Re: Hot-standby/Reporting database.

On Sun, Mar 20, 2011 at 10:38 PM, jonathan ferguson <jdpf(at)hoozinga(dot)com> wrote:
> hi.
>
> On Mar 20, 2011, at 9:05 PM, Jaime Casanova wrote:
>
>> On Sun, Mar 20, 2011 at 8:04 PM, Jaime Casanova <jaime(at)2ndquadrant(dot)com> wrote:
>>> On Fri, Mar 18, 2011 at 4:25 PM, John P Weatherman
>>> <jweatherman91(at)alumni(dot)wfu(dot)edu> wrote:
>>>> Has anyone run into this already and have an idea for a work around?  I
>>>> am primarily an Oracle guy and in that environment I would set up a
>>>> second DB with database links to the hot standby,
>>>
>>> you can use the contrib module dblink for this:
>>> http://www.postgresql.org/docs/9.0/static/dblink.html
>>
>> obviously you need to install the module libraries in both the master
>> and the slave and the sql functions that create objects in master only
>
> Perhaps I'm in the clueless n00b category, here, but how does:
>
> "dblink [] a module which supports connections to other PostgreSQL databases from within a database session."
>

your question had 2 parts...
> 1) real time data replication (which the hot standby does) and

for this one hot standby is almost that. ok, you have a little delay
but is that delay that bad that you can't use it?
if it is, 9.1 will ship with synchronous replication

> 2) the ability to create temporary tables as part of their reporting jobs
> (which is expressly prohibited in a hot standby.

AFAIU, the links you say you create in Oracle are for this... no?
that's why i suggest using dblink...
if i misunderstood how you use the database links, please explain...

--
Jaime Casanova         www.2ndQuadrant.com
Professional PostgreSQL: Soporte y capacitación de PostgreSQL