Hi,
> in pg_stat_user_tables, not since the last time ANALYZE run, but you have the
> number of reltuples from pg_class that is used to calculate the ratio.
So how does the autovacuum daemon to determine the number of obsolete row since its last work on a specific table ?
> Everything is relative to the size of the table, what is the content of
> pg_class for the second kind of tables ? (relpages/reltuples)
relname | last_autoanalyse | reltuples | relpages | n_tup_ins + n_tup_upd + n_tup_del
| (from pg_stat_all_tables) | | | (from pg_stat_all_tables)
| | | |
table1 | 2012-10-15 21:12:04.362647+02 | 2.53296e+06 | 50789 | 28542483
table2 | 2012-10-15 18:27:30.181583+02 | 2.57144e+06 | 54833 | 31127737
table3 | 2012-10-15 13:13:09.569758+02 | 2.58646e+06 | 63208 | 27167944
table4 | 2012-10-15 12:05:39.094652+02 | 2.58655e+06 | 63181 | 27160409
table5 | 2012-10-15 12:05:32.347738+02 | 2.58611e+06 | 57272 | 27171661
table6 | 2012-10-15 10:43:40.558078+02 | 2.57925e+06 | 63145 | 27155875
table7 | 2012-10-15 10:40:25.469625+02 | 2.59551e+06 | 57254 | 27161876
table8 | 2012-10-15 10:25:10.776152+02 | 2.56987e+06 | 63147 | 27177907
table9 | 2012-10-15 09:30:17.407181+02 | 2.5923e+06 | 57231 | 27152953
table10 | 2012-10-15 08:29:25.427207+02 | 2.56237e+06 | 63085 | 31770581
table11 | 2012-10-15 08:26:14.550742+02 | 2.56503e+06 | 63065 | 31769609
table12 | 2012-10-15 06:39:01.737914+02 | 2.57833e+06 | 63023 | 31767543
table13 | 2012-10-15 06:35:36.43683+02 | 2.57217e+06 | 63047 | 31759954
table14 | 2012-10-15 05:08:02.858623+02 | 2.59068e+06 | 57119 | 27154555
table15 | 2012-10-15 04:40:48.695142+02 | 2.57449e+06 | 63002 | 31768034
table16 | 2012-10-10 07:51:49.625226+02 | 2.48584e+06 | 54014 | 22110223
table17 | 2012-10-09 10:03:36.649331+02 | 2.58047e+06 | 63742 | 23423257
table18 | 2012-10-09 10:02:13.549362+02 | 2.57854e+06 | 63585 | 23364035
table19 | 2012-10-09 09:18:55.447895+02 | 2.58851e+06 | 63516 | 23375180
Regards, Baptiste Lhoste