many thanks for the help.
i found the instructions on a mailing list via a search of the web.
i eventually added a btree index but was under the (probably mistaken!)
impression that hash indexes were better for varchar values.
once again many thanks,
Steve
"Tom Lane" <tgl@sss.pgh.pa.us> wrote in message
news:26967.1003331234@sss.pgh.pa.us...
> "Steve Brett" <steve.brett@e-mis.com> writes:
> > i followed the instructions very carefully and get the following error
when
> > i try to run pg_upgrade ..
>
> pg_upgrade hasn't worked since 7.0. Where did you find instructions
> that told you to run it?
>
> >> I added a hash index to a varchar value and when i vacuumed i got the
> >> following error:
> >>
> >> Index customer_ha_hash: NUMBER OF INDEX' TUPLES (9176) IS NOT THE SAME
AS
> >> HEAP' (9181).
>
> If you have any rows that contain NULL in the indexed column, then this
> result isn't very surprising, because hash indexes don't index nulls.
> (Current sources have been fixed not to issue the cross-check notice
> message for hash indexes, btw.)
>
> Personally I'd advise not bothering with hash indexes; use a plain btree
> index instead. Does more, works better, doesn't have concurrency
> problems.
>
> regards, tom lane
>
> ---------------------------(end of broadcast)---------------------------
> TIP 3: if posting/reading through Usenet, please send an appropriate
> subscribe-nomail command to majordomo@postgresql.org so that your
> message can get through to the mailing list cleanly
