docs: Enhance the pg_stat_checkpointer view documentation.

MasaoFujii · MasaoFujii · commit a19f83f87966 · 2024-10-01T02:01:57.000+09:00
This commit updates the documentation for the pg_stat_checkpointer view to clarify what kind of checkpoints or restartpoints each counter tracks. This makes it easier to understand the meaning of each counter. Previously, the num_requested description included "backend," which could be misleading since requests come from other sources as well. This commit also removes "backend" from the description of num_requested, to avoid confusion. Author: Fujii Masao Reviewed-by: Anton A. Melnikov Discussion: https://postgr.es/m/4640258e-d959-4cf0-903c-cd02389c3e05@oss.nttdata.com
diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
@@ -3051,10 +3051,7 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        <structfield>num_timed</structfield> <type>bigint</type>
       </para>
       <para>
-       Number of scheduled checkpoints due to timeout.
-       Note that checkpoints may be skipped if the server has been idle
-       since the last one, and this value counts both completed and
-       skipped checkpoints
+       Number of scheduled checkpoints due to timeout
       </para></entry>
      </row>
 
@@ -3063,7 +3060,7 @@ description | Waiting for a newly initialized WAL file to reach durable storage
        <structfield>num_requested</structfield> <type>bigint</type>
       </para>
       <para>
-       Number of backend requested checkpoints
+       Number of requested checkpoints
       </para></entry>
      </row>
 
@@ -3146,6 +3143,18 @@ description | Waiting for a newly initialized WAL file to reach durable storage
    </tgroup>
   </table>
 
+  <para>
+   Checkpoints may be skipped if the server has been idle since the last one.
+   <structfield>num_timed</structfield> and
+   <structfield>num_requested</structfield> count both completed and skipped
+   checkpoints, while <structfield>num_done</structfield> tracks only
+   the completed ones.  Similarly, restartpoints may be skipped
+   if the last replayed checkpoint record is already the last restartpoint.
+   <structfield>restartpoints_timed</structfield> and
+   <structfield>restartpoints_req</structfield> count both completed and
+   skipped restartpoints, while <structfield>restartpoints_done</structfield>
+   tracks only the completed ones.
+  </para>
  </sect2>
 
  <sect2 id="monitoring-pg-stat-wal-view">
