ -183,37 +183,6 @@ while True:
 
                         node_remote_devices_check &= bool(ret)
 
-                # for node in db_controller.get_storage_nodes_by_cluster_id(snode.cluster_id):
-                #     if node.status != StorageNode.STATUS_ONLINE or node.get_id() == snode.get_id():
-                #         continue
-                #     for dev in node.nvme_devices:
-                #         if dev.status == NVMeDevice.STATUS_ONLINE:
-                #             if dev.get_id() not in connected_devices:
-                #                 if not dev.alceml_bdev:
-                #                     logger.error(f"device alceml bdev not found!, {dev.get_id()}")
-                #                     continue
-                #                 logger.info(f"connecting to online device: {dev.get_id()}")
-                #                 name = f"remote_{dev.alceml_bdev}"
-                #                 bdev_name = f"{name}n1"
-                #                 if rpc_client.bdev_nvme_controller_list(name):
-                #                     logger.info(f"detaching {name} from {snode.get_id()}")
-                #                     rpc_client.bdev_nvme_detach_controller(name)
-                #                     time.sleep(1)
-                #
-                #                 logger.info(f"Connecting {name} to {snode.get_id()}")
-                #                 ret = rpc_client.bdev_nvme_attach_controller_tcp(
-                #                     name, dev.nvmf_nqn, dev.nvmf_ip,
-                #                     dev.nvmf_port)
-                #                 if ret:
-                #                     logger.info(f"Successfully connected to device: {dev.get_id()}")
-                #                     dev.remote_bdev = bdev_name
-                #                     snode = db_controller.get_storage_node_by_id(snode.get_id())
-                #                     snode.remote_devices.append(dev)
-                #                     snode.write_to_db()
-                #                     distr_controller.send_dev_status_event(dev, NVMeDevice.STATUS_ONLINE, snode)
-                #                 else:
-                #                     logger.error(f"Failed to connect to device: {dev.get_id()}")
-
                 connected_jms = []
                 if snode.jm_device and snode.jm_device.get_id():
                     jm_device = snode.jm_device
@@ -224,8 +193,6 @@ while True:
                     else:
                         logger.info(f"Checking jm bdev: {jm_device.jm_bdev} ... not found")
 
-                    # node_devices_check &= ret
-
                 if snode.enable_ha_jm:
                     logger.info(f"Node remote JMs: {len(snode.remote_jm_devices)}")
                     for remote_device in snode.remote_jm_devices:
@@ -237,8 +204,12 @@ while True:
 
                     for jm_id in snode.jm_ids:
                         if jm_id not in connected_jms:
-                            node_remote_devices_check = False
-                            break
+                            for nd in db_controller.get_storage_nodes():
+                                if nd.jm_device and nd.jm_device.get_id() == jm_id:
+                                    if nd.status == StorageNode.STATUS_ONLINE:
+                                        node_remote_devices_check = False
+                                    break
+
                     if snode.lvstore_stack_secondary_1:
                         primary_node = db_controller.get_storage_node_by_id(snode.lvstore_stack_secondary_1)
                         if primary_node:

{sbcli_dev-17.1.8 → sbcli_dev-17.2.0}/simplyblock_core/services/lvol_monitor.py

@@ -95,10 +95,6 @@ while True:
 
                 for lvol in db_controller.get_lvols_by_node_id(snode.get_id()):
 
-                    if lvol.io_error:
-                        logger.debug(f"Skipping LVol health check because of io_error {lvol.get_id()}")
-                        continue
-
                     if lvol.status == LVol.STATUS_IN_CREATION:
                         continue
 

{sbcli_dev-17.1.8 → sbcli_dev-17.2.0}/simplyblock_core/services/main_distr_event_collector.py

@@ -69,6 +69,10 @@ def process_device_event(event):
                 logger.info(f"Setting device to read-only")
                 device_controller.device_set_read_only(device_obj.get_id())
 
+            elif event.message == 'error_write_cannot_allocate':
+                logger.info(f"Setting device to cannot_allocate")
+                device_controller.device_set_state(device_obj.get_id(), NVMeDevice.STATUS_CANNOT_ALLOCATE)
+
             else:
                 logger.info(f"Setting device to unavailable")
                 device_controller.device_set_unavailable(device_obj.get_id())

{sbcli_dev-17.1.8 → sbcli_dev-17.2.0}/simplyblock_core/storage_node_ops.py

@@ -1938,7 +1938,7 @@ def list_storage_nodes(is_json, cluster_id=None):
             "Status": node.status,
             "Health": node.health_check,
             "Up time": uptime,
-            "CPU": f"{len(utils.hexa_to_cpu_list(node.spdk_cpu_mask))} {format(int(node.spdk_cpu_mask, 16), f'#0{node.cpu+2}b')}",
+            "CPU": f"{len(utils.hexa_to_cpu_list(node.spdk_cpu_mask))} {node.spdk_cpu_mask}",
             "MEM": utils.humanbytes(node.spdk_mem),
             "SPDK P": node.rpc_port,
             "LVOL P": node.lvol_subsys_port,
@@ -2812,7 +2812,10 @@ def recreate_lvstore_on_sec(snode):
         node.lvstore_status = "in_creation"
         node.write_to_db()
 
-        lvol_list = db_controller.get_lvols_by_node_id(node.get_id())
+        lvol_list = []
+        for lv in db_controller.get_lvols_by_node_id(node.get_id()):
+            if lv.status not in [LVol.STATUS_IN_DELETION, LVol.STATUS_IN_CREATION]:
+                lvol_list.append(lv)
 
         ### 1- create distribs and raid
         ret, err = _create_bdev_stack(snode, node.lvstore_stack, primary_node=node)
@@ -2891,7 +2894,10 @@ def recreate_lvstore(snode):
     sec_node = db_controller.get_storage_node_by_id(snode.secondary_node_id)
     sec_node_api = SNodeClient(sec_node.api_endpoint)
 
-    lvol_list = db_controller.get_lvols_by_node_id(snode.get_id())
+    lvol_list = []
+    for lv in db_controller.get_lvols_by_node_id(snode.get_id()):
+        if lv.status not in [LVol.STATUS_IN_DELETION, LVol.STATUS_IN_CREATION]:
+            lvol_list.append(lv)
 
     prim_node_suspend = False
     if sec_node: