[Fix](warmup) Fix coredump in CloudTablet::complete_rowset_segment_warmup due to capture by reference (#56395)

bobhan1 · web-flow · commit 1e738f46a302 · 2025-09-25T15:58:48.000+08:00
### What problem does this PR solve? introduced in #54611 ``` (gdb) bt #0 __GI___pthread_sigmask (how=2, newmask=<optimized out>, oldmask=0x0) at ./nptl/pthread_sigmask.c:43 #1 0x00007f7aa6b5d71e in PosixSignals::chained_handler(int, siginfo*, void*) [clone .part.0] () from /usr/lib/jvm/java-17-openjdk-amd64/lib/server/libjvm.so #2 0x00007f7aa6b5e206 in JVM_handle_linux_signal () from /usr/lib/jvm/java-17-openjdk-amd64/lib/server/libjvm.so #3 <signal handler called> #4 0x000055e20050d443 in std::_Hashtable<doris::RowsetId, std::pair<doris::RowsetId const, doris::CloudTablet::RowsetWarmUpInfo>, std::allocator<std::pair<doris::RowsetId const, doris::CloudTablet::RowsetWarmUpInfo> >, std::__detail::_Select1st, std::equal_to<doris::RowsetId>, std::hash<doris::RowsetId>, std::__detail::_Mod_range_hashing, std::__detail::_Default_ranged_hash, std::__detail::_Prime_rehash_policy, std::__detail::_Hashtable_traits<true, false, true> >::_M_find_before_node (this=0x7f79e69acde0, __bkt=4186920012728959759, __k=..., __code=11200260987994981938) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/hashtable.h:2205 #5 std::_Hashtable<doris::RowsetId, std::pair<doris::RowsetId const, doris::CloudTablet::RowsetWarmUpInfo>, std::allocator<std::pair<doris::RowsetId const, doris::CloudTablet::RowsetWarmUpInfo> >, std::__detail::_Select1st, std::equal_to<doris::RowsetId>, std::hash<doris::RowsetId>, std::__detail::_Mod_range_hashing, std::__detail::_Default_ranged_hash, std::__detail::_Prime_rehash_policy, std::__detail::_Hashtable_traits<true, false, true> >::_M_locate (this=this@entry=0x7f79e69acde0, __k=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/hashtable.h:2283 #6 0x000055e200505a8d in std::_Hashtable<doris::RowsetId, std::pair<doris::RowsetId const, doris::CloudTablet::RowsetWarmUpInfo>, std::allocator<std::pair<doris::RowsetId const, doris::CloudTablet::RowsetWarmUpInfo> >, std::__detail::_Select1st, std::equal_to<doris::RowsetId>, std::hash<doris::RowsetId>, std::__detail::_Mod_range_hashing, std::__detail::_Default_ranged_hash, std::__detail::_Prime_rehash_policy, std::__detail::_Hashtable_traits<true, false, true> >::find (this=0x7f79e69acde0, __k=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/hashtable.h:1929 #7 std::unordered_map<doris::RowsetId, doris::CloudTablet::RowsetWarmUpInfo, std::hash<doris::RowsetId>, std::equal_to<doris::RowsetId>, std::allocator<std::pair<doris::RowsetId const, doris::CloudTablet::RowsetWarmUpInfo> > >::contains (this=0x7f79e69acde0, __x=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/unordered_map.h:999 #8 doris::CloudTablet::complete_rowset_segment_warmup (this=<optimized out>, rowset_id=..., status=..., segment_num=1, inverted_idx_num=0) at /home/zcp/repo_center/doris_master/doris/be/src/cloud/cloud_tablet.cpp:1692 #9 0x000055e2005338ba in doris::CloudWarmUpManager::handle_jobs()::$_0::operator()(doris::Status) const (this=0x7f79e8bebad0, st=...) at /home/zcp/repo_center/doris_master/doris/be/src/cloud/cloud_warm_up_manager.cpp:243 #10 std::__invoke_impl<void, doris::CloudWarmUpManager::handle_jobs()::$_0&, doris::Status>(std::__invoke_other, doris::CloudWarmUpManager::handle_jobs()::$_0&, doris::Status&&) (__f=..., __args=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/invoke.h:63 #11 std::__invoke_r<void, doris::CloudWarmUpManager::handle_jobs()::$_0&, doris::Status>(doris::CloudWarmUpManager::handle_jobs()::$_0&, doris::Status&&) (__fn=..., __args=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/invoke.h:113 #12 std::_Function_handler<void (doris::Status), doris::CloudWarmUpManager::handle_jobs()::$_0>::_M_invoke(std::_Any_data const&, doris::Status&&) (__functor=..., __args=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:292 #13 0x000055e200533310 in std::function<void (doris::Status)>::operator()(doris::Status) const (this=0x7f795a3ff1d0, __args=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:593 #14 doris::CloudWarmUpManager::submit_download_tasks(std::filesystem::__cxx11::path, long, std::shared_ptr<doris::io::FileSystem>, long, std::shared_ptr<bthread::CountdownEvent>, bool, std::function<void (doris::Status)>)::$_0::operator()(doris::Status) const ( this=0x7f79f0710b60, st=...) at /home/zcp/repo_center/doris_master/doris/be/src/cloud/cloud_warm_up_manager.cpp:149 #15 std::__invoke_impl<void, doris::CloudWarmUpManager::submit_download_tasks(std::filesystem::__cxx11::path, long, std::shared_ptr<doris::io::FileSystem>, long, std::shared_ptr<bthread::CountdownEvent>, bool, std::function<void (doris::Status)>)::$_0&, doris::Status>(std::__invoke_other, doris::CloudWarmUpManager::submit_download_tasks(std::filesystem::__cxx11::path, long, std::shared_ptr<doris::io::FileSystem>, long, std::shared_ptr<bthread::CountdownEvent>, bool, std::function<void (doris::Status)>)::$_0&, doris::Status&&) ( __f=..., __args=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/invoke.h:63 #16 std::__invoke_r<void, doris::CloudWarmUpManager::submit_download_tasks(std::filesystem::__cxx11::path, long, std::shared_ptr<doris::io::FileSystem>, long, std::shared_ptr<bthread::CountdownEvent>, bool, std::function<void (doris::Status)>)::$_0&, doris::Status>(doris::CloudWarmUpManager::submit_download_tasks(std::filesystem::__cxx11::path, long, std::shared_ptr<doris::io::FileSystem>, long, std::shared_ptr<bthread::CountdownEvent>, bool, std::function<void (doris::Status)>)::$_0&, doris::Status&&) (__fn=..., __args=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/invoke.h:113 #17 std::_Function_handler<void (doris::Status), doris::CloudWarmUpManager::submit_download_tasks(std::filesystem::__cxx11::path, long, std::shared_ptr<doris::io::FileSystem>, long, std::shared_ptr<bthread::CountdownEvent>, bool, std::function<void (doris::Status)>)::$_0>::_M_invoke(std::_Any_data const&, doris::Status&&) (__functor=..., __args=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:292 #18 0x000055e200469cb1 in std::function<void (doris::Status)>::operator()(doris::Status) const (this=0x7f796941cee0, __args=...) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:593 #19 doris::io::FileCacheBlockDownloader::download_segment_file (this=<optimized out>, meta=...) at /home/zcp/repo_center/doris_master/doris/be/src/io/cache/block_file_cache_downloader.cpp:297 #20 0x000055e1fbfcbe85 in doris::ThreadPool::dispatch_thread (this=0x7f7995a1da00) at /home/zcp/repo_center/doris_master/doris/be/src/util/threadpool.cpp:614 #21 0x000055e1fbfc0fac in std::function<void ()>::operator()() const (this=0x7f796941cee0) at /usr/local/ldb-toolchain-v0.26/bin/../lib/gcc/x86_64-pc-linux-gnu/15/include/g++-v15/bits/std_function.h:593 #22 doris::Thread::supervise_thread (arg=0x7f79f13de010) at /home/zcp/repo_center/doris_master/doris/be/src/util/thread.cpp:460 #23 0x00007f7aa5bf8ac3 in start_thread (arg=<optimized out>) at ./nptl/pthread_create.c:442 #24 0x00007f7aa5c8a850 in __closefrom_fallback (from=1685031872, dirfd_fallback=<optimized out>) at ../sysdeps/unix/sysv/linux/closefrom_fallback.c:45 #25 0x0000000000000000 in ?? () ``` Problem Summary: ### Release note None ### Check List (For Author) - Test  - [ ] 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  - Behavior changed: - [ ] No. - [ ] Yes.  - Does this need documentation? - [ ] No. - [ ] Yes.  ### Check List (For Reviewer who merge this PR) - [ ] Confirm the release note - [ ] Confirm test cases - [ ] Confirm document - [ ] Add branch pick label
diff --git a/be/src/cloud/cloud_warm_up_manager.cpp b/be/src/cloud/cloud_warm_up_manager.cpp
@@ -154,7 +154,7 @@ void CloudWarmUpManager::submit_download_tasks(io::Path path, int64_t file_size,
                         .is_dryrun = config::enable_reader_dryrun_when_download_file_cache,
                         .is_warmup = true},
                 .download_done =
-                        [&](Status st) {
+                        [&, done_cb = std::move(done_cb)](Status st) {
                             if (done_cb) done_cb(st);
                             if (!st) {
                                 LOG_WARNING("Warm up error ").error(st);
