From: Bruce Momjian <bruce@momjian.us>
Date: Thu, 17 Apr 2014 14:52:32 +0000 (-0400)
Subject: docs: adjustments for streaming standbys that disconnect frequently
X-Git-Tag: REL9_4_BETA1~152
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=52e757420fa98a76015c2c88432db94269f3e8f4;p=postgresql

docs: adjustments for streaming standbys that disconnect frequently

Document problems when disconnection causes loss of hot_standby_feedback
and suggest adjusting max_standby_archive_delay and
max_standby_streaming_delay.

Initial patch by Marko Tiikkaja, adjustments by me
---

diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml
index ecb51c4b1a..39d7467ddd 100644
--- a/doc/src/sgml/high-availability.sgml
+++ b/doc/src/sgml/high-availability.sgml
@@ -1889,9 +1889,14 @@ if (!triggered)
     situation will be no worse than if the standby queries were running
     directly on the primary server, and you are still getting the benefit of
     off-loading execution onto the standby.
-    <varname>max_standby_archive_delay</> must be kept large in this case,
-    because delayed WAL files might already contain entries that conflict with
-    the desired standby queries.
+    If standby servers connect and disconnect frequently, you
+    might want to make adjustments to handle the period when
+    <varname>hot_standby_feedback</> feedback is not being provided.
+    For example, consider increasing <varname>max_standby_archive_delay</>
+    so that queries are not rapidly cancelled by conflicts in WAL archive
+    files during disconnected periods.  You should also consider increasing
+    <varname>max_standby_streaming_delay</> to avoid rapid cancellations
+    by newly-arrived streaming WAL entries after reconnection.
    </para>
 
    <para>