Re: pg_dump / pg_dumpall / memory issues

We're running the 7.3 series. We are getting better backup performance
than 7.2 indeed. Looking at pg_dump.c -- it seems to use the COPY
command so perhaps throttling in the code may not be the best solution.
I did see the notes about this that Tom mentioned.

> So I'm wondering if the simple solution might be to either use a slower
> box / network connection / throttled port on the backup box, or just
> backup into another database since the copies into the other machine
> probably slow things down enough to render less of a load on the server
> being backed up.

Hmmm... that might be an interesting solution. We do have a slower
standby DB, that would be excellent for that purpose. It would be an
added incentive too, because the standby DB would be hot after backup.
I'm gonna give this a shot and report back. I guess we can do a
pg_dumpall from the standby DB as soon as the main DB has finished
backing up too!

Regards
- Ericson Smith
eric(at)did-it(dot)com

scott.marlowe wrote:

>What version of pgsql are you running? I've gotten MUCH better backup
>restore performance on 7.3 series than I did with 7.2. I can backup 1 gig
>of data in about 10 minutes across 100 Base Tx network with a pipe like
>so:
>
>pg_dump -h hostname databasename | psql databasename
>
>It took something like 30 minutes to an hour before to do this in 7.2.x.
>
>(My box is a dual PIII 750 with 1.5 gig ram, and a 10KRPM UWScsi drive for
>the database seperate from the system.)
>
>
>
>
>