Hi
Sorry, i was out
The first try was:
create database dummy1;
create table dummy ... and filled with 500 records
create database dummy2;
create table dummy ... and filled with 500 records
connecting to dummy1, opening a transaction and issued an update
begin;
update dummy set f1='achterbahn';
then opened a second console and connected to dummy2:
dummy2=# select count(*) from dummy;
count
-------
5000
(1 row)
dummy2=# SELECT relpages, reltuples FROM pg_class WHERE relname ='dummy';
relpages | reltuples
----------+-----------
160 | 5000
(1 row)
updated the table several times - to generate dead tuples:
dummy2=# SELECT relpages, reltuples FROM pg_class WHERE relname ='dummy';
relpages | reltuples
----------+-----------
326 | 30000
(1 row)
dummy2=# vacuum;
VACUUM
dummy2=# SELECT relpages, reltuples FROM pg_class WHERE relname ='dummy';
relpages | reltuples
----------+-----------
326 | 30000
(1 row)
dummy2=# select count(*) from dummy;
count
-------
5000
(1 row)
dummy2=# vacuum full;
--------------------------------
vacuum was in waiting state as long the transaction in dummy1 was
opened. After committing the transaction the vacuum full was carried out.
-------------------------------
VACUUM
dummy2=# SELECT relpages, reltuples FROM pg_class WHERE relname ='dummy';
relpages | reltuples
----------+-----------
326 | 30000
(1 row)
running on pg 8.1.4 on Fedora 5
Thanks
Conni
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
