diff --git a/docs/reference/modules/cluster/disk_allocator.asciidoc b/docs/reference/modules/cluster/disk_allocator.asciidoc
index 765f249f20e50..314b4b877737b 100644
--- a/docs/reference/modules/cluster/disk_allocator.asciidoc
+++ b/docs/reference/modules/cluster/disk_allocator.asciidoc
@@ -69,7 +69,7 @@ PUT /twitter/_settings
 
 `cluster.routing.allocation.disk.include_relocations`::
 
-    deprecated[7.5, Future versions will always account for relocations.]
+    deprecated:[7.5.0, Future versions will always account for relocations.]
     Defaults to +true+, which means that Elasticsearch will take into account
     shards that are currently being relocated to the target node when computing
     a node's disk usage. Taking relocating shards' sizes into account may,