repmgr: cluster check commands - non-zero exit code if node(s) unavailable

Return ERR_CLUSTER_CHECK if one or nodes was not reachable.

Implements GitHub #447.

HISTORY

@@ -1,4 +1,4 @@
-4.0.6   2018-06-??
+4.0.6   2018-06-14
         repmgr: (witness register) prevent registration of a witness server with the
  		  same name as an existing node (Ian)
 		repmgr: (standby follow) check node has actually connected to new primary
@@ -13,6 +13,8 @@
 		  GitHub #442 (Ian)
 		repmgr: when using --dry-run, force log level to INFO to ensure output
 		  will always be displayed; GitHub #441 (Ian)
+        repmgr: (cluster matrix/crosscheck) return non-zero exit code if node
+           connection issues detected; GitHub #447 (Ian)
 		repmgrd: ensure local node is counted as quorum member; GitHub #439 (Ian)
 
 4.0.5   2018-05-02

doc/appendix-release-notes.sgml

@@ -28,6 +28,40 @@
       for more details.
 	</para>
 
+    <sect2>
+      <title>Usability enhancements</title>
+
+      <para>
+        <itemizedlist>
+          <listitem>
+            <para>
+              <command><link linkend="repmgr-cluster-crosscheck">repmgr cluster crosscheck</link></command> and
+              <command><link linkend="repmgr-cluster-matrix">repmgr cluster matrix</link></command>:
+              return non-zero exit code if node connection issues detected (GitHub #447)
+            </para>
+          </listitem>
+
+          <listitem>
+            <para>
+               <command><link linkend="repmgr-standby-clone">repmgr standby clone</link></command>:
+			   Improve handling of external configuration file copying,  including consideration in
+			   <option>--dry-run</option> check
+			   (GitHub #443)
+           </para>
+          </listitem>
+
+          <listitem>
+            <para>
+              When using <option>--dry-run</option>, force log level to <literal>INFO</literal>
+			  to ensure output will always be displayed
+			  (GitHub #441)
+           </para>
+          </listitem>
+
+        </itemizedlist>
+      </para>
+    </sect2>
+
     <sect2>
       <title>Bug fixes</title>
       <para>
@@ -51,15 +85,6 @@
           </listitem>
 
 
-          <listitem>
-            <para>
-               <command><link linkend="repmgr-standby-clone">repmgr standby clone</link></command>:
-			   Improve handling of external configuration file copying,  including consideration in
-			   <option>--dry-run</option> check
-			   (GitHub #443)
-           </para>
-          </listitem>
-
           <listitem>
             <para>
                <command><link linkend="repmgr-standby-clone">repmgr standby clone</link></command>:
@@ -84,14 +109,6 @@
            </para>
           </listitem>
 
-          <listitem>
-            <para>
-              When using <option>--dry-run</option>, force log level to <literal>INFO</literal>
-			  to ensure output will always be displayed
-			  (GitHub #441)
-           </para>
-          </listitem>
-
           <listitem>
             <para>
               <application>repmgrd</application>: ensure local node is counted as quorum member

doc/repmgr-cluster-crosscheck.sgml

@@ -38,5 +38,34 @@
       and therefore determine the state of outbound connections from that node.
     </para>
   </refsect1>
+
+  <refsect1>
+    <title>Exit codes</title>
+    <para>
+      Following exit codes can be emitted by <command>repmgr cluster crosscheck</command>:
+    </para>
+    <variablelist>
+
+      <varlistentry>
+        <term><option>SUCCESS (0)</option></term>
+        <listitem>
+          <para>
+            The check completed successfully and all nodes are reachable.
+          </para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
+        <term><option>ERR_CLUSTER_CHECK (25)</option></term>
+        <listitem>
+          <para>
+            One or more nodes could not be reached.
+          </para>
+        </listitem>
+      </varlistentry>
+
+   </variablelist>
+  </refsect1>
+
 </refentry>
 

doc/repmgr-cluster-matrix.sgml

@@ -97,5 +97,35 @@
     useful result.
   </para>
   </refsect1>
+
+
+  <refsect1>
+    <title>Exit codes</title>
+    <para>
+      Following exit codes can be emitted by <command>repmgr cluster matrix</command>:
+    </para>
+    <variablelist>
+
+      <varlistentry>
+        <term><option>SUCCESS (0)</option></term>
+        <listitem>
+          <para>
+            The check completed successfully and all nodes are reachable.
+          </para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
+        <term><option>ERR_CLUSTER_CHECK (25)</option></term>
+        <listitem>
+          <para>
+            One or more nodes could not be reached.
+          </para>
+        </listitem>
+      </varlistentry>
+
+   </variablelist>
+  </refsect1>
+
 </refentry>
 

doc/repmgr-standby-switchover.sgml

@@ -199,7 +199,7 @@
   <refsect1>
     <title>Exit codes</title>
     <para>
-      Following exit codes can be emitted by <literal>repmgr standby switchover</literal>:
+      Following exit codes can be emitted by <command>repmgr standby switchover</command>:
     </para>
     <variablelist>
 
@@ -227,7 +227,7 @@
           <para>
             The switchover was executed but a problem was encountered.
             Typically this means the former primary could not be reattached
-            as a standby.
+            as a standby. Check preceding log messages for more information.
           </para>
         </listitem>
       </varlistentry>

errcode.h

@@ -46,5 +46,6 @@
 #define ERR_SWITCHOVER_INCOMPLETE 22
 #define ERR_FOLLOW_FAIL 23
 #define ERR_REJOIN_FAIL 24
+#define ERR_CLUSTER_CHECK 25
 
 #endif							/* _ERRCODE_H_ */

repmgr-action-cluster.c

@@ -569,6 +569,8 @@ do_cluster_crosscheck(void)
 
 	t_node_status_cube **cube;
 
+	bool		error_found = false;
+
 	n = build_cluster_crosscheck(&cube, &name_length);
 	if (runtime_options.output_mode == OM_CSV)
 	{
@@ -648,9 +650,11 @@ do_cluster_crosscheck(void)
 				{
 					case -2:
 						c = '?';
+						error_found = true;
 						break;
 					case -1:
 						c = 'x';
+						error_found = true;
 						break;
 					case 0:
 						c = '*';
@@ -689,6 +693,11 @@ do_cluster_crosscheck(void)
 
 		free(cube);
 	}
+
+	if (error_found == true)
+	{
+		exit(ERR_CLUSTER_CHECK);
+	}
 }
 
 
@@ -704,6 +713,8 @@ do_cluster_matrix()
 
 	t_node_matrix_rec **matrix_rec_list;
 
+	bool		error_found = false;
+
 	n = build_cluster_matrix(&matrix_rec_list, &name_length);
 
 	if (runtime_options.output_mode == OM_CSV)
@@ -742,9 +753,11 @@ do_cluster_matrix()
 				{
 					case -2:
 						c = '?';
+						error_found = true;
 						break;
 					case -1:
 						c = 'x';
+						error_found = true;
 						break;
 					case 0:
 						c = '*';
@@ -770,6 +783,11 @@ do_cluster_matrix()
 	}
 
 	free(matrix_rec_list);
+
+	if (error_found == true)
+	{
+		exit(ERR_CLUSTER_CHECK);
+	}
 }