DOC: improved instructions for removing and restoring nodes

Author: Liudmila Mantrova

doc/src/sgml/multimaster.sgml

@@ -968,7 +968,7 @@ multimaster.conn_strings = 'dbname=mydb user=myuser host=node1,
       </listitem>
       <listitem>
         <para>
-          Start <productname>PostgreSQL</productname> on the new node:
+          Start <productname>&project;</productname> on the new node:
         </para>
         <programlisting>
 pg_ctl -D <replaceable>datadir</replaceable> -l <replaceable>pg.log</replaceable> start
@@ -1017,10 +1017,16 @@ SELECT mtm.stop_node(3);
 </programlisting>
     <para>
       This command excludes node 3 from the cluster and stops replication to
-      this node. While the WAL lag between the node and the current cluster state 
+      this node.
+    </para>
+    <para>While the WAL lag between the node and the current cluster state 
       is less than the <varname>multimaster.max_recovery_lag</varname> value,
-      you can restore the node using the <function>mtm.recover_node</function> function.
-      For details, see <xref linkend="multimaster-restoring-a-node-manually">.
+      you can restore the node by running the following command:
+<programlisting>
+SELECT mtm.recover_node(3);
+</programlisting>
+      Otherwise, follow the procedure described in
+      <xref linkend="multimaster-restoring-a-node-manually">.
     </para>
     <note>
     <para>
@@ -1031,21 +1037,40 @@ SELECT mtm.stop_node(3);
     </para>
     </note>
     <para>
-      To permanently drop the node from the cluster, run the
-      <literal>mtm.stop_node()</literal> function with the <literal>drop_slot</literal> parameter
-      set to <literal>true</literal>:
+      To permanently drop the node from the cluster:
+    </para>
+    <orderedlist>
+     <listitem>
+      <para>Run the <literal>mtm.stop_node()</literal> function with
+      the <literal>drop_slot</literal> parameter set to <literal>true</literal>:
     </para>
     <programlisting>
 SELECT mtm.stop_node(3, true);
 </programlisting>
-    <para>
+     <para>
       This disables replication slots for node 3 on all cluster nodes and stops replication to
-      this node. If you would like to return the node to the cluster, you will have to add it
-      as a new node. For details, see <xref linkend="multimaster-adding-new-nodes-to-the-cluster">.
+      this node.
+     </para>
+    </listitem>
+    <listitem>
+     <para>Adjust <xref linkend="multimaster-node-id"> and
+     <xref linkend="multimaster-conn-strings"> settings in
+     <filename>postgresql.conf</filename> on the remaining
+     cluster nodes to reflect the new state of the cluster.
+     </para>
+    </listitem>
+    <listitem>
+     <para>Edit the <filename>pg_hba.conf</filename> file on the remaining cluster
+     nodes to disable replication to the removed node, if required.
+     </para>
+    </listitem>
+    </orderedlist>
+    <para>If you would like to return the node to the cluster later, you will have to add it
+      as a new node, as explained in <xref linkend="multimaster-adding-new-nodes-to-the-cluster">.
     </para>
   </sect3>
   <sect3 id="multimaster-restoring-a-node-manually">
-    <title>Restoring a Cluster Node</title>
+    <title>Restoring a Cluster Node Manually</title>
     <para>
       The <filename>multimaster</filename> extension can <link linkend="multimaster-failure-detection-and-recovery">automatically restore</link> a failed node if the WAL is available for the time when the node was disconnected from the cluster. However, if the data updates on the alive nodes exceed the allowed WAL size specified in the <literal>multimaster.max_recovery_lag</literal> variable, automatic recovery is impossible. In this case, you can manually restore the failed node.
       </para>
@@ -1077,7 +1102,19 @@ pg_basebackup -D <replaceable>datadir</replaceable> -h node1 -x
       </listitem>
       <listitem>
         <para>
-          Start <productname>PostgreSQL</productname> on the restored node:
+         On the restored node, update the <varname>multimaster.node_id</varname>
+         setting to the value this node used to have before the failure.
+        </para>
+      </listitem>
+      <listitem>
+        <para>
+         Make sure replication is enabled between the restored node
+         and the rest of the cluster.
+        </para>
+      </listitem>
+      <listitem>
+        <para>
+          Start <productname>&project;</productname> on the restored node:
         </para>
         <programlisting>
 pg_ctl -D <replaceable>datadir</replaceable> -l <replaceable>pg.log</replaceable> start