Sophie: postgresql9.6-docs-9.6.15-1.mga6 noarch

postgresql9.6-docs-9.6.15-1.mga6.noarch.rpm

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<HTML
><HEAD
><TITLE
>WAL Configuration</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.79"><LINK
REV="MADE"
HREF="mailto:pgsql-docs@postgresql.org"><LINK
REL="HOME"
TITLE="PostgreSQL 9.6.15 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="Reliability and the Write-Ahead Log"
HREF="wal.html"><LINK
REL="PREVIOUS"
TITLE="Asynchronous Commit"
HREF="wal-async-commit.html"><LINK
REL="NEXT"
TITLE="WAL Internals"
HREF="wal-internals.html"><LINK
REL="STYLESHEET"
TYPE="text/css"
HREF="stylesheet.css"><META
HTTP-EQUIV="Content-Type"
CONTENT="text/html; charset=ISO-8859-1"><META
NAME="creation"
CONTENT="2019-08-11T16:06:51"></HEAD
><BODY
CLASS="SECT1"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="4"
ALIGN="center"
VALIGN="bottom"
><A
HREF="index.html"
>PostgreSQL 9.6.15 Documentation</A
></TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
TITLE="Asynchronous Commit"
HREF="wal-async-commit.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="wal.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="60%"
ALIGN="center"
VALIGN="bottom"
>Chapter 30. Reliability and the Write-Ahead Log</TD
><TD
WIDTH="20%"
ALIGN="right"
VALIGN="top"
><A
TITLE="WAL Internals"
HREF="wal-internals.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="WAL-CONFIGURATION"
>30.4. <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> Configuration</A
></H1
><P
>   There are several <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
>-related configuration parameters that
   affect database performance. This section explains their use.
   Consult <A
HREF="runtime-config.html"
>Chapter 19</A
> for general information about
   setting server configuration parameters.
  </P
><P
>   <I
CLASS="FIRSTTERM"
>Checkpoints</I
>
   are points in the sequence of transactions at which it is guaranteed
   that the heap and index data files have been updated with all
   information written before that checkpoint.  At checkpoint time, all
   dirty data pages are flushed to disk and a special checkpoint record is
   written to the log file.  (The change records were previously flushed
   to the <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> files.)
   In the event of a crash, the crash recovery procedure looks at the latest
   checkpoint record to determine the point in the log (known as the redo
   record) from which it should start the REDO operation.  Any changes made to
   data files before that point are guaranteed to be already on disk.
   Hence, after a checkpoint, log segments preceding the one containing
   the redo record are no longer needed and can be recycled or removed. (When
   <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> archiving is being done, the log segments must be
   archived before being recycled or removed.)
  </P
><P
>   The checkpoint requirement of flushing all dirty data pages to disk
   can cause a significant I/O load.  For this reason, checkpoint
   activity is throttled so that I/O begins at checkpoint start and completes
   before the next checkpoint is due to start; this minimizes performance
   degradation during checkpoints.
  </P
><P
>   The server's checkpointer process automatically performs
   a checkpoint every so often.  A checkpoint is begun every <A
HREF="runtime-config-wal.html#GUC-CHECKPOINT-TIMEOUT"
>checkpoint_timeout</A
> seconds, or if
   <A
HREF="runtime-config-wal.html#GUC-MAX-WAL-SIZE"
>max_wal_size</A
> is about to be exceeded,
   whichever comes first.
   The default settings are 5 minutes and 1 GB, respectively.
   If no WAL has been written since the previous checkpoint, new checkpoints
   will be skipped even if <TT
CLASS="VARNAME"
>checkpoint_timeout</TT
> has passed.
   (If WAL archiving is being used and you want to put a lower limit on how
   often files are archived in order to bound potential data loss, you should
   adjust the <A
HREF="runtime-config-wal.html#GUC-ARCHIVE-TIMEOUT"
>archive_timeout</A
> parameter rather than the
   checkpoint parameters.)
   It is also possible to force a checkpoint by using the SQL
   command <TT
CLASS="COMMAND"
>CHECKPOINT</TT
>.
  </P
><P
>   Reducing <TT
CLASS="VARNAME"
>checkpoint_timeout</TT
> and/or
   <TT
CLASS="VARNAME"
>max_wal_size</TT
> causes checkpoints to occur
   more often. This allows faster after-crash recovery, since less work
   will need to be redone. However, one must balance this against the
   increased cost of flushing dirty data pages more often. If
   <A
HREF="runtime-config-wal.html#GUC-FULL-PAGE-WRITES"
>full_page_writes</A
> is set (as is the default), there is
   another factor to consider. To ensure data page consistency,
   the first modification of a data page after each checkpoint results in
   logging the entire page content. In that case,
   a smaller checkpoint interval increases the volume of output to the WAL log,
   partially negating the goal of using a smaller interval,
   and in any case causing more disk I/O.
  </P
><P
>   Checkpoints are fairly expensive, first because they require writing
   out all currently dirty buffers, and second because they result in
   extra subsequent WAL traffic as discussed above.  It is therefore
   wise to set the checkpointing parameters high enough so that checkpoints
   don't happen too often.  As a simple sanity check on your checkpointing
   parameters, you can set the <A
HREF="runtime-config-wal.html#GUC-CHECKPOINT-WARNING"
>checkpoint_warning</A
>
   parameter.  If checkpoints happen closer together than
   <TT
CLASS="VARNAME"
>checkpoint_warning</TT
> seconds,
   a message will be output to the server log recommending increasing
   <TT
CLASS="VARNAME"
>max_wal_size</TT
>.  Occasional appearance of such
   a message is not cause for alarm, but if it appears often then the
   checkpoint control parameters should be increased. Bulk operations such
   as large <TT
CLASS="COMMAND"
>COPY</TT
> transfers might cause a number of such warnings
   to appear if you have not set <TT
CLASS="VARNAME"
>max_wal_size</TT
> high
   enough.
  </P
><P
>   To avoid flooding the I/O system with a burst of page writes,
   writing dirty buffers during a checkpoint is spread over a period of time.
   That period is controlled by
   <A
HREF="runtime-config-wal.html#GUC-CHECKPOINT-COMPLETION-TARGET"
>checkpoint_completion_target</A
>, which is
   given as a fraction of the checkpoint interval.
   The I/O rate is adjusted so that the checkpoint finishes when the
   given fraction of
   <TT
CLASS="VARNAME"
>checkpoint_timeout</TT
> seconds have elapsed, or before
   <TT
CLASS="VARNAME"
>max_wal_size</TT
> is exceeded, whichever is sooner.
   With the default value of 0.5,
   <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> can be expected to complete each checkpoint
   in about half the time before the next checkpoint starts.  On a system
   that's very close to maximum I/O throughput during normal operation,
   you might want to increase <TT
CLASS="VARNAME"
>checkpoint_completion_target</TT
>
   to reduce the I/O load from checkpoints.  The disadvantage of this is that
   prolonging checkpoints affects recovery time, because more WAL segments
   will need to be kept around for possible use in recovery.  Although
   <TT
CLASS="VARNAME"
>checkpoint_completion_target</TT
> can be set as high as 1.0,
   it is best to keep it less than that (perhaps 0.9 at most) since
   checkpoints include some other activities besides writing dirty buffers.
   A setting of 1.0 is quite likely to result in checkpoints not being
   completed on time, which would result in performance loss due to
   unexpected variation in the number of WAL segments needed.
  </P
><P
>   On Linux and POSIX platforms <A
HREF="runtime-config-wal.html#GUC-CHECKPOINT-FLUSH-AFTER"
>checkpoint_flush_after</A
>
   allows to force the OS that pages written by the checkpoint should be
   flushed to disk after a configurable number of bytes.  Otherwise, these
   pages may be kept in the OS's page cache, inducing a stall when
   <TT
CLASS="LITERAL"
>fsync</TT
> is issued at the end of a checkpoint.  This setting will
   often help to reduce transaction latency, but it also can have an adverse
   effect on performance; particularly for workloads that are bigger than
   <A
HREF="runtime-config-resource.html#GUC-SHARED-BUFFERS"
>shared_buffers</A
>, but smaller than the OS's page cache.
  </P
><P
>   The number of WAL segment files in <TT
CLASS="FILENAME"
>pg_xlog</TT
> directory depends on
   <TT
CLASS="VARNAME"
>min_wal_size</TT
>, <TT
CLASS="VARNAME"
>max_wal_size</TT
> and
   the amount of WAL generated in previous checkpoint cycles. When old log
   segment files are no longer needed, they are removed or recycled (that is,
   renamed to become future segments in the numbered sequence). If, due to a
   short-term peak of log output rate, <TT
CLASS="VARNAME"
>max_wal_size</TT
> is
   exceeded, the unneeded segment files will be removed until the system
   gets back under this limit. Below that limit, the system recycles enough
   WAL files to cover the estimated need until the next checkpoint, and
   removes the rest. The estimate is based on a moving average of the number
   of WAL files used in previous checkpoint cycles. The moving average
   is increased immediately if the actual usage exceeds the estimate, so it
   accommodates peak usage rather than average usage to some extent.
   <TT
CLASS="VARNAME"
>min_wal_size</TT
> puts a minimum on the amount of WAL files
   recycled for future usage; that much WAL is always recycled for future use,
   even if the system is idle and the WAL usage estimate suggests that little
   WAL is needed.
  </P
><P
>   Independently of <TT
CLASS="VARNAME"
>max_wal_size</TT
>,
   <A
HREF="runtime-config-replication.html#GUC-WAL-KEEP-SEGMENTS"
>wal_keep_segments</A
> + 1 most recent WAL files are
   kept at all times. Also, if WAL archiving is used, old segments can not be
   removed or recycled until they are archived. If WAL archiving cannot keep up
   with the pace that WAL is generated, or if <TT
CLASS="VARNAME"
>archive_command</TT
>
   fails repeatedly, old WAL files will accumulate in <TT
CLASS="FILENAME"
>pg_xlog</TT
>
   until the situation is resolved. A slow or failed standby server that
   uses a replication slot will have the same effect (see
   <A
HREF="warm-standby.html#STREAMING-REPLICATION-SLOTS"
>Section 26.2.6</A
>).
  </P
><P
>   In archive recovery or standby mode, the server periodically performs
   <I
CLASS="FIRSTTERM"
>restartpoints</I
>,
   which are similar to checkpoints in normal operation: the server forces
   all its state to disk, updates the <TT
CLASS="FILENAME"
>pg_control</TT
> file to
   indicate that the already-processed WAL data need not be scanned again,
   and then recycles any old log segment files in the <TT
CLASS="FILENAME"
>pg_xlog</TT
>
   directory.
   Restartpoints can't be performed more frequently than checkpoints in the
   master because restartpoints can only be performed at checkpoint records.
   A restartpoint is triggered when a checkpoint record is reached if at
   least <TT
CLASS="VARNAME"
>checkpoint_timeout</TT
> seconds have passed since the last
   restartpoint, or if WAL size is about to exceed
   <TT
CLASS="VARNAME"
>max_wal_size</TT
>. However, because of limitations on when a
   restartpoint can be performed, <TT
CLASS="VARNAME"
>max_wal_size</TT
> is often exceeded
   during recovery, by up to one checkpoint cycle's worth of WAL.
   (<TT
CLASS="VARNAME"
>max_wal_size</TT
> is never a hard limit anyway, so you should
   always leave plenty of headroom to avoid running out of disk space.)
  </P
><P
>   There are two commonly used internal <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> functions:
   <CODE
CLASS="FUNCTION"
>XLogInsertRecord</CODE
> and <CODE
CLASS="FUNCTION"
>XLogFlush</CODE
>.
   <CODE
CLASS="FUNCTION"
>XLogInsertRecord</CODE
> is used to place a new record into
   the <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> buffers in shared memory. If there is no
   space for the new record, <CODE
CLASS="FUNCTION"
>XLogInsertRecord</CODE
> will have
   to write (move to kernel cache) a few filled <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
>
   buffers. This is undesirable because <CODE
CLASS="FUNCTION"
>XLogInsertRecord</CODE
>
   is used on every database low level modification (for example, row
   insertion) at a time when an exclusive lock is held on affected
   data pages, so the operation needs to be as fast as possible.  What
   is worse, writing <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> buffers might also force the
   creation of a new log segment, which takes even more
   time. Normally, <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> buffers should be written
   and flushed by an <CODE
CLASS="FUNCTION"
>XLogFlush</CODE
> request, which is
   made, for the most part, at transaction commit time to ensure that
   transaction records are flushed to permanent storage. On systems
   with high log output, <CODE
CLASS="FUNCTION"
>XLogFlush</CODE
> requests might
   not occur often enough to prevent <CODE
CLASS="FUNCTION"
>XLogInsertRecord</CODE
>
   from having to do writes.  On such systems
   one should increase the number of <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> buffers by
   modifying the <A
HREF="runtime-config-wal.html#GUC-WAL-BUFFERS"
>wal_buffers</A
> parameter.  When
   <A
HREF="runtime-config-wal.html#GUC-FULL-PAGE-WRITES"
>full_page_writes</A
> is set and the system is very busy,
   setting <TT
CLASS="VARNAME"
>wal_buffers</TT
> higher will help smooth response times
   during the period immediately following each checkpoint.
  </P
><P
>   The <A
HREF="runtime-config-wal.html#GUC-COMMIT-DELAY"
>commit_delay</A
> parameter defines for how many
   microseconds a group commit leader process will sleep after acquiring a
   lock within <CODE
CLASS="FUNCTION"
>XLogFlush</CODE
>, while group commit
   followers queue up behind the leader.  This delay allows other server
   processes to add their commit records to the WAL buffers so that all of
   them will be flushed by the leader's eventual sync operation.  No sleep
   will occur if <A
HREF="runtime-config-wal.html#GUC-FSYNC"
>fsync</A
> is not enabled, or if fewer
   than <A
HREF="runtime-config-wal.html#GUC-COMMIT-SIBLINGS"
>commit_siblings</A
> other sessions are currently
   in active transactions; this avoids sleeping when it's unlikely that
   any other session will commit soon.  Note that on some platforms, the
   resolution of a sleep request is ten milliseconds, so that any nonzero
   <TT
CLASS="VARNAME"
>commit_delay</TT
> setting between 1 and 10000
   microseconds would have the same effect.  Note also that on some
   platforms, sleep operations may take slightly longer than requested by
   the parameter.
  </P
><P
>   Since the purpose of <TT
CLASS="VARNAME"
>commit_delay</TT
> is to allow the
   cost of each flush operation to be amortized across concurrently
   committing transactions (potentially at the expense of transaction
   latency), it is necessary to quantify that cost before the setting can
   be chosen intelligently.  The higher that cost is, the more effective
   <TT
CLASS="VARNAME"
>commit_delay</TT
> is expected to be in increasing
   transaction throughput, up to a point.  The <A
HREF="pgtestfsync.html"
><SPAN
CLASS="APPLICATION"
>pg_test_fsync</SPAN
></A
> program can be used to measure the average time
   in microseconds that a single WAL flush operation takes.  A value of
   half of the average time the program reports it takes to flush after a
   single 8kB write operation is often the most effective setting for
   <TT
CLASS="VARNAME"
>commit_delay</TT
>, so this value is recommended as the
   starting point to use when optimizing for a particular workload.  While
   tuning <TT
CLASS="VARNAME"
>commit_delay</TT
> is particularly useful when the
   WAL log is stored on high-latency rotating disks, benefits can be
   significant even on storage media with very fast sync times, such as
   solid-state drives or RAID arrays with a battery-backed write cache;
   but this should definitely be tested against a representative workload.
   Higher values of <TT
CLASS="VARNAME"
>commit_siblings</TT
> should be used in
   such cases, whereas smaller <TT
CLASS="VARNAME"
>commit_siblings</TT
> values
   are often helpful on higher latency media.  Note that it is quite
   possible that a setting of <TT
CLASS="VARNAME"
>commit_delay</TT
> that is too
   high can increase transaction latency by so much that total transaction
   throughput suffers.
  </P
><P
>   When <TT
CLASS="VARNAME"
>commit_delay</TT
> is set to zero (the default), it
   is still possible for a form of group commit to occur, but each group
   will consist only of sessions that reach the point where they need to
   flush their commit records during the window in which the previous
   flush operation (if any) is occurring.  At higher client counts a
   <SPAN
CLASS="QUOTE"
>"gangway effect"</SPAN
> tends to occur, so that the effects of group
   commit become significant even when <TT
CLASS="VARNAME"
>commit_delay</TT
> is
   zero, and thus explicitly setting <TT
CLASS="VARNAME"
>commit_delay</TT
> tends
   to help less.  Setting <TT
CLASS="VARNAME"
>commit_delay</TT
> can only help
   when (1) there are some concurrently committing transactions, and (2)
   throughput is limited to some degree by commit rate; but with high
   rotational latency this setting can be effective in increasing
   transaction throughput with as few as two clients (that is, a single
   committing client with one sibling transaction).
  </P
><P
>   The <A
HREF="runtime-config-wal.html#GUC-WAL-SYNC-METHOD"
>wal_sync_method</A
> parameter determines how
   <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> will ask the kernel to force
   <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> updates out to disk.
   All the options should be the same in terms of reliability, with
   the exception of <TT
CLASS="LITERAL"
>fsync_writethrough</TT
>, which can sometimes
   force a flush of the disk cache even when other options do not do so.
   However, it's quite platform-specific which one will be the fastest.
   You can test the speeds of different options using the <A
HREF="pgtestfsync.html"
><SPAN
CLASS="APPLICATION"
>pg_test_fsync</SPAN
></A
> program.
   Note that this parameter is irrelevant if <TT
CLASS="VARNAME"
>fsync</TT
>
   has been turned off.
  </P
><P
>   Enabling the <A
HREF="runtime-config-developer.html#GUC-WAL-DEBUG"
>wal_debug</A
> configuration parameter
   (provided that <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> has been
   compiled with support for it) will result in each
   <CODE
CLASS="FUNCTION"
>XLogInsertRecord</CODE
> and <CODE
CLASS="FUNCTION"
>XLogFlush</CODE
>
   <ACRONYM
CLASS="ACRONYM"
>WAL</ACRONYM
> call being logged to the server log. This
   option might be replaced by a more general mechanism in the future.
  </P
></DIV
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="wal-async-commit.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
><A
HREF="wal-internals.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>Asynchronous Commit</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="wal.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>WAL Internals</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>