Re: Possible bug in vacuum redo - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Possible bug in vacuum redo
Date
Msg-id 5492.1009154684@sss.pgh.pa.us
Whole thread Raw
In response to Re: Possible bug in vacuum redo  (Hiroshi Inoue <Inoue@tpf.co.jp>)
List pgsql-hackers
Hiroshi Inoue <Inoue@tpf.co.jp> writes:
> I don't think it's preferable either. However there's
> no problem unless there's an application which handle
> the tuples containing the t_ctid link.

What about READ COMMITTED mode?  EvalPlanQual uses the t_ctid field
to find the updated version of the row.  If the t_ctid is wrong,
you might get an elog(), or you might miss the row you should have
updated, or possibly you might update a row that you shouldn't have.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Hiroshi Inoue
Date:
Subject: Re: Possible bug in vacuum redo
Next
From: Hiroshi Inoue
Date:
Subject: Re: Possible bug in vacuum redo