Re: TRUNCATE+COPY optimization and --jobs=1 in pg_restore - Mailing list pgsql-hackers

From Takahiro Itagaki
Subject Re: TRUNCATE+COPY optimization and --jobs=1 in pg_restore
Date
Msg-id 20100210132455.47F4.52131E4D@oss.ntt.co.jp
Whole thread Raw
In response to Re: TRUNCATE+COPY optimization and --jobs=1 in pg_restore  (Andrew Dunstan <andrew@dunslane.net>)
List pgsql-hackers
Andrew Dunstan <andrew@dunslane.net> wrote:

> Takahiro-san is suggesting there is a case for doing the optimisation in 
> non-parallel mode. But if we do that, is there still a case for 
> --single-transaction?

I think --single-transaction is useful to restore data into non-empty
databases. A normal restore ignores errors, but it might make database
inconsistent state. So, we'd better keep --single-transaction option
to support all-or-nothing restore.

Regards,
---
Takahiro Itagaki
NTT Open Source Software Center




pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: TRUNCATE+COPY optimization and --jobs=1 in pg_restore
Next
From: Takahiro Itagaki
Date:
Subject: pg_restore --single-transaction and --clean