Hi Emanuele,
I was testing the configuration of shared gridmapdir between CE, SE. I
am also testing the configuration of shared Worker Nodes between two
testbeds: LCG-1 and CrossGrid. This is because several CrossGrid sites
are also participating in LCG-1 and they are interested in sharing
computing farm between different CEs. So I am the man from CrossGrid
responsible for finding such a solution. So during my work on this I
accidentialy did a mess in gridmapdir, namely some users' DNs were
hardlinked to not existing inodes and globus was unable to map them...
Sorry for troubles.
Piotr
Emanuele LEONARDI wrote:
> Hi Piotr.
>
> Access to your site is back to normal. What was the problem with the
> gridmapdir?
>
> Thanks
> Emanuele
|