diff --git a/test/units/TEST-07-PID1.delegate-namespaces.sh b/test/units/TEST-07-PID1.delegate-namespaces.sh
index 9bd9691197..210635ebbc 100755
--- a/test/units/TEST-07-PID1.delegate-namespaces.sh
+++ b/test/units/TEST-07-PID1.delegate-namespaces.sh
@@ -35,11 +35,6 @@ testcase_network() {
     systemd-run -p PrivateUsersEx=self -p PrivateNetwork=yes -p DelegateNamespaces=net --wait --pipe -- ip link add veth1 type veth peer name veth2
 }
 
-testcase_cgroup() {
-    (! systemd-run -p PrivateUsersEx=self -p ProtectControlGroupsEx=private --wait --pipe -- sh -c 'echo 0 >/sys/fs/cgroup/cgroup.pressure')
-    systemd-run -p PrivateUsersEx=self -p ProtectControlGroupsEx=private -p DelegateNamespaces=cgroup --wait --pipe -- sh -c 'echo 0 >/sys/fs/cgroup/cgroup.pressure'
-}
-
 testcase_pid() {
     # MountAPIVFS=yes always bind mounts child mounts of APIVFS filesystems, which means /proc/sys is always read-only
     # so we can't write to it when running in a container.