[sabhijatya]

I am creating a HPC cluster using Rocks Cluster 6.2 for creating test-bed for one of R & D experiments. The cluster consists of 4 nodes. 2 of them are "HP ProLiant DL580 Generation 5 (G5)" one of which is formed as front-node and other is compute node. Other 2 nodes are "Intel® Server System R2000GZ Family" both of which are used as compute node. All the servers have hardware raid 1+0 with 2 SAS SFF hdd.


The rocks creates front node perfectly fine with automatic partition scheme and adds first compute node of HP server. But while adding Intel server as compute node it gives the error: "Could not allocate requested partitions, partitioning failed, could not allocate partitions as primary partitions, not enough space left to create partition for /boot".

As of now I have come across that the storage controller of both HP and intel servers are different. HP servers created the partition:
/dev/cciss/c0d0p****

But Intel server does as:
/dev/md*****

So is it possible that errors coming is due to different storage controllers if so, what will be the solution? How to go about to solve the issue?

Thanks,
Amarjeet Sharma