In message
<4e79ea8c2ba6694eb6a87686c3a9a76908a8a...@afs_exchange.afs.afsi.com>,
Mark Eastwood <[email protected]> writes
I suspect the Distributed file is not the problem, but rather bad
dictionaries. Do all the dictionaries have a data-type defined (A&S
types attr 6, D&I types attr 8)? A common issue I've encountered is when
a dictionary is right justified but attribute contains non-numeric data.
Following up from this, go to PickWiki, to the Universe page, and then
look at the page there that talks about ODBC. It will point you at a
bunch of utilities, written specifically to go through warning/cleaning
your data. I learnt the hard way that HS.SCRUB had a habit of crashing
and it was difficult to debug why - programs should NEVER crash when fed
duff data and HS.SCRUB is (or was) an egregious example of a program
that does exactly that :-(
Oh - yes I don't think you can run HS.SCRUB on the master distributed
file, you have to run it on the parts. But read the docu. And it could
easily NOT be a duff dictionary, but a duff VOC :-( Read that wiki page!
Lastly, are you using ODBC? If not, and you're planning to, don't
bother. ODBC is now a layer on top of something else (OleDB?), and you
should use that instead. It'll be quicker, easier, nicer, generally
better all round.
Cheers,
Wol
--
Anthony W. Youngman <[email protected]>
'Yings, yow graley yin! Suz ae rikt dheu,' said the blue man, taking the
thimble. 'What *is* he?' said Magrat. 'They're gnomes,' said Nanny. The man
lowered the thimble. 'Pictsies!' Carpe Jugulum, Terry Pratchett 1998
Visit the MaVerick web-site - <http://www.maverick-dbms.org> Open Source Pick
_______________________________________________
U2-Users mailing list
[email protected]
http://listserver.u2ug.org/mailman/listinfo/u2-users
