Re: [COMMITTERS] pgsql: Assert that we don't invent relfilenodes ortype OIDs in binary

On Tue, Jun 13, 2017 at 03:10:16PM -0400, Tom Lane wrote:
> Bruce Momjian <bruce@momjian.us> writes:
> > On Tue, Jun 13, 2017 at 11:52:04AM -0700, Andres Freund wrote:
> >> I'm not sure what you're actually concerned about here?
>
> > I am concerned a non-assert build will not error out, but if no one else
> > is concerned about that, I am fine.
>
> I think in a production situation, we actually don't want it to error
> out.  The odds are fairly good that the run would complete successfully
> (ie, the potential OID collision never actually materializes).  So all
> we're doing is converting a possible failure into an unavoidable one.
>
> Where we want to hear about the problem is in development.  So really
> an Assert is the right thing.

OK, but my point is that all the other places, which seems similar,
error out in production.

--
  Bruce Momjian  <bruce@momjian.us>        http://momjian.us
  EnterpriseDB                             http://enterprisedb.com

+ As you are, so once was I.  As I am, so you will be. +
+                      Ancient Roman grave inscription +