To clarify my original post, the point at which we are tripping is in the installation. The implication is that the master/server and the execute hosts need access to some common shared disk space in order to exchange semaphores etc. The data directories and analysis applications *are* already set up across our cluster as identical mirrors of the same disk image. I don't want to use NFS for that as the overhead of loading blast databases across NFS every time I want to search is too great to contemplate. We just create on one node, then sync out to all the other nodes for that. It seems to me having sat and watched/commented as we tried the installation yesterday, that the master installation went just fine, but that the execute host installation seemed to be looking for directories that the master installation had created (specifically $GRIDENGINEROOT/default). We reached that point at 5:30 yesterday and David is off today so we haven't had time to look at it further but if someone can point me in the right direction here, we can save ourselves some time next week. Later, Andy ------------- Yada, yada, yada... The information contained in this e-mail (including any attachments) is confidential and is intended for the use of the addressee only. The opinions expressed within this e-mail (including any attachments) are the opinions of the sender and do not necessarily constitute those of Roslin Institute (Edinburgh) ("the Institute") unless specifically stated by a sender who is duly authorised to do so on behalf of the Institute.