[ 
https://issues.apache.org/jira/browse/KUDU-2727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17142356#comment-17142356
 ] 

ASF subversion and git services commented on KUDU-2727:
-------------------------------------------------------

Commit da1c66b61c4924a4726aaeb171e79cc0b03ca65c in kudu's branch 
refs/heads/master from Alexey Serbin
[ https://gitbox.apache.org/repos/asf?p=kudu.git;h=da1c66b ]

[consensus] KUDU-2727 lock-free CheckLeadershipAndBindTerm()

This addresses the following lock contention issue outlined in KUDU-2727:

  
tids=[93293,93284,93285,93286,93287,93288,93289,93290,93291,93292,93304,93294,93295,93296,93297,93298,93299,93300,93301,93302,93303,93313,93322,93321,93320,93319,93318,93317,93316,93315,93314,93283,93312,93311,93310,93309,93308,93307,93306,93305]
      0x7f61b79fc5e0 <unknown>
           0x1ec35f4 base::internal::SpinLockDelay()
           0x1ec347c base::SpinLock::SlowLock()
            0xb7deb8 
kudu::consensus::RaftConsensus::CheckLeadershipAndBindTerm()
            0xaab010 kudu::tablet::TransactionDriver::ExecuteAsync()
            0xaa344c kudu::tablet::TabletReplica::SubmitWrite()
            0x928fb0 kudu::tserver::TabletServiceImpl::Write()
           0x1d2e8d9 kudu::rpc::GeneratedServiceIf::Handle()
           0x1d2efd9 kudu::rpc::ServicePool::RunThread()
           0x1ea4a84 kudu::Thread::SuperviseThread()
      0x7f61b79f4e25 start_thread
      0x7f61b5cd234d __clone
  tids=[93383]
      0x7f61b79fc5e0 <unknown>
      0x7f61b79f8cf2 __pthread_cond_timedwait
           0x1dfcfa9 kudu::ConditionVariable::WaitUntil()
            0xb73bc7 kudu::consensus::RaftConsensus::UpdateReplica()
            0xb75128 kudu::consensus::RaftConsensus::Update()
            0x92c5d1 kudu::tserver::ConsensusServiceImpl::UpdateConsensus()
           0x1d2e8d9 kudu::rpc::GeneratedServiceIf::Handle()
           0x1d2efd9 kudu::rpc::ServicePool::RunThread()
           0x1ea4a84 kudu::Thread::SuperviseThread()
      0x7f61b79f4e25 start_thread
      0x7f61b5cd234d __clone

In addition, with this patch RaftConsensus::DumpStatusHtml() no longer
blocks Raft consensus activity and isn't blocked by it either.

The inspiration for this update came from Yugabyte-DB:
  
https://github.com/yugabyte/yugabyte-db/commit/7b3271e4a0f34ff3c3f6e8961288db9cee335461

Change-Id: I934ae3035d893fd850afe27d96f8dd6612c9ffbd
Reviewed-on: http://gerrit.cloudera.org:8080/16034
Tested-by: Alexey Serbin <aser...@cloudera.com>
Reviewed-by: Andrew Wong <aw...@cloudera.com>


> Contention on the Raft consensus lock can cause tablet service queue overflows
> ------------------------------------------------------------------------------
>
>                 Key: KUDU-2727
>                 URL: https://issues.apache.org/jira/browse/KUDU-2727
>             Project: Kudu
>          Issue Type: Improvement
>          Components: consensus, tserver
>            Reporter: William Berkeley
>            Assignee: Alexey Serbin
>            Priority: Major
>              Labels: performance, scalability
>
> Here's stacks illustrating the phenomenon:
> {noformat}
>   tids=[2201]
>         0x379ba0f710 <unknown>
>            0x1fb951a base::internal::SpinLockDelay()
>            0x1fb93b7 base::SpinLock::SlowLock()
>             0xb4e68e kudu::consensus::Peer::SignalRequest()
>             0xb9c0df kudu::consensus::PeerManager::SignalRequest()
>             0xb8c178 kudu::consensus::RaftConsensus::Replicate()
>             0xaab816 kudu::tablet::TransactionDriver::Prepare()
>             0xaac0ed kudu::tablet::TransactionDriver::PrepareTask()
>            0x1fa37ed kudu::ThreadPool::DispatchThread()
>            0x1f9c2a1 kudu::Thread::SuperviseThread()
>         0x379ba079d1 start_thread
>         0x379b6e88fd clone
>   tids=[4515]
>         0x379ba0f710 <unknown>
>            0x1fb951a base::internal::SpinLockDelay()
>            0x1fb93b7 base::SpinLock::SlowLock()
>             0xb74c60 kudu::consensus::RaftConsensus::NotifyCommitIndex()
>             0xb59307 kudu::consensus::PeerMessageQueue::NotifyObserversTask()
>             0xb54058 
> _ZN4kudu8internal7InvokerILi2ENS0_9BindStateINS0_15RunnableAdapterIMNS_9consensus16PeerMessageQueueEFvRKSt8functionIFvPNS4_24PeerMessageQueueObserverEEEEEEFvPS5_SC_EFvNS0_17UnretainedWrapperIS5_EEZNS5_34NotifyObserversOfCommitIndexChangeElEUlS8_E_EEESH_E3RunEPNS0_13BindStateBaseE
>            0x1fa37ed kudu::ThreadPool::DispatchThread()
>            0x1f9c2a1 kudu::Thread::SuperviseThread()
>         0x379ba079d1 start_thread
>         0x379b6e88fd clone
>   tids=[22185,22194,22193,22188,22187,22186]
>         0x379ba0f710 <unknown>
>            0x1fb951a base::internal::SpinLockDelay()
>            0x1fb93b7 base::SpinLock::SlowLock()
>             0xb8bff8 
> kudu::consensus::RaftConsensus::CheckLeadershipAndBindTerm()
>             0xaaaef9 kudu::tablet::TransactionDriver::ExecuteAsync()
>             0xaa3742 kudu::tablet::TabletReplica::SubmitWrite()
>             0x92812d kudu::tserver::TabletServiceImpl::Write()
>            0x1e28f3c kudu::rpc::GeneratedServiceIf::Handle()
>            0x1e2986a kudu::rpc::ServicePool::RunThread()
>            0x1f9c2a1 kudu::Thread::SuperviseThread()
>         0x379ba079d1 start_thread
>         0x379b6e88fd clone
>   tids=[22192,22191]
>         0x379ba0f710 <unknown>
>            0x1fb951a base::internal::SpinLockDelay()
>            0x1fb93b7 base::SpinLock::SlowLock()
>            0x1e13dec kudu::rpc::ResultTracker::TrackRpc()
>            0x1e28ef5 kudu::rpc::GeneratedServiceIf::Handle()
>            0x1e2986a kudu::rpc::ServicePool::RunThread()
>            0x1f9c2a1 kudu::Thread::SuperviseThread()
>         0x379ba079d1 start_thread
>         0x379b6e88fd clone
>   tids=[4426]
>         0x379ba0f710 <unknown>
>            0x206d3d0 <unknown>
>            0x212fd25 google::protobuf::Message::SpaceUsedLong()
>            0x211dee4 
> google::protobuf::internal::GeneratedMessageReflection::SpaceUsedLong()
>             0xb6658e kudu::consensus::LogCache::AppendOperations()
>             0xb5c539 kudu::consensus::PeerMessageQueue::AppendOperations()
>             0xb5c7c7 kudu::consensus::PeerMessageQueue::AppendOperation()
>             0xb7c675 
> kudu::consensus::RaftConsensus::AppendNewRoundToQueueUnlocked()
>             0xb8c147 kudu::consensus::RaftConsensus::Replicate()
>             0xaab816 kudu::tablet::TransactionDriver::Prepare()
>             0xaac0ed kudu::tablet::TransactionDriver::PrepareTask()
>            0x1fa37ed kudu::ThreadPool::DispatchThread()
>            0x1f9c2a1 kudu::Thread::SuperviseThread()
>         0x379ba079d1 start_thread
>         0x379b6e88fd clone
> {noformat}
> {{kudu::consensus::RaftConsensus::CheckLeadershipAndBindTerm()}} needs to 
> take the lock to check the term and the Raft role. When many RPCs come in for 
> the same tablet, the contention can hog service threads and cause queue 
> overflows on busy systems.
> Yugabyte switched their equivalent lock to be an atomic that allows them to 
> read the term and role wait-free.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to