Re: BUG #16031: Group by returns duplicate groups - Mailing list pgsql-bugs

From Andrew Gierth
Subject Re: BUG #16031: Group by returns duplicate groups
Date
Msg-id 87o8z0lc6i.fsf@news-spur.riddles.org.uk
Whole thread Raw
In response to RE: BUG #16031: Group by returns duplicate groups  (David Raymond <David.Raymond@tomtom.com>)
List pgsql-bugs
>>>>> "David" == David Raymond <David.Raymond@tomtom.com> writes:

 David> Here's what it's reporting for an error:

 David> testing=# select bt_index_check('big_table_name_id_1_id_2_idx', true);
 David> ERROR:  high key invariant violated for index "big_table_name_id_1_id_2_idx"
 David> DETAIL:  Index tid=(86990,140) points to heap tid=(139313,57) page lsn=0/0.
 David> Time: 65695.359 ms (01:05.695)
 David> testing=# select bt_index_parent_check('big_table_name_id_1_id_2_idx', true);
 David> ERROR:  down-link lower bound invariant violated for index "big_table_name_id_1_id_2_idx"
 David> DETAIL:  Parent block=43131 child index tid=(43197,9) parent page lsn=0/0.
 David> Time: 1697.205 ms (00:01.697)
 David> testing=#

Based on the other response, it looks like something is fundamentally
broken with regard to sorting this dataset, and since btree index build
works by sorting, the index corruption is probably just another symptom
of the real problem rather than a cause in itself.

-- 
Andrew (irc:RhodiumToad)



pgsql-bugs by date:

Previous
From: "David G. Johnston"
Date:
Subject: Re: psql: FATAL: database “” does not exist
Next
From: PG Bug reporting form
Date:
Subject: BUG #16032: pg_basebackup when running on Windows doesn't clean up on failure correctly