[MITgcm-devel] http://mitgcm.org/testing.html

[Martin Losch]

Thanks Ed,

I think I know what the problem was (although it's still only a guess): 
disk space. MITgcm and testreport produce enough junk (object files, 
data files etc) that 2.8GB of disk space is not enough. The job dies 
just before it is complete (last two experiments or so). I guess I'll 
have to find a scratch disk somewhere.
Thanks again,
Martin

On Jun 6, 2005, at 2:40 PM, Ed Hill wrote:

> On Mon, 2005-06-06 at 09:01 +0200, Martin Losch wrote:
>> Hi Ed,
>>
>> I have a con-job that runs testreport on two of our Sun-machines 
>> (which
>> go by the innovative names "model" and "rays1") every Sunday morning 
>> at
>> 2am (8pm on Saturday in Boston). Normally I try to remember to check 
>> on
>> Mondays by inspecting http://mitgcm.org/testing.html whether these 
>> jobs
>> were successful. For some time now, the "rays1"-job doesn't appear on
>> the list on http://mitgcm.org/testing.html anymore, although I cannot
>> see why it should have failed on my end.
>> What is the half-life of the test results on testing.html? If I send a
>> testreport on Sunday, should it appear on the list on Monday 
>> regardless
>> of the length of the list? I'll keep looking for problems at my end 
>> ...
>
> Hi Martin,
>
> Test results should appear on our site within an hour of the email 
> being
> received since the update scripts are run every hour (at :20 past).
> Things that could be going wrong are most likely email delivery
> failures.  I'll look into the logs on our side and see what we've
> received in the past week or so.  And you can test email delivery on
> your side by adding a few extra emails to the testreport list:
>
>   ./testreport ...other options... \
>     -a 'edhill at mitgcm.org other at someplace.com martin at work.edu ...'
>
> and then confirming whether or not they get delivered.
>
> Also, our testing page gets "refreshed" every month and the last 
> testing
> page for the old month goes into a "testing archives" that is linked
> from the main testing page (in the middle):
>
>   http://mitgcm.org/testing.html
>   http://mitgcm.org/testing/results/
>
> The archives show that your Sun machines "rays1" and "model" were 
> indeed
> working through most of April & May.
>
> Ed
>
> -- 
> Edward H. Hill III, PhD
> office:  MIT Dept. of EAPS;  Rm 54-1424;  77 Massachusetts Ave.
>              Cambridge, MA 02139-4307
> emails:  eh3 at mit.edu                ed at eh3.com
> URLs:    http://web.mit.edu/eh3/    http://eh3.com/
> phone:   617-253-0098
> fax:     617-253-4464
>
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel