sbcli-dev 3.8.40__zip → 3.8.42__zip

{sbcli_dev-3.8.40 → sbcli_dev-3.8.42}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 3.8.40
+Version: 3.8.42
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-3.8.40 → sbcli_dev-3.8.42}/env_var

@@ -1,5 +1,5 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
-SIMPLY_BLOCK_VERSION=3.8.40
+SIMPLY_BLOCK_VERSION=3.8.42
 
 
 SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:main

{sbcli_dev-3.8.40 → sbcli_dev-3.8.42}/sbcli_dev.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 3.8.40
+Version: 3.8.42
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-3.8.40 → sbcli_dev-3.8.42}/simplyblock_core/controllers/tasks_controller.py

@@ -54,22 +54,20 @@ def _add_task(function_name, cluster_id, node_id, device_id,
 
 
 def add_device_mig_task(device_id):
-    # device = db_controller.get_storage_devices(device_id)
-    # for node in db_controller.get_storage_nodes_by_cluster_id(device.cluster_id):
-    #     for lvol_id in node.lvols:
-    #         _add_task(JobSchedule.FN_DEV_MIG, device.cluster_id, node.get_id(), device.get_id(),
-    #                   max_retry=0, function_params={'lvol_id': lvol_id})
+    device = db_controller.get_storage_devices(device_id)
+    for node in db_controller.get_storage_nodes_by_cluster_id(device.cluster_id):
+        for lvol_id in node.lvols:
+            _add_task(JobSchedule.FN_DEV_MIG, device.cluster_id, node.get_id(), device.get_id(),
+                      max_retry=0, function_params={'lvol_id': lvol_id})
     return True
 
 
 def add_device_to_auto_restart(device):
-    return True
-    # return _add_task(JobSchedule.FN_DEV_RESTART, device.cluster_id, device.node_id, device.get_id())
+    return _add_task(JobSchedule.FN_DEV_RESTART, device.cluster_id, device.node_id, device.get_id())
 
 
 def add_node_to_auto_restart(node):
-    return True
-    # return _add_task(JobSchedule.FN_NODE_RESTART, node.cluster_id, node.get_id(), "")
+    return _add_task(JobSchedule.FN_NODE_RESTART, node.cluster_id, node.get_id(), "")
 
 
 def list_tasks(cluster_id):
@@ -137,20 +135,20 @@ def get_active_node_mig_task(cluster_id, node_id):
 
 
 def add_device_failed_mig_task(device_id):
-    # device = db_controller.get_storage_devices(device_id)
-    # for node in db_controller.get_storage_nodes_by_cluster_id(device.cluster_id):
-    #     for lvol_id in node.lvols:
-    #         _add_task(JobSchedule.FN_FAILED_DEV_MIG, device.cluster_id, node.get_id(), device.get_id(),
-    #                   max_retry=0, function_params={'lvol_id': lvol_id})
+    device = db_controller.get_storage_devices(device_id)
+    for node in db_controller.get_storage_nodes_by_cluster_id(device.cluster_id):
+        for lvol_id in node.lvols:
+            _add_task(JobSchedule.FN_FAILED_DEV_MIG, device.cluster_id, node.get_id(), device.get_id(),
+                      max_retry=0, function_params={'lvol_id': lvol_id})
     return True
 
 
 def add_new_device_mig_task(device_id):
-    # device = db_controller.get_storage_devices(device_id)
-    # for node in db_controller.get_storage_nodes_by_cluster_id(device.cluster_id):
-    #     for lvol_id in node.lvols:
-    #         _add_task(JobSchedule.FN_NEW_DEV_MIG, device.cluster_id, node.get_id(), device.get_id(),
-    #                   max_retry=0, function_params={'lvol_id': lvol_id})
+    device = db_controller.get_storage_devices(device_id)
+    for node in db_controller.get_storage_nodes_by_cluster_id(device.cluster_id):
+        for lvol_id in node.lvols:
+            _add_task(JobSchedule.FN_NEW_DEV_MIG, device.cluster_id, node.get_id(), device.get_id(),
+                      max_retry=0, function_params={'lvol_id': lvol_id})
     return True
 
 

{sbcli_dev-3.8.40 → sbcli_dev-3.8.42}/simplyblock_core/distr_controller.py

@@ -74,7 +74,8 @@ def disconnect_device(device):
 def get_distr_cluster_map(snodes, target_node):
     map_cluster = {}
     map_prob = []
-    for snode in snodes:
+    local_node_index = 0
+    for index, snode in enumerate(snodes):
         dev_map = {}
         dev_w_map = []
         node_w = 0
@@ -88,6 +89,7 @@ def get_distr_cluster_map(snodes, target_node):
             if snode.get_id() == target_node.get_id():
                 name = dev.alceml_bdev
                 dev_status = dev.status
+                local_node_index = index
             else:
                 for dev2 in target_node.remote_devices:
                     if dev2.get_id() == dev.get_id():
@@ -118,7 +120,8 @@ def get_distr_cluster_map(snodes, target_node):
         "UUID_node_target": "",
         "timestamp": datetime.datetime.now().isoformat("T", "seconds")+'Z',
         "map_cluster": map_cluster,
-        "map_prob": map_prob
+        "map_prob": map_prob,
+        'ppln1': local_node_index
     }
     return cl_map
 

{sbcli_dev-3.8.40 → sbcli_dev-3.8.42}/simplyblock_core/rpc_client.py

@@ -385,10 +385,11 @@ class RPCClient:
             "block_size": block_size,
             "chunk_size": chunk_size,
             "pba_page_size": pba_page_size,
-            "ha_is_non_leader": ha_is_non_leader,
         }
         if jm_vuid > 0:
             params["jm_vuid"] = jm_vuid
+            params["ha_is_non_leader"] = ha_is_non_leader
+
         if ha_comm_addrs:
             params['ha_comm_addrs'] = ha_comm_addrs
             params['ha_inode_self'] = ha_inode_self

{sbcli_dev-3.8.40 → sbcli_dev-3.8.42}/simplyblock_core/services/main_distr_event_collector.py

@@ -3,7 +3,7 @@ import threading
 import time
 
 
-from simplyblock_core import constants, kv_store, utils, rpc_client
+from simplyblock_core import constants, kv_store, utils, rpc_client, distr_controller
 from simplyblock_core.controllers import events_controller, device_controller, lvol_events
 from simplyblock_core.models.lvol_model import LVol
 
@@ -25,7 +25,6 @@ def process_device_event(event):
         storage_id = event.storage_id
 
         device = None
-        device_node = None
         for node in db_controller.get_storage_nodes():
             for dev in node.nvme_devices:
                 if dev.cluster_device_order == storage_id:
@@ -35,7 +34,6 @@ def process_device_event(event):
                         return
 
                     device = dev
-                    device_node = node
                     break
 
         if not device:
@@ -44,60 +42,63 @@ def process_device_event(event):
             return
 
         device_id = device.get_id()
-        if event.message == 'SPDK_BDEV_EVENT_REMOVE':
-            if device.node_id == node_id:
-                logger.info(f"Removing storage id: {storage_id} from node: {node_id}")
-                device_controller.device_remove(device_id)
-            else:
-                logger.info(f"Removing remote storage id: {storage_id} from node: {node_id}")
-                new_remote_devices = []
-                rpc_client = RPCClient(device_node.mgmt_ip, device_node.rpc_port,
-                                       device_node.rpc_username, device_node.rpc_password)
-                for rem_dev in device_node.remote_devices:
-                    if rem_dev.get_id() == device.get_id():
-                        ctrl_name = rem_dev.remote_bdev[:-2]
-                        rpc_client.bdev_nvme_detach_controller(ctrl_name)
-                    else:
-                        new_remote_devices.append(rem_dev)
-                device_node.remote_devices = new_remote_devices
-                device_node.write_to_db(db_controller.kv_store)
-
-        elif event.message in ['error_write', 'error_unmap']:
-            logger.info(f"Setting device to read-only")
-            device_controller.device_set_io_error(device_id, True)
-            device_controller.device_set_read_only(device_id)
+        node = db_controller.get_storage_node_by_id(node_id)
+        if device.node_id != node_id:
+            logger.info(f"Removing remote storage id: {storage_id} from node: {node_id}")
+            new_remote_devices = []
+            rpc_client = RPCClient(node.mgmt_ip, node.rpc_port,
+                                   node.rpc_username, node.rpc_password)
+            for rem_dev in node.remote_devices:
+                if rem_dev.get_id() == device.get_id():
+                    ctrl_name = rem_dev.remote_bdev[:-2]
+                    rpc_client.bdev_nvme_detach_controller(ctrl_name)
+                else:
+                    new_remote_devices.append(rem_dev)
+            node.remote_devices = new_remote_devices
+            node.write_to_db(db_controller.kv_store)
+            distr_controller.send_cluster_map_to_node(node)
+
         else:
-            logger.info(f"Setting device to unavailable")
-            device_controller.device_set_io_error(device_id, True)
-            device_controller.device_set_unavailable(device_id)
+            if event.message == 'SPDK_BDEV_EVENT_REMOVE':
+                if device.node_id == node_id:
+                    logger.info(f"Removing storage id: {storage_id} from node: {node_id}")
+                    device_controller.device_remove(device_id)
+
+            elif event.message in ['error_write', 'error_unmap']:
+                logger.info(f"Setting device to read-only")
+                device_controller.device_set_io_error(device_id, True)
+                device_controller.device_set_read_only(device_id)
+            else:
+                logger.info(f"Setting device to unavailable")
+                device_controller.device_set_io_error(device_id, True)
+                device_controller.device_set_unavailable(device_id)
 
         event.status = 'processed'
 
 
 def process_lvol_event(event):
     if event.message in ["error_open", 'error_read', "error_write", "error_unmap"]:
-        vuid = event.object_dict['vuid']
-        lvol = None
+        # vuid = event.object_dict['vuid']
+        node_id = event.node_id
+        lvols = []
         for lv in db_controller.get_lvols():  # pass
-            if lv.vuid == vuid:
-                lvol = lv
-                break
+            if lv.node_id == node_id:
+                lvols.append(lv)
 
-        if not lvol:
-            logger.error(f"LVol with vuid {vuid} not found")
-            event.status = 'lvol_not_found'
+        if not lvols:
+            logger.error(f"LVols on node {node_id} not found")
+            event.status = 'lvols_not_found'
         else:
-            lvol.io_error = True
-            if lvol.status == LVol.STATUS_ONLINE:
-                logger.info("Setting LVol to offline")
-                old_status = lvol.status
-                lvol.status = LVol.STATUS_OFFLINE
-                lvol.write_to_db(db_controller.kv_store)
-                lvol_events.lvol_status_change(lvol, lvol.status, old_status, caused_by="monitor")
-                lvol_events.lvol_io_error_change(lvol, True, False, caused_by="monitor")
-                event.status = 'processed'
-            else:
-                event.status = 'skipped'
+            for lvol in lvols:
+                if lvol.status == LVol.STATUS_ONLINE:
+                    logger.info("Setting LVol to offline")
+                    lvol.io_error = True
+                    old_status = lvol.status
+                    lvol.status = LVol.STATUS_OFFLINE
+                    lvol.write_to_db(db_controller.kv_store)
+                    lvol_events.lvol_status_change(lvol, lvol.status, old_status, caused_by="monitor")
+                    lvol_events.lvol_io_error_change(lvol, True, False, caused_by="monitor")
+            event.status = 'processed'
     else:
         logger.error(f"Unknown LVol event message: {event.message}")
         event.status = "event_unknown"