Hi Marco,

I wrote up, as far as I could remember, the steps I took to install the latest version of the site services at Glasgow. This is at

http://ppewww.ph.gla.ac.uk/~caitrian/log/log-03-08-06.txt

It's a rather rough step-by-step guide and mostly repeats what is on the DQ2 wiki but hopefully will be useful for you. I found the 0.2.11 release to install much more smoothly than previous releases.

I have some other logs (written mostly for my own benefit, to help me remember what I did in particular circumstances), at http://ppewww.ph.gla.ac.uk/~caitrian/log/

I also have a script (which is still work-in-progress and very untidy!), called TagQueryTool.py, which I put at

http://ppewww.ph.gla.ac.uk/~caitrian/tag/TagQueryTool.py

in case you are interested in seeing it. It does the following:

1) queries tag DB with query given on command line (I need to add to the command line the facility to pass the collection name) and returns list of guid 2) Finds which datasets contain each guid (currently picks the 1st in the list, if there are more than 1) 3) sorts guids according to which dataset they are in 4) Finds which sites these datasets are at 5) generates new DQ2 dataset with only the files from the query, registers it at all the sites where it is (partially) present, and subscribes it to a site specified by the -s option on the command line 6) If specified on command line with option -l, it writes out a list of the LFNs corresponding to the query result 7) Likewise, option -g makes it write out a list of the guids 8) Option -p makes it parse the LFC which contains the PFNs and write out an XML file catalogue of the files (stolen from the dq2_poolFCjobO script!).

Step 1) is done with the POOL Collection utilities, steps 2-5 use the DQ2 API directly. As I said, it's still very rough, but it gives an idea of how you can use DQ2 for this kind of thing.

Let me know if you need any more information,

cheers,

Caitriana

-- RobGardner - 07 Aug 2006