** The SAM MPI tests are raising alarms from this morning, as agreed last week **
Update received from Isabel Campos (MPI Task Force) last week here summarised.
The current situation is the following:
there are 90 sites which publish the MPI-START tag, 88 are tested by SAM and 2 other sites (IFCA and RAL) are not tested because the way they publish the SubCluster info.
Of those sites: 69 working fine (67 at SAM + IFCA + RAL)
20 errors
2 maintenance
which gives a 76% of sites passing the tests (75% if we don't count the sites out of SAM)
For all the sites with errors a ticket in GGUS has been opened and most of them are working actively on finding a solution. There is a guide with the list of errors found and possible solutions for them at http://wiki.ifca.es/e-ciencia/index.php/MPI_Errors
Documentation for MPI Support in EGEE:
https://twiki.cern.ch/twiki/bin/view/EGEE/MpiTools
More information about errors in the MPI knowledge DB:
http://wiki.ifca.es/e-ciencia/index.php/MPI_Errors