Actually, you don't actually have to redo the source data mappings or metadata if you create a Darwin Core Archive from the original data in the test instance of the IPT and then load that into a new resource in the production instance of the IPT afterwards. On Sun, Jan 23, 2011 at 2:36 PM, Tim Robertson (GBIF) <trobertson@gbif.org> wrote:

Dear David

Thanks for your questions.  You'll find answers inline:

...

On Fri, Jan 7, 2011 at 10:48 PM, Tim Robertson <trobertson@gbif.org> wrote:

...

As we are approaching the 2.0 General Availability release, RC3 allows users to configure the IPT to register against the GBIF live registry and register and publish data to the GBIF network.  On a new installation, after the data directory is selected, you may choose “production use”, which will ensure the IPT communicates with the LIVE GBIF registry.  Once selected, this cannot be altered so please be vigilant in your selection.  If your intention is to test and evaluate, please do not select production installation.

My intention is to test and evaluate. Sorry if this has an obvious answer but just to be safe I prefer to ask before.

I am following these instructions here: http://code.google.com/p/gbif-providertoolkit/wiki/IPT2ManualNotes#IPT_setup...

I have chosen "NOT production use". Beyond that point, the Quick Reference Guide says:

"Click on the button labeled 'Continue' to open the IPT Administration page (...) Before adding data resources to the IPT, the administrator must (...) set the GBIF registration options, and associate the IPT with an organisation. The Organisations button is disabled by default until the GBIF registration options have been set".

As I am just testing the software, my question is: in a "NOT production" installation, can I safely test all the steps above?

Yes.  You may do whatever you wish to in testing when you have selected that mode.

Please see the user manual which is nearing completion for further information: http://code.google.com/p/gbif-providertoolkit/wiki/IPT2ManualNotes

...

 Is it possible to set the GBIF registration options and associate to an organisation in "not production mode"? Is there a live GBIF registry plus another "test GBIF registry" for "not production" IPT installations?

That is correct.  There are 2 registries, and by choosing not to run in production mode, the IPT will operation against a test registry, and therefore no "noise" will appear on the GBIF network due to test registrations.

...

In this case I suppose there is no problem in using a server address/port which is not the same I will use when going into a production server. Right?

That is correct, but please be aware that there is no way to take a non production IPT into a Live production mode.  You will need to reinstall and map the data again. This was necessary because of the complexities in doing subsequent synchronisations with respect to the datasets and organisations which might differ between registries.

I hope this helps.

Best wishes, Tim

...

Thanks in advance for your answers David

-- David García San León Herbario SANT Universidade de Santiago de Compostela (Spain)

_______________________________________________ IPT mailing list IPT@lists.gbif.org http://lists.gbif.org/mailman/listinfo/ipt

David, maybe Im too late in this game and its all set by now. But if I understand your intention correctly and you simply want to copy a resource configuration (all the metadata, the sql sources, the uploaded files and the entire mapping incl value translation) you should be able to do that by simply copying the resource specific configuration folder in your datadir. This should NOT be done for already registered resources and its not an officially supported feature, but all information needed for a resource is in such a folder IPT_DATA_DIR/resources/YOU_RESOURCE_NAME Make sure your target IPT data dir has the same extensions and users installed. We have an enhancement issue logged for a future version to officially support this feature at some stage via the IPT interface: http://code.google.com/p/gbif-providertoolkit/issues/detail?id=538 thanks. Markus On Jan 25, 2011, at 21:30, Herbario SANT wrote:

Thanks a lot for your replies, Tim and John.

John, with "you don't actually have to redo the source data mappings or metadata if you create a Darwin Core Archive from the original data", you mean an archive containing not only metadata, but also all our database data?

Or perhaps this kind of file may content not a database dump, but a copy of the database type, db server address, db user/password, db fields mapping to DwC and so on?

I am mostly interested in serving herbarium data (preserved specimens). So, are we talking of a small file size (mostly filled in with institution and collection metadata, but not with specimens data) or a huge file (containing a static copy of all our specimen labels)?

Since our data are highly dynamic, the second file will soon be outdated. So I am mostly interested if the first option is possible (a small IPT backup file, not for backing up all our data, just backing up the IPT installation/metadata).

Perhaps there is an example of such a "backup IPT installation archive" I can see somewhere? Or instructions on howto create it

Thanks

-- David García San León Herbario SANT Facultade de Farmacia - Laboratorio de Botánica Universidade de Santiago de Compostela 15782 - Galicia (Spain) _______________________________________________ IPT mailing list IPT@lists.gbif.org http://lists.gbif.org/mailman/listinfo/ipt