The tools in this directory help find out which computing elements support CVMFS for a given VO. Here is a manual on how to use the set of scripts.

1. Run the 1_launch.sh script: this will submit one job on each CE supporting the VO. The job consists of test.sh and test_ce.jdl.
Output: _jobids.txt = IDs of jobs run, _celist.txt = list of CEs retrieved from BDII (lcg-infosites)

2. The day after, run 2_retrieve_output.sh, this retrieves the output of all jobs whom IDs are listed in _jobids.txt. When output is successful, file _jobids_ce_dir.txt is appended with job id, CE name and output directory.
Output: _jobids_ce_dir.txt
Output: jobs output in /tmp/jobOutput/...

3. Run 3_process_output.sh: this analyzes job outputs and creates files:
Output: vo_alice.txt, vo_atlas.txt, vo_biomed.txt, vo_mice.txt, vo_no_biomed.txt
vo_biomed.txt = list of sites that support biomed and CVMFS
vo_no_biomed.txt = list of sites that support CVMFS but not biomed

4. Run 4_list_CEs_biomed_CVMFS.sh: this script analyzes file vo_biomed.txt and builds the list of CEs from sites that support biomed and provide CVMFS.
Output: _CEs_biomed_CVMFS.txt

5. Finally, 5_script tag_CEs.sh adds a tag to all CEs retrieved in the previous step by reading file _CEs_biomed_CVMFS.txt.
To do this you need to have a proxy certificate with role lcgadmin. You initialize it like this:
  $ voms-proxy-init --voms biomed:/biomed/Role=lcgadmin
Output: 5_tag_CEs.sh_failed.log = CEs that could not be tagged

IMPORTANT: ARC CEs do not support the lcg_tag command, thus this script will fail for them.
To fix this: 
- select ARC CEs using VAPOR (https://operations-portal.egi.eu/vapor/vapor-voSupportingResources?vo=biomed), column 'Impl. name' = 'ARC-CE'
- submit a ticket to the sites asking them to manually add the tag for us.

6. To double check which CEs were successfully tagged, use look_for_cvmfs.jdl. Run the command:
   glite-wms-job-list-match -a look_for_cvmfs.jdl