Re: An isolation test for SERIALIZABLE READ ONLY DEFERRABLE

On Thu, Jan 12, 2017 at 4:09 AM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> On Tue, Jan 10, 2017 at 11:41 PM, Michael Paquier
> <michael(dot)paquier(at)gmail(dot)com> wrote:
>> Do you think that expanding the wait query by default could be
>> intrusive for the other tests? I am wondering about such a white list
>> to generate false positives for the existing tests, including
>> out-of-core extensions, as all the tests now rely only on
>> pg_blocking_pids().
>
> It won't affect anything unless running at transaction isolation level
> serializable with the "read only deferrable" option.

Indeed as monitoring.sgml says. And that's now very likely close to
zero. It would be nice to add a comment in the patch to just mention
that. In short, I withdraw my concerns about this patch, the addition
of a test for repeatable read outweights the tweaks done in the
isolation tester. I am marking this as ready for committer, I have not
spotted issues with it.
--
Michael