Re: monitoring CREATE INDEX [CONCURRENTLY]

On Mon, 25 Mar 2019 at 22:23, Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>
wrote:

> Here's v6 of this patch. I have rebased on top of today's CLUSTER
> monitoring, as well as on table AM commits. The latter caused a bit of
> trouble, as now the number of blocks processed by a scan is not as easy
> to get as before; I added a new entry point heapscan_get_blocks_done on
> heapam.c to help with that. (I suppose this will need some fixups later
> on.)
>
> I removed the "M of N" phase labels that Robert didn't like; those were
> suggested by Rahila and upvoted by Amit L. I'm of two minds about
> those. If you care about those and want them back, please speak up.
>
> I see value in reporting those numbers because it gives user insight into
where
we are at in the whole process without having to refer to documentation or
code.
Besides here also we are reporting facts as we follow for other metrics.

I agree that it will be most effective if the phases are carried out in
succession
which is not the case every time and its relevance varies for each command
as mentioned upthread by Alvaro and Robert. But I feel as long as we have in
the documentation that some phases overlap, some are mutually exclusive
hence
may be skipped etc. reporting `phase number versus total phases` does
provide
valuable information.
We are able to give user a whole picture in addition to reporting progress
within phases.

Thank you,
--
Rahila Syed
Performance Engineer
2ndQuadrant
http://www.2ndQuadrant.com <http://www.2ndquadrant.com/>
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services