[Go-essp-tech] [esg-node-dev] Re: CMIP5 Datanode : data access services.

Mon May 2 16:04:28 MDT 2011

 Hi,

In the context of this thread, that version is fine.

On 5/2/11 1:49 PM, Sébastien Denvil wrote:
> Hi,
>
> very useful link Roland. Reading it I was not able to figure out if
> the bug comes from the client or the server?
>
> This wiki points netcdf-4.1.2-beta2. The 4.1.2 has been released:
> http://www.unidata.ucar.edu/software/netcdf/release-notes-4.1.2.html
> Do you know if the released library contains the C API Security
> Extension for Earth System Grid?
>
> On the data node side : our production data node is using thredds
> Version 4.2.20100908.2246. The 4.2.5 thredds version released
> 2011-03-02 includes opendap ESG access control :
> ftp://ftp.unidata.ucar.edu/pub/thredds/4.2/CHANGES.txt
> Do you know if this is  a required version for the data node.
>
> Gavin : what is the installer version you suggest us to use.
> v1.0.3-14-g485f969 or a later one?
>
> Thanks.
> Sébastien
>
> On 02/05/2011 17:51, Roland Schweitzer wrote:
>> Hi,
>>
>> Here are some detailed notes on client access.  The netCDF library
>> target that contains the security infrastructure has still not been
>> officially release and in our most recent testing still contains some
>> bugs.
>>
>> I'll update this page after we get our builds finalized with the
>> netCDF library release.
>>
>> http://esgf.org/wiki/ClientAccessToESGFOPeNDAPServers
>>
>> (I had  to login into the Wiki to see it.  Maybe Gavin can link to
>> this page and make it more public.  I don't know how to manage the
>> access in this Wiki.)
>>
>> Roland
>>
>> On 05/02/2011 10:14 AM, Estanislao Gonzalez wrote:
>>> Hi Sébastien,
>>>
>>> there is a requirement about the NetCDF library version that work
>>> with ESGF. I don't recall it now (Phil, which version was it?), but
>>> it means that not every nco can connect to the data node. The
>>> OpenDAP connector must provide the required certificate and follow
>>> redirects for it to work.
>>>
>>> Thanks,
>>> Estani
>>>
>>> Am 02.05.2011 16:09, schrieb Sébastien Denvil:
>>>>  Thanks Luca.
>>>>
>>>>
>>>> On 02/05/2011 15:38, Cinquini, Luca (3880) wrote:
>>>>> Hi Sebastian,
>>>>>     the ESGF security infrastructure can be used for opendap
>>>>> access - infact, this is already working at JPL. Once opendap
>>>>> access is configured with security, users can use either a browser
>>>>> or an esg-enabled client to access datasets.
>>>>
>>>> Great. At least with opendap they can access from the browser a
>>>> region if they know array indices that cover it. If they have nco
>>>> operator with opendap client enable they should in theory be able
>>>> to extract their area using physical lat/lon. Not everyone has made
>>>> this nco compilation.
>>>>
>>>> The good thing with netCDF subsets is that it works with longitude
>>>> and latitude from the browser.... Lot of browser addict out there.
>>>>
>>>>> I will try to write some documentation on the esgf.org wiki either
>>>>> this evening or tomorrow, and then let you try it out...
>>>>>
>>>>> thanks for raising the issue, I meant to start an email
>>>>> conversation on this but never quite got to it...
>>>>
>>>> That's already big fun to setup model / run them / produce &
>>>> distribute those data. We tried hardly in the past and was
>>>> discourage by the difficulty and the fact that we did not know how
>>>> to find out precisely what was supposed to work and what was not
>>>> (given the day we install the stack). In that case it's not easy to
>>>> investigate and to find a way to contribute without being in the
>>>> development main stream.
>>>>
>>>> Glad to hear it should be smooth install now. We will read your
>>>> notes and try it out with confidence.
>>>>
>>>> Cheers.
>>>> Sébastien
>>>>
>>>> ps : any further comments about gridFTP or netcdf subsets? If you
>>>> say it should work we will try and report, but we need this green
>>>> light to try.
>>>>
>>>>> Luca
>>>>>
>>>>> On May 2, 2011, at 7:32 AM, Sébastien Denvil wrote:
>>>>>
>>>>>>   Dear all,
>>>>>>
>>>>>> we received increasing demand asking for regional model boundary
>>>>>> conditions from CMIP5 simulations. Those datasets are heavy and
>>>>>> therefore "wget" is not my favourite answer. I would like this
>>>>>> answer to
>>>>>> be base on a ESG-F solution and not a home made one.
>>>>>>
>>>>>> In order to build a meaningful response I would like to know what
>>>>>> are
>>>>>> the plan regarding the datanode to have :
>>>>>> - openDAP enable with CMIP5 filters.
>>>>>> - netCDF subsets enable with CMIP5 filters.
>>>>>>      * openDAP and/or netCDF subsets will be very useful for those
>>>>>> focusing on a specific region (Africa OR Asia for example). They
>>>>>> could
>>>>>> then extract only the area of interest.
>>>>>>
>>>>>> - gridFTP enable with CMIP5 filters.
>>>>>>      *gridFTP will be very useful for those focusing on many regions
>>>>>> and/or doing the CORDEX exercise but without a limited area
>>>>>> model. It's
>>>>>> very likely that they will download the whole datasets.
>>>>>>
>>>>>> My understanding is that gridFTP is not fully functional at this
>>>>>> point
>>>>>> of time and that openDAP and netCDF subsets are not yet secure by
>>>>>> the
>>>>>> ESG-F security layer.
>>>>>>
>>>>>> A corollary coming from those questions : Is there a
>>>>>> documentation we
>>>>>> could refer to if we want to update a production data node?
>>>>>>
>>>>>> Please refine my understanding.
>>>>>>
>>>>>> Thanks.
>>>>>> Sébastien
>>>>>>
>>>>>> -- 
>>>>>> Sébastien Denvil
>>>>>> IPSL, Pôle de modélisation du climat
>>>>>> UPMC, Case 101, 4 place Jussieu,
>>>>>> 75252 Paris Cedex 5
>>>>>>
>>>>>> Tour 45-55 2ème étage Bureau 209
>>>>>> Tel: 33 1 44 27 21 10
>>>>>> Fax: 33 1 44 27 39 02
>>>>>>
>>>>>>
>>>>>> <smime.p7s>_______________________________________________
>>>>>> GO-ESSP-TECH mailing list
>>>>>> GO-ESSP-TECH at ucar.edu
>>>>>> http://mailman.ucar.edu/mailman/listinfo/go-essp-tech
>>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> GO-ESSP-TECH mailing list
>>>> GO-ESSP-TECH at ucar.edu
>>>> http://mailman.ucar.edu/mailman/listinfo/go-essp-tech
>>>
>>>
>>> -- 
>>> Estanislao Gonzalez
>>>
>>> Max-Planck-Institut für Meteorologie (MPI-M)
>>> Deutsches Klimarechenzentrum (DKRZ) - German Climate Computing Centre
>>> Room 108 - Bundesstrasse 45a, D-20146 Hamburg, Germany
>>>
>>> Phone:   +49 (40) 46 00 94-126
>>> E-Mail:  gonzalez at dkrz.de 
>>>
>>>
>>> _______________________________________________
>>> GO-ESSP-TECH mailing list
>>> GO-ESSP-TECH at ucar.edu
>>> http://mailman.ucar.edu/mailman/listinfo/go-essp-tech
>>
>>
>> _______________________________________________
>> GO-ESSP-TECH mailing list
>> GO-ESSP-TECH at ucar.edu
>> http://mailman.ucar.edu/mailman/listinfo/go-essp-tech
>
>
> -- 
> Sébastien Denvil
> IPSL, Pôle de modélisation du climat
> UPMC, Case 101, 4 place Jussieu,
> 75252 Paris Cedex 5
>
> Tour 45-55 2ème étage Bureau 209
> Tel: 33 1 44 27 21 10
> Fax: 33 1 44 27 39 02

-- 
Gavin M. Bell
Lawrence Livermore National Labs
--

 "Never mistake a clear view for a short distance."
       	       -Paul Saffo

(GPG Key - http://rainbow.llnl.gov/dist/keys/gavin.asc)

 A796 CE39 9C31 68A4 52A7  1F6B 66B7 B250 21D5 6D3E

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/go-essp-tech/attachments/20110502/ba1b7113/attachment.html 