Slow metrics calucation on production

[peterdudfield]

Currently production database takes ~4 hours to run calculating ME, comapred to 30 mins on dev. This could be for a number of reasons.

Useful postgres tools

SELECT relname, n_dead_tup FROM pg_stat_user_tables;
SELECT relname, last_vacuum, last_autovacuum FROM pg_stat_user_tables;
SELECT reloptions FROM pg_class;
SELECT * from pg_settings where category like 'Autovacuum';

although there is nothing obvious from this

[gilbertgong]

This looks like an internal issue but I'd love to work on this, I've done a ton of performance optimization in my life, including sql queries and postgres/mysql tuning.