Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> Tom Lane wrote:
>> ... but everybody else is locked out completely, because the ALTER takes
>> an exclusive lock on the table. It's a bit misleading to describe that
>> as a local change.
> The pre-8.1 method was to UPDATE pg_class.reltriggers = 0. Would that
> allow concurrent access?
It did not lock the table per se. (Whether that was a safe behavior
at all, I'm too tired to work out at the moment.)
regards, tom lane