PG_UPGRADE (1)

SYNOPSIS

pg_upgrade [ -f filename ] old_data_dir

DESCRIPTION

pg_upgrade is a utility for upgrading from a previous PostgreSQL release without reloading all the data. Not all Postgres release transitions can be handled this way; check the release notes for details on your installation.

"UPGRADING POSTGRES WITH PG_UPGRADE"
Back up your existing data directory, preferably by making a complete dump with pg_dumpall.

Do

% pg_dumpall -s >db.out

to dump out your old database's table definitions without any data.

Stop the old postmaster and all backends.

Rename (using mv) your old pgsql data/ directory to data.old/.

Do

% make install

to install the new binaries.

Run initdb to create a new template1 database containing the system tables for the new release.

Start the new postmaster. (Note: it is critical that no users connect to the database until the upgrade is complete. You may wish to start the postmaster without -i and/or alter pg_hba.conf temporarily.)

Change your working directory to the pgsql main directory, and type:

% pg_upgrade -f db.out data.old

The program will do some checking to make sure everything is properly configured, and will run your db.out script to recreate all the databases and tables you had, but with no data. It will then physically move the data files containing non-system tables and indexes from data.old/ into the proper data/ subdirectories, replacing the empty data files created during the db.out script.

Carefully examine the contents of the upgraded database. If you detect problems, you'll need to recover by restoring from your full pg_dump backup.

The upgraded database will be in an un-vacuumed state. You will probably want to run a VACUUM ANALYZE before beginning production work.

Restart the postmaster and/or restore your old pg_hba.conf if needed to allow user logins. You can delete the data.old/ directory when you are finished.