frontend stack overflow when calling to_stream_prost on a complex query
#15916
Assignees
Milestone
Comments
9 tasks
|
reproduce CREATE TABLE supplier (s_suppkey INT, s_name CHARACTER VARYING, s_address CHARACTER VARYING, s_nationkey INT, s_phone CHARACTER VARYING, s_acctbal NUMERIC, s_comment CHARACTER VARYING, PRIMARY KEY (s_suppkey));
CREATE TABLE part (p_partkey INT, p_name CHARACTER VARYING, p_mfgr CHARACTER VARYING, p_brand CHARACTER VARYING, p_type CHARACTER VARYING, p_size INT, p_container CHARACTER VARYING, p_retailprice NUMERIC, p_comment CHARACTER VARYING, PRIMARY KEY (p_partkey));
CREATE TABLE partsupp (ps_partkey INT, ps_suppkey INT, ps_availqty INT, ps_supplycost NUMERIC, ps_comment CHARACTER VARYING, PRIMARY KEY (ps_partkey, ps_suppkey));
CREATE TABLE customer (c_custkey INT, c_name CHARACTER VARYING, c_address CHARACTER VARYING, c_nationkey INT, c_phone CHARACTER VARYING, c_acctbal NUMERIC, c_mktsegment CHARACTER VARYING, c_comment CHARACTER VARYING, PRIMARY KEY (c_custkey));
CREATE TABLE orders (o_orderkey BIGINT, o_custkey INT, o_orderstatus CHARACTER VARYING, o_totalprice NUMERIC, o_orderdate DATE, o_orderpriority CHARACTER VARYING, o_clerk CHARACTER VARYING, o_shippriority INT, o_comment CHARACTER VARYING, PRIMARY KEY (o_orderkey));
CREATE TABLE lineitem (l_orderkey BIGINT, l_partkey INT, l_suppkey INT, l_linenumber INT, l_quantity NUMERIC, l_extendedprice NUMERIC, l_discount NUMERIC, l_tax NUMERIC, l_returnflag CHARACTER VARYING, l_linestatus CHARACTER VARYING, l_shipdate DATE, l_commitdate DATE, l_receiptdate DATE, l_shipinstruct CHARACTER VARYING, l_shipmode CHARACTER VARYING, l_comment CHARACTER VARYING, PRIMARY KEY (l_orderkey, l_linenumber));
CREATE TABLE nation (n_nationkey INT, n_name CHARACTER VARYING, n_regionkey INT, n_comment CHARACTER VARYING, PRIMARY KEY (n_nationkey));
CREATE TABLE region (r_regionkey INT, r_name CHARACTER VARYING, r_comment CHARACTER VARYING, PRIMARY KEY (r_regionkey));
CREATE TABLE person (id BIGINT, name CHARACTER VARYING, email_address CHARACTER VARYING, credit_card CHARACTER VARYING, city CHARACTER VARYING, state CHARACTER VARYING, date_time TIMESTAMP, extra CHARACTER VARYING, PRIMARY KEY (id));
CREATE TABLE auction (id BIGINT, item_name CHARACTER VARYING, description CHARACTER VARYING, initial_bid BIGINT, reserve BIGINT, date_time TIMESTAMP, expires TIMESTAMP, seller BIGINT, category BIGINT, extra CHARACTER VARYING, PRIMARY KEY (id));
CREATE TABLE bid (auction BIGINT, bidder BIGINT, price BIGINT, channel CHARACTER VARYING, url CHARACTER VARYING, date_time TIMESTAMP, extra CHARACTER VARYING);
CREATE TABLE alltypes1 (c1 BOOLEAN, c2 SMALLINT, c3 INT, c4 BIGINT, c5 REAL, c6 DOUBLE, c7 NUMERIC, c8 DATE, c9 CHARACTER VARYING, c10 TIME, c11 TIMESTAMP, c13 INTERVAL, c14 STRUCT<a INT>, c15 INT[], c16 CHARACTER VARYING[]);
CREATE TABLE alltypes2 (c1 BOOLEAN, c2 SMALLINT, c3 INT, c4 BIGINT, c5 REAL, c6 DOUBLE, c7 NUMERIC, c8 DATE, c9 CHARACTER VARYING, c10 TIME, c11 TIMESTAMP, c13 INTERVAL, c14 STRUCT<a INT>, c15 INT[], c16 CHARACTER VARYING[]);
CREATE MATERIALIZED VIEW m6 AS SELECT t_5.r_comment AS col_0, ((SMALLINT '1') * (((FLOAT '0') + (FLOAT '113')) * (INTERVAL '-604800'))) AS col_1, t_5.r_comment AS col_2 FROM region AS t_5 WHERE false GROUP BY t_5.r_comment;
CREATE MATERIALIZED VIEW stream_query AS
SELECT
tumble_10.c6 AS col_0,
tumble_10.c11 AS col_1,
(TIMESTAMP '2024-03-26 08:35:02') AS col_2
FROM
(
(
SELECT
t_0.c6 AS col_0,
((SMALLINT '232')) AS col_1,
t_1.col_2 AS col_2,
t_0.c6 AS col_3
FROM
(
alltypes2 AS t_0
RIGHT JOIN m6 AS t_1 ON t_0.c9 = t_1.col_0
)
LEFT JOIN (
SELECT
sq_6.col_2 AS col_0,
sq_6.col_2 AS col_1,
(INT '818') AS col_2
FROM
(
SELECT
TIME '08:33:59' AS col_0,
((FLOAT '852')) AS col_1,
TIMESTAMP '2024-03-26 08:34:13' AS col_2,
(INT '973') AS col_3
FROM
(
SELECT
tumble_2.state AS col_0,
(
coalesce(
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
NULL,
(SMALLINT '-32768'),
NULL,
NULL
)
) AS col_1,
tumble_2.state AS col_2,
tumble_2.state AS col_3
FROM
tumble(person, person.date_time, INTERVAL '5') AS tumble_2 FULL
JOIN tumble(alltypes1, alltypes1.c11, INTERVAL '17') AS tumble_3 ON tumble_2.credit_card = tumble_3.c9
WHERE
tumble_3.c1
GROUP BY
tumble_2.state
HAVING
true
) AS sq_4
LEFT JOIN bid AS t_5 ON sq_4.col_0 = t_5.url
) AS sq_6 FULL
JOIN nation AS t_7 ON sq_6.col_3 = t_7.n_nationkey
AND true
WHERE
false
GROUP BY
sq_6.col_2
HAVING
false
) AS sq_8 ON t_0.c11 = sq_8.col_1
AND true
WHERE
t_0.c1
GROUP BY
t_0.c6,
t_1.col_2
HAVING
true
) AS sq_9 FULL
JOIN tumble(alltypes1, alltypes1.c11, INTERVAL '18') AS tumble_10 ON sq_9.col_2 = tumble_10.c9
AND true
)
LEFT JOIN tumble(bid, bid.date_time, INTERVAL '3') AS tumble_11 ON tumble_10.c4 = tumble_11.price
AND tumble_10.c1
WHERE
false
GROUP BY
tumble_10.c6,
tumble_10.c11; |
|
The |
|
It's not the fault of fragmentor. The stack overflows during the recursive calls of risingwave/src/frontend/src/optimizer/plan_node/mod.rs Lines 722 to 726 in 2afbb6f It is also not a bug of infinite recursion. When I increase the stack size to 1GB (with env var Some possible improvements:
|
BugenZhao
changed the title
to_stream_prost on a complex query
4 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
https://buildkite.com/risingwavelabs/pull-request/builds/45532#018e79dc-9596-4f36-9aed-c57fabef65ce
It seems to be caused by
SET RW_FORCE_TWO_PHASE_AGG TO TRUE CREATE MATERIALIZED VIEW stream_query AS SELECT tumble_10.c6 AS col_0, tumble_10.c11 AS col_1, (TIMESTAMP '2024-03-26 08:35:02') AS col_2 FROM ((SELECT t_0.c6 AS col_0, ((SMALLINT '232')) AS col_1, t_1.col_2 AS col_2, t_0.c6 AS col_3 FROM (alltypes2 AS t_0 RIGHT JOIN m6 AS t_1 ON t_0.c9 = t_1.col_0) LEFT JOIN (SELECT sq_6.col_2 AS col_0, sq_6.col_2 AS col_1, (INT '818') AS col_2 FROM (SELECT TIME '08:33:59' AS col_0, ((FLOAT '852')) AS col_1, TIMESTAMP '2024-03-26 08:34:13' AS col_2, (INT '973') AS col_3 FROM (SELECT tumble_2.state AS col_0, (coalesce(NULL, NULL, NULL, NULL, NULL, NULL, NULL, (SMALLINT '-32768'), NULL, NULL)) AS col_1, tumble_2.state AS col_2, tumble_2.state AS col_3 FROM tumble(person, person.date_time, INTERVAL '5') AS tumble_2 FULL JOIN tumble(alltypes1, alltypes1.c11, INTERVAL '17') AS tumble_3 ON tumble_2.credit_card = tumble_3.c9 WHERE tumble_3.c1 GROUP BY tumble_2.state HAVING true) AS sq_4 LEFT JOIN bid AS t_5 ON sq_4.col_0 = t_5.url) AS sq_6 FULL JOIN nation AS t_7 ON sq_6.col_3 = t_7.n_nationkey AND true WHERE false GROUP BY sq_6.col_2 HAVING false) AS sq_8 ON t_0.c11 = sq_8.col_1 AND true WHERE t_0.c1 GROUP BY t_0.c6, t_1.col_2 HAVING true) AS sq_9 FULL JOIN tumble(alltypes1, alltypes1.c11, INTERVAL '18') AS tumble_10 ON sq_9.col_2 = tumble_10.c9 AND true) LEFT JOIN tumble(bid, bid.date_time, INTERVAL '3') AS tumble_11 ON tumble_10.c4 = tumble_11.price AND tumble_10.c1 WHERE false GROUP BY tumble_10.c6, tumble_10.c11The text was updated successfully, but these errors were encountered: