[sqlsmith] ERROR: failed to build any %d-way joins

Started by Andreas Seltenreichover 10 years ago3 messages
#1Andreas Seltenreich
seltenreich@gmx.de

Hi,

there's a 1/1e6 chance that a sqlsmith query on the regression db of
master (c124cef) fails with

ERROR: failed to build any {4..8}-way joins

They all appear to work fine on REL9_5_STABLE.

sample query:

select 1 from
information_schema.collations as rel_60113935
inner join information_schema.sql_sizing_profiles as rel_60113936
on (rel_60113935.pad_attribute = rel_60113936.sizing_name )
inner join information_schema.foreign_tables as rel_60113937
on (rel_60113936.profile_id = rel_60113937.foreign_table_catalog )
right join information_schema.constraint_table_usage as rel_60113938
on (rel_60113935.collation_schema = rel_60113938.table_catalog )
left join public.btree_tall_tbl as rel_60113939
on (rel_60113936.required_value = rel_60113939.id )
where rel_60113938.table_name is not NULL;

regards,
Andreas

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

#2Andreas Seltenreich
seltenreich@gmx.de
In reply to: Andreas Seltenreich (#1)
Re: [sqlsmith] ERROR: failed to build any %d-way joins

I wrote:

They all appear to work fine on REL9_5_STABLE.

oops, that branch was slightly stale. Updating it with the latest
planner changes (8dccf03..0853388), the queries fail there as well.

regards,
Andreas

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

#3Tom Lane
tgl@sss.pgh.pa.us
In reply to: Andreas Seltenreich (#1)
Re: [sqlsmith] ERROR: failed to build any %d-way joins

Andreas Seltenreich <seltenreich@gmx.de> writes:

there's a 1/1e6 chance that a sqlsmith query on the regression db of
master (c124cef) fails with
ERROR: failed to build any {4..8}-way joins

Looks like I broke that while trying to fix one of your previous
reports :-(. I think it's all better now, but your continued testing
is much appreciated.

regards, tom lane

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers