OSM

deploy.sh

@@ -356,6 +356,12 @@ if ! kubectl apply -f "$SCRIPT_DIR/system/metallb-config.yaml"; then
 fi
 echo "MetalLB installation and configuration complete."
 
+# Install Longhorn
+echo "Installing Longhorn..."
+helm install longhorn longhorn/longhorn --namespace longhorn-system --create-namespace --set persistence.defaultClass=true
+echo "Waiting for Longhorn pods to be ready..."
+kubectl rollout status deployment/longhorn-manager -n longhorn-system
+
 # Install Contour Ingress controller (used for routing external traffic to services)
 if [ "$K8S_CONTROLLER" == 1 ]; then
   echo "Installing Contour Ingress controller..."

docs-source/content/500-System.md

@@ -164,7 +164,7 @@ source ~/.bashrc
 ```bash
 cd ~ # Change to the home directory
 sudo apt update && sudo apt upgrade -y
-sudo apt install -y curl git unzip htop ufw aria2
+sudo apt install -y curl git unzip htop ufw aria2 open-iscsi
 git clone https://github.com/WorldHistoricalGazetteer/place.git
 ```
 

vespa/repository/api/ingestion/streamer.py

@@ -190,22 +190,18 @@ async def _parse_xml_stream(self, stream):
         Asynchronous parser for XML streams.
         """
         def parse():
-            for line in stream:
-                try:
-                    # Parse XML into a dictionary
-                    doc = xmltodict.parse(line)
+            # Use xmltodict's streaming mode to process XML elements one by one
+            try:
+                for doc in xmltodict.parse(stream, items=("item", self.item_path), stream=True):
                     yield doc
-                except Exception as e:
-                    self.logger.error(f"Failed to parse XML line: {line}. Error: {e}")
-                    continue
+            except Exception as e:
+                self.logger.error(f"Failed to parse XML stream. Error: {e}")
+                raise
 
-        for item in await asyncio.to_thread(parse):
+        # Run parsing in a separate thread to avoid blocking the event loop
+        async for item in asyncio.to_thread(parse):
             yield item
 
-        # Run synchronous parsing in a thread and asynchronously yield results
-        for elem_data in await asyncio.to_thread(parse):
-            yield elem_data
-
     def _split_triple(self, line):
         parts = line.rstrip(' .').split(' ', 2)
 

vespa/values.yaml

@@ -105,9 +105,9 @@ resources:
       memory: "4G"
   api:
     requests:
-      memory: "4G"
+      memory: "1G"
     limits:
-      memory: "8G"
+      memory: "4G"
 
 common:
   initContainer: |