In response to Ow Mun Heng <Ow.Mun.Heng@xxxxxxx>: > > Even with the regular vacuuming and even a vacuum full ( on my test DB) > I still see that perhaps something is wrong (from the below) > > (I got this gem from the mailling list archives) > hmxmms=> SELECT > c.relname, > c.reltuples::bigint as rowcnt, > pg_stat_get_tuples_inserted(c.oid) AS inserted, > pg_stat_get_tuples_updated(c.oid) AS updated, > pg_stat_get_tuples_deleted(c.oid) AS deleted > FROM pg_class c > WHERE c.relkind = 'r'::"char" > GROUP BY c.oid, c.relname, c.reltuples > HAVING pg_stat_get_tuples_updated(c.oid) + > pg_stat_get_tuples_deleted(c.oid) > 1000 > ORDER BY pg_stat_get_tuples_updated(c.oid) + > pg_stat_get_tuples_deleted(c.oid) DESC; > relname | rowcnt | inserted | updated | deleted > -----------------------+----------+----------+---------+---------- > tst_r | 11971691 | 0 | 0 | 22390528 <-- > pg_statistic | 1465 | 280 | 7716 | 153 > dr_ns | 2305571 | 1959 | 0 | 1922 > pg_attribute | 3787 | 1403 | 184 | 1292 > > No matter how many times I vacuum/full the deleted number still doesn't > go down. Are you sure you're interpreting that number correctly? I took it to mean a counter of the number of delete operations since server start. -- Bill Moran http://www.potentialtech.com ---------------------------(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
