| From: | "scott(dot)marlowe" <scott(dot)marlowe(at)ihs(dot)com> |
|---|---|
| To: | "Ed L(dot)" <pggeneral(at)bluepolka(dot)net> |
| Cc: | <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: Backup alternatives |
| Date: | 2002-10-09 16:48:50 |
| Message-ID: | Pine.LNX.4.33.0210091047270.28259-100000@css120.ihs.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Wed, 9 Oct 2002, Ed L. wrote:
>
> I am curious to know if anyone has any viable live backup procedures other
> than pg_dump/pg_dumpall when you have a continuously running server with
> continuous query activity? The motivation is to find a live alternative
> faster than pg_dump, which presently takes about 45 minutes on some of my
> larger databases. I assume any kind of file copy backup is not workable
> due to the complexities of mvcc, caching, etc.
There is an ongoing project to provide "Point in Time Recovery" (PITR for
short) which does exactly this. Basically, it stores the WAL files
sequentially off the box, and then can use them and the last backup to
rebuild your database should your main box catch on fire, burn down your
hosting center, and initiate an infopocalypse.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Markus Gieppner | 2002-10-09 16:55:26 | High-end PGSQL / Business Intelligence |
| Previous Message | scott.marlowe | 2002-10-09 16:46:09 | Re: access time performance problem |