Well - it hadn't run on any table in over 24 hours (according to
pg_stat_user_tables). My tables are constantly being inserted into and
deleted from, and the autovacuum settings are pretty aggressive. I also
had not seen the autovac process running in the past 24 hours. (although
I wasn't watchin
"Jeremy Haile" <[EMAIL PROTECTED]> writes:
> No tables have been vacuumed or analyzed today. I had thought that this
> problem was due to my pg_autovacuum changes, but perhaps not. I
> restarted PostgreSQL (in production - yikes) About a minute after being
> restarted, the autovac process fired
Jeremy Haile wrote:
Also, are other auto-vacuums and auto-analyzes showing up in the
pg_stats table? Maybe it's a stats system issue.
No tables have been vacuumed or analyzed today. I had thought that this
problem was due to my pg_autovacuum changes, but perhaps not. I
restarted Postgre
> Unless it's just a bug, my only guess is that autovacuum may be getting
> busy at times (vacuuming large tables for example) and hasn't had a
> chance to even look at that table for a while, and by the time it gets
> to it, there have been tens of thousands of inserts. Does that sounds
> pla
Jeremy Haile wrote:
I changed the table-specific settings so that the ANALYZE base threshold
was 5000 and the ANALYZE scale factor is 0. According to the documented
formula: analyze threshold = analyze base threshold + analyze scale
factor * number of tuples, I assumed that this would cause the
Some of my very large tables (10 million rows) need to be analyzed by
autovacuum on a frequent basis. Rather than specifying this as a
percentage of table size + base threshold, I wanted to specify it as an
explicit number of rows.
I changed the table-specific settings so that the ANALYZE base th
