On Thu, May 13, 2010 at 8:16 PM, Scott Marlowe <scott.marlowe@xxxxxxxxx> wrote:
On Thu, May 13, 2010 at 9:50 AM, Wang, Mary Y <mary.y.wang@xxxxxxxxxx> wrote:That will work. So will pg_dumpall | gzip > filename.gz if you want
> Hi,
>
> I'm running on Postgres 8.3.8. My system admin is ready to set up a cron job for a daily database backup.
> By reading the documentation over here: http://www.postgresql.org/docs/8.3/static/backup-dump.html#BACKUP-DUMP-ALL (I only found the documentation for 8.3.10), and it looks like pg_dumpall > outfile is the best choice. I'd like to ask the community to reconfirm.
compression.
My preference is to use
pg_dumpall -g > globals.bak
pg_dump -Fc database > database.bak
pg_dump -Fc database1 > database1.bak
etc..
The major reason for this is because this allows selective restores. If something goes wrong with my 'users' table in 'database1' :
Using the pg_dumpall strategy
- uncompress large text file
- manually cut 'users' table from large text file
- restore
With the pg_dump -Fc strategy
- pg_restore -t users -d database1 < database1.bak
The pg_dump -Fc strategy also allows you to leverage the new, parallel restore feature (-j option to pg_restore) in 8.4.
--Scott
What's more important is that you first prove your backups are working
by restoring them elsewhere, then check every month or so to make sure
they're still happening.
--
Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-general
