Automated backups for PG running on Windows Server?
Quick question: I have installed PG 8.2.x on a customer's system and we will be storing a lot of critical data on it. It needs to be backed up maybe twice a day. It's very easy to back data up with the pgadmin tool (great tool btw) but we need an automated solution that we can configure. This is all running on a Windows 2003 server. Any suggestions? We're perfectly open to commercial or non-commercial solutions.
Ideally this should be able to do a hot backup, where it gets a valid snapshot even as the server is running. I don't know that, in these days of super-cheap disks, we need to worry about incremental backup. If the entire dataset is 1gb (which it isn't) then a large current hard drive can hold a full years worth of dumps, so incremental backup doesn't seem relevant.
Btw, this pgadmin was pretty cool. I've only ever used PG on Linux until now and I had never seen a tool like that. I was impressed with how easy it was to install PG, get it running, and then use this pgadmin tool to control everything. And I'm a total newbie on Windows. This was the first time I have ever looked at Windows 2003 Server.
____________________________________________________________________________________
Never Miss an Email
Stay connected with Yahoo! Mail on your mobile. Get started!
http://mobile.yahoo.com/services?promote=mail
Vacuum Joe wrote:
Quick question: I have installed PG 8.2.x on a customer's system and
we will be storing a lot of critical data on it. It needs to be
backed up maybe twice a day. It's very easy to back data up with the
pgadmin tool (great tool btw) but we need an automated solution that
we can configure. This is all running on a Windows 2003 server. Any
suggestions? We're perfectly open to commercial or non-commercial
solutions.
Use "Scheduled Tasks" to schedule the run of a .BAT file, from which you
run pg_dump.
Ideally this should be able to do a hot backup, where it gets a valid
snapshot even as the server is running. I don't know that, in these
days of super-cheap disks, we need to worry about incremental backup.
If the entire dataset is 1gb (which it isn't) then a large current
hard drive can hold a full years worth of dumps, so incremental
backup doesn't seem relevant.
pg_dump should do that fine. It doesn't do incremental, but it does hot
backup.
Btw, this pgadmin was pretty cool. I've only ever used PG on Linux
until now and I had never seen a tool like that. I was impressed
with how easy it was to install PG, get it running, and then use this
pgadmin tool to control everything. And I'm a total newbie on
Windows. This was the first time I have ever looked at Windows 2003
Server.
pgadmin works fine on Linux as well, you know.. ;-)
//Magnus