pg_upgradecluster - upgrade an existing PostgreSQL cluster to a new major version.
pg_upgradecluster [-v newversion] version name [newdatadir]
pg_upgradecluster upgrades an existing PostgreSQL server cluster (i. e. a collection of databases served by a postmaster instance) to a new version specified by newversion (default: latest available version). The configuration files of the old version are copied to the new cluster.
The cluster of the old version will be configured to use a previously unused port since the upgraded one will use the original port. The old cluster is not automatically removed. After upgrading, please verify that the new cluster indeed works as expected; if so, you should remove the old cluster with pg_dropcluster(8). Please note that the old cluster is set to "manual" startup mode, in order to avoid inadvertently changing it; this means that it will not be started automatically on system boot, and you have to use pg_ctlcluster(8) to start/stop it. See section "STARTUP CONTROL" in pg_createcluster(8) for details.
The newdatadir argument can be used to specify a non-default data directory of the upgraded cluster. It is passed to pg_createcluster. If not specified, this defaults to /var/lib/postgresql/version/name.
Like --locale, but only sets the locale in the specified category.
Some PostgreSQL extensions like PostGIS need metadata in auxiliary tables which
must not be upgraded from the old version, but rather initialized for the new
version before copying the table data. For this purpose, extensions (as well as
administrators, of course) can drop upgrade hook scripts into
/etc/postgresql-common/pg_upgradecluster.d/
. Scripts in that
directory will be called with the following arguments:
<old version> <cluster name> <new version> <phase>
Phases:
The scripts are called as the user who owns the database.
pg_createcluster(8), pg_dropcluster(8), pg_lsclusters(1), pg_wrapper(1)
Martin Pitt <mpitt@debian.org>