On 5/10/05, stephen booth <stephenbooth.uk_at_gmail.com> wrote:
> On 5/10/05, Oracle <all_about_oracle_at_hotpop.com> wrote:
> > and most of database name under my care is either orcl or prod
> >=3D20
>=20
> Oh, joy. =3D20
>=20
> Yeah, I've got that problem where I work. Databases (indeed servers
> as well) tended to be set up by people in the local departments they
> tended to just give them what ever name made sense to them not
> thinking that maybe someone from outside that department might have to
> support them some day.
>=20
> We've got a lot of databases with names like prod, test, uat &c along
> with a fair few named marvin, ford, zaphod, thomas, fatcon, rover,
> pugwash, blackpig &c. We do have a standard for naming servers but
> that's not helpful as it was designed by someone who only dealt with
> file and print servers scattered accross a geographically diverse
> organisation so the name relates to where the server is, not what it
> does. The upshot of this (coupled with the frequent office moves) is
> that from a server name you cannot tell what it does, which
> application it supports or where it is now but you do know where it
> was when it was initially built.
>=20
> I'm currently in the process of putting together a standards document
> for Oracle databases and am including a passage stating that databases
> shall have meaningful names relating to their purpose.
>=20
> Stephen
Stephen,
I believe that Sean O'Neil put together such a paper years ago that
was for win32 land.
It might serve as a good framework.
I also recall seeing a paper named "Putting OFA on Steriods" by Bill
Burke and Byron Pearce that was geared toward *nix.
If you need their contact info, let me know.
Paul
--=20
#/etc/init.d/init.cssd stop
-- f=3Dma, divide by 1, convert to moles.
--
http://www.freelists.org/webpage/oracle-l
Received on Tue May 10 2005 - 11:23:36 CDT
