[ 
https://issues.apache.org/jira/browse/KUDU-1292?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Todd Lipcon resolved KUDU-1292.
-------------------------------
       Resolution: Fixed
         Assignee: Todd Lipcon
    Fix Version/s: 1.1.0

I'm guessing I fixed this way back in 1.1 in 
1eb24183a540f4e3bbbc8a399e440ecf905f6129, considering I committed it a couple 
days after the discussion in this bug, and we haven't seen these issues on the 
flaky dashboard in recent memory.

> CHECK failure in replica due to invalid term advance
> ----------------------------------------------------
>
>                 Key: KUDU-1292
>                 URL: https://issues.apache.org/jira/browse/KUDU-1292
>             Project: Kudu
>          Issue Type: Bug
>          Components: consensus
>    Affects Versions: 0.6.0
>            Reporter: Todd Lipcon
>            Assignee: Todd Lipcon
>            Priority: Major
>             Fix For: 1.1.0
>
>         Attachments: raft_consensus-itest.txt.gz
>
>
> In looping RaftConsensusITest.TestChurnyElections, I hit a CHECK failure:
> {code}
> F0113 08:09:59.060348   161 raft_consensus_state.cc:578] Check failed: 
> _s.ok() Bad status: Corruption: New operation's term is not >= than the 
> previous op's term. Current: 520.837. Previous: 522.836
> *** Check failure stack trace: ***
>     @     0x7fe35c8e768d  google::LogMessage::Fail() at ??:0
>     @     0x7fe35c8e96fc  google::LogMessage::SendToLog() at ??:0
>     @     0x7fe35c8e71e9  google::LogMessage::Flush() at ??:0
>     @     0x7fe35c8ea08f  google::LogMessageFatal::~LogMessageFatal() at ??:0
>     @     0x7fe35f4242aa  
> kudu::consensus::ReplicaState::AdvanceCommittedIndexUnlocked() at ??:0
>     @     0x7fe35f40de6a  kudu::consensus::RaftConsensus::UpdateReplica() at 
> ??:0
>     @     0x7fe35f40d6d0  kudu::consensus::RaftConsensus::Update() at ??:0
>     @     0x7fe36030eb6d  
> kudu::tserver::ConsensusServiceImpl::UpdateConsensus() at ??:0
>     @     0x7fe35f21893a  kudu::consensus::ConsensusServiceIf::Handle() at 
> ??:0
>     @     0x7fe35f0e7336  kudu::rpc::ServicePool::RunThread() at ??:0
>     @     0x7fe35f0e92fe  boost::_mfi::mf0<>::operator()() at ??:0
>     @     0x7fe35f0e926b  boost::_bi::list1<>::operator()<>() at ??:0
>     @     0x7fe35f0e9214  boost::_bi::bind_t<>::operator()() at ??:0
>     @     0x7fe35f0e903a  
> boost::detail::function::void_function_obj_invoker0<>::invoke() at ??:0
>     @     0x7fe35f0a1552  boost::function0<>::operator()() at ??:0
>     @     0x7fe35d00bfea  kudu::Thread::SuperviseThread() at ??:0
>     @           0x42140a  __tsan_thread_start_func at ??:0
>     @     0x7fe35d548182  start_thread at ??:0
>     @     0x7fe35b8ac47d  clone at ??:0
>     @              (nil)  (unknown)
> {code}
> which seems like it might be serious.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to