提交 3bf080da 编写于 作者: B Bruce Momjian

> OK, well as we wait on the fix for the stats system, let me submit my

> patch for pg_autovacuum.  This patch assumes that the stats system will
> be fixed so that all inserts, updates and deletes performed on shared
> tables reguardless of what database those commands were executed from,
> will show up in the stats shown in each database.

I had to make a further change to this to take quotes off the 'last
ANALYZE' in order for it to not overquote the relation name, so
there's a _little_ work left to get it to play well.

I have deployed it onto several boxes that should be doing some
vacuuming over the weekend, and it is now certainly hitting pg_
tables.

I would like to present a CVS-oriented patch; unfortunately, I had to
change the indentation patterns when editing some of it :-(.  The
following _may_ be good; not sure...

Matthew T. O'Connor
Christopher Browne
上级 b041d3e3
......@@ -2,66 +2,97 @@ pg_autovacuum README
--------------------
pg_autovacuum is a libpq client program that monitors all the
databases associated with a postgresql server. It uses the stats
databases associated with a PostgreSQL server. It uses the statistics
collector to monitor insert, update and delete activity.
When a table exceeds its insert or delete threshold (more detail
on thresholds below) then that table will be vacuumed or analyzed.
When a table exceeds a insert or delete threshold (for more detail on
thresholds, see "Vacuum and Analyze" below) then that table will be
vacuumed and/or analyzed.
This allows postgresql to keep the fsm and table statistics up to
date, and eliminates the need to schedule periodic vacuums.
This allows PostgreSQL to keep the FSM (Free Space Map) and table
statistics up to date, and eliminates the need to schedule periodic
vacuums.
The primary benefit of pg_autovacuum is that the FSM and table
statistic information are updated as needed. When a table is actively
changing, pg_autovacuum will perform the necessary vacuums and
analyzes, whereas if a table remains static, no cycles will be wasted
performing unnecessary vacuums/analyzes.
statistic information are updated more nearly as frequently as needed.
When a table is actively changing, pg_autovacuum will perform the
VACUUMs and ANALYZEs that such a table needs, whereas if a table
remains static, no cycles will be wasted performing this
unnecessarily.
A secondary benefit of pg_autovacuum is that it ensures that a
database wide vacuum is performed prior to xid wraparound. This is an
database wide vacuum is performed prior to XID wraparound. This is an
important, if rare, problem, as failing to do so can result in major
data loss.
data loss. (See the section in the _Administrator's Guide_ entitled
"Preventing transaction ID wraparound failures" for more details.)
KNOWN ISSUES:
-------------
pg_autovacuum has been tested under Redhat Linux (by me) and Solaris (by
Christopher B. Browne) and all known bugs have been resolved. Please report
any problems to the hackers list.
pg_autovacuum does not get started automatically by either the postmaster or
by pg_ctl. Along the sames lines, when the postmaster exits no one tells
pg_autovacuum. The result is that at the start of the next loop,
pg_autovacuum fails to connect to the server and exits. Any time it fails
to connect pg_autovacuum exits.
pg_autovacuum requires that the stats system be enabled and reporting row
level stats. The overhead of the stats system has been shown to be
significant under certain workloads. For instance a tight loop of queries
performing "select 1" was nearly 30% slower with stats enabled. However,
in practice with more realistic workloads, the stats system overhead is
usually nominal.
pg_autovacuum has been tested under Redhat Linux (by me) and Debian
GNU/Linux and Solaris (by Christopher B. Browne) and all known bugs
have been resolved. Please report any problems to the hackers list.
pg_autovacuum requires that the statistics system be enabled and
reporting row level stats. The overhead of the stats system has been
shown to be significant costly under certain workloads. For instance,
a tight loop of queries performing "select 1" was found to run nearly
30% slower when stats were enabled. However, in practice, with more
realistic workloads, the stats system overhead is usually nominal.
pg_autovacuum does not get started automatically by either the
postmaster or by pg_ctl. Similarly, when the postmaster exits, no one
tells pg_autovacuum. The result of that is that at the start of the
next loop, pg_autovacuum will fail to connect to the server and
exit(). Any time it fails to connect pg_autovacuum exit()s.
While pg_autovacuum can manage vacuums for as many databases as you
may have tied to a particular PostgreSQL postmaster, it can only
connect to a single PostgreSQL postmaster. Thus, if you have multiple
postmasters on a particular host, you will need multiple pg_autovacuum
instances, and they have no way, at present, to coordinate between one
another to ensure that they do not concurrently vacuum big tables.
TODO:
-----
At present, there are no sample scripts to automatically start up
pg_autovacuum along with the database. It would be desirable to have
a SysV script to start up pg_autovacuum after PostgreSQL has been
started.
Some users have expressed interest in making pg_autovacuum more
configurable so that certain tables known to be inactive could be
excluded from being vacuumed. It would probably make sense to
introduce this sort of functionality by providing arguments to specify
the database and schema in which to find a configuration table.
INSTALL:
--------
As of postgresql v7.4 pg_autovacuum is included in the main source tree
under contrib. Therefore you just make && make install (similar to most other
contrib modules) and it will be installed for you.
As of postgresql v7.4 pg_autovacuum is included in the main source
tree under contrib. Therefore you merely need to "make && make
install" (similar to most other contrib modules) and it will be
installed for you.
If you are using an earlier version of postgresql just uncompress the tar.gz
into the contrib directory and modify the contrib/Makefile to include the pg_autovacuum
directory. pg_autovacuum will then be made as part of the standard
postgresql install.
If you are using an earlier version of PostgreSQL, uncompress the
tar.gz file into the contrib directory and modify the contrib/Makefile
to include the pg_autovacuum directory. pg_autovacuum will then be
built as part of the standard postgresql install.
make sure that the folowing are set in postgresql.conf
make sure that the following are set in postgresql.conf:
stats_start_collector = true
stats_row_level = true
start up the postmaster, then execute the pg_autovacuum executable.
Start up the postmaster, then execute the pg_autovacuum executable.
If you have a script that automatically starts up the PostgreSQL
instance, you might add in, after that, something similar to the
following:
sleep 10 # To give the database some time to start up
$PGBINS/pg_autovacuum -D -s $SBASE -S $SSCALE ... [other arguments]
Command line arguments:
-----------------------
......@@ -69,7 +100,7 @@ Command line arguments:
pg_autovacuum has the following optional arguments:
-d debug: 0 silent, 1 basic info, 2 more debug info, etc...
-D dameonize: Detach from tty and run in background.
-D daemonize: Detach from tty and run in background.
-s sleep base value: see "Sleeping" below.
-S sleep scaling factor: see "Sleeping" below.
-v vacuum base threshold: see Vacuum and Analyze.
......@@ -80,18 +111,18 @@ pg_autovacuum has the following optional arguments:
-U username: Username pg_autovacuum will use to connect with, if not
specified the current username is used.
-P password: Password pg_autovacuum will use to connect with.
-H host: host name or IP to connect too.
-H host: host name or IP to connect to.
-p port: port used for connection.
-h help: list of command line options.
All arguments have default values defined in pg_autovacuum.h. At the
time of writing they are:
Numerous arguments have default values defined in pg_autovacuum.h. At
the time of writing they are:
-d 1
-v 1000
-V 2
-a 500 (half of -v is not specified)
-A 1 (half of -v is not specified)
-a 500 (half of -v if not specified)
-A 1 (half of -v if not specified)
-s 300 (5 minutes)
-S 2
......@@ -99,13 +130,14 @@ time of writing they are:
Vacuum and Analyze:
-------------------
pg_autovacuum performs either a vacuum analyze or just analyze depending
on the quantity and type of table activity (insert, update, or delete):
pg_autovacuum performs either a VACUUM ANALYZE or just ANALYZE
depending on the mixture of table activity (insert, update, or
delete):
- If the number of (inserts + updates + deletes) > AnalyzeThreshold, then
only an analyze is performed.
- If the number of (deletes + updates ) > VacuumThreshold, then a
- If the number of (deletes + updates) > VacuumThreshold, then a
vacuum analyze is performed.
deleteThreshold is equal to:
......@@ -115,26 +147,27 @@ insertThreshold is equal to:
analyze_base_value + (analyze_scaling_factor * "number of tuples in the table")
The AnalyzeThreshold defaults to half of the VacuumThreshold since it
represents a much less expensive operation (approx 5%-10% of vacuum), and
running it more often should not substantially degrade system performance.
represents a much less expensive operation (approx 5%-10% of vacuum),
and running ANALYZE more often should not substantially degrade system
performance.
Sleeping:
---------
pg_autovacuum sleeps for a while after it is done checking all the
databases. It does this in order to limit the amount of system
resources it consumes. This also allows the system administrator to
resources it consumes. This allows the system administrator to
configure pg_autovacuum to be more or less aggressive.
Reducing the sleep time will cause pg_autovacuum to respond more
quickly to changes, whether they be database addition/removal, table
addition/removal, or just normal table activity.
On the other hand, setting pg_autovaccum to sleep values to agressivly
(for too short a period of time) can have a negative effect on server
performance. If a table gets vacuumed 5 times during the course of a
large update, this is likely to take much longer than if the table was
vacuumed only once, at the end.
On the other hand, setting pg_autovacuum to sleep values too
aggressively (to too short periods of time) can have a negative effect
on server performance. For instance, if a table gets vacuumed 5 times
during the course of a large set of updates, this is likely to take a
lot more work than if the table was vacuumed just once, at the end.
The total time it sleeps is equal to:
......@@ -142,15 +175,17 @@ The total time it sleeps is equal to:
loop"
Note that timing measurements are made in seconds; specifying
"pg_vacuum -s 1" means pg_autovacuum could poll the database upto 60 times
minute. In a system with large tables where vacuums may run for several
minutes, longer times between vacuums are likely to be appropriate.
"pg_vacuum -s 1" means pg_autovacuum could poll the database up to 60
times minute. In a system with large tables where vacuums may run for
several minutes, rather longer times between vacuums are likely to be
appropriate.
What pg_autovacuum monitors:
----------------------------
pg_autovacuum dynamically generates a list of all databases and tables that
exist on the server. It will dynamically add and remove databases and
tables that are removed from the database server while pg_autovacuum is
running. Overhead is fairly small per object. For example: 10 databases
with 10 tables each appears to less than 10k of memory on my Linux box.
pg_autovacuum dynamically generates a list of all databases and tables
that exist on the server. It will dynamically add and remove
databases and tables that are removed from the database server while
pg_autovacuum is running. Overhead is fairly small per object. For
example: 10 databases with 10 tables each appears to less than 10k of
memory on my Linux box.
......@@ -118,6 +118,12 @@ init_table_info(PGresult *res, int row, db_info * dbi)
new_tbl->reltuples = atoi(PQgetvalue(res, row, PQfnumber(res, "reltuples")));
new_tbl->relpages = atoi(PQgetvalue(res, row, PQfnumber(res, "relpages")));
log_entry(PQgetvalue(res, row, PQfnumber(res, "relisshared")));
if (strcmp("t", PQgetvalue(res, row, PQfnumber(res, "relisshared"))))
new_tbl->relisshared = 0;
else
new_tbl->relisshared = 1;
new_tbl->analyze_threshold =
args->analyze_base_threshold + args->analyze_scaling_factor * new_tbl->reltuples;
new_tbl->vacuum_threshold =
......@@ -213,7 +219,7 @@ update_table_list(db_info * dbi)
* both remove tables from the list that no longer exist and add
* tables to the list that are new
*/
res = send_query(query_table_stats(dbi), dbi);
res = send_query((char *) TABLE_STATS_QUERY, dbi);
t = PQntuples(res);
/*
......@@ -353,7 +359,7 @@ print_table_info(tbl_info * tbl)
{
sprintf(logbuffer, " table name: %s.%s", tbl->dbi->dbname, tbl->table_name);
log_entry(logbuffer);
sprintf(logbuffer, " relfilenode: %i", tbl->relfilenode);
sprintf(logbuffer, " relfilenode: %i; relisshared: %i", tbl->relfilenode, tbl->relisshared);
log_entry(logbuffer);
sprintf(logbuffer, " reltuples: %i; relpages: %i", tbl->reltuples, tbl->relpages);
log_entry(logbuffer);
......@@ -688,19 +694,7 @@ print_db_info(db_info * dbi, int print_tbl_list)
/* End of DB List Management Function */
/* Begninning of misc Functions */
char *
query_table_stats(db_info * dbi)
{
if (!strcmp(dbi->dbname, "template1")) /* Use template1 to
* monitor the system
* tables */
return (char *) TABLE_STATS_ALL;
else
return (char *) TABLE_STATS_USER;
}
/* Beginning of misc Functions */
/* Perhaps add some test to this function to make sure that the stats we need are available */
PGconn *
......@@ -753,6 +747,9 @@ send_query(const char *query, db_info * dbi)
if (NULL == dbi->conn)
return NULL;
if (args->debug >= 4)
log_entry(query);
res = PQexec(dbi->conn, query);
if (!res)
......@@ -964,7 +961,7 @@ main(int argc, char *argv[])
int j = 0,
loops = 0;
/* int numInserts, numDeletes, */
/* int numInserts, numDeletes, */
int sleep_secs;
Dllist *db_list;
Dlelem *db_elem,
......@@ -1055,7 +1052,7 @@ main(int argc, char *argv[])
if (0 == xid_wraparound_check(dbs));
{
res = send_query(query_table_stats(dbs), dbs); /* Get an updated
res = send_query(TABLE_STATS_QUERY, dbs); /* Get an updated
* snapshot of this dbs
* table stats */
for (j = 0; j < PQntuples(res); j++)
......@@ -1087,7 +1084,11 @@ main(int argc, char *argv[])
*/
if ((tbl->curr_vacuum_count - tbl->CountAtLastVacuum) >= tbl->vacuum_threshold)
{
snprintf(buf, sizeof(buf), "VACUUM ANALYZE \"%s\"", tbl->table_name);
/* if relisshared = t and database != template1 then only do an analyze */
if((tbl->relisshared > 0) && (strcmp("template1",dbs->dbname)))
snprintf(buf, sizeof(buf), "ANALYZE %s", tbl->table_name);
else
snprintf(buf, sizeof(buf), "VACUUM ANALYZE %s", tbl->table_name);
if (args->debug >= 1)
{
sprintf(logbuffer, "Performing: %s", buf);
......@@ -1101,7 +1102,7 @@ main(int argc, char *argv[])
}
else if ((tbl->curr_analyze_count - tbl->CountAtLastAnalyze) >= tbl->analyze_threshold)
{
snprintf(buf, sizeof(buf), "ANALYZE \"%s\"", tbl->table_name);
snprintf(buf, sizeof(buf), "ANALYZE %s", tbl->table_name);
if (args->debug >= 1)
{
sprintf(logbuffer, "Performing: %s", buf);
......
......@@ -34,8 +34,8 @@
#define VACUUM_ANALYZE 0
#define ANALYZE_ONLY 1
#define TABLE_STATS_ALL "select a.relfilenode,a.relname,a.relnamespace,a.relpages,a.reltuples,b.schemaname,b.n_tup_ins,b.n_tup_upd,b.n_tup_del from pg_class a, pg_stat_all_tables b where a.relfilenode=b.relid"
#define TABLE_STATS_USER "select a.relfilenode,a.relname,a.relnamespace,a.relpages,a.reltuples,b.schemaname,b.n_tup_ins,b.n_tup_upd,b.n_tup_del from pg_class a, pg_stat_user_tables b where a.relfilenode=b.relid"
#define TABLE_STATS_QUERY "select a.relfilenode,a.relname,a.relnamespace,a.relpages,a.relisshared,a.reltuples,b.schemaname,b.n_tup_ins,b.n_tup_upd,b.n_tup_del from pg_class a, pg_stat_all_tables b where a.relfilenode=b.relid and a.relkind = 'r'"
#define FRONTEND
#define PAGES_QUERY "select relfilenode,reltuples,relpages from pg_class where relfilenode=%i"
#define FROZENOID_QUERY "select oid,age(datfrozenxid) from pg_database where datname = 'template1'"
......@@ -86,6 +86,7 @@ struct tableinfo
*table_name;
int relfilenode,
reltuples,
relisshared,
relpages;
long analyze_threshold,
vacuum_threshold;
......@@ -132,7 +133,6 @@ static int check_stats_enabled(db_info * dbi);
static PGconn *db_connect(db_info * dbi);
static void db_disconnect(db_info * dbi);
static PGresult *send_query(const char *query, db_info * dbi);
static char *query_table_stats(db_info * dbi);
/* Other Generally needed Functions */
static void daemonize(void);
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册