Depends on what you mean by okay.  Every time the statistics change, I 
have to do a lot of hand checking of results.

If the test cases are needed for release, check them in.  If they are 
post release issues, I'd rather that they were not checked in this close 
to release (that's what I've been trying to say for the last month, 
incidentally), but that means extra work out there to make sure that you 
don't break cases that you've written.  The reason for that preference 
is that we are trying to put out RC3 on 9/15 (last I heard, but plans 
seem to be pretty loose with SPH's illness).  Today is 9/14.  If there 
is a change in the test system and a code change, I can't rerun this 
test set and diff the results, because the control group has changed.  
Changing the test system after QA complete (July 10 for this release) 
breaks manageability for testing GMAT -- we end up testing both GMAT and 
the test system, increasing risk on a fully tested build.

This piece has been hugely mishandled in this release, in part because 
of the accelerated schedule.  The impact has mostly hit out here rather 
than at Goddard for this round (since the focus is the Linux build by 
Sep 30), so that might not be very obvious on the east coast.

I think we ought to be handling this dichotomy (release test cases vs 
other new cases being written) differently.  If the test system 
repository were in Git, for example, it would be much simpler to do this 
type of thing using a branching strategy, once people became used to 
working with branches.

- DJC


On 09/14/2016 07:56 AM, Slojkowski, Steven E. (GSFC-595.0)[OMITRON] wrote:
> I'll take a look at my cases. In the meantime, does this mean it's ok for me 
> to check in a few new test cases?
>
> -----Original Message-----
> From: Darrel J. Conway [mailto:[email protected]]
> Sent: Wednesday, September 14, 2016 10:51 AM
> To: Cooley, D S. (GSFC-5950); Slojkowski, Steven E. (GSFC-595.0)[OMITRON]; 
> [email protected]
> Subject: Test results: 2016-09-14 
> (Catsclaw/RHEL-Linux/GmatConsole-64/M2016a/GCC4.8.5)
>
> Results from last night's run
>

-- 
Darrel J. Conway, Ph.D.      Thinking Systems, Inc.
Senior Scientist and CEO     6441 N Camino Libby
Phone: (623) 298-4530        Tucson, AZ 85718
FAX:   (520) 232-2533        www.thinksysinc.com
Cell:  (520) 425-3626        [email protected]


------------------------------------------------------------------------------
_______________________________________________
Gmat-buildtest mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/gmat-buildtest

Reply via email to