Re: 20220221-Clarification regarding PostgeSQL DB backup

[Daevor The Devoted <dollien@xxxxxxxxx>]

Hi Karthick

I'm a little rusty on PG (it's been about a decade since I last worked intensely with it), but I seem to remember that restoring with indexes can be very slow. The faster approach is to restore without indexes, and then create the indexes once the data restore is complete.

My knowledge may be outdated, so best to check with others that have more recent knowledge. However, it's probably worth investigating in the meantime.

Okay, I just had a quick look at the documentation, and it seems there is an option to run multiple concurrent jobs for the time-consuming parts of a restore (which includes index creation):

-j number-of-jobs

see: https://www.postgresql.org/docs/current/app-pgrestore.html (or check the docs for whatever PG version you have)

Good luck!

~~Na-iem Dollie

On Mon, Feb 21, 2022 at 6:59 AM Techsupport <techsupport@xxxxxxxxxxx> wrote:

Hi Team,                

 

We have used PG_DUMP to take backup of particular database from the PostgreSQL Server. It takes too long to restore the databases , which has  Half Billion records (almost 8 Hour)

 

My primary need is to make the Differential and Incremental backup on the Windows Server. When I search there is a tool Barman and PG_BackRest is available. But that will be supported only for Linux based servers only

 

Please suggest me a tool to make differential backup

 

Thanks,

Karthick Ramu