Just upgraded local grid node server / vstudio2013 to NCRunch NCrunch_VS2013_2.13.0.1 along with 1 external grid node server (other 3 grid node server remain on 2.10).
Running tests went very slow (the 3 grid nodes not upgraded should already have the "spin up time" done since I did not upgrade those) but I chose to restart all grid nodes via vstudio, then my local grid node service crashed with below stacktrace and had to be started manually. (InvalidOperationException).
Code:
> System.ServiceProcess.dll!System.ServiceProcess.ServiceController.GetServiceHandle(int desiredAccess) Unknown
System.ServiceProcess.dll!System.ServiceProcess.ServiceController.Start(string[] args) Unknown
nCrunch.GridNode.Recycler.exe!nCrunch.GridNode.Recycler.Program.Main(string[] args) Unknown
2 other grid nodes (one upgraded, one not) crashed as well according to event viewer (had to start them manually).
Can easily repro it by setting engine mode to run all tests, when nodes are processing I switch back to "Run all tests manually" but before engines settles I restart every grid node one by one.
The Azure nodes never crashes though (I miss a version info column in the "Server Information" for the grid nodes).