On Dec 27, 2012, at 12:46 PM, Tom Lane wrote:
> Or you could run contrib/pgstattuple's pgstattuple() function every so
> often --- it will report the uncommitted tuples as "dead", which is
> inaccurate, but you'd be able to see how fast the number is increasing.
That's exactly the kind of thing I
Scott Ribe writes:
> Is there any way to get some insight into the progress of:
> insert into foo select distinct on (...) from bar where...
Watching the physical size of the foo table might be close enough.
Or you could run contrib/pgstattuple's pgstattuple() function every so
often --- it will
