On Thu, 17 May 2007, Tom Lane wrote:
> Raf <rafiq@joshua.dreamthought.com> writes:
>> On Thu, 17 May 2007, Tom Lane wrote:
>> Even though there were 0 rows associated with the table's oid in
>> pg_depend and pg_triggers ?
>
> Oh, OK, you want zero then. I had misread it to indicate that only one
> of four triggers had gone missing.
Well, the counter was set at 4. The triggers were sloni related, and I
think running Sloni removal may have removed the other triggers without
updating the trigger count. Why, I don't know? I was left with the one
dangling trigger, which was manually removed and I then updated the
counter to represent the state of pg_deps and pg_triggers.
Thanks,
Raf