Tests run, pass or fail, and nothing left in Processing Queue, yet Ncrunch Test window fails to update with correct Symbol. All points to that NCrunch is finished with it's work, but I still see the darned Timer symbol:
And this happens irrespective how I run tests. Both when NCrunch runs because of impacted tests, and both when I right click that test group and Run Tests.
Thanks for sending through the bug report. Was NCrunch enabled when you sent this? The log file was blank, which suggests the engine was either completely crashed out or otherwise disabled.
If you're receiving OutOfMemoryExceptions in the NCrunch engine process itself, then this would likely be because of one of the following:
1. NCrunch is set to run using an x86 process host and the resource demands by the engine have exceeded the 32-bit memory limit.
2. NCrunch is set to run its engine inside the IDE and the IDE is overloaded (happens easily under VS2017 with other packages installed).
3. Your machine is out of physical memory and swap space.
I restarted Visual Studio, and put the "Use CPU Architecture" from "Autodectect" to "x64".
It seems that on Restart of Visual Studio, the icon issue disappeared. I think it's related to the habit of mine to let the machine just idle on lock at all times.
I'm not sure if the most recent bug report with this issue was yours, but in the likely event that it is, it contained a useful stack trace:
System.Exception: Unknown DataRow parameter type: System.RuntimeType
at nCrunch.Module.MSTest.Integration.MSTestResolvedTest.convertStringToType(String stringValue, String typeName, Type parameterType)
at nCrunch.Module.MSTest.Integration.MSTestResolvedTest.DecodeAndResolve(ReflectedAssembly assemblyContainingTest, String encodedTest)
at nCrunch.Module.MSTest.Integration.MSTestTest..ctor(MSTestNames names, ReflectedAssembly assembly, ReflectedType fixtureType, PendingTest pendingTest, MSTestInterceptingContext testContext, Boolean considerInconclusiveTestsAsPassing)
at nCrunch.Module.MSTest.Integration.MSTestFixture.<>c__DisplayClass2_0.<.ctor>b__0(PendingTest t)
at System.Linq.Enumerable.WhereSelectArrayIterator`2.MoveNext()
at System.Collections.Generic.List`1..ctor(IEnumerable`1 collection)
at nCrunch.Module.MSTest.Integration.MSTestFixture..ctor(MSTestNames names, ReflectedAssembly assembly, TestName fixtureTest, IList`1 tests, MSTestInterceptingContext testContext, Boolean considerInconclusiveTestsAsPassing)
at nCrunch.Module.MSTest.Integration.MSTestRunner.runTestsWithinAssembly(IList`1 testsToRun, TestTaskOutput testOutput)
at nCrunch.Common.PerformanceTracking.PerfTracker.TrackUnreliableActivity(String name, Action activity)
at nCrunch.Common.PerformanceTracking.PerfTracker.TryTrackUnreliableActivity(String name, Action activity)
at nCrunch.Module.MSTest.Integration.MSTestRunner.RunTests(IList`1 testsToRun, TestTaskOutput testOutput)
at nCrunch.TestExecution.TestExecutionMonitor.PerformMonitoredTestExecution(Action testExecutionAction)
at nCrunch.TestExecution.TestRunnerThread.()
This is a hole in NCrunch's MSTest DataRow support. Somewhere in your test project you're making use of a DataRow with an unusual parameter type that NCrunch can't seem to handle. If you can get me the signature for the test that's causing it (unfortunately the log doesn't tell me which one), I should be able to investigate further.
I'm not sure if it's possible to get this kind of thing working under NCrunch, since it would involve the test's meta structure consisting of a user type (a bad thing when it comes to test uniqueness/identification). But the error can certainly be improved.
I submitted another bug report. One of those cases where I write multiple tests after each other, and at some point NCrunch just fails to properly update the status of some of the tests.
The test clearly passes, at least in Debug mode, but it has Dark green breadcrumb, Green timer symbol, and state 'Running (Passed last run)'
Would you be interested in trying the build below? I've attempted to implement a fix for this problem. Note that you'll also need to set your 'Framework Utilisation type for MSTest' setting to 'DynamicAnalysis'.
Kaerwek wrote:
Did you do you attempt to do something about that other issue?
From what I could establish in the reports you submitted, these looked to be tied to the same thing (exceptions thrown during test discovery/execution). Are you seeing any problems with the new build?
It sure seems much more stable now. It seems to refresh the statuses correctly, haven't checked it very deeply, but earlier every now and then it seemed to report things wrong.
Kaerwek wrote:It sure seems much more stable now. It seems to refresh the statuses correctly, haven't checked it very deeply, but earlier every now and then it seemed to report things wrong.
Please do let me know if you manage to pin down any problems. As far as I know, all results reporting in this build should be stable now for your use case.
