Re: Solving the OID-collision problem

On Wed, 2005-08-10 at 14:42 -0400, Greg Stark wrote:
> Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
>
> > Just to chime in --- I have been surprised how _few_ complaints we have
> > gotten about oid wraparound hitting system table oid conflicts. I agree
> > that telling people to retry their CREATE statements isn't really an
> > ideal solution, and the idea of looping to find a free oid is a good one.
>
> So in a world where all user tables have OIDs I can see this happening quite
> easily. A reasonably large database could easily have 4 billion records
> inserted in a reasonable amount of time.
>
> But with no OIDs on user tables it must take a really long time for this to
> happen. I mean, even if you have thousands of tables you would have to go
> through thousands (many thousands even) of dump/reload cycles before you push
> oid to 4 billion.
>
> Perhaps just a periodic warning starting when the OID counter hits, say, 2
> billion telling people to dump/reload their database before it gets to 4
> billion would be enough?
>
> All this stuff about retrying OIDs is cool and if someone wants to go and do
> it I wouldn't say they shouldn't. But it seems like a lot of effort to avoid a
> situation that I'm unclear will ever arise.
>
> A warning could more easily be backpatched to versions that defaulted to OIDs
> on user tables too.

I agree with everything you just said.

I think its a non-issue for 8.1+, but an important one for many earlier
users. We *can* ask people to upgrade, but if they have not, there is
usually a good reason. If we force them, they may upgrade to another
RDBMS...

Best Regards, Simon Riggs