On Thu, 11 Nov 2004 13:56:48 -0500, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> Were they both VACUUM ANALYZEs? There's a known gotcha that if you
> ANALYZE the same table concurrently in two different sessions, one
> of them can fail this way when it goes to update the pg_statistic
> entries. (Harmless, but annoying.) I don't know of any cause for
> this message from a plain VACUUM though.
>
> regards, tom lane
>
That's what it is I have cron jobs running over each other during heavy load.
Thanks
