sbcli-dev 3.9.12__zip → 4.0.1__zip

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 3.9.12
+Version: 4.0.1
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

sbcli_dev-4.0.1/env_var

@@ -0,0 +1,7 @@
+SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
+SIMPLY_BLOCK_VERSION=4.0.1
+
+
+SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:main-dev
+
+DOCKER_USER=hamdysimplyblock

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/sbcli_dev.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: sbcli-dev
-Version: 3.9.12
+Version: 4.0.1
 Summary: CLI for managing SimplyBlock cluster
 Home-page: https://www.simplyblock.io/
 Author: Hamdy

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/constants.py

@@ -11,7 +11,7 @@ LOG_WEB_DEBUG = False
 
 INSTALL_DIR = os.path.dirname(os.path.realpath(__file__))
 
-NODE_MONITOR_INTERVAL_SEC = 5
+NODE_MONITOR_INTERVAL_SEC = 10
 DEVICE_MONITOR_INTERVAL_SEC = 5
 STAT_COLLECTOR_INTERVAL_SEC = 60*5  # 5 minutes
 LVOL_STAT_COLLECTOR_INTERVAL_SEC = 5
@@ -57,7 +57,7 @@ SIMPLY_BLOCK_CLI_NAME = "sbcli-dev"
 TASK_EXEC_INTERVAL_SEC = 30
 TASK_EXEC_RETRY_COUNT = 8
 
-SIMPLY_BLOCK_SPDK_ULTRA_IMAGE = "simplyblock/spdk:feature-journal-ha-latest"
+SIMPLY_BLOCK_SPDK_ULTRA_IMAGE = "simplyblock/spdk:main-latest"
 
 GELF_PORT = 12201
 

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/controllers/device_controller.py

@@ -660,11 +660,9 @@ def add_device(device_id):
         else:
             node.remote_devices.append(device_obj)
         node.write_to_db(db_controller.kv_store)
-        distr_controller.send_cluster_map_to_node(node)
         time.sleep(1)
 
-    # send cluster map to current node:
-    distr_controller.send_cluster_map_to_node(snode)
+    distr_controller.send_dev_status_event(device_obj, device_obj.status)
 
     device_events.device_create(device_obj)
 

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/controllers/tasks_controller.py

@@ -7,6 +7,7 @@ import uuid
 from simplyblock_core import kv_store, constants, utils
 from simplyblock_core.controllers import tasks_events, device_controller
 from simplyblock_core.models.job_schedule import JobSchedule
+from simplyblock_core.models.storage_node import StorageNode
 
 logger = logging.getLogger()
 db_controller = kv_store.DBController()
@@ -155,6 +156,8 @@ 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):
+        if node.status == StorageNode.STATUS_REMOVED:
+            continue
         for bdev in node.lvstore_stack:
             if bdev['type'] == "bdev_distr":
                 _add_task(JobSchedule.FN_FAILED_DEV_MIG, device.cluster_id, node.get_id(), device.get_id(),

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/controllers/tasks_events.py

@@ -25,13 +25,16 @@ def task_create(task, caused_by=ec.CAUSED_BY_CLI):
 
 
 def task_updated(task, caused_by=ec.CAUSED_BY_CLI):
-    _task_event(task, f"Task updated: {task.uuid}", caused_by, ec.EVENT_STATUS_CHANGE)
+    return
+    # _task_event(task, f"Task updated: {task.uuid}", caused_by, ec.EVENT_STATUS_CHANGE)
 
 
 def task_status_change(task, new_state, old_status, caused_by=ec.CAUSED_BY_CLI):
-    _task_event(task, f"task status changed from: {old_status} to: {new_state}", caused_by, ec.EVENT_STATUS_CHANGE)
+    return
+    # _task_event(task, f"task status changed from: {old_status} to: {new_state}", caused_by, ec.EVENT_STATUS_CHANGE)
 
 
 def task_canceled(task, caused_by=ec.CAUSED_BY_CLI):
-    _task_event(task, f"Task canceled: {task.uuid}", caused_by, ec.EVENT_STATUS_CHANGE)
+    return
+    # _task_event(task, f"Task canceled: {task.uuid}", caused_by, ec.EVENT_STATUS_CHANGE)
 

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/distr_controller.py

@@ -24,7 +24,7 @@ def send_node_status_event(node, node_status):
     logger.debug(node_status_event)
     snodes = db_controller.get_storage_nodes_by_cluster_id(node.cluster_id)
     for node in snodes:
-        if node.status != node.STATUS_ONLINE:
+        if node.status not in [StorageNode.STATUS_ONLINE, StorageNode.STATUS_SUSPENDED]:
             continue
         logger.info(f"Sending to: {node.get_id()}")
         rpc_client = RPCClient(node.mgmt_ip, node.rpc_port, node.rpc_username, node.rpc_password, timeout=10)
@@ -43,7 +43,7 @@ def send_dev_status_event(device, dev_status):
     logger.debug(node_status_event)
     snodes = db_controller.get_storage_nodes_by_cluster_id(device.cluster_id)
     for node in snodes:
-        if node.status != StorageNode.STATUS_ONLINE:
+        if node.status not in [StorageNode.STATUS_ONLINE, StorageNode.STATUS_SUSPENDED]:
             continue
         logging.debug(f"Sending event updates, device: {storage_ID}, status: {dev_status}, node: {node.get_id()}")
         rpc_client = RPCClient(node.mgmt_ip, node.rpc_port, node.rpc_username, node.rpc_password, timeout=5, retry=2)

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/services/health_check_service.py

@@ -54,7 +54,7 @@ while True:
         for snode in snodes:
             logger.info("Node: %s, status %s", snode.get_id(), snode.status)
 
-            if snode.status not in [StorageNode.STATUS_ONLINE, StorageNode.STATUS_UNREACHABLE]:
+            if snode.status not in [StorageNode.STATUS_ONLINE, StorageNode.STATUS_UNREACHABLE, StorageNode.STATUS_SUSPENDED]:
                 logger.info(f"Node status is: {snode.status}, skipping")
                 set_node_health_check(snode, False)
                 for dev in snode.nvme_devices:

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/services/main_distr_event_collector.py

@@ -150,43 +150,45 @@ def process_event(event_id):
 def start_event_collector_on_node(node_id):
     logger.info(f"Starting Distr event collector on node: {node_id}")
 
-    while True:
-
-        snode = db_controller.get_storage_node_by_id(node_id)
-        client = rpc_client.RPCClient(
-            snode.mgmt_ip,
-            snode.rpc_port,
-            snode.rpc_username,
-            snode.rpc_password,
-            timeout=10, retry=2)
-
-        try:
-            events = client.distr_status_events_discard_then_get(0, constants.DISTR_EVENT_COLLECTOR_NUM_OF_EVENTS)
-            if events:
-                logger.info(f"Found events: {len(events)}")
-                event_ids = []
-                for ev in events:
-                    logger.debug(ev)
-                    ev_id = events_controller.log_distr_event(snode.cluster_id, snode.get_id(), ev)
-                    event_ids.append(ev_id)
-
-                for eid in event_ids:
-                    logger.info(f"Processing event: {eid}")
-                    process_event(eid)
-
-                logger.info(f"Discarding events: {len(events)}")
-                client.distr_status_events_discard_then_get(len(events), 0)
-            else:
-                logger.info("no events found, sleeping")
-
-        except Exception as e:
-            logger.error("Failed to process distr events")
-            logger.exception(e)
+    snode = db_controller.get_storage_node_by_id(node_id)
+    client = rpc_client.RPCClient(
+        snode.mgmt_ip,
+        snode.rpc_port,
+        snode.rpc_username,
+        snode.rpc_password,
+        timeout=10, retry=2)
 
+    while True:
+        page = 1
+        while True:
+            try:
+                events = client.distr_status_events_discard_then_get(
+                    0, constants.DISTR_EVENT_COLLECTOR_NUM_OF_EVENTS * page)
+                if events:
+                    logger.info(f"Found events: {len(events)}")
+                    event_ids = []
+                    for ev in events:
+                        logger.debug(ev)
+                        ev_id = events_controller.log_distr_event(snode.cluster_id, snode.get_id(), ev)
+                        event_ids.append(ev_id)
+
+                    for eid in event_ids:
+                        logger.info(f"Processing event: {eid}")
+                        process_event(eid)
+
+                    logger.info(f"Discarding events: {len(events)}")
+                    client.distr_status_events_discard_then_get(len(events), 0)
+                    page *= 10
+                else:
+                    logger.info("no events found, sleeping")
+                    break
+            except Exception as e:
+                logger.error("Failed to process distr events")
+                logger.exception(e)
+                break
         time.sleep(constants.DISTR_EVENT_COLLECTOR_INTERVAL_SEC)
 
 
-
 threads_maps = {}
 
 while True:

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/services/spdk_http_proxy_server.py

@@ -8,6 +8,7 @@ import socket
 import ssl
 import sys
 
+from http.server import HTTPServer
 from http.server import ThreadingHTTPServer
 from http.server import BaseHTTPRequestHandler
 
@@ -127,7 +128,8 @@ def run_server(host, port, user, password, cert=None):
 
     try:
         ServerHandler.key = key
-        httpd = ThreadingHTTPServer((host, port), ServerHandler)
+        # httpd = ThreadingHTTPServer((host, port), ServerHandler)
+        httpd = HTTPServer((host, port), ServerHandler)
         httpd.timeout = TIMEOUT
         if cert is not None:
             httpd.socket = ssl.wrap_socket(httpd.socket, certfile=cert, server_side=True)

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/services/tasks_runner_restart.py

@@ -172,12 +172,13 @@ def task_runner_node(task):
         task.write_to_db(db_controller.kv_store)
         return False
 
+
     # shutting down node
     logger.info(f"Shutdown node {node.get_id()}")
     ret = storage_node_ops.shutdown_storage_node(node.get_id(), force=True)
     if ret:
         logger.info(f"Node shutdown succeeded")
-    time.sleep(3)
+    time.sleep(60)
 
     # resetting node
     logger.info(f"Restart node {node.get_id()}")

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_core/storage_node_ops.py

@@ -1288,9 +1288,9 @@ def delete_storage_node(node_id):
 
     snode.remove(db_controller.kv_store)
 
-    for lvol in db_controller.get_lvols(snode.cluster_id):
-        logger.info(f"Sending cluster map to LVol: {lvol.get_id()}")
-        send_cluster_map(lvol.get_id())
+    for node in db_controller.get_storage_nodes_by_cluster_id(snode.cluster_id):
+        logger.info(f"Sending cluster map to node: {node.get_id()}")
+        send_cluster_map(node.get_id())
 
     storage_events.snode_delete(snode)
     logger.info("done")
@@ -1683,6 +1683,10 @@ def restart_storage_node(
         logger.info("Connecting to remote JMs")
         snode.remote_jm_devices = _connect_to_remote_jm_devs(snode)
 
+    logger.info(f"Sending device status event")
+    for dev in snode.nvme_devices:
+        distr_controller.send_dev_status_event(dev, dev.status)
+
     logger.info("Setting node status to Online")
     set_node_status(node_id, StorageNode.STATUS_ONLINE)
 
@@ -1697,16 +1701,12 @@ def restart_storage_node(
             node.remote_jm_devices = _connect_to_remote_jm_devs(node)
         node.write_to_db(kv_store)
 
-        logger.info(f"Sending cluster map to node {node.get_id()}")
-        distr_controller.send_cluster_map_to_node(node)
 
     logger.info("Starting migration tasks")
     for dev in snode.nvme_devices:
         if dev.status != NVMeDevice.STATUS_ONLINE:
-            logger.info(f"Device is not online: {dev.get_id()}, status: {dev.status}")
+            logger.debug(f"Device is not online: {dev.get_id()}, status: {dev.status}")
             continue
-
-        # distr_controller.send_dev_status_event(dev, NVMeDevice.STATUS_ONLINE)
         tasks_controller.add_device_mig_task(dev.get_id())
 
     # Create distribs, raid0, and lvstore and expose lvols to the fabrics

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_web/blueprints/snode_ops_k8s.py

@@ -352,6 +352,7 @@ def spdk_process_start():
             'SPDK_CPU_MASK': spdk_cpu_mask,
             'SPDK_MEM': spdk_mem,
             'MEM_GEGA': spdk_mem_gega,
+            'MEM2_GEGA': spdk_mem_gega+4,
             'SERVER_IP': data['server_ip'],
             'RPC_PORT': data['rpc_port'],
             'RPC_USERNAME': data['rpc_username'],

{sbcli_dev-3.9.12 → sbcli_dev-4.0.1}/simplyblock_web/templates/storage_deploy_spdk.yaml.j2

@@ -90,9 +90,9 @@ spec:
         resources:
           limits:
             hugepages-2Mi: {{ MEM_GEGA }}Gi
-            memory: 4Gi
+            memory: {{ MEM2_GEGA }}Gi
           requests:
-            memory: 4Gi
+            hugepages-2Mi: {{ MEM_GEGA }}Gi
 
       - name: spdk-proxy-container
         image: {{ SIMPLYBLOCK_DOCKER_IMAGE }}

sbcli_dev-3.9.12/env_var

@@ -1,7 +0,0 @@
-SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
-SIMPLY_BLOCK_VERSION=3.9.12
-
-
-SIMPLY_BLOCK_DOCKER_IMAGE=simplyblock/simplyblock:main
-
-DOCKER_USER=hamdysimplyblock