adding the buffer size is not wrong but may cause confusing information

when the buffer is cleaned after a buffer flush which is not then
available in Solr since that is waiting for a commit. In such cases the
counter would run backwards which is prevented by ignoring the buffer
size.

source/net/yacy/cora/federate/solr/connector/ConcurrentUpdateSolrConnector.java

@@ -158,7 +158,7 @@ public class ConcurrentUpdateSolrConnector implements SolrConnector {
 
     @Override
     public long getSize() {
-        return Math.max(this.metadataCache.size(), this.connector.getSize() + this.docBuffer.size());
+        return Math.max(this.metadataCache.size(), this.connector.getSize());
     }
 
     @Override