1
0
mirror of https://github.com/postgres/postgres.git synced 2025-06-23 14:01:44 +03:00

Avoid amcheck inline compression false positives.

The previous tacit assumption that index_form_tuple() hides differences
in the TOAST state of its input datums was wrong.  Normalize input
varlena datums by decompressing compressed values, and forming a new
index tuple for fingerprinting using uncompressed inputs.  The final
normalized representation may actually be compressed once again within
index_form_tuple(), though that shouldn't matter.  When the original
tuple is found to have no datums that are compressed inline, fingerprint
the original tuple directly.

Normalization avoids false positive reports of corruption in certain
cases.  For example, the executor can apply toasting with some inline
compression to an entire heap tuple because its input has a single
external TOAST pointer.  Varlena datums for other attributes that are
not particularly good candidates for inline compression can be
compressed in the heap tuple in passing, without the representation of
the same values in index tuples ever receiving concomitant inline
compression.

Add a test case to recreate the issue in a simpler though less realistic
way: by exploiting differences in pg_attribute.attstorage between heap
and index relations.

This bug was discovered by me during testing of an upcoming set of nbtree
enhancements.  It was also independently reported by Andreas Kunert, as
bug #15597.  His test case was rather more realistic than the one I
ended up using.

Bug: #15597
Discussion: https://postgr.es/m/CAH2-WznrVd9ie+TTJ45nDT+v2nUt6YJwQrT9SebCdQKtAvfPZw@mail.gmail.com
Discussion: https://postgr.es/m/15597-294e5d3e7f01c407@postgresql.org
Backpatch: 11-, where heapallindexed verification was introduced.
This commit is contained in:
Peter Geoghegan
2019-02-06 15:54:17 -08:00
parent 45ae2031ec
commit 2f54166668
3 changed files with 162 additions and 23 deletions

View File

@ -88,10 +88,26 @@ DELETE FROM delete_test_table WHERE a > 10;
VACUUM delete_test_table;
SELECT bt_index_parent_check('delete_test_table_pkey', true);
--
-- BUG #15597: must not assume consistent input toasting state when forming
-- tuple. Bloom filter must fingerprint normalized index tuple representation.
--
CREATE TABLE toast_bug(buggy text);
ALTER TABLE toast_bug ALTER COLUMN buggy SET STORAGE plain;
-- pg_attribute entry for toasty.buggy will have plain storage:
CREATE INDEX toasty ON toast_bug(buggy);
-- Whereas pg_attribute entry for toast_bug.buggy now has extended storage:
ALTER TABLE toast_bug ALTER COLUMN buggy SET STORAGE extended;
-- Insert compressible heap tuple (comfortably exceeds TOAST_TUPLE_THRESHOLD):
INSERT INTO toast_bug SELECT repeat('a', 2200);
-- Should not get false positive report of corruption:
SELECT bt_index_check('toasty', true);
-- cleanup
DROP TABLE bttest_a;
DROP TABLE bttest_b;
DROP TABLE bttest_multi;
DROP TABLE delete_test_table;
DROP TABLE toast_bug;
DROP OWNED BY bttest_role; -- permissions
DROP ROLE bttest_role;