Re: [HACKERS] Default index tablespace

2008-01-25 Thread Tom Lane
Peter Eisentraut <[EMAIL PROTECTED]> writes: > What about a feature to set a default tablespace just for indexes? Well, for one thing it would break existing dumps, since pg_dump expects that "SET default_tablespace" works for indexes and tables both. This has been suggested and rejected multiple

Re: [HACKERS] Default index tablespace

2008-01-25 Thread Zdenek Kotala
Peter Eisentraut wrote: What about a feature to set a default tablespace just for indexes? I have been told that this was originally proposed when tablespaces where designed, but did not end up being implemented. Does anyone recall the details? I have had people ask me about this feature.

Re: [HACKERS] Default index tablespace

2008-01-25 Thread Simon Riggs
On Fri, 2008-01-25 at 13:01 +0100, Peter Eisentraut wrote: > What about a feature to set a default tablespace just for indexes? Why would we want this particular idea? I'm not against it specifically, but it seems like there are many similar ideas, so we should think about what we are trying to

Re: [HACKERS] Default index tablespace

2008-01-25 Thread Hans-Juergen Schoenig
On Jan 25, 2008, at 1:01 PM, Peter Eisentraut wrote: What about a feature to set a default tablespace just for indexes? I have been told that this was originally proposed when tablespaces where designed, but did not end up being implemented. Does anyone recall the details? I have had p

[HACKERS] Default index tablespace

2008-01-25 Thread Peter Eisentraut
What about a feature to set a default tablespace just for indexes? I have been told that this was originally proposed when tablespaces where designed, but did not end up being implemented. Does anyone recall the details? I have had people ask me about this feature. -- Peter Eisentraut http: