We faced some issues on our clustered production environment like query latency, slow down and much more. We are unable to find out any reason behind that. We checked the monitoring system and it looks good to us. However, we have found some strange unknown error information in ML log which is given here:
2016-11-10 00:00:10.089 Error: Semaphore::post: SVC-SEMPOST: Semaphore post error: ReleaseSemaphore: The handle is invalid.
2016-11-10 00:00:10.089 Error:
2016-11-10 00:00:10.089 Error: svc::Semaphore::post()
2016-11-10 00:00:10.089 Error: xdmp::TransactionSection::commit()
2016-11-10 00:00:10.089 Error: xdmp::TransactionSection::end()
2016-11-10 00:00:10.089 Error: xdmp::UsageMeterTask::run()
2016-11-10 00:00:10.089 Error: svc::PooledThread::run()
2016-11-10 00:00:10.089 Error: svc::Thread::top()
2016-11-10 00:00:10.089 Error: runThread()
2016-11-10 00:00:10.089 Error: _callthreadstartex()
2016-11-10 00:00:10.089 Error: _threadstartex()
2016-11-10 00:00:10.089 Error: BaseThreadInitThunk()
2016-11-10 00:00:10.089 Error: RtlUserThreadStart()
2016-11-10 00:05:44.425 Error: XDQPServerSession::run: SVC-SOCSHUT: Socket shutdown error: shutdown : Socket operation on nonsocket
2016-11-10 00:06:00.509 Error: XDQPServerSession::run: SVC-SOCSHUT: Socket shutdown error: shutdown : Socket operation on nonsocket
We restarted the server and resolved it. Now, it is working fine. But, we are still unable to identify these errors. If you have any idea about these errors, please share with us so that we can take any necessary action to avoid it in future.
