[fix](pipeline) Crashing caused by repeated spill operations#56755
Merged
yiguolei merged 1 commit intoapache:masterfrom Oct 16, 2025
Merged
[fix](pipeline) Crashing caused by repeated spill operations#56755yiguolei merged 1 commit intoapache:masterfrom
yiguolei merged 1 commit intoapache:masterfrom
Conversation
Contributor
|
Thank you for your contribution to Apache Doris. Please clearly describe your PR:
|
Member
Author
|
run buildall |
TPC-DS: Total hot run time: 190430 ms |
ClickBench: Total hot run time: 30.01 s |
BE UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
mrhhsg
force-pushed
the
fix_spill_core
branch
from
7444984 to
bfdd1be
Compare
Gabriel39
reviewed
Oct 10, 2025
be/src/pipeline/pipeline_task.h
Outdated
| _blocked_dep = dependency; | ||
| { | ||
| std::unique_lock<std::mutex> lc(_dependency_lock); | ||
| _blocked_dep = dependency; |
Contributor
There was a problem hiding this comment.
If you want operations on _blocked_dep are atomic, you'd better to use a atomic variable instead of a lock guarded
mrhhsg
force-pushed
the
fix_spill_core
branch
from
bfdd1be to
b14db1a
Compare
Member
Author
|
run buildall |
TPC-DS: Total hot run time: 190522 ms |
ClickBench: Total hot run time: 30.35 s |
mrhhsg
force-pushed
the
fix_spill_core
branch
from
b14db1a to
69ad960
Compare
Member
Author
|
run buildall |
TPC-DS: Total hot run time: 190412 ms |
ClickBench: Total hot run time: 30.52 s |
Contributor
BE UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
mrhhsg
force-pushed
the
fix_spill_core
branch
from
69ad960 to
2b796d5
Compare
Member
Author
|
run buildall |
mrhhsg
force-pushed
the
fix_spill_core
branch
from
2b796d5 to
0726fd4
Compare
Member
Author
|
run buildall |
TPC-DS: Total hot run time: 190493 ms |
ClickBench: Total hot run time: 29.73 s |
Contributor
BE UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
yiguolei
reviewed
Oct 14, 2025
be/src/pipeline/dependency.h
Outdated
| if (it == _blocked_task.end()) { | ||
| return; | ||
| } | ||
| _blocked_task.erase(it); |
Contributor
There was a problem hiding this comment.
直接在这里调用pipeline task的api，去更新pipeline task的block dep吧，不要再外面改了。
因为我们设置pipeline task的block dep 也是在这个_task_lock 里完成的。
Status blocked(Dependency* dependency, std::unique_lockstd::mutex& /* dep_lock */) {
DCHECK_EQ(_blocked_dep, nullptr) << "task: " << debug_string();
_blocked_dep = dependency;
return _state_transition(PipelineTask::State::BLOCKED);
}
把这个代码改改，这样都在一个lock 里完成比较保险
yiguolei
reviewed
Oct 14, 2025
be/src/pipeline/pipeline_task.cpp
Outdated
| _need_to_revoke_memory = true; | ||
| _spill_context = spill_context; | ||
|
|
||
| auto* dep = _blocked_dep; |
Contributor
There was a problem hiding this comment.
这个逻辑放在这里可能不对。
即使838 行，我们remove 了当前的dep，但是可能下一个时刻一个正在running 的task 还是会更新一个dep
这个逻辑，得在 pipeline task的execute 方法里执行才行。
mrhhsg
force-pushed
the
fix_spill_core
branch
from
0726fd4 to
7b9e98e
Compare
Member
Author
|
run buildall |
github-actions
bot
removed
the
approved
mrhhsg
force-pushed
the
fix_spill_core
branch
from
7b9e98e to
6a7e007
Compare
Member
Author
|
run buildall |
ClickBench: Total hot run time: 31 s |
mrhhsg
force-pushed
the
fix_spill_core
branch
from
6a7e007 to
e12705e
Compare
mrhhsg
force-pushed
the
fix_spill_core
branch
from
e12705e to
d662433
Compare
Member
Author
|
run buildall |
ClickBench: Total hot run time: 30.38 s |
BE UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
1 similar comment
Contributor
BE Regression && UT Coverage ReportIncrement line coverage Increment coverage report
|
yiguolei
approved these changes
Oct 16, 2025
Contributor
|
PR approved by at least one committer and no changes requested. |
github-actions
bot
added
the
approved
Gabriel39
approved these changes
Oct 16, 2025
github-actions bot
pushed a commit
that referenced
this pull request
Oct 16, 2025
### What problem does this PR solve? ```text #0 __GI___pthread_sigmask (how=2, newmask=<optimized out>, oldmask=0x0) at ./nptl/pthread_sigmask.c:43 #1 0x00007f91fd6c471e in PosixSignals::chained_handler(int, siginfo*, void*) [clone .part.0] () from /usr/lib/jvm/java-17-openjdk-amd64/lib/server/libjvm.so #2 0x00007f91fd6c5206 in JVM_handle_linux_signal () from /usr/lib/jvm/java-17-openjdk-amd64/lib/server/libjvm.so #3 <signal handler called> #4 __gnu_cxx::__exchange_and_add (__mem=0xe, __val=-1) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/ext/atomicity.h:68 #5 __gnu_cxx::__exchange_and_add_dispatch (__mem=0xe, __val=-1) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/ext/atomicity.h:103 #6 std::_Sp_counted_base<(__gnu_cxx::_Lock_policy)2>::_M_weak_release (this=0x2) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/shared_ptr_base.h:211 #7 std::__weak_count<(__gnu_cxx::_Lock_policy)2>::~__weak_count (this=<optimized out>) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/shared_ptr_base.h:1168 #8 std::__weak_ptr<doris::QueryContext, (__gnu_cxx::_Lock_policy)2>::~__weak_ptr (this=<optimized out>) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/shared_ptr_base.h:2003 #9 doris::QueryTaskController::revoke_memory()::$_1::~$_1() (this=<optimized out>) at /root/doris/be/src/runtime/workload_management/query_task_controller.cpp:168 #10 std::_Function_base::_Base_manager<doris::QueryTaskController::revoke_memory()::$_1>::_M_destroy(std::_Any_data&, std::integral_constant<bool, false>) (__victim=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:177 #11 std::_Function_base::_Base_manager<doris::QueryTaskController::revoke_memory()::$_1>::_M_manager(std::_Any_data&, std::_Any_data const&, std::_Manager_operation) (__dest=..., __op=std::__destroy_functor, __source=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:205 #12 std::_Function_handler<void (doris::pipeline::SpillContext*), doris::QueryTaskController::revoke_memory()::$_1>::_M_manager(std::_Any_data&, std::_Any_data const&, std::_Manager_operation) (__dest=..., __source=..., __op=std::__destroy_functor) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:284 #13 0x000055f37fba60c5 in std::_Function_base::~_Function_base (this=0x7f8dbb0c9fe0) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:246 #14 doris::pipeline::SpillContext::~SpillContext (this=0x7f91536f0150) at /root/doris/be/src/pipeline/exec/spill_utils.h:57 #15 0x000055f384259e4b in std::_Sp_counted_base<(__gnu_cxx::_Lock_policy)2>::_M_release (this=0x7f91536f0140) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/shared_ptr_base.h:345 #16 std::__shared_count<(__gnu_cxx::_Lock_policy)2>::operator= (this=0x7f8d61302518, __r=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/shared_ptr_base.h:1088 #17 std::__shared_ptr<doris::pipeline::SpillContext, (__gnu_cxx::_Lock_policy)2>::operator= (this=0x7f8d61302510) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/shared_ptr_base.h:1530 #18 std::shared_ptr<doris::pipeline::SpillContext>::operator= (this=0x7f8d61302510) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/shared_ptr.h:413 #19 doris::pipeline::PipelineTask::revoke_memory (this=0x7f8d61302490, spill_context=...) at /root/doris/be/src/pipeline/pipeline_task.cpp:812 #20 0x000055f37fba3c2d in doris::QueryTaskController::revoke_memory (this=<optimized out>) at /root/doris/be/src/runtime/workload_management/query_task_controller.cpp:185 #21 0x000055f37fb99d08 in doris::WorkloadGroupMgr::handle_single_query_ (this=<optimized out>, requestor=..., size_to_reserve=size_to_reserve@entry=1024000, time_in_queue=time_in_queue@entry=27, paused_reason=...) at /root/doris/be/src/runtime/workload_group/workload_group_manager.cpp:820 #22 0x000055f37fb981d0 in doris::WorkloadGroupMgr::handle_paused_queries (this=0x7f9141d8a800) at /root/doris/be/src/runtime/workload_group/workload_group_manager.cpp:381 #23 0x000055f37eb97297 in doris::Daemon::memory_maintenance_thread (this=0x7ffe63cad730) at /root/doris/be/src/common/daemon.cpp:354 #24 0x000055f37fd812fc in std::function<void ()>::operator()() const (this=0x7f8dbb0c9fe0) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:593 #25 doris::Thread::supervise_thread (arg=0x7f914ba25e10) at /root/doris/be/src/util/thread.cpp:460 #26 0x00007f91fc75fac3 in start_thread (arg=<optimized out>) at ./nptl/pthread_create.c:442 #27 0x00007f91fc7f1850 in __closefrom_fallback (from=1674236160, dirfd_fallback=<optimized out>) at ../sysdeps/unix/sysv/linux/closefrom_fallback.c:45 #28 0x0000000000000000 in ?? () ``` Related PR: #xxx Problem Summary: ### Release note None ### Check List (For Author) - Test <!-- At least one of them must be included. --> - [ ] Regression test - [ ] Unit Test - [ ] Manual test (add detailed scripts or steps below) - [ ] No need to test or manual test. Explain why: - [ ] This is a refactor/code format and no logic has been changed. - [ ] Previous test can cover this change. - [ ] No code files have been changed. - [ ] Other reason <!-- Add your reason? --> - Behavior changed: - [ ] No. - [ ] Yes. <!-- Explain the behavior change --> - Does this need documentation? - [ ] No. - [ ] Yes. <!-- Add document PR link here. eg: apache/doris-website#1214 --> ### Check List (For Reviewer who merge this PR) - [ ] Confirm the release note - [ ] Confirm test cases - [ ] Confirm document - [ ] Add branch pick label <!-- Add branch pick label that this PR should merge into -->
Closed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What problem does this PR solve?
Related PR: #xxx
Problem Summary:
Release note
None
Check List (For Author)
Test
Behavior changed:
Does this need documentation?
Check List (For Reviewer who merge this PR)