Re: BUG #5727: Indexes broken in streaming replication - Mailing list pgsql-bugs

From Simon Riggs
Subject Re: BUG #5727: Indexes broken in streaming replication
Date
Msg-id 1288127762.1587.470.camel@ebony
Whole thread Raw
In response to Re: BUG #5727: Indexes broken in streaming replication  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-bugs
On Tue, 2010-10-26 at 14:08 -0400, Tom Lane wrote:
> Heikki Linnakangas <heikki.linnakangas@enterprisedb.com> writes:
> >> * Operations on hash indexes are not presently WAL-logged, so replay will not update these indexes. Hash indexes
willnot be used for query plans during recovery. 
>
> > The initial patch indeed had a special-case in the planner to ignore
> > hash indexes during hot standby, but it was left out because the lack of
> > WAL-logging is a general problem with hash indexes, not a hot standby
> > issue.
>
> Yeah, and also the index would still be broken after the slave exits hot
> standby and becomes live; so that hack didn't cure the problem anyway.

OK, that's a good argument.

--
 Simon Riggs           http://www.2ndQuadrant.com/books/
 PostgreSQL Development, 24x7 Support, Training and Services

pgsql-bugs by date:

Previous
From: Heikki Linnakangas
Date:
Subject: Re: Recovery bug
Next
From: "Dobes Vandermeer"
Date:
Subject: BUG #5728: Unexpected behavior comparing result of age() to an interval