database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-30 11:03:19 +03:00
Code Activity

Incorporate README.fsync into main documentation body

Browse Source
This commit is contained in:
Peter Eisentraut
2000-07-16 14:47:57 +00:00
parent b4c315ba9e
commit 81fd7532a9
2 changed files with 34 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
doc/src/sgml/runtime.sgml
View File

@ -1,5 +1,5 @@
<!--
$Header: /cvsroot/pgsql/doc/src/sgml/runtime.sgml,v 1.14 2000/07/15 21:35:47 petere Exp $
$Header: /cvsroot/pgsql/doc/src/sgml/runtime.sgml,v 1.15 2000/07/16 14:47:57 petere Exp $
-->
<Chapter Id="runtime">
@ -846,11 +846,39 @@ env PGOPTIONS='--geqo=off' psql
<term>FSYNC (<type>boolean</type>)</term>
<listitem>
<para>
When this is on (default), an <function>fsync()</function>
call is done after each transaction. Turning this off
increases performance but an operating system crash or power
outage might cause data corruption. (Note that a crash of
<productname>Postgres</productname> itself is not affected.)
If this is option is on, the <productname>Postgres</> backend
will use the <function>fsync()</> system call in several
places to make sure that updates are physically written to
disk and will not hang around in the write caches. This
increases the chance that a database installation will still
be usable after a operating system or hardware crashes by a
large amount. (Crashes of the database server itself do
<emphasis>not</> affect this consideration.)
</para>
<para>
However, this operation severely slows down
<productname>Postgres</>, because at all those points it has
to block and wait for the operating system to flush the
buffers. Without <function>fsync</>, the operating system is
allowed to do its best in buffering, sorting, and delaying
writes, so this can be a <emphasis>very</> big perfomance
increase. However, if the system crashes, parts of the data of
a transaction that has already been committed -- according to
the information on disk -- will still hang around in memory.
Inconsistent data (i.e., data corruption) is therefore likely
to occur.
</para>
<para>
This option is the subject of an eternal debate in the
<productname>Postgres</> user and developer communities. Some
always leave it off, some turn it off only for bulk loads,
where there is a clear restart point if something goes wrong,
some leave it on just to be on the safe side. Because it is
the safe side, on is also the default. If you trust your
operating system, your utility company, and your hardware, you
might want to disable it.
</para>
</listitem>
</varlistentry>