Support remote management of cluster

aiopslab/observer/prometheus/prometheus-pvc.yml

@@ -0,0 +1,11 @@
+apiVersion: v1
+kind: PersistentVolumeClaim
+metadata:
+  name: prometheus-pvc
+spec:
+  accessModes:
+  - ReadWriteOnce
+  resources:
+    requests:
+      storage: 10Gi
+  storageClassName: openebs-hostpath

aiopslab/observer/prometheus/prometheus/values.yaml

@@ -456,7 +456,7 @@ server:
     ##   set, choosing the default provisioner.  (gp2 on AWS, standard on
     ##   GKE, AWS & OpenStack)
     ## yinfang
-    storageClass: "local-storage"
+    storageClass: "openebs-hostpath"
 
     ## Prometheus server data Persistent Volume Binding Mode
     ## If defined, volumeBindingMode: <volumeBindingMode>
@@ -483,7 +483,7 @@ server:
     ## Persistent Volume Name
     ## Useful if Persistent Volumes have been provisioned in advance and you want to use a specific one
     ## yinfang
-    volumeName: "prometheus-pv"
+    #volumeName: "prometheus-pv" # Changed in order to use PVC
 
   emptyDir:
     ## Prometheus server emptyDir volume size limit

aiopslab/orchestrator/orchestrator.py

@@ -4,6 +4,7 @@
 """Orchestrator class that interfaces with the agent and the environment."""
 
 from aiopslab.service.helm import Helm
+from aiopslab.service.kubectl import KubeCtl
 from aiopslab.session import Session
 from aiopslab.orchestrator.problems.registry import ProblemRegistry
 from aiopslab.orchestrator.parser import ResponseParser
@@ -23,6 +24,7 @@ def __init__(self):
         self.sprint = SessionPrint()
         self.execution_start_time = None
         self.execution_end_time = None
+        self.kubectl = KubeCtl()
 
     def init_problem(self, problem_id: str):
         """Initialize a problem instance for the agent to solve.
@@ -42,6 +44,17 @@ def init_problem(self, problem_id: str):
         self.session.set_problem(prob, pid=problem_id)
         self.session.set_agent(self.agent_name)
 
+        print('Setting up OpenEBS...')
+        self.kubectl.exec_command(
+            f"kubectl apply -f https://openebs.github.io/charts/openebs-operator.yaml"
+        )
+        self.kubectl.exec_command(
+            "kubectl patch storageclass openebs-hostpath -p '{\"metadata\": {\"annotations\":{\"storageclass.kubernetes.io/is-default-class\":\"true\"}}}'"
+        )
+        # Wait for openebs to be ready
+        self.kubectl.wait_for_state('openebs', 'Running')
+        print('OpenEBS setup completed.')
+
         # Setup and deploy Prometheus
         self.prometheus = Prometheus()
         self.prometheus.deploy()

aiopslab/service/helm.py

@@ -35,12 +35,7 @@ def install(**args):
         )
         dependency_output, dependency_error = dependency_process.communicate()
 
-        if version:
-            command = f"helm install {release_name} {chart_path} -n {namespace} --version {version}"
-        else:
-            command = f"helm install {release_name} {chart_path} -n {namespace}"
-
-        command = f"helm install {release_name} {chart_path} -n {namespace}"
+        command = f"helm install {release_name} {chart_path} -n {namespace} --create-namespace"
         process = subprocess.Popen(command, shell=True, stdout=subprocess.PIPE)
         output, error = process.communicate()
 

aiopslab/service/metadata/prometheus.json

@@ -13,5 +13,5 @@
         "chart_path": "observer/prometheus/prometheus/", 
         "namespace": "observe"
     },
-    "PersistentVolumeConfig": "observer/prometheus/prometheus-pv.yml" 
+    "PersistentVolumeClaimConfig": "observer/prometheus/prometheus-pvc.yml" 
 }

aiopslab/service/telemetry/prometheus.py

@@ -17,7 +17,7 @@ def __init__(self):
         self.name = None
         self.namespace = None
         self.helm_configs = {}
-        self.pv_config_file = None
+        self.pvc_config_file = None
 
         self.load_service_json()
 
@@ -39,8 +39,8 @@ def load_service_json(self):
                 chart_path = self.helm_configs["chart_path"]
                 self.helm_configs["chart_path"] = str(BASE_DIR / chart_path)
 
-        self.pv_config_file = os.path.join(
-            BASE_DIR, metadata.get("PersistentVolumeConfig")
+        self.pvc_config_file = os.path.join(
+            BASE_DIR, metadata.get("PersistentVolumeClaimConfig")
         )
 
     def get_service_json(self) -> dict:
@@ -73,8 +73,8 @@ def deploy(self):
         self._delete_pv()
         Helm.uninstall(**self.helm_configs)
 
-        if self.pv_config_file:
-            pv_name = self._get_pv_name_from_file(self.pv_config_file)
+        if self.pvc_config_file:
+            pv_name = self._get_pv_name_from_file(self.pvc_config_file)
             if not self._pv_exists(pv_name):
                 self._apply_pv()
 
@@ -87,19 +87,19 @@ def teardown(self):
         """Teardown the metric collector deployment."""
         Helm.uninstall(**self.helm_configs)
 
-        if self.pv_config_file:
+        if self.pvc_config_file:
             self._delete_pv()
 
     def _apply_pv(self):
         """Apply the PersistentVolume configuration."""
-        print(f"Applying PersistentVolume from {self.pv_config_file}")
+        print(f"Applying PersistentVolume from {self.pvc_config_file}")
         KubeCtl().exec_command(
-            f"kubectl apply -f {self.pv_config_file} -n {self.namespace}"
+            f"kubectl apply -f {self.pvc_config_file} -n {self.namespace}"
         )
 
     def _delete_pv(self):
         """Delete the PersistentVolume and associated PersistentVolumeClaim."""
-        pv_name = self._get_pv_name_from_file(self.pv_config_file)
+        pv_name = self._get_pv_name_from_file(self.pvc_config_file)
         result = KubeCtl().exec_command(f"kubectl get pv {pv_name} --ignore-not-found")
 
         if result: