Re: BUG #14526: no unique or exclusion constraint matching the ON CONFLICT

On Tue, Feb 7, 2017 at 3:46 PM, Tiago Babo <tiago(dot)babo(at)gmail(dot)com> wrote:
> I don't really understand how PostgreSQL handles indexes, but would it be
> possible that the INDEX is being used/updated at that moment and so the
> INSERT doesn't know it exists? Can concurrency also be a problem?

Anything is possible, I suppose, but that seems very unlikely to be a
factor here. Any problem like this occurs in codepaths that only
consider metadata -- the definition of indexes themselves, underlying
types, and so on. This presumably never changes here; you aren't
creating and dropping indexes on the table in question around the time
you see problems.

--
Peter Geoghegan