Hi, Tom,
thanks for your reply, the problem is that our app use temp tables a lot so the pg_attribute getting "bloated",
i've noticed that during business hours autovacuum vacuum is not being triggered on pg_attribute,
but during night time it triggered every two hours, both vacuum and analyze, maybe you know why?
my autovacuum conf is:
autovacuum_max_workers = 16 # max number of autovacuum subprocesses
# (change requires restart)
autovacuum_naptime = 51min # time between autovacuum runs
autovacuum_vacuum_threshold = 500 # min number of row updates before
# vacuum
autovacuum_analyze_threshold = 500 # min number of row updates before
# analyze
autovacuum_vacuum_scale_factor = 0.002 # fraction of table size before vacuum
autovacuum_analyze_scale_factor = 0.01 # fraction of table size before analyze
2018-04-02 18:18 GMT+03:00 Tom Lane <tgl@xxxxxxxxxxxxx>:
Artem Tomyuk <admin@xxxxxxxxxxxxxx> writes:
> Is there any "queue" in postgres where tables that marked to be vacuumed or
> analyzed can be viewed?
No, there's no queue as such. You can look at the contents of
pg_stat_all_tables, particularly the n_dead_tup and n_mod_since_analyze
columns, to get a sense of which tables are likely to seem highest
priority to the autovacuum code.
regards, tom lane
