To follow up, ZFS snapshots (appear to) offer a great solution to the
problem I posed a couple of weeks ago, and avoid any hacking around with
misuse of tablespaces.
My goal was to have a database with a 100GB table and a 600GB table, and to
routinely and efficiently clone the 100GB table and its
Thanks for such quick and helpful answers! My plan sounds probably better
to avoid, but if it turns out to be necessary, you all gave me some helpful
avenues and things to look out for.
Best,
Jack
On Tue, Aug 14, 2018 at 1:06 PM, Stephen Frost wrote:
> Greetings,
>
> * Jack Cushman
On 2018-08-14 11:57:38 -0400, Jack Cushman wrote:
> I have a large database of text, with a 600GB table and a 100GB table
> connected
> by a join table. They both see occasional updates throughout the week. Once a
> week I want to "cut a release," meaning I will clone just the 100GB table and
>
Tom Lane writes:
> Jack Cushman writes:
>
>> Now here's the situation where I want to do what Christophe said not to do:
>> :)
>
>> I have a large database of text, with a 600GB table and a 100GB table
>> connected by a join table. They both see occasional updates throughout the
>> week. Once a
Greetings,
* Jack Cushman (jcush...@gmail.com) wrote:
> I have a large database of text, with a 600GB table and a 100GB table
> connected by a join table. They both see occasional updates throughout the
> week. Once a week I want to "cut a release," meaning I will clone just the
> 100GB table and
Jack Cushman writes:
> Now here's the situation where I want to do what Christophe said not to do:
> :)
> I have a large database of text, with a 600GB table and a 100GB table
> connected by a join table. They both see occasional updates throughout the
> week. Once a week I want to "cut a
Hi --
I'm wondering whether, in my specific situation, it would be safe to copy a
database cluster's data folder, and bring up the copy for read access,
without copying a tablespace linked from it. My situation (described below)
involves a database with a 100GB table and a 600GB table where I
