@@ -320,20 +320,14 @@ NET STOP postgresql-9.0
320320 <title>Prepare for standby server upgrades</title>
321321
322322 <para>
323- If you are upgrading standby servers (as outlined in section <xref
324- linkend="pgupgrade-step-replicas">) , verify that the old standby
323+ If you are upgrading standby servers using methods outlined in section <xref
324+ linkend="pgupgrade-step-replicas">, verify that the old standby
325325 servers are caught up by running <application>pg_controldata</>
326326 against the old primary and standby clusters. Verify that the
327327 <quote>Latest checkpoint location</> values match in all clusters.
328328 (There will be a mismatch if old standby servers were shut down
329329 before the old primary.)
330330 </para>
331-
332- <para>
333- Also, if upgrading standby servers, change <varname>wal_level</>
334- to <literal>replica</> in the <filename>postgresql.conf</> file on
335- the new primary cluster.
336- </para>
337331 </step>
338332
339333 <step>
@@ -423,12 +417,18 @@ pg_upgrade.exe
423417 <para>
424418 If you used link mode and have Streaming Replication (see <xref
425419 linkend="streaming-replication">) or Log-Shipping (see <xref
426- linkend="warm-standby">) standby servers, follow these steps to
427- upgrade them. You will not be running <application>pg_upgrade</> on
420+ linkend="warm-standby">) standby servers, you can follow these steps to
421+ quickly upgrade them. You will not be running <application>pg_upgrade</> on
428422 the standby servers, but rather <application>rsync</> on the primary.
429- Do not start any servers yet. If you did <emphasis>not</> use link
430- mode, skip the instructions in this section and simply recreate the
431- standby servers.
423+ Do not start any servers yet.
424+ </para>
425+
426+ <para>
427+ If you did <emphasis>not</> use link mode, do not have or do not
428+ want to use <application>rsync</>, or want an easier solution, skip
429+ the instructions in this section and simply recreate the standby
430+ servers once <application>pg_upgrade</> completes and the new primary
431+ is running.
432432 </para>
433433
434434 <substeps>
@@ -448,7 +448,7 @@ pg_upgrade.exe
448448 <para>
449449 Make sure the new standby data directories do <emphasis>not</>
450450 exist or are empty. If <application>initdb</> was run, delete
451- the standby server data directories.
451+ the standby servers' new data directories.
452452 </para>
453453 </step>
454454
@@ -474,9 +474,10 @@ pg_upgrade.exe
474474 <title>Save configuration files</title>
475475
476476 <para>
477- Save any configuration files from the standbys you need to keep,
478- e.g. <filename>postgresql.conf</>, <literal>recovery.conf</>,
479- as these will be overwritten or removed in the next step.
477+ Save any configuration files from the old standbys' data
478+ directories you need to keep, e.g. <filename>postgresql.conf</>,
479+ <literal>recovery.conf</>, because these will be overwritten or
480+ removed in the next step.
480481 </para>
481482 </step>
482483
@@ -507,6 +508,12 @@ rsync --archive --delete --hard-links --size-only /opt/PostgreSQL/9.5/data \
507508 /opt/PostgreSQL/9.6/data standby.example.com:/opt/PostgreSQL
508509</programlisting>
509510
511+ You can verify what the command will do using
512+ <application>rsync</>'s <option>--dry-run</> option. While
513+ <application>rsync</> must be run on the primary for at least one
514+ standby, it is possible to run <application>rsync</> on an upgraded
515+ standby to upgrade other standbys, as long as the upgraded standby
516+ has not been started.
510517 </para>
511518
512519 <para>
0 commit comments