PubDB V4 and zipped-dataset catalogues are now being fully deployed at CERN, and there are significant changes happening in the set of published catalogues. These should be transparent to normal users, but in case they are not you might want to be aware of them, or to notify your users.

The details:

- all data published in PubDBV3 is being converted to the 'new style' catalogues needed by the latest version of CRAB.

- these new catalogues are being published in PubDB V4, and unpublished from PubDB V3.

- the new catalogues are being systematically updated to refer to the zipped files instead of unzipped files.

- the old catalogues are not being removed, users who have jobs or scripts that refer to them directly will still be able to use them. Until...

- access to the unzipped files will be blocked soon. This was already announced to happen some time ago, and did for some of the data, but had to be reversed because of this delay in moving to the new publication scheme.

- access to the old catalogues will also be blocked soon.

New catalogues are available for much of the data that exists at CERN, and the rest will become available over the next few days. During that time, users should not have any break in access to data. If they are using a catalogue for which access is blocked they can switch immediately to a newer catalogue for the same dataset.

This is a 'point of no return' step, so people should report problems to me promptly, but check the documentation first (http://cmsdoc.cern.ch/cms/production/www/html/general/ZippedDatasets.html).

The target is for full migration of catalogues (and retirement of the old ones) by the end of next week, blocking access to the unzipped data will take a bit longer than that.

Cheers, Tony.

-- MariaSpiropulu - 12 Oct 2005