Thread: pgsql: Add regression test for 04ae11f62e643e07c411c4935ea6af46cb112aa9
Add regression test for 04ae11f62e643e07c411c4935ea6af46cb112aa9. The code in this area needs further revision, and it would be best not to re-break the things we've already fixed. Per a gripe from Tom Lane. Branch ------ master Details ------- http://git.postgresql.org/pg/commitdiff/8c1d9d56e9a00680a035b8b333a98ea16b121eb7 Modified Files -------------- src/test/regress/expected/select_parallel.out | 48 +++++++++++++++++++++++++++ src/test/regress/sql/select_parallel.sql | 18 ++++++++++ 2 files changed, 66 insertions(+)
Robert Haas <rhaas@postgresql.org> writes:
> Add regression test for 04ae11f62e643e07c411c4935ea6af46cb112aa9.
BTW, the buildfarm thinks this test fails a lot of the time, and
running "make installcheck-parallel" a few times here agrees: it
falls over more than 20% of the time. Don't know why it would be
probabilistic, but the alleged protection against running
parallel-restricted functions in workers apparently doesn't work
at all.
regards, tom lane
On Fri, Jun 17, 2016 at 9:20 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> Robert Haas <rhaas@postgresql.org> writes:
> > Add regression test for 04ae11f62e643e07c411c4935ea6af46cb112aa9.
>
> BTW, the buildfarm thinks this test fails a lot of the time, and
> running "make installcheck-parallel" a few times here agrees: it
> falls over more than 20% of the time. Don't know why it would be
> probabilistic,
+ if (found) then
+ raise 'parallel restricted function run in worker';
+ end if;
>
> Robert Haas <rhaas@postgresql.org> writes:
> > Add regression test for 04ae11f62e643e07c411c4935ea6af46cb112aa9.
>
> BTW, the buildfarm thinks this test fails a lot of the time, and
> running "make installcheck-parallel" a few times here agrees: it
> falls over more than 20% of the time. Don't know why it would be
> probabilistic,
>
I also got failure with this test on my m/c:
select parallel_restricted(unique1) from tenk1
where stringu1 = 'GRAAAA' order by 1;
ERROR: parallel restricted function run in worker
CONTEXT: PL/pgSQL function parallel_restricted(integer) line 5 at RAISE
Isn't it possible that the below statement used in parallel restricted function fails in leader backend itself:
+ perform * from pg_stat_activity where client_port is null;+ if (found) then
+ raise 'parallel restricted function run in worker';
+ end if;
On Fri, Jun 17, 2016 at 12:55 AM, Amit Kapila <amit.kapila16@gmail.com> wrote: > Isn't it possible that the below statement used in parallel restricted > function fails in leader backend itself: > + perform * from pg_stat_activity where client_port is null; > + if (found) then > + raise 'parallel restricted function run in worker'; > + end if; Yeah, I can't believe I managed to be dumb enough to think that was going to work. Clearly, I shouldn't be trusted with anything more dangerous than a dull pair of scissors. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
Re: pgsql: Add regression test for 04ae11f62e643e07c411c4935ea6af46cb112aa9
From
Andres Freund
Date:
On 2016-06-17 08:41:31 -0400, Robert Haas wrote: > On Fri, Jun 17, 2016 at 12:55 AM, Amit Kapila <amit.kapila16@gmail.com> wrote: > > Isn't it possible that the below statement used in parallel restricted > > function fails in leader backend itself: > > + perform * from pg_stat_activity where client_port is null; > > + if (found) then > > + raise 'parallel restricted function run in worker'; > > + end if; > > Yeah, I can't believe I managed to be dumb enough to think that was > going to work. > > Clearly, I shouldn't be trusted with anything more dangerous than a > dull pair of scissors. Dull scissors actually cause much harder to sow up / heal wounds, than sharp ones ... :P