Hacktoberfest 2026: le issue che i maintainer hanno segnato per ottobre, aperte e adatte ai principianti. Sfoglia le issue Hacktoberfest

[Bug] GPORCA: “could not find MotionState” for correlated EXISTS with LIMIT and a scalar subquery

Aperta
  • Qualcosa non va?
#2,221 0 commenti 0 reazioni 1 assegnatario Vedi su GitHub

I maintainer di solito rispondono entro 2 giorni

@Alena0704 ci sta già lavorando.

Dal 29/10/2026.

  • #2223 di @Alena0704 — aperta

Valutazione

Questa issue non è ancora stata valutata.

Descrizione

type: Bug
Apache Cloudberry version

ANY

What happened

A valid SELECT fails with SQLSTATE XX000 when using GPORCA. The same query succeeds and returns zero rows with the Postgres optimizer.
The query combines a cross join, a correlated EXISTS ... LIMIT 1, and a scalar aggregate subquery in the EXISTS target list. The reproducer requires only one distributed table containing one row.

What you think should happen instead

No response

How to reproduce
CREATE TEMP TABLE motion_repro (id integer) DISTRIBUTED BY (id);
INSERT INTO motion_repro VALUES (1);
ANALYZE motion_repro;

SET optimizer = on;

SELECT y.id
FROM motion_repro x CROSS JOIN motion_repro y
WHERE EXISTS (
    SELECT (SELECT count(*) FROM motion_repro)
    FROM motion_repro z
    WHERE y.id IS NULL
    LIMIT 1
);

Expected results: Zero rows: the only value of y.id is 1, so the correlated predicate is false.
Actual result:

ERROR: XX000: could not find MotionState for slice 4 in executor tree
       (execUtils.c:2310)
LOCATION: getMotionState, execUtils.c:2310

With optimizer=off, the query returns zero rows successfully. Replacing the EXISTS target list with SELECT 1 also returns zero rows successfully with GPORCA.
full sql reproduction:

\set ON_ERROR_STOP on
\set VERBOSITY verbose
SET statement_timeout = '10s';
CREATE TEMP TABLE motion_repro (id integer) DISTRIBUTED BY (id);
INSERT INTO motion_repro VALUES (1);
ANALYZE motion_repro;

SET optimizer = off;
EXPLAIN (COSTS OFF)
SELECT y.id FROM motion_repro x CROSS JOIN motion_repro y
WHERE EXISTS (SELECT (SELECT count(*) FROM motion_repro)
              FROM motion_repro z WHERE y.id IS NULL LIMIT 1);
SELECT y.id FROM motion_repro x CROSS JOIN motion_repro y
WHERE EXISTS (SELECT (SELECT count(*) FROM motion_repro)
              FROM motion_repro z WHERE y.id IS NULL LIMIT 1);

SET optimizer = on;
EXPLAIN (COSTS OFF)
SELECT y.id FROM motion_repro x CROSS JOIN motion_repro y
WHERE EXISTS (SELECT (SELECT count(*) FROM motion_repro)
              FROM motion_repro z WHERE y.id IS NULL LIMIT 1);
\set ON_ERROR_STOP off
SELECT y.id FROM motion_repro x CROSS JOIN motion_repro y
WHERE EXISTS (SELECT (SELECT count(*) FROM motion_repro)
              FROM motion_repro z WHERE y.id IS NULL LIMIT 1);
\echo SQLSTATE=:SQLSTATE

-- Equivalent control: an EXISTS target list does not affect existence.
\set ON_ERROR_STOP on
SELECT y.id FROM motion_repro x CROSS JOIN motion_repro y
WHERE EXISTS (SELECT 1
              FROM motion_repro z WHERE y.id IS NULL LIMIT 1);
DROP TABLE motion_repro;

output:

SET
CREATE TABLE
INSERT 0 1
ANALYZE
SET
                          QUERY PLAN
---------------------------------------------------------------
 Gather Motion 3:1  (slice1; segments: 3)
   ->  Nested Loop Semi Join
         ->  Nested Loop
               ->  Broadcast Motion 3:3  (slice2; segments: 3)
                     ->  Seq Scan on motion_repro x
               ->  Seq Scan on motion_repro y
                     Filter: (id IS NULL)
         ->  Materialize
               ->  Broadcast Motion 3:3  (slice3; segments: 3)
                     ->  Seq Scan on motion_repro z
 Optimizer: Postgres query optimizer
(11 rows)

 id
----
(0 rows)

SET
                                       QUERY PLAN
-----------------------------------------------------------------------------------------
 Gather Motion 3:1  (slice1; segments: 3)
   ->  Nested Loop
         Join Filter: true
         ->  Seq Scan on motion_repro y
               Filter: ((id IS NULL) AND (SubPlan 2))
               SubPlan 2
                 ->  Result
                       ->  Limit
                             ->  Result
                                   One-Time Filter: (y.id IS NULL)
                                   ->  Materialize
                                         ->  Broadcast Motion 3:3  (slice3; segments: 3)
                                               ->  Seq Scan on motion_repro z
         ->  Materialize
               ->  Broadcast Motion 3:3  (slice2; segments: 3)
                     ->  Seq Scan on motion_repro x
 Optimizer: GPORCA
(17 rows)

SQLSTATE=XX000
psql:/home/alena/rel2-fuzz-20261029/triage-20261029/repro.sql:25: ERROR:  XX000: could not find MotionState for slice 4 in executor tree (execUtils.c:2310)  (entry db 127.0.0.1:25460 pid=71013) (execUtils.c:2310)
LOCATION:  getMotionState, execUtils.c:2310
 id
----
(0 rows)

DROP TABLE
Operating System

ANY

Anything else

No response

Are you willing to submit PR?
  • Yes, I am willing to submit a PR!
Code of Conduct
Lingua principale
C
Stelle
1.4k
Fork
265
Merge medio
4g 15m
PR unite (30g)
68

Preparare l'ambiente

Come iniziare

  1. Leggi tutta la issue e poi la guida ai contributi del progetto.
  2. Commenta sulla issue per dire che te ne occupi tu — evita che due persone facciano lo stesso lavoro.
  3. Fai un fork del repository e lavora su un branch.
  4. Apri una pull request che faccia riferimento al numero della issue.

Altre issue di apache/cloudberry

Tutte le issue di apache/cloudberry

Issue simili

Altre issue su C

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.