On Mon, Oct 5, 2015 at 8:50 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> That's certainly something worth looking at, but I think it's probably
> more complicated than that. If you just write "WHERE x IN (1,2,3,4)",
> that gets folded to a ScalarArrayOp with a single array constant, which
> the existing code would deal with just fine. We need to identify
> situations where that's not the case but yet we shouldn't distinguish.
I have a certain amount of sympathy for doing that kind of thing, but
significantly less than I have for cases with many failed queries,
which is why I focused on that.
> In any case, that's just a marginal tweak for one class of query.
Sometimes narrow cases are also important and representative cases. I
don't care if someone has thrashing type issues with
pg_stat_statements when they're doing something really odd that calls
into question the purpose of using it to begin with [1]. The two
classes of queries we talked about (1. Many aborted data integration
transactions, and 2. A variable number of constants) are interesting
because a reasonable person could have those cases, and run into
trouble with pg_stat_statements as a consequence.
> I suspect the right fix for the core problem is the one Peter mentioned
> in passing earlier, namely make it possible to do garbage collection
> without having to slurp the entire file into memory at once. It'd be
> slower, without a doubt, but we could continue to use the existing code
> path unless the file gets really large.
While it would be nice to not lose query texts on OOM, that's not my
primary concern. My primary concern is infrequent garbage collection.
This fix certainly isn't going to help with the fact that garbage
collection can be stalled for far too long in at least
quasi-reasonable cases.
[1] http://www.postgresql.org/message-id/52E9D98A.4000007@lab.ntt.co.jp
--
Peter Geoghegan