Re: Autovacuum behavior

[John Scalia <jayknowsunix@xxxxxxxxx>]

Thanks OM, I'll try setting that.

On Thu, Jul 30, 2015 at 11:44 PM, Om Prakash Jaiswal <op12om@xxxxxxxxxxx> wrote:

track_count = on;

It is missing.

Regards

Om Prakash

On Friday, 31 July 2015 12:21 AM, John Scalia <jayknowsunix@xxxxxxxxx> wrote:

Hi all,

The autovacuum settings for a 9.4.2 database are shown below, I'm not absolutely certain if I missed anything:

autovacuum = on

log_autovacuum_min_duration = 100

autovacuum_max_workers = 15

autovacuum_naptime = 10min

#autovacuum_vacuum_threshold = 50

autovacuum_analyze_threshold = 80

autovacuum_vacuum_scale_factor = 0.1

autovacuum_analyze_scale_factor = 0.2

autovacuum_freeze_max_age = 100000000

autovacuum_vacuum_cost_delay = 20ms

autovacuum_vacuum_cost_limit = -1

vacuum_freeze_min_age =   5000000

vacuum_freeze_table_age = 2500000

But, when I examine pg_stat_all_tables, I'm seeing a lot of tables where n_dead_tup is still a lot greater than n_live_tup. Mind you, these are all fairly small tables. I'm also seeing that the last_autovacuum ran about 11:22 AM CDT this morning.I would think the tables where there were no live tuples and a bunch of dead_tuples would have been vacuumed after 11:22 AM to clear the dead ones. Am I missing something?

--

Jay