This is an automated email from the ASF dual-hosted git repository.
yiguolei pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/doris.git
The following commit(s) were added to refs/heads/master by this push:
new 40b3be8197d [fix](agg) prevent core dump in ColumnArray::size_at when
handling type mismatch in streaming_agg_serialize_to_column (#50001)
40b3be8197d is described below
commit 40b3be8197da0043e7cfb2808427d2e2d7a12dd1
Author: lw112 <131352377+felixw...@users.noreply.github.com>
AuthorDate: Wed Apr 16 14:49:33 2025 +0800
[fix](agg) prevent core dump in ColumnArray::size_at when handling type
mismatch in streaming_agg_serialize_to_column (#50001)
### What problem does this PR solve?
Problem Summary:
core
```
*** is nereids: 1 ***
*** tablet id: 0 ***
*** Aborted at 1744342233 (unix time) try "date -d @1744342233" if you are
using GNU date ***
*** Current BE git commitID: 2503281 ***
*** SIGSEGV unknown detail explain (@0x0) received by PID 681431 (TID
683727 OR 0x7fcd153a0700) from PID 0; stack trace: ***
0# doris::signal::(anonymous namespace)::FailureSignalHandler(int,
siginfo_t*, void*) in /usr/local/service/doris/lib/be/doris_be
1# os::Linux::chained_handler(int, siginfo*, void*) in
/usr/local/jdk/jre/lib/amd64/server/libjvm.so
2# JVM_handle_linux_signal in /usr/local/jdk/jre/lib/amd64/server/libjvm.so
3# signalHandler(int, siginfo*, void*) in
/usr/local/jdk/jre/lib/amd64/server/libjvm.so
4# 0x00007FD26771E400 in /lib64/libc.so.6
5# doris::vectorized::ColumnArray::insert_from(doris::vectorized::IColumn
const&, unsigned long) in /usr/local/service/doris/lib/be/doris_be
6#
doris::vectorized::ColumnNullable::insert_from(doris::vectorized::IColumn
const&, unsigned long) in /usr/local/service/doris/lib/be/doris_be
7#
doris::vectorized::AggregateFunctionCollect<doris::vectorized::AggregateFunctionArrayAggData<void>,
std::integral_constant<bool, false>, std::integral_constant<bool, true>
>::streaming_agg_serialize_to_column(doris::vectorized::IColumn const**,
COW<doris::vectorized::IColumn>::mutable_ptr<doris::vectorized::IColumn>&,
unsigned long, doris::vectorized::Arena*) const in
/usr/local/service/doris/lib/be/doris_be
8#
doris::vectorized::AggFnEvaluator::streaming_agg_serialize_to_column(doris::vectorized::Block*,
COW<doris::vectorized::IColumn>::mutable_ptr<doris::vectorized::IColumn>&,
unsigned long, doris::vectorized::Arena*) in
/usr/local/service/doris/lib/be/doris_be
9#
_ZNSt8__detail9__variant17__gen_vtable_implINS0_12_Multi_arrayIPFNS0_21__deduce_visit_resultIN5doris6StatusEEEONS4_10vectorized8OverloadIJZNS4_8pipeline22StreamingAggLocalState28_pre_agg_with_serialized_keyEPNS7_5BlockESC_E3$_0ZNSA_28_pre_agg_with_serialized_keyESC_SC_E3$_1EEERSt7variantIJNS7_16MethodSerializedI9PHHashMapINS4_9StringRefEPc11DefaultHashISK_vELb0EEEENS7_15MethodOneNumberIhSJ_IhSL_SM_IhvELb0EEEENSQ_ItSJ_ItSL_SM_ItvELb0EEEENSQ_IjSJ_IjSL_9HashCRC32IjELb0EEEENSQ_ImSJ_Im
[...]
10#
doris::pipeline::StreamingAggLocalState::_pre_agg_with_serialized_key(doris::vectorized::Block*,
doris::vectorized::Block*) in /usr/local/service/doris/lib/be/doris_be
11#
doris::pipeline::StreamingAggLocalState::do_pre_agg(doris::vectorized::Block*,
doris::vectorized::Block*) in /usr/local/service/doris/lib/be/doris_be
12# doris::pipeline::StreamingAggOperatorX::push(doris::RuntimeState*,
doris::vectorized::Block*, bool) const in
/usr/local/service/doris/lib/be/doris_be
13#
doris::pipeline::StatefulOperatorX<doris::pipeline::StreamingAggLocalState>::get_block(doris::RuntimeState*,
doris::vectorized::Block*, bool*) in /usr/local/service/doris/lib/be/doris_be
14#
doris::pipeline::OperatorXBase::get_block_after_projects(doris::RuntimeState*,
doris::vectorized::Block*, bool*) in /usr/local/service/doris/lib/be/doris_be
15# doris::pipeline::PipelineXTask::execute(bool*) in
/usr/local/service/doris/lib/be/doris_be
16# doris::pipeline::TaskScheduler::_do_work(unsigned long) in
/usr/local/service/doris/lib/be/doris_be
17# doris::ThreadPool::dispatch_thread() in
/usr/local/service/doris/lib/be/doris_be
18# doris::Thread::supervise_thread(void*) in
/usr/local/service/doris/lib/be/doris_be
19# start_thread in /lib64/libpthread.so.0
20# __clone in /lib64/libc.so.6
```
---
.../aggregate_function_collect.h | 3 +-
.../data/query_p0/aggregate/array_agg.out | Bin 59884 -> 59963 bytes
.../suites/query_p0/aggregate/array_agg.groovy | 52 ++++++++++++++++-----
3 files changed, 43 insertions(+), 12 deletions(-)
diff --git a/be/src/vec/aggregate_functions/aggregate_function_collect.h
b/be/src/vec/aggregate_functions/aggregate_function_collect.h
index ab60737cd7c..aff008dbf3c 100644
--- a/be/src/vec/aggregate_functions/aggregate_function_collect.h
+++ b/be/src/vec/aggregate_functions/aggregate_function_collect.h
@@ -885,7 +885,8 @@ public:
col_src.get_nested_column());
vec.insert_from(vec_src, i);
} else if constexpr (std::is_same_v<Data,
AggregateFunctionArrayAggData<void>>) {
- to_nested_col.insert_from(col_src.get_nested_column(), i);
+ auto& vec = col_null->get_nested_column();
+ vec.insert_from(col_src.get_nested_column(), i);
} else {
using ColVecType = ColumnVectorOrDecimal<typename
Data::ElementType>;
auto& vec = assert_cast<ColVecType&,
TypeCheckOnRelease::DISABLE>(
diff --git a/regression-test/data/query_p0/aggregate/array_agg.out
b/regression-test/data/query_p0/aggregate/array_agg.out
index f34157154cb..c20e7ce9623 100644
Binary files a/regression-test/data/query_p0/aggregate/array_agg.out and
b/regression-test/data/query_p0/aggregate/array_agg.out differ
diff --git a/regression-test/suites/query_p0/aggregate/array_agg.groovy
b/regression-test/suites/query_p0/aggregate/array_agg.groovy
index 56f40f11920..c78f84394fd 100644
--- a/regression-test/suites/query_p0/aggregate/array_agg.groovy
+++ b/regression-test/suites/query_p0/aggregate/array_agg.groovy
@@ -38,7 +38,7 @@ suite("array_agg") {
"disable_auto_compaction" = "false",
"enable_single_replica_compaction" = "false"
);
- """
+ """
sql """
CREATE TABLE `test_array_agg_int` (
@@ -57,7 +57,7 @@ suite("array_agg") {
"disable_auto_compaction" = "false",
"enable_single_replica_compaction" = "false"
);
- """
+ """
sql """
CREATE TABLE `test_array_agg_decimal` (
@@ -78,7 +78,7 @@ suite("array_agg") {
"disable_auto_compaction" = "false",
"enable_single_replica_compaction" = "false"
);
- """
+ """
sql """
insert into `test_array_agg` values
@@ -108,8 +108,8 @@ suite("array_agg") {
(7, "LC", "V7_3"),
(7, "LC", NULL),
(7, NULL, "V7_3");
- """
-
+ """
+
sql """
insert into `test_array_agg_int` values
(1, "alex",NULL,NULL),
@@ -137,7 +137,7 @@ suite("array_agg") {
(7, "LC", "V7_3",NULL),
(7, "LC", NULL,NULL),
(7, NULL, "V7_3",NULL);
- """
+ """
sql """
insert into `test_array_agg_decimal` values
@@ -158,7 +158,7 @@ suite("array_agg") {
(5, "LC", "V5_3",NULL,NULL,"alexcoco662"),
(7, "", NULL,NULL,NULL,"alexcoco1"),
(8, "", NULL,0,NULL,"alexcoco2");
- """
+ """
order_qt_sql1 """
SELECT count(id), array_agg(`label_name`) FROM `test_array_agg` GROUP BY
`id` order by id;
@@ -197,7 +197,7 @@ suite("array_agg") {
"""
- // test for bucket 10
+ // test for bucket 10
sql """ CREATE TABLE `test_array_agg1` (
`id` int(11) NOT NULL,
`label_name` varchar(32) default null,
@@ -243,7 +243,7 @@ suite("array_agg") {
(7, "LC", NULL),
(7, NULL, "V7_3");
"""
-
+
order_qt_sql11 """
SELECT count(id), size(array_agg(`label_name`)) FROM `test_array_agg`
GROUP BY `id` order by id;
"""
@@ -314,7 +314,7 @@ suite("array_agg") {
k1 BIGINT ,
k4 ipv4 ,
k6 ipv6 ,
- s string
+ s string
) DISTRIBUTED BY HASH(k1) BUCKETS 1 PROPERTIES("replication_num" =
"1");
"""
sql """ insert into test_array_agg_ip values(1,123,34141,"0.0.0.123") ,
(2,3114,318903,"0.0.0.123") ,
(3,7832131,192837891738927931231,"2001:0DB8:AC10:FE01:FEED:BABE:CAFE:F00D"),(4,null,null,"2001:0DB8:AC10:FE01:FEED:BABE:CAFE:F00D");
"""
@@ -323,8 +323,38 @@ suite("array_agg") {
qt_select """select array_sort(array_agg(k4)),array_sort(array_agg(k6))
from test_array_agg_ip """
sql "DROP TABLE `test_array_agg`"
- sql "DROP TABLE `test_array_agg1`"
+ sql "DROP TABLE `test_array_agg1`"
sql "DROP TABLE `test_array_agg_int`"
sql "DROP TABLE `test_array_agg_decimal`"
sql "DROP TABLE `test_array_agg_ip`"
+
+
+ sql """ drop table if exists test_user_tags;"""
+
+ sql """
+ CREATE TABLE test_user_tags (
+ k1 varchar(150) NULL,
+ k2 varchar(150) NULL,
+ k3 varchar(150) NULL,
+ k4 array<varchar(150)> NULL,
+ k5 array<varchar(150)> NULL,
+ k6 datetime NULL
+ ) ENGINE=OLAP
+ UNIQUE KEY(k1, k2, k3)
+ DISTRIBUTED BY HASH(k2) BUCKETS 3
+ PROPERTIES ("replication_allocation" = "tag.location.default: 1");
+ """
+
+ sql """
+ INSERT INTO test_user_tags VALUES
+ ('corp001', 'wx001', 'vip', ['id1', 'id2'], ['tag1', 'tag2'],
'2023-01-01 10:00:00'),
+ ('corp001', 'wx001', 'level', ['id3'], ['tag3'], '2023-01-01
10:00:00'),
+ ('corp002', 'wx002', 'vip', ['id4', 'id5'], ['tag4', 'tag5'],
'2023-01-02 10:00:00');
+ """
+ sql "SET spill_streaming_agg_mem_limit = 1024;"
+ sql "SET enable_spill = true;"
+
+ qt_select """ SELECT k1,array_agg(k5) FROM test_user_tags group by k1
order by k1; """
+
+ sql "UNSET VARIABLE ALL;"
}
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org
For additional commands, e-mail: commits-h...@doris.apache.org
