Re: alter_table regression test problem

Kevin Grittner <kgrittn(at)ymail(dot)com> wrote:

> In checking things out with CLOBBER_CACHE_ALWAYS, I was getting
> this problem, which seems to be unrelated to my changes:

On a CLOBBER_CACHE_ALWAYS build I did a fresh initdb, started the
cluster, and immediately tested this query on both the postgres and
template1 databases, with the same result:

SELECT
    *
FROM (
    SELECT
        oid, reltablespace, relfilenode, relname,
        pg_filenode_relation(reltablespace,
pg_relation_filenode(oid)) mapped_oid
    FROM pg_class
    WHERE relkind IN ('r', 'i', 'S', 't', 'm')
    ) mapped
WHERE (mapped_oid != oid OR mapped_oid IS NULL);
 oid  | reltablespace | relfilenode |   relname    | mapped_oid
------+---------------+-------------+--------------+------------
 2619 |             0 |       11828 | pg_statistic | 2139062143
(1 row)

That makes for a pretty simple test for git bisect, even if
everything including initdb is painfully slow with
CLOBBER_CACHE_ALWAYS.

--
Kevin Grittner
EDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company