Follow-up Comment #2, bug #87960 (project xrootd): Hi Sorry for the confusion on the same physical machine (two nics) standalone dataserver atlas8 (eth0) atlas21 (eth2) has a 19 TB xfs file system There is a test interactive machine (ascint1y). I have checked that all xrootdfs were removed before repeating the exercise. from ascint1y I mount atlas21.hep.anl.gov I have the system setup so ascint1y, ascint2 and ascint3 are clustered xrootd storage. Ultimately frm will copy files from atlas8(atlas21) to the ascint* nodes automatically. Users will use the storage on the ascint* nodes. _______________________________________________________ Reply to this item at: <http://savannah.cern.ch/bugs/?87960> _______________________________________________ Message sent via/by LCG Savannah http://savannah.cern.ch/