Re: [platform-releng-dev] What's wrong the Mac tests?

[David M Williams <david_williams@xxxxxxxxxx>]

I have a theory, now, at least. Related
to -XstartOnFirstThread. As I say in the bug, not sure if this was something
I/we recently broke, or if just the change to the newer machine/settings
cause it to show up now where it did not before? 

https://bugs.eclipse.org/bugs/show_bug.cgi?id=392130

I think its worth a re-spin of the I-build,
though, admittedly, 10% experimental. (But, even if doesn't fix immediate
problem, its still better coding, IMHO). 

(I will likely kill the N-build mac
tests to make sure the respin Mac tests finish by morning)

Advice/education welcome, 





From:      
 David M Williams/Raleigh/IBM@IBMUS
To:      
 "Eclipse platform
release engineering list." <platform-releng-dev@xxxxxxxxxxx>,

Date:      
 10/16/2012 07:23 PM
Subject:    
   [platform-releng-dev]
What's wrong the Mac tests?
Sent by:    
   platform-releng-dev-bounces@xxxxxxxxxxx

---

I do not know. 

The handful of "DNFs" seemed to have started after bug 391737
was fixed (though, it was not actually fixed).  
https://bugs.eclipse.org/bugs/show_bug.cgi?id=391737

But I can find no obvious related error messages in our logs. They do not
appear to be the typical "time out" errors that usually cause
DNFs, but something else causing Java to exit abnormally. 

Hence, if owners of the tests see anything unusual that I need to fix,
let me know. Otherwise, I may just try random things like restarting the
slave (having webmasters reboot) or similar and try running the I-build
tests again. 

Thanks, 
_______________________________________________
platform-releng-dev mailing list
platform-releng-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/platform-releng-dev