Add explanation for fencing policy line (#238)

jsc#DOCTEAM-439
bsc#1193441

xml/article_nfs_storage.xml

@@ -199,15 +199,14 @@
        Create the file <filename>/etc/drbd.d/nfs.res</filename> with the
         following contents:
       </para>
-      <!--<remark>toms 2016-07-25: TODO bsc#981560</remark>-->
 <screen>resource nfs {
    device /dev/drbd0; <co xml:id="co-ha-quick-nfs-drbd-device"/>
    disk   /dev/nfs/work; <co xml:id="co-ha-quick-nfs-drbd-disk"/>
    meta-disk internal; <co xml:id="co-ha-quick-nfs-drbd-metadisk"/>
 
    net {
       protocol  C; <co xml:id="co-ha-quick-nfs-drbd-protocol"/>
-      fencing resource-and-stonith;
+      fencing resource-and-stonith; <co xml:id="co-ha-quick-nfs-fencing-policy"/>
    }
 
    handlers { <co xml:id="co-ha-quick-nfs-fencing-handlers"/>
@@ -251,6 +250,12 @@
          it has reached all disks, be they local or remote.
         </para>
        </callout>
+       <callout arearefs="co-ha-quick-nfs-fencing-policy">
+        <para>
+         Specifies the fencing policy. For clusters with a &stonith; device
+         configured, use <literal>resource-and-stonith</literal>.
+        </para>
+       </callout>
        <callout arearefs="co-ha-quick-nfs-fencing-handlers">
         <para>
          Enables resource-level fencing. If the DRBD replication link