test: avoid potential data inconsistency on zapping disk

Data inconsistency might happen if the disk is accessed just after disk
zapping because direct I/O is not synchronous by itself.

Signed-off-by: Satoru Takeuchi <satoru.takeuchi@gmail.com>

tests/scripts/multi-node/build-rook.sh

@@ -50,7 +50,7 @@ function purge_ceph_vms {
     vagrant ssh "$i" -c "cat << 'EOF' > /tmp/purge-ceph.sh
     sudo rm -rf /var/lib/rook
     for disk in \$(sudo blkid | awk '/ROOK/ {print \$1}' | sed 's/[0-9]://' | uniq); do
-    sudo dd if=/dev/zero of=\$disk bs=1M count=20 oflag=direct
+    sudo dd if=/dev/zero of=\$disk bs=1M count=20 oflag=direct,dsync
     done
 EOF"
     vagrant ssh "$i" -c "bash /tmp/purge-ceph.sh"