We recently upgraded from 8.3.something to 9.0.1. With 9.0.1, we have a huge spike in vacuums every 8 days only on one of our DB servers. We go from approx 20 vacuums every 5 minutes to 350 per 5 minutes. This lasts for several hours, then stops. I have attached a graph that shows the occurrence. I am assuming that it needs to vacuum all of my tables to avoid some sort of wrap around counter. I am wondering what is the best way to make autovacuum spread this out so it will not be quite a big of a hit. I we did not see this with 8.3 and I kept the setting the same after the upgrade.
Here are my autovacuum settings: autovacuum = on log_autovacuum_min_duration = 0 autovacuum_max_workers = 5 autovacuum_naptime = 1min autovacuum_vacuum_threshold = 100 autovacuum_analyze_threshold = 100 autovacuum_vacuum_scale_factor = 0.1 autovacuum_analyze_scale_factor = 0.05 autovacuum_freeze_max_age = 200000000 autovacuum_vacuum_cost_delay = 20ms autovacuum_vacuum_cost_limit = -1 Any suggestions would be appreciated. Thanks, George Woodring -- iGLASS Networks www.iglass.net
<<attachment: staticGraph.do.png>>
-- Sent via pgsql-general mailing list (pgsql-general@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
