Add request ID tracking to SPDK HTTP proxy

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023-2025 simplyblock GmbH
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

docker/Dockerfile_base

@@ -38,3 +38,4 @@ RUN pip3 install setuptools --upgrade
 COPY requirements.txt requirements.txt
 
 RUN pip3 install -r requirements.txt
+

docs/talos.md

@@ -19,26 +19,12 @@ kubectl label namespace simplyblock \
   --overwrite
 ```
 
-
-Patch the host machine so that OpenEBS could work
-
 Create a machine config patch with the contents below and save as patch.yaml
 ```
 cat > patch.yaml <<'EOF'
 machine:
   sysctls:
     vm.nr_hugepages: "1024"
-  nodeLabels:
-    openebs.io/engine: mayastor
-  kubelet:
-    extraMounts:
-      - destination: /var/openebs/local
-        type: bind
-        source: /var/openebs/local
-        options:
-          - rbind
-          - rshared
-          - rw
 EOF
 
 talosctl -e <endpoint ip/hostname> -n <node ip/hostname> patch mc -p @patch.yaml

simplyblock_core/cluster_ops.py

@@ -371,8 +371,6 @@ def create_cluster(blk_size, page_size_in_blocks, cli_pass,
 
     cluster.write_to_db(db_controller.kv_store)
 
-    qos_controller.add_class("Default", 100, cluster.get_id())
-
     cluster_events.cluster_create(cluster)
 
     mgmt_node_ops.add_mgmt_node(dev_ip, mode, cluster.uuid)
@@ -459,6 +457,7 @@ def add_cluster(blk_size, page_size_in_blocks, cap_warn, cap_crit, prov_cap_warn
     cluster.strict_node_anti_affinity = strict_node_anti_affinity
 
     default_cluster = clusters[0]
+    cluster.mode = default_cluster.mode
     cluster.db_connection = default_cluster.db_connection
     cluster.grafana_secret = monitoring_secret if default_cluster.mode == "kubernetes" else default_cluster.grafana_secret
     cluster.grafana_endpoint = default_cluster.grafana_endpoint
@@ -1176,9 +1175,13 @@ def update_cluster(cluster_id, mgmt_only=False, restart=False, spdk_image=None,
         for service in cluster_docker.services.list():
             if image_parts in service.attrs['Spec']['Labels']['com.docker.stack.image'] or \
             "simplyblock" in service.attrs['Spec']['Labels']['com.docker.stack.image']:
-                logger.info(f"Updating service {service.name}")
-                service.update(image=service_image, force_update=True)
-                service_names.append(service.attrs['Spec']['Name'])
+                if service.name == "app_CachingNodeMonitor":
+                    logger.info(f"Removing service {service.name}")
+                    service.remove()
+                else:
+                    logger.info(f"Updating service {service.name}")
+                    service.update(image=service_image, force_update=True)
+                    service_names.append(service.attrs['Spec']['Name'])
 
         if "app_SnapshotMonitor" not in service_names:
             logger.info("Creating snapshot monitor service")
@@ -1191,6 +1194,18 @@ def update_cluster(cluster_id, mgmt_only=False, restart=False, spdk_image=None,
                 networks=["host"],
                 constraints=["node.role == manager"]
             )
+
+        if "app_TasksRunnerLVolSyncDelete" not in service_names:
+            logger.info("Creating lvol sync delete service")
+            cluster_docker.services.create(
+                image=service_image,
+                command="python simplyblock_core/services/tasks_runner_sync_lvol_del.py",
+                name="app_TasksRunnerLVolSyncDelete",
+                mounts=["/etc/foundationdb:/etc/foundationdb"],
+                env=["SIMPLYBLOCK_LOG_LEVEL=DEBUG"],
+                networks=["host"],
+                constraints=["node.role == manager"]
+            )
         logger.info("Done updating mgmt cluster")
 
     elif cluster.mode == "kubernetes":

simplyblock_core/constants.py

@@ -133,7 +133,8 @@ def get_config_var(name, default=None):
 LVOL_NVME_CONNECT_NR_IO_QUEUES=3
 LVOL_NVME_KEEP_ALIVE_TO=10
 LVOL_NVME_KEEP_ALIVE_TO_TCP=7
-LVOL_NVMF_PORT_START=int(os.getenv('LVOL_NVMF_PORT_START', 9100))
+LVOL_NVMF_PORT_ENV = os.getenv("LVOL_NVMF_PORT_START", "")
+LVOL_NVMF_PORT_START = int(LVOL_NVMF_PORT_ENV) if LVOL_NVMF_PORT_ENV else 9100
 QPAIR_COUNT=32
 CLIENT_QPAIR_COUNT=3
 NVME_TIMEOUT_US=8000000
@@ -224,4 +225,4 @@ def get_config_var(name, default=None):
 
 qos_class_meta_and_migration_weight_percent = 25
 
-MIG_PARALLEL_JOBS = 16
+MIG_PARALLEL_JOBS = 64

simplyblock_core/controllers/tasks_controller.py

@@ -70,6 +70,11 @@ def _add_task(function_name, cluster_id, node_id, device_id,
         if task_id:
             logger.info(f"Task found, skip adding new task: {task_id}")
             return False
+    elif function_name == JobSchedule.FN_LVOL_SYNC_DEL:
+        task_id = get_lvol_sync_del_task(cluster_id, node_id, function_params['lvol_bdev_name'])
+        if task_id:
+            logger.info(f"Task found, skip adding new task: {task_id}")
+            return False
 
     task_obj = JobSchedule()
     task_obj.uuid = str(uuid.uuid4())
@@ -386,3 +391,21 @@ def get_jc_comp_task(cluster_id, node_id, jm_vuid=0):
                 if jm_vuid and "jm_vuid" in task.function_params and task.function_params["jm_vuid"] == jm_vuid:
                     return task.uuid
     return False
+
+
+def add_lvol_sync_del_task(cluster_id, node_id, lvol_bdev_name):
+    return _add_task(JobSchedule.FN_LVOL_SYNC_DEL, cluster_id, node_id, "",
+                     function_params={"lvol_bdev_name": lvol_bdev_name}, max_retry=10)
+
+def get_lvol_sync_del_task(cluster_id, node_id, lvol_bdev_name=None):
+    tasks = db.get_job_tasks(cluster_id)
+    for task in tasks:
+        if task.function_name == JobSchedule.FN_LVOL_SYNC_DEL and task.node_id == node_id :
+            if task.status != JobSchedule.STATUS_DONE and task.canceled is False:
+                if lvol_bdev_name:
+                    if "lvol_bdev_name" in task.function_params and task.function_params["lvol_bdev_name"] == lvol_bdev_name:
+                        return task.uuid
+                else:
+                    return task.uuid
+    return False
+

simplyblock_core/env_var

@@ -1,6 +1,6 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
-SIMPLY_BLOCK_VERSION=19.2.23
+SIMPLY_BLOCK_VERSION=19.2.25
 
-SIMPLY_BLOCK_DOCKER_IMAGE=public.ecr.aws/simply-block/simplyblock:main
+SIMPLY_BLOCK_DOCKER_IMAGE=public.ecr.aws/simply-block/simplyblock:main-lvol-sync-delete
 SIMPLY_BLOCK_SPDK_ULTRA_IMAGE=public.ecr.aws/simply-block/ultra:main-latest
 

simplyblock_core/models/cluster.py

@@ -45,7 +45,7 @@ class Cluster(BaseModel):
     distr_npcs: int = 0
     enable_node_affinity: bool = False
     grafana_endpoint: str = ""
-    mode: str = ""
+    mode: str = "docker"
     grafana_secret: str = ""
     contact_point: str = ""
     ha_type: str = "single"

simplyblock_core/models/job_schedule.py

@@ -22,6 +22,7 @@ class JobSchedule(BaseModel):
     FN_BALANCING_AFTER_DEV_REMOVE = "balancing_on_dev_rem"
     FN_BALANCING_AFTER_DEV_EXPANSION = "balancing_on_dev_add"
     FN_JC_COMP_RESUME = "jc_comp_resume"
+    FN_LVOL_SYNC_DEL = "lvol_sync_del"
 
     canceled: bool = False
     cluster_id: str = ""

simplyblock_core/models/storage_node.py

@@ -102,7 +102,6 @@ class StorageNode(BaseNodeObject):
     hublvol: HubLVol = None  # type: ignore[assignment]
     active_tcp: bool = True
     active_rdma: bool = False
-    lvol_sync_del_queue: List[str] = []
 
     def rpc_client(self, **kwargs):
         """Return rpc client to this node

simplyblock_core/rpc_client.py

@@ -379,11 +379,11 @@ def create_lvol(self, name, size_in_mib, lvs_name, lvol_priority_class=0, ndcs=0
             "clear_method": "unmap",
             "lvol_priority_class": lvol_priority_class,
         }
-        # if ndcs or npcs:
-        #     params.update({
-        #         'ndcs' : ndcs,
-        #         'npcs' : npcs,
-        #     })
+        if ndcs or npcs:
+            params.update({
+                'ndcs' : ndcs,
+                'npcs' : npcs,
+            })
         return self._request("bdev_lvol_create", params)
 
     def delete_lvol(self, name, del_async=False):
@@ -922,7 +922,7 @@ def distr_migration_status(self, name):
         params = {"name": name}
         return self._request("distr_migration_status", params)
 
-    def distr_migration_failure_start(self, name, storage_ID, qos_high_priority=False, job_size=1024, jobs=4):
+    def distr_migration_failure_start(self, name, storage_ID, qos_high_priority=False, job_size=64, jobs=64):
         params = {
             "name": name,
             "storage_ID": storage_ID,
@@ -935,7 +935,7 @@ def distr_migration_failure_start(self, name, storage_ID, qos_high_priority=Fals
             params["jobs"] = jobs
         return self._request("distr_migration_failure_start", params)
 
-    def distr_migration_expansion_start(self, name, qos_high_priority=False, job_size=1024, jobs=4):
+    def distr_migration_expansion_start(self, name, qos_high_priority=False, job_size=64, jobs=64):
         params = {
             "name": name,
         }

simplyblock_core/scripts/charts/Chart.yaml

@@ -26,11 +26,6 @@ dependencies:
     version: "25.18.0"
     repository: "https://prometheus-community.github.io/helm-charts"
     condition: monitoring.enabled
-  - name: openebs
-    version: 3.9.0 
-    repository: https://openebs.github.io/charts
-    alias: openebs
-    condition: openebs.enabled
   - name: ingress-nginx
     version: 4.10.1
     repository: "https://kubernetes.github.io/ingress-nginx"

simplyblock_core/scripts/charts/templates/csi-hostpath-driverinfo.yaml

@@ -0,0 +1,24 @@
+apiVersion: storage.k8s.io/v1
+kind: CSIDriver
+metadata:
+  name: hostpath.csi.k8s.io
+  labels:
+    app.kubernetes.io/instance: hostpath.csi.k8s.io
+    app.kubernetes.io/part-of: csi-driver-host-path
+    app.kubernetes.io/name: hostpath.csi.k8s.io
+    app.kubernetes.io/component: csi-driver
+spec:
+  # Supports persistent and ephemeral inline volumes.
+  volumeLifecycleModes:
+  - Persistent
+  - Ephemeral
+  # To determine at runtime which mode a volume uses, pod info and its
+  # "csi.storage.k8s.io/ephemeral" entry are needed.
+  podInfoOnMount: true
+  # No attacher needed.
+  attachRequired: false
+  storageCapacity: false
+  # Kubernetes may use fsGroup to change permissions and ownership 
+  # of the volume to match user requested fsGroup in the pod's SecurityPolicy
+  fsGroupPolicy: File
+