diff options
Diffstat (limited to 'doc/src/sgml/monitoring.sgml')
| -rw-r--r-- | doc/src/sgml/monitoring.sgml | 70 |
1 files changed, 49 insertions, 21 deletions
diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 78c1bdbec3..002134c9ac 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -1,5 +1,5 @@ <!-- -$Header: /cvsroot/pgsql/doc/src/sgml/monitoring.sgml,v 1.15 2002/11/11 20:14:03 petere Exp $ +$Header: /cvsroot/pgsql/doc/src/sgml/monitoring.sgml,v 1.16 2003/03/20 03:34:55 momjian Exp $ --> <chapter id="monitoring"> @@ -146,12 +146,13 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re <important> <para> Since the variables <varname>STATS_COMMAND_STRING</varname>, - <varname>STATS_BLOCK_LEVEL</varname>, - and <varname>STATS_ROW_LEVEL</varname> - default to <literal>false</>, no statistics are actually collected - in the default configuration. You must turn one or more of them on - before you will get useful results from the statistical display - functions. + <varname>STATS_BLOCK_LEVEL</varname>, and + <varname>STATS_ROW_LEVEL</varname> default to <literal>false</>, + very few statistics are collected in the default + configuration. Enabling one or more of these configuration + variables will significantly enhance the amount of useful data + produced by the statistics collector, at the expense of + additional run-time overhead. </para> </important> @@ -205,11 +206,15 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re <tbody> <row> <entry><structname>pg_stat_activity</></entry> - <entry>One row per server process, showing process <acronym>ID</>, database, - user, and current query. The current query column is only available - to superusers; for others it reads as null. (Note that because of - the collector's reporting delay, current query will only be up-to-date - for long-running queries.)</entry> + <entry>One row per server process, showing process + <acronym>ID</>, database, user, current query, and the time at + which the current query began execution. The columns that report + data on the current query are only available if the + <varname>STATS_COMMAND_STRING</varname> configuration option has + been enabled. Furthermore, these columns can only be accessed by + superusers; to other users, they always appear NULL. (Note that + because of the collector's reporting delay, current query will + only be up-to-date for long-running queries.)</entry> </row> <row> @@ -333,10 +338,20 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re </para> <para> - The <structname>pg_statio_</> views are primarily useful to determine - the effectiveness of the buffer cache. When the number of actual disk - reads is much smaller than the number of buffer hits, then the cache - is satisfying most read requests without invoking a kernel call. + The <structname>pg_statio_</> views are primarily useful to + determine the effectiveness of the buffer cache. When the number + of actual disk reads is much smaller than the number of buffer + hits, then the cache is satisfying most read requests without + invoking a kernel call. However, these statistics do not give the + entire story: due to the way in which <productname>PostgreSQL</> + handles disk I/O, data that is not in the + <productname>PostgreSQL</> buffer cache may still reside in the + kernel's I/O cache, and may therefore still be fetched without + requiring a physical read. Users interested in obtaining more + detailed information on <productname>PostgreSQL</> I/O behavior are + advised to use the <productname>PostgreSQL</> statistics collector + in combination with operating system utilities that allow insight + into the kernel's handling of I/O. </para> <para> @@ -401,7 +416,7 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re <entry><function>pg_stat_get_db_blocks_hit</function>(<type>oid</type>)</entry> <entry><type>bigint</type></entry> <entry> - Number of disk block requests found in cache for database + Number of disk block fetch requests found in cache for database </entry> </row> @@ -478,7 +493,7 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re <entry><type>set of integer</type></entry> <entry> Set of currently active backend IDs (from 1 to N where N is the - number of active backends). See usage example below. + number of active backends). See usage example below </entry> </row> @@ -518,15 +533,27 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re <entry><function>pg_stat_get_backend_activity</function>(<type>integer</type>)</entry> <entry><type>text</type></entry> <entry> - Current query of backend process (NULL if caller is not superuser) + Current query of backend process (NULL if caller is not + superuser, or <varname>STATS_COMMAND_STRING</varname> is not enabled) </entry> </row> <row> + <entry><function>pg_stat_get_backend_activity_start</function>(<type>integer</type>)</entry> + <entry><type>text</type></entry> + <entry> + The time at which the specified backend's currently executing query was + initiated (NULL if caller is not superuser, or + <varname>STATS_COMMAND_STRING</varname> is not enabled) + </entry> + </row> + + + <row> <entry><function>pg_stat_reset</function>()</entry> <entry><type>boolean</type></entry> <entry> - Reset all currently collected statistics. + Reset all currently collected statistics </entry> </row> </tbody> @@ -535,7 +562,8 @@ postgres: <replaceable>user</> <replaceable>database</> <replaceable>host</> <re <note> <para> - Blocks_fetched minus blocks_hit gives the number of kernel + <literal>blocks_fetched</literal> minus + <literal>blocks_hit</literal> gives the number of kernel <function>read()</> calls issued for the table, index, or database; but the actual number of physical reads is usually lower due to kernel-level buffering. |