Hi Julien,
>> But I would have prefered this new feature to work the same way with or
>> without track_planning activated ;o(
> Definitely, but fixing the issue in pgss (ignoring planner calls when
> we don't have a query text) means that pgss won't give an exhaustive
> view of activity anymore, so a fix in IVM would be a better solution.
> Let's wait and see if Nagata-san and other people involved in that
> have an opinion on it.
It seems IVM team does not consider this point as a priority ...
We should not wait for them, if we want to keep a chance to be
included in PG13.
So we have to make this feature more robust, an assert failure being
considered as a severe regression (even if this is not coming from pgss).
I like the idea of adding a check for a non-zero queryId in the new
pgss_planner_hook() (zero queryid shouldn't be reserved for
utility_statements ?).
Fixing the corner case where a query (with no sql text) can be planned
without being parsed is an other subject that should be resolved in an
other thread.
This kind of query was ignored in pgss, it should be ignored in pgss with
planning counters.
Any thoughts ?
Regards
PAscal
--
Sent from: https://www.postgresql-archive.org/PostgreSQL-hackers-f1928748.html
