Thank you for the discussion. I'm on Postgres 8.4, and the hardware between Slony master and slave is identical, as is the autovacuum config. We do have transactions that fail to commit, transactions that roll back. I'm glad to have some idea of the cause of the difference in table size between Slony Master and Slave. If disk usage on the Master goes over 75% before my upgrade money is approved, I will try Bill Moran's suggestion of doing a practice vacuum run on the Slave, and then we'll take a maintenance window to VACUUM, VACUUM FULL, REINDEX; or CLUSTER on the master. THANK YOU! This is a super-helpful list. I really appreciate the positive energy in the PostgreSQL community. I'm looking forward to helping out at the PostgresSQL booth at the So Cal Linux Expo later this month. Thank you! Aleksey -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
