Re: Corrupt RTREE index

-----Original Message-----
From: pgsql-general-owner@postgresql.org
[mailto:pgsql-general-owner@postgresql.org] On Behalf Of Greg Stark
Sent: Tuesday, December 14, 2004 8:49 PM
To: pgsql-general@postgresql.org
Subject: Re: [GENERAL] Corrupt RTREE index


Scott Marlowe <smarlowe@g2switchworks.com> writes:

> IS this same issue true for hash or GiST indexes?

I think that's true, afaik rtree, GiST, and hash are all not WAL-logged.

> On Tue, 2004-12-14 at 13:49, Dann Corbit wrote:
> > I suggest a warning (if there is not already one generated) on
create
> > index for rtree indexes so that users know that they are not fully
> > supported.

I'm not sure what he means by "supported" though. I'm getting all the
support
I'm paying for, plus a whole lot more.
>>
By "supported" I mean the operations against the index are logged, so
that if someone kicks the plug out of the wall on my PostgreSQL database
and I walk over and plug it back in, I can rely on my btree indexes but
all bets are off for hash, rtree and gist indexes when the server
restarts.

Or perhaps I misunderstand the repercussions of index types not being
included in the WAL.
<<