Re: oids as primary keys? - Mailing list pgsql-novice

From Michael Fuhr
Subject Re: oids as primary keys?
Date
Msg-id 20050415160452.GA83778@winnie.fuhr.org
Whole thread Raw
In response to oids as primary keys?  ("Cima" <ruel.cima@facinf.uho.edu.cu>)
List pgsql-novice
On Fri, Apr 15, 2005 at 03:42:40PM -0700, Cima wrote:
>
> someone has drawn my attention to the fact that oids may not be such a good
> idea to set as a primary key in a table. i have designed a relativley large
> database and defined oids as primary keys. i would like your opinions or
> recomendations on this.

See "Object Identifier Types" in the "Data Types" chapter of the
documentation:

http://www.postgresql.org/docs/8.0/interactive/datatype-oid.html

"The oid type is currently implemented as an unsigned four-byte
integer.  Therefore, it is not large enough to provide database-wide
uniqueness in large databases, or even in large individual tables.
So, using a user-created table's OID column as a primary key is
discouraged.  OIDs are best used only for references to system
tables."

See also "What is an OID?  What is a TID?" in the FAQ:

http://www.postgresql.org/docs/faqs.FAQ.html#4.12

"OIDs are autotomatically assigned unique 4-byte integers that are
unique across the entire installation.  However, they overflow at
4 billion, and then the OIDs start being duplicated."

"To uniquely number columns in user tables, it is best to use SERIAL
rather than OIDs because SERIAL sequences are unique only within a
single table and are therefore less likely to overflow.  SERIAL8
is available for storing eight-byte sequence values."

--
Michael Fuhr
http://www.fuhr.org/~mfuhr/

pgsql-novice by date:

Previous
From: Frank Bax
Date:
Subject: Re: oids as primary keys?
Next
From: "Cima"
Date:
Subject: file oids