Hi,
I have a database with a lot of schemas; now one schema has been added, re-integrating a former separate database (of
course,there are good reasons to do so). This "new" schema contains very much data which is rarely changed; thus, it
wouldbe sufficient to save it much less often than the rest of the database (e.g. once a month).
So what I'd like to do is do the daily dump with this schema (i.e. its data) dropped.
I read the "Disabling and enabling constraints and triggers to make pg_restore work" thread
(http://groups.google.de/group/pgsql.general/browse_thread/thread/83cc7097eb765220);yes, I can use the database
superuseraccount.
So, are there any contraindications to do create a working backup like so:
- dump the schemas
- "update pg_catalog.pg_class set relchecks=0;"
- dump all the data minus the one schema to be left out
- "update pg_catalog.pg_class set relchecks=1;"
The data of the missing schema would be restored separately; I'd have to support this with scripts both for Linux and
Windows.
--
Tia,
Tobias
