Fix triplicate API calls on GET commands

This was never exposed before, so expose it for use in other functions
being built.

api-daemon/pvcapid/flaskapi.py

@@ -597,7 +597,7 @@ class API_Status(Resource):
         Set the cluster maintenance mode
         ---
         tags:
-          - node
+          - root
         parameters:
           - in: query
             name: state
@@ -622,6 +622,45 @@ class API_Status(Resource):
 api.add_resource(API_Status, "/status")
 
 
+# /status/primary_node
+class API_Status_Primary(Resource):
+    def get(self):
+        """
+        Return the name of the current primary node.
+        ---
+        tags:
+          - root
+        responses:
+          200:
+            description: OK
+            schema:
+              type: object
+              properties:
+                primary_node:
+                  type: string
+                  description: The name of the current primary node
+          204:
+            description: No content
+            schema:
+              type: object
+              properties:
+                primary_node:
+                  type: string
+                  description: An empty response; there is not currently a primary node, try again later
+        """
+        primary_node = get_primary_node()
+        if primary_node is None:
+            retdata = None
+            retcode = 204
+        else:
+            retdata = {"primary_node": primary_node}
+            retcode = 200
+        return retdata, retcode
+
+
+api.add_resource(API_Status_Primary, "/status/primary_node")
+
+
 # /metrics
 class API_Metrics(Resource):
     def get(self):
@@ -838,6 +877,39 @@ class API_Faults(Resource):
         ---
         tags:
           - faults
+        definitions:
+          - schema:
+              type: object
+              id: fault
+              properties:
+                id:
+                  type: string
+                  description: The ID of the fault
+                  example: "10ae144b78b4cc5fdf09e2ebbac51235"
+                first_reported:
+                  type: date
+                  description: The first time the fault was reported
+                  example: "2023-12-01 16:47:59.849742"
+                last_reported:
+                  type: date
+                  description: The last time the fault was reported
+                  example: "2023-12-01 17:39:45.188398"
+                acknowledged_at:
+                  type: date
+                  description: The time the fault was acknowledged, or empty if not acknowledged
+                  example: "2023-12-01 17:50:00.000000"
+                status:
+                  type: string
+                  description: The current state of the fault, either "new" or "ack" (acknowledged)
+                  example: "new"
+                health_delta:
+                  type: integer
+                  description: The health delta (amount it reduces cluster health from 100%) of the fault
+                  example: 25
+                message:
+                  type: string
+                  description: The textual description of the fault
+                  example: "Node hv1 was at 40% (psur@-10%, psql@-50%) <= 50% health"
         parameters:
           - in: query
             name: sort_key
@@ -857,37 +929,7 @@ class API_Faults(Resource):
             schema:
               type: array
               items:
-                type: object
-                id: fault
-                properties:
-                  id:
-                    type: string
-                    description: The ID of the fault
-                    example: "10ae144b78b4cc5fdf09e2ebbac51235"
-                  first_reported:
-                    type: date
-                    description: The first time the fault was reported
-                    example: "2023-12-01 16:47:59.849742"
-                  last_reported:
-                    type: date
-                    description: The last time the fault was reported
-                    example: "2023-12-01 17:39:45.188398"
-                  acknowledged_at:
-                    type: date
-                    description: The time the fault was acknowledged, or empty if not acknowledged
-                    example: "2023-12-01 17:50:00.000000"
-                  status:
-                    type: string
-                    description: The current state of the fault, either "new" or "ack" (acknowledged)
-                    example: "new"
-                  health_delta:
-                    type: integer
-                    description: The health delta (amount it reduces cluster health from 100%) of the fault
-                    example: 25
-                  message:
-                    type: string
-                    description: The textual description of the fault
-                    example: "Node hv1 was at 40% (psur@-10%, psql@-50%) <= 50% health"
+                $ref: '#/definitions/fault'
         """
         return api_helper.fault_list(sort_key=reqargs.get("sort_key", "last_reported"))
 
@@ -948,8 +990,6 @@ class API_Faults_Element(Resource):
             schema:
               type: array
               items:
-                type: object
-                id: fault
                 $ref: '#/definitions/fault'
         """
         return api_helper.fault_list(limit=fault_id)
@@ -1625,7 +1665,7 @@ class API_VM_Root(Resource):
                         descrpition: Unix timestamp of the snapshot
                       age:
                         type: string
-                        description: Human-readable age of the snapshot in the largest viable unit: seconds, minutes, hours, days
+                        description: Human-readable age of the snapshot in the largest viable unit (seconds, minutes, hours, days)
                       rbd_snapshots:
                         type: array
                         items:

client-cli/pvc/lib/common.py

@@ -158,9 +158,10 @@ def call_api(
                     if response.status_code in retry_on_code:
                         failed = True
                         continue
+                    break
                 except requests.exceptions.ConnectionError:
                     failed = True
-                    pass
+                    continue
             if failed:
                 error = f"Code {response.status_code}" if response else "Timeout"
                 raise requests.exceptions.ConnectionError(

daemon-common/ceph.py

@@ -1015,23 +1015,27 @@ def add_snapshot(zkhandler, pool, volume, name, zk_only=False):
                 ),
             )
 
-    # 2. Add the snapshot to Zookeeper
+    # 2. Get snapshot stats
+    retcode, stdout, stderr = common.run_os_command(
+        "rbd info --format json {}/{}@{}".format(pool, volume, name)
+    )
+    snapstats = stdout
+
+    # 3. Add the snapshot to Zookeeper
     zkhandler.write(
         [
             (("snapshot", f"{pool}/{volume}/{name}"), ""),
-            (("snapshot.stats", f"{pool}/{volume}/{name}"), "{}"),
+            (("snapshot.stats", f"{pool}/{volume}/{name}"), snapstats),
         ]
     )
 
-    # 3. Update the count of snapshots on this volume
+    # 4. Update the count of snapshots on this volume
     volume_stats_raw = zkhandler.read(("volume.stats", f"{pool}/{volume}"))
     volume_stats = dict(json.loads(volume_stats_raw))
-    # Format the size to something nicer
     volume_stats["snapshot_count"] = volume_stats["snapshot_count"] + 1
-    volume_stats_raw = json.dumps(volume_stats)
     zkhandler.write(
         [
-            (("volume.stats", f"{pool}/{volume}"), volume_stats_raw),
+            (("volume.stats", f"{pool}/{volume}"), json.dumps(volume_stats)),
         ]
     )
 
@@ -1174,6 +1178,11 @@ def get_list_snapshot(zkhandler, target_pool, target_volume, limit=None, is_fuzz
             continue
         if target_volume and volume_name != target_volume:
             continue
+        snapshot_stats = json.loads(
+            zkhandler.read(
+                ("snapshot.stats", f"{pool_name}/{volume_name}/{snapshot_name}")
+            )
+        )
         if limit:
             try:
                 if re.fullmatch(limit, snapshot_name):
@@ -1182,13 +1191,19 @@ def get_list_snapshot(zkhandler, target_pool, target_volume, limit=None, is_fuzz
                             "pool": pool_name,
                             "volume": volume_name,
                             "snapshot": snapshot_name,
+                            "stats": snapshot_stats,
                         }
                     )
             except Exception as e:
                 return False, "Regex Error: {}".format(e)
         else:
             snapshot_list.append(
-                {"pool": pool_name, "volume": volume_name, "snapshot": snapshot_name}
+                {
+                    "pool": pool_name,
+                    "volume": volume_name,
+                    "snapshot": snapshot_name,
+                    "stats": snapshot_stats,
+                }
             )
 
     return True, sorted(snapshot_list, key=lambda x: str(x["snapshot"]))