database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-31 22:04:40 +03:00
Code Activity

Introduce wal_level GUC to explicitly control if information needed for

Browse Source 
archival or hot standby should be WAL-logged, instead of deducing that from
other options like archive_mode. This replaces recovery_connections GUC in
the primary, where it now has no effect, but it's still used in the standby
to enable/disable hot standby.

Remove the WAL-logging of "unlogged operations", like creating an index
without WAL-logging and fsyncing it at the end. Instead, we keep a copy of
the wal_mode setting and the settings that affect how much shared memory a
hot standby server needs to track master transactions (max_connections,
max_prepared_xacts, max_locks_per_xact) in pg_control. Whenever the settings
change, at server restart, write a WAL record noting the new settings and
update pg_control. This allows us to notice the change in those settings in
the standby at the right moment, they used to be included in checkpoint
records, but that meant that a changed value was not reflected in the
standby until the first checkpoint after the change.

Bump PG_CONTROL_VERSION and XLOG_PAGE_MAGIC. Whack XLOG_PAGE_MAGIC back to
the sequence it used to follow, before hot standby and subsequent patches
changed it to 0x9003.
This commit is contained in:
Heikki Linnakangas
2010-04-28 16:10:43 +00:00
parent a2de4826e9
commit 9b8a73326e
22 changed files with 340 additions and 224 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
doc/src/sgml/backup.sgml
View File

@ -1,4 +1,4 @@
<!-- $PostgreSQL: pgsql/doc/src/sgml/backup.sgml,v 2.152 2010/04/20 00:26:06 rhaas Exp $ -->
<!-- $PostgreSQL: pgsql/doc/src/sgml/backup.sgml,v 2.153 2010/04/28 16:10:39 heikki Exp $ -->
<chapter id="backup">
<title>Backup and Restore</title>
@ -689,8 +689,7 @@ archive_command = 'test ! -f /mnt/server/archivedir/%f &amp;&amp; cp %p /mnt/ser
</para>
<para>
When <varname>archive_mode</> is <literal>off</> and <xref
linkend="guc-max-wal-senders"> is zero some SQL commands
When <varname>wal_level</> is <literal>minimal</> some SQL commands
are optimized to avoid WAL logging, as described in <xref
linkend="populate-pitr">. If archiving or streaming replication were
turned on during execution of one of these statements, WAL would not

64
doc/src/sgml/config.sgml
View File

@ -1,4 +1,4 @@
<!-- $PostgreSQL: pgsql/doc/src/sgml/config.sgml,v 1.270 2010/04/26 10:51:59 rhaas Exp $ -->
<!-- $PostgreSQL: pgsql/doc/src/sgml/config.sgml,v 1.271 2010/04/28 16:10:39 heikki Exp $ -->
<chapter Id="runtime-config">
<title>Server Configuration</title>
@ -1353,6 +1353,45 @@ SET ENABLE_SEQSCAN TO OFF;
<title>Settings</title>
<variablelist>
<varlistentry id="guc-wal-level" xreflabel="wal_level">
<term><varname>wal_level</varname> (<type>enum</type>)</term>
<indexterm>
<primary><varname>wal_level</> configuration parameter</primary>
</indexterm>
<listitem>
<para>
<varname>wal_level</> determines how much information is written
to the WAL. The default value is <literal>minimal</>, which writes
only minimal information needed to recover from a crash or immediate
shutdown. <literal>archive</> adds logging required for WAL archiving,
and <literal>hot_standby</> further adds information required to run
read-only queries on a standby server.
This parameter can only be set at server start.
</para>
<para>
In <literal>minimal</> level, WAL-logging of some bulk operations, like
<command>CREATE INDEX</>, <command>CLUSTER</> and <command>COPY</> on
a table that was created or truncated in the same transaction can be
safely skipped, which can make those operations much faster (see
<xref linkend="populate-pitr">). But minimal WAL does not contain
enough information to reconstruct the data from a base backup and the
WAL logs, so at least <literal>archive</> level must be used to enable
WAL archiving (<xref linkend="guc-archive-mode">) and streaming
replication.
</para>
<para>
In <literal>hot_standby</> level, the same information is logged as
with <literal>archive</>, plus information needed to reconstruct
the status of running transactions from the WAL. To enable read-only
queries on a standby server, <varname>wal_level</> must be set to
<literal>hot_standby</> on the primary. It is thought that there is
little measurable difference in performance from using
<literal>hot_standby</> level over <literal>archive</>, so feedback
is welcome if any production impacts are noticeable.
</para>
</listitem>
</varlistentry>
<varlistentry id="guc-fsync" xreflabel="fsync">
<indexterm>
<primary><varname>fsync</> configuration parameter</primary>
@ -1726,7 +1765,9 @@ SET ENABLE_SEQSCAN TO OFF;
<varname>archive_mode</> and <varname>archive_command</> are
separate variables so that <varname>archive_command</> can be
changed without leaving archiving mode.
This parameter can only be set at server start.
This parameter can only be set at server start. <varname>wal_level</>
must be set to <literal>archive</> or <literal>hot_standby</> to
enable <varname>archive_mode</>.
</para>
</listitem>
</varlistentry>
@ -1818,7 +1859,9 @@ SET ENABLE_SEQSCAN TO OFF;
Specifies the maximum number of concurrent connections from standby
servers (i.e., the maximum number of simultaneously running WAL sender
processes). The default is zero. This parameter can only be set at
server start.
server start. <varname>wal_level</> must be set to <literal>archive</>
or <literal>hot_standby</> to allow connections from standby
connections.
</para>
</listitem>
</varlistentry>
@ -1884,16 +1927,11 @@ SET ENABLE_SEQSCAN TO OFF;
</indexterm>
<listitem>
<para>
Parameter has two roles. During recovery, specifies whether or not
you can connect and run queries to enable <xref linkend="hot-standby">.
During normal running, specifies whether additional information is written
to WAL to allow recovery connections on a standby server that reads
WAL data generated by this server. The default value is
<literal>on</literal>. It is thought that there is little
measurable difference in performance from using this feature, so
feedback is welcome if any production impacts are noticeable.
It is likely that this parameter will be removed in later releases.
This parameter can only be set at server start.
Specifies whether or not you can connect and run queries during
recovery, for <xref linkend="hot-standby">. The default value is
<literal>on</literal>.
This parameter can only be set at server start. It only has effect
during archive recovery or in standby mode.
</para>
</listitem>
</varlistentry>

8
doc/src/sgml/high-availability.sgml
View File

@ -1,4 +1,4 @@
<!-- $PostgreSQL: pgsql/doc/src/sgml/high-availability.sgml,v 1.63 2010/04/26 19:09:25 momjian Exp $ -->
<!-- $PostgreSQL: pgsql/doc/src/sgml/high-availability.sgml,v 1.64 2010/04/28 16:10:40 heikki Exp $ -->
<chapter id="high-availability">
<title>High Availability, Load Balancing, and Replication</title>
@ -1593,9 +1593,9 @@ LOG: database system is ready to accept read only connections
</programlisting>
Consistency information is recorded once per checkpoint on the primary, as long
as <varname>recovery_connections</> is enabled on the primary. It is not possible
as <varname>wal_level</> is set to <literal>hot_standby</> on the primary. It is not possible
to enable recovery connections on the standby when reading WAL written during the
period that <varname>recovery_connections</> was disabled on the primary.
period that <varname>wal_level</> was not set to <literal>hot_standby</> on the primary.
Reaching a consistent state can also be delayed in the presence
of both of these conditions:
@ -1842,7 +1842,7 @@ LOG: database system is ready to accept read only connections
</para>
<para>
On the primary, parameters <varname>recovery_connections</> and
On the primary, parameters <varname>wal_level</> and
<varname>vacuum_defer_cleanup_age</> can be used.
<varname>max_standby_delay</> has no effect if set on the primary.
</para>

45
doc/src/sgml/perform.sgml
View File

@ -1,4 +1,4 @@
<!-- $PostgreSQL: pgsql/doc/src/sgml/perform.sgml,v 1.76 2010/04/20 00:26:06 rhaas Exp $ -->
<!-- $PostgreSQL: pgsql/doc/src/sgml/perform.sgml,v 1.77 2010/04/28 16:10:40 heikki Exp $ -->
<chapter id="performance-tips">
<title>Performance Tips</title>
@ -835,10 +835,9 @@ SELECT * FROM x, y, a, b, c WHERE something AND somethingelse;
<command>TRUNCATE</command> command. In such cases no WAL
needs to be written, because in case of an error, the files
containing the newly loaded data will be removed anyway.
However, this consideration does not apply when
<xref linkend="guc-archive-mode"> is on or streaming replication
is allowed (i.e., <xref linkend="guc-max-wal-senders"> is more
than or equal to one), as all commands must write WAL in that case.
However, this consideration only applies when
<xref linkend="guc-wal-level"> is <literal>minimal</> as all commands
must write WAL otherwise.
</para>
</sect2>
@ -910,29 +909,27 @@ SELECT * FROM x, y, a, b, c WHERE something AND somethingelse;
</sect2>
<sect2 id="populate-pitr">
<title>Turn off <varname>archive_mode</varname> and streaming replication</title>
<title>Disable WAL archival and streaming replication</title>
<para>
When loading large amounts of data into an installation that uses
WAL archiving or streaming replication, you might want to disable
archiving (turn off the <xref linkend="guc-archive-mode">
configuration variable) and replication (zero the
<xref linkend="guc-max-wal-senders"> configuration variable)
while loading. It might be
faster to take a new base backup after the load has completed
than to process a large amount of incremental WAL data.
But note that changing either of these variables requires
a server restart.
WAL archiving or streaming replication, it might be faster to take a
new base backup after the load has completed than to process a large
amount of incremental WAL data. You might want to disable archiving
and streaming replication while loading, by setting
<xref linkend="guc-wal-level"> to <literal>minimal</>,
<xref linkend="guc-archive-mode"> <literal>off</>, and
<xref linkend="guc-max-wal-senders"> to zero).
But note that changing these settings requires a server restart.
</para>
<para>
Aside from avoiding the time for the archiver or WAL sender to
process the WAL data,
doing this will actually make certain commands faster, because they
are designed not to write WAL at all if <varname>archive_mode</varname>
is off and <varname>max_wal_senders</varname> is zero. (They can
guarantee crash safety more cheaply by doing an
<function>fsync</> at the end than by writing WAL.)
are designed not to write WAL at all if <varname>wal_level</varname>
is <literal>minimal</>. (They can guarantee crash safety more cheaply
by doing an <function>fsync</> at the end than by writing WAL.)
This applies to the following commands:
<itemizedlist>
<listitem>
@ -1014,10 +1011,12 @@ SELECT * FROM x, y, a, b, c WHERE something AND somethingelse;
</listitem>
<listitem>
<para>
If using WAL archiving, consider disabling it during the restore.
To do that, turn off <varname>archive_mode</varname> before loading the
dump script, and afterwards turn it back on
and take a fresh base backup.
If using WAL archiving or streaming replication, consider disabling
them during the restore. To do that, set <varname>arcive_mode</> off,
<varname>wal_level</varname> to <literal>minimal</>, and
<varname>max_wal_senders</> zero before loading the dump script,
and afterwards set them back to the right values and take a fresh
base backup.
</para>
</listitem>
<listitem>