diff --git a/scripts/refresh-sharding-and-restart b/scripts/refresh-sharding-and-restart
index 4d29c706a7..bb82f769be 100755
--- a/scripts/refresh-sharding-and-restart
+++ b/scripts/refresh-sharding-and-restart
@@ -25,5 +25,5 @@ fi
 # clients getting into reload loops ending in crashing on 500 response
 # while Django is restarting.  For this reason it's important to
 # reload nginx only after Django and Tornado.
-"$(dirname "$0")/restart-server" --skip-client-reloads --tornado-reshard
+"$(dirname "$0")/restart-server" --tornado-reshard
 service nginx reload
diff --git a/scripts/restart-server b/scripts/restart-server
index f36dce046b..f65f10c572 100755
--- a/scripts/restart-server
+++ b/scripts/restart-server
@@ -303,7 +303,12 @@ if (action == "start" or args.less_graceful) and not args.only_django:
         logging.info("Starting workers")
         subprocess.check_call(["supervisorctl", "start", *workers])
 
-if has_application_server() and not args.skip_client_reloads and not args.only_django:
+if (
+    has_application_server()
+    and not args.skip_client_reloads
+    and not args.only_django
+    and not args.tornado_reshard
+):
     # All of the servers have been (re)started; now enqueue events in
     # the Tornado servers to tell clients to reload.
     logging.info("Sending reload events to clients in the background")