diff --git a/azure/resize-disk.sh b/azure/resize-disk.sh
index d6eebc5..5f5430c 100755
--- a/azure/resize-disk.sh
+++ b/azure/resize-disk.sh
@@ -1,36 +1,51 @@
 # This script is used to attach data disk
 # to for example node destined to be postgres replica
 
 # node's name defined by the user
 nodename=${1-"dbreplica1"}
 
 # type
 type=${2-"db"}
 
-# resource group
-resource_group="euwest-${type}"
+# disk's size in gb
+disk_size=${3-2048}
 
 # not a choice
+resource_prefix="euwest"
 location=westeurope
+resource_group="euwest-${type}"
 
-# disk's size in gb
-disk_size=${3-2048}
-
-# type of node (worker, db, etc...)
+full_nodename="${nodename}-${resource_prefix}"
+
+# Depending on the types, we compute the resource group
+# worker, db, storage have dedicated shared resource group
+# other can be specifically tailored for them
+if [ $type = 'worker' ]; then   # for workers, it's a shared resource
+    resource_group="${resource_prefix}-${type}s"
+elif [ $type = 'db' ]; then     # for dbs as well
+    resource_group="${resource_prefix}-${type}"
+    full_nodename="${nodename}"
+elif [ $type = 'storage' ]; then
+    resource_group="${resource_prefix}-server"
+else # for other node types (webapp), that is specifically tailored for
+    resource_group="${resource_prefix}-${nodename}"
+fi
+
+# disk's name
 disk_name="${nodename}_pgdata0"
 
 # deallocate the vm (stop, deallocate)
 cmd="az vm deallocate \
     --resource-group ${resource_group} \
-    --name ${nodename}"
+    --name ${full_nodename}"
 echo $cmd
 $cmd
 
 # update the disk
 cmd="az disk update \
    --resource-group ${resource_group} \
    --name ${disk_name} \
    --size-gb ${disk_size}"
 
 echo $cmd
 $cmd