java NullPointerException with R2017b when closing simulink project

I need help tracking down the source of a java error that appears in R2017b, but not in R2016b.

Closing a Simulink project when a Model is open generates errors like this:

Exception in thread "Simulink Project Thread: 48" java.lang.NullPointerException
    at com.mathworks.toolbox.slproject.extensions.dependency.problems.managers.ProblemManagerBase.updateProblems(ProblemManagerBase.java:130)
    at com.mathworks.toolbox.slproject.extensions.dependency.problems.managers.ProblemManagerBase.access$500(ProblemManagerBase.java:41)
    at com.mathworks.toolbox.slproject.extensions.dependency.problems.managers.ProblemManagerBase$UpdateRunnable.run(ProblemManagerBase.java:242)
    at com.mathworks.toolbox.shared.computils.threads.CoalescingExecutor$1.run(CoalescingExecutor.java:62)
    at com.mathworks.toolbox.shared.computils.threads.WrappingExecutorService$WrappedRunnable.run(WrappingExecutorService.java:74)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
    at java.lang.Thread.run(Thread.java:745)

My project architecture is fairly complex: I have several Simulink projects which all have a reference to a common Simulink project. Besides a block library, the common project modifies my simulink environment by:

• adding custom menus on the model editor window
• selecting a set of model advisor options
• storing a configuration set in a data dictionary, that all models reference

I've tried eliminating components of the referenced project (the error does not occur for projects which don't reference this project). But I can't narrow it down.

[Sharan Paramasivam Murugesan]

1. It appears to be an unknown issue. Without reproduction files, it is hard to identify the root cause of this issue. Have you tried reproducing this with a sample parent project and a referenced project? This should help you narrow down the particular action that is causing this exception.

2. There should be a java log (java.log.xxxx) created when there is a java exception. It would be really useful if you could retrieve the java log file and attach it here. For info on how to retrieve this file, refer to https://www.mathworks.com/matlabcentral/answers/100816-how-do-i-locate-the-crash-dump-files-generated-by-matlab

3. If you are using GitHub for source control of these projects, can you try "https://github.com" as the remote address instead of "[e-mail]"

[John Harris]

Thanks, Sharan!

Unfortunately, I'm striking out on all these suggestions. Yes, I've tried narrowing it down to a relatively simple architecture, but that did not bring me to a more specific scenario. I may try to generate a smaller example that I could send in as a working test case.

There are no java.log.xxxx files on the disk, and I caused the error then used the extractCrashLogs function, but the only extracted logs were matlab_crash_dump logs from days ago.

And lastly, we're not using any form of git - just TFS.

I'll report again if the problem persists in R2018a...Thanks!