[DocDB] Crash Observed in TransactionParticipant::Impl::Abort

[shamanthchandra-yb]

Jira Link: DB-14948

Description

Version: 2.25.1.0-b203

Packed toggle off/on stress testcase failed because of:

* thread #1, name = 'yb-tserver', stop reason = signal SIGABRT
  * frame #0: 0x00007f4ecfd66acf libc.so.6`raise + 271
    frame #1: 0x00007f4ecfd39ea5 libc.so.6`abort + 295
    frame #2: 0x00005606b0891403 yb-server`abort_message + 195
    frame #3: 0x00005606b0890f9c yb-server`demangling_terminate_handler() + 268
    frame #4: 0x00005606b0890c66 yb-server`std::__terminate(void (*)()) + 6
    frame #5: 0x00005606b0892bab yb-server`__cxxabiv1::failed_throw(__cxxabiv1::__cxa_exception*) + 27
    frame #6: 0x00005606b0892b3f yb-server`__cxa_throw + 111
    frame #7: 0x00005606ae47cd1e yb-server`std::__1::__throw_bad_weak_ptr[abi:ue170006]() at shared_ptr.h:137:5
    frame #8: 0x00005606af954382 yb-server`yb::tablet::TransactionParticipant::Impl::Abort(yb::StronglyTypedUuid<yb::TransactionId_Tag> const&, std::__1::function<void (yb::Result<yb::TransactionStatusResult>)>) [inlined] std::__1::shared_ptr<yb::tablet::RunningTransaction>::shared_ptr[abi:ue170006]<yb::tablet::RunningTransaction, void>(this=<unavailable>, __r=std::__1::weak_ptr<yb::tablet::RunningTransaction>::element_type @ 0x0000162400000001) at shared_ptr.h:704:13
    frame #9: 0x00005606af954334 yb-server`yb::tablet::TransactionParticipant::Impl::Abort(yb::StronglyTypedUuid<yb::TransactionId_Tag> const&, std::__1::function<void (yb::Result<yb::TransactionStatusResult>)>) [inlined] std::__1::enable_shared_from_this<yb::tablet::RunningTransaction>::shared_from_this[abi:ue170006](this=0x00001624cd5da018) at shared_ptr.h:1954:17
    frame #10: 0x00005606af954334 yb-server`yb::tablet::TransactionParticipant::Impl::Abort(yb::StronglyTypedUuid<yb::TransactionId_Tag> const&, std::__1::function<void (yb::Result<yb::TransactionStatusResult>)>) [inlined] yb::tablet::RunningTransaction::Abort(this=0x00001624cd5da018, client=0x00001624fd7d7f10, callback=yb::TransactionStatusCallback @ 0x00007f4babc39700, lock=0x00007f4babc396e0)>, std::__1::unique_lock<std::__1::mutex>*) at running_transaction.cc:200:34
    frame #11: 0x00005606af953ccf yb-server`yb::tablet::TransactionParticipant::Impl::Abort(this=<unavailable>, id=<unavailable>, callback=<unavailable>)>) at transaction_participant.cc:707:45
    frame #12: 0x00005606af95d7d4 yb-server`yb::tablet::TransactionParticipant::StopActiveTxnsPriorTo(yb::HybridTime, std::__1::chrono::time_point<yb::CoarseMonoClock, std::__1::chrono::duration<long long, std::__1::ratio<1l, 1000000000l>>>, yb::StronglyTypedUuid<yb::TransactionId_Tag>*) at transaction_participant.cc:1355:7
    frame #13: 0x00005606af95d3e5 yb-server`yb::tablet::TransactionParticipant::StopActiveTxnsPriorTo(this=<unavailable>, cutoff=<unavailable>, deadline=yb::CoarseTimePoint @ 0x00007f4babc398b8, exclude_txn_id=<unavailable>) at transaction_participant.cc:2700:17
    frame #14: 0x00005606afc07a35 yb-server`yb::tserver::TabletServiceAdminImpl::AlterSchema(this=0x00001624fb36c020, req=0x00001624fc146320, resp=0x00001624fc1463d0, context=<unavailable>) at tablet_service.cc:1022:65
    frame #15: 0x00005606afd264b2 yb-server`std::__1::__function::__func<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3, std::__1::allocator<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3>, void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator()(std::__1::shared_ptr<yb::rpc::InboundCall>&&) [inlined] yb::tserver::TabletServerAdminServiceIf::InitMethods(this=<unavailable>, req=<unavailable>, resp=<unavailable>, rpc_context=RpcContext @ 0x00007f4babc39c80)::$_3::operator()(std::__1::shared_ptr<yb::rpc::InboundCall>) const::'lambda'(yb::tablet::ChangeMetadataRequestPB const*, yb::tserver::ChangeMetadataResponsePB*, yb::rpc::RpcContext)::operator()(yb::tablet::ChangeMetadataRequestPB const*, yb::tserver::ChangeMetadataResponsePB*, yb::rpc::RpcContext) const at tserver_admin.service.cc:473:9
    frame #16: 0x00005606afd2647a yb-server`std::__1::__function::__func<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3, std::__1::allocator<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3>, void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator()(std::__1::shared_ptr<yb::rpc::InboundCall>&&) at local_call.h:126:7
    frame #17: 0x00005606afd26054 yb-server`std::__1::__function::__func<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3, std::__1::allocator<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3>, void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator()(std::__1::shared_ptr<yb::rpc::InboundCall>&&) [inlined] yb::tserver::TabletServerAdminServiceIf::InitMethods(this=<unavailable>, call=<unavailable>)::$_3::operator()(std::__1::shared_ptr<yb::rpc::InboundCall>) const at tserver_admin.service.cc:471:7
    frame #18: 0x00005606afd25fc5 yb-server`std::__1::__function::__func<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3, std::__1::allocator<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3>, void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator()(std::__1::shared_ptr<yb::rpc::InboundCall>&&) [inlined] decltype(__f=<unavailable>, __args=<unavailable>)::$_3&>()(std::declval<std::__1::shared_ptr<yb::rpc::InboundCall>>())) std::__1::__invoke[abi:ue170006]<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3&, std::__1::shared_ptr<yb::rpc::InboundCall>>(yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3&, std::__1::shared_ptr<yb::rpc::InboundCall>&&) at invoke.h:340:25
    frame #19: 0x00005606afd25fa4 yb-server`std::__1::__function::__func<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3, std::__1::allocator<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3>, void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator()(std::__1::shared_ptr<yb::rpc::InboundCall>&&) [inlined] void std::__1::__invoke_void_return_wrapper<void, true>::__call[abi:ue170006]<yb::tserver::TabletServerAdminServiceIf::InitMethods(__args=<unavailable>, __args=<unavailable>)::$_3&, std::__1::shared_ptr<yb::rpc::InboundCall>>(yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3&, std::__1::shared_ptr<yb::rpc::InboundCall>&&) at invoke.h:415:5
    frame #20: 0x00005606afd25fa4 yb-server`std::__1::__function::__func<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3, std::__1::allocator<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3>, void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator()(std::__1::shared_ptr<yb::rpc::InboundCall>&&) [inlined] std::__1::__function::__alloc_func<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3, std::__1::allocator<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3>, void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator(this=<unavailable>, __arg=<unavailable>)[abi:ue170006](std::__1::shared_ptr<yb::rpc::InboundCall>&&) at function.h:192:16
    frame #21: 0x00005606afd25fa4 yb-server`std::__1::__function::__func<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3, std::__1::allocator<yb::tserver::TabletServerAdminServiceIf::InitMethods(scoped_refptr<yb::MetricEntity> const&)::$_3>, void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator(this=<unavailable>, __arg=<unavailable>)(std::__1::shared_ptr<yb::rpc::InboundCall>&&) at function.h:363:12
    frame #22: 0x00005606afd28f6f yb-server`yb::tserver::TabletServerAdminServiceIf::Handle(std::__1::shared_ptr<yb::rpc::InboundCall>) [inlined] std::__1::__function::__value_func<void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator(this=<unavailable>, __args=nullptr)[abi:ue170006](std::__1::shared_ptr<yb::rpc::InboundCall>&&) const at function.h:517:16
    frame #23: 0x00005606afd28f50 yb-server`yb::tserver::TabletServerAdminServiceIf::Handle(std::__1::shared_ptr<yb::rpc::InboundCall>) [inlined] std::__1::function<void (std::__1::shared_ptr<yb::rpc::InboundCall>)>::operator(this=<unavailable>, __arg=nullptr)(std::__1::shared_ptr<yb::rpc::InboundCall>) const at function.h:1168:12
    frame #24: 0x00005606afd28f50 yb-server`yb::tserver::TabletServerAdminServiceIf::Handle(this=<unavailable>, call=<unavailable>) at tserver_admin.service.cc:411:3
    frame #25: 0x00005606af79e140 yb-server`yb::rpc::ServicePoolImpl::Handle(this=0x00001624fb924240, incoming=<unavailable>) at service_pool.cc:269:19
    frame #26: 0x00005606af6b7e7f yb-server`yb::rpc::InboundCall::InboundCallTask::Run(this=<unavailable>) at inbound_call.cc:317:13
    frame #27: 0x00005606af7ad9b3 yb-server`yb::rpc::(anonymous namespace)::Worker::Execute(this=0x000016255f97d340) at thread_pool.cc:115:15
    frame #28: 0x00005606b0094193 yb-server`yb::Thread::SuperviseThread(void*) [inlined] std::__1::__function::__value_func<void ()>::operator(this=0x00001625bc4ed6e0)[abi:ue170006]() const at function.h:517:16
    frame #29: 0x00005606b009417d yb-server`yb::Thread::SuperviseThread(void*) [inlined] std::__1::function<void ()>::operator(this=0x00001625bc4ed6e0)() const at function.h:1168:12
    frame #30: 0x00005606b009417d yb-server`yb::Thread::SuperviseThread(arg=0x00001625bc4ed680) at thread.cc:895:3
    frame #31: 0x00007f4ecfb001ca libpthread.so.0`start_thread + 234
    frame #32: 0x00007f4ecfd51e73 libc.so.6`__clone + 67

Issue Type

kind/bug

Warning: Please confirm that this issue does not contain any sensitive information

• I confirm this issue does not contain any sensitive information.

[rthallamko3]

Per @basavaraj29 , looks like the root cause is the below

void RunningTransaction::Abort(client::YBClient* client,
                               TransactionStatusCallback callback,
                               std::unique_lock<std::mutex>* lock) {
  ...
  lock->unlock();    // we release the lock, the underlying object could be destroyed, as we don't hold a shared ref.
  ...
  context_.rpcs_.RegisterAndStart(
      client::AbortTransaction(
          ..., 
          [status_tablet, self = shared_from_this(), weak_context = context_.RetainWeak()]( // trying to access shared ref
              const Status& status, const tserver::AbortTransactionResponsePB& response) {
            auto context_lock = weak_context.lock();
            if (!context_lock) {
              return;
            }
            self->AbortReceived(status_tablet, status, response);
          }),
      &abort_handle_);
}

This should instead have been

auto shared_ref = shared_from_this();
lock->unlock();
// can safely use shared_ref below.