[yt-dev] [enzo-dev] testing: current gold standard on Amazon

[Michael Kuhlen]

Hi guys [yt-dev cc'd]

Thanks for looking into this. I've determined that the errors occur because
of a change in yt. Changeset 6d71f5517955 changes how the final simulation
time/redshift is set, and (I guess) this results in a mismatch between the
gold standard on Amazon and the locally generated final output. It's
unclear to me whether the yt change should be reverted, or if the Enzo gold
standard needs to be regenerated. Probably the latter.

Mike


On Wed, Jan 9, 2013 at 4:14 AM, Nathan Goldbaum <nathan12343 at gmail.com>wrote:

> Hey Mike and Cameron,
>
> This is confusing.  I just ran the quick suite on my desktop in Santa Cruz
> using the latest changeset (a3149b6944ef).  Enzo was compiled using gcc on
> OS X 10.6.  I see no failures due to missing answers and all tests
> sucessfully run (although a number of tests fail, I guess since I was
> running with compiler optimizations).  The test_results.txt file for my
> test is pasted here: http://pastebin.com/n4nZvEt1
>
> I invoked the test runner using the following command: ./test_runner.py
> --suite=quick -o ~/Desktop/enzo_tests
>
> It looks like we're running the same number of tests - the only difference
> is that you were unable to find the answer tests on the amazon cloud.
>
> Have you tried rerunning the test suite?  My concern is that amazon may
> have been experiencing a hiccup when you tried to run the tests, preventing
> you from downloading some of the test results files.  I guess that
> Cameron's replication of your test failures means that amazon may not be
> the culprit.
>
> Cheers,
>
> Nathan
>
>
> On 1/9/13 11:58 AM, Cameron Hummels wrote:
>
>> Hi Mike,
>>
>> I'm not sure what is going on here.  Normally, YTNoOldAnswer arises when
>> your simulation runs to a different output than the one which generated the
>> gold standard.  Unfortunately, when I look at AdiabaticExpansion with a new
>> and an old run, they both seem to go to DD0016 (consistent with your runs).
>>  Unlike, Nathan, I *do* get the errors that you list, when I pull and run
>> on a3149b6944ef on OS X using gcc.  Will continue to talk to Nathan on this
>> (we're at the same conference) and we'll hopefully have something on this
>> soon.
>>
>> Cameron
>>
>> On Tue, Jan 8, 2013 at 5:19 PM, Michael Kuhlen <mqk at astro.berkeley.edu<mailto:
>> mqk at astro.berkeley.edu**>> wrote:
>>
>>     Hi all
>>
>>     Is something wrong with the current Enzo gold standard on Amazon?
>>     I'm getting YTNoOldAnswer errors for AdiabaticExpansion,
>>     ExtremeAdvectionTest, MHDZeldovichPancake, and ZeldovichPancake.
>>
>>     $ cat version.txt
>>     Enzo: 301413545861 (week-of-code)
>>     yt: 1ff31ebd14c4
>>
>>     test_runner.log (stdout and stderr): http://pastebin.com/GtYzEQbT
>>
>>     test_results.txt: http://pastebin.com/DbD6Ke77
>>
>>
>>     Mike
>>
>>     --     **************************************************************
>> ***********
>>     *                       *
>>     *  Dr. Michael Kuhlen              Theoretical Astrophysics Center  *
>>     *  email: mqk at astro.berkeley.edu <mailto:mqk at astro.berkeley.edu**>
>>     UC Berkeley                      *
>>     *  cell phone: (831) 588-1468 <tel:%28831%29%20588-1468>
>> B-116 Hearst Field Annex # 3411  *
>>
>>     *  skype username: mikekuhlen      Berkeley, CA 94720               *
>>     *                       *
>>     ****************************************************************
>> *********
>>     --     You received this message because you are subscribed to the
>> Google
>>     Groups "enzo-dev" group.
>>     To post to this group, send email to enzo-dev at googlegroups.com
>>     <mailto:enzo-dev at googlegroups.**com <enzo-dev at googlegroups.com>>.
>>
>>     To unsubscribe from this group, send email to
>>     enzo-dev+unsubscribe@**googlegroups.com<enzo-dev%2Bunsubscribe at googlegroups.com>
>>     <mailto:enzo-dev%**2Bunsubscribe at googlegroups.com<enzo-dev%252Bunsubscribe at googlegroups.com>
>> **>.
>>
>>     For more options, visit this group at
>>     http://groups.google.com/**group/enzo-dev?hl=en<http://groups.google.com/group/enzo-dev?hl=en>
>> .
>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "enzo-dev" group.
>> To post to this group, send email to enzo-dev at googlegroups.com.
>> To unsubscribe from this group, send email to enzo-dev+unsubscribe@**
>> googlegroups.com <enzo-dev%2Bunsubscribe at googlegroups.com>.
>> For more options, visit this group at http://groups.google.com/**
>> group/enzo-dev?hl=en <http://groups.google.com/group/enzo-dev?hl=en>.
>>
>
> --
> You received this message because you are subscribed to the Google Groups
> "enzo-dev" group.
> To post to this group, send email to enzo-dev at googlegroups.com.
> To unsubscribe from this group, send email to enzo-dev+unsubscribe@**
> googlegroups.com <enzo-dev%2Bunsubscribe at googlegroups.com>.
> For more options, visit this group at http://groups.google.com/**
> group/enzo-dev?hl=en <http://groups.google.com/group/enzo-dev?hl=en>.
>
>


-- 
*********************************************************************
*                                                                   *
*  Dr. Michael Kuhlen              Theoretical Astrophysics Center  *
*  email: mqk at astro.berkeley.edu   UC Berkeley                      *
*  cell phone: (831) 588-1468      B-116 Hearst Field Annex # 3411  *
*  skype username: mikekuhlen      Berkeley, CA 94720               *
*                                                                   *
*********************************************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.spacepope.org/pipermail/yt-dev-spacepope.org/attachments/20130109/a94dd4c1/attachment.htm>