none
Unhandled exception when closing the HPC Cluster Manager

    Question

  • Hi all,

    Every time I close the HPC Cluster Manager I get the following error message:

    Unhandled exception - A task was cancelled

    I have had a look in the log files, and this appears to be the related log entry:

    08/02/2018    09:59:23.493    e    HpcTrace    26604    24996    Current Application Domain UnhandledException event invoked: System.Threading.Tasks.TaskCanceledException: A task was canceled...   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)..   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)..   at System.Windows.Threading.DispatcherOperation.Wait(TimeSpan timeout)..   at System.Windows.Threading.Dispatcher.InvokeImpl(DispatcherOperation operation, CancellationToken cancellationToken, TimeSpan timeout)..   at System.Windows.Threading.Dispatcher.Invoke(Action callback, DispatcherPriority priority, CancellationToken cancellationToken, TimeSpan timeout)..   at MS.Internal.WeakEventTable.OnShutDown()..   at MS.Internal.ShutDownListener.HandleShutDown(Object sender, EventArgs e) 

    None of my tasks have been cancelled, and I cannot find any other signs that anything has 'stopped'. I am running HPC Pack 2016 (5.1.6086.0) on Windows Server 2016 Standard.

    If anyone can offer me a solution to this that'd be great, many thanks!

    Thursday, 2 August 2018 10:29 AM

Answers

  • Hi, this is known issue that's caused by a .net related security update that's related to WPF. And this won't impact any of the HPC Cluster service or jobs. 

    We already have a fix for this issue but will only be shipped with other fixes in our next QFE (For HPC Pack 2016 Update 1 and HPC Pack 2012 R2 Update 3)

    Thus at this moment, you could just safely ignore this error dialog


    Qiufang Shi

    Friday, 3 August 2018 3:13 AM

All replies

  • Hi, this is known issue that's caused by a .net related security update that's related to WPF. And this won't impact any of the HPC Cluster service or jobs. 

    We already have a fix for this issue but will only be shipped with other fixes in our next QFE (For HPC Pack 2016 Update 1 and HPC Pack 2012 R2 Update 3)

    Thus at this moment, you could just safely ignore this error dialog


    Qiufang Shi

    Friday, 3 August 2018 3:13 AM
  • I'm getting this behavior as well - only I've identified the issue as repeatable using the Job Manager, and looking at the monitoring graphs.  After clicking Retry, the app closes and reopens, and displays the graphs with no issue.  If I click on anything else, then come back to the graph, the error reappears, with the same behavior.

    I'm running HPC Pack 2016 U2 on Win16.


    Mike Molenda

    Friday, 8 February 2019 3:44 PM
  • Bump - any help would be appreciated.  Does anyone else have this issue?  It's not cluster-breaking, but it is annoying.

    Mike Molenda


    Tuesday, 12 February 2019 5:40 PM
  • Hi Mike,

    The issue is a known issue in HPC Pack 2016 U2 and has been fixed in the QFE KB4481650 (version 5.2.6291) . Please upgrade and check it.

    Regards,

    Yutong Sun

    Wednesday, 13 February 2019 1:41 AM