gpfs: adding a new nsd server to a cluster

Posted by alessandra on Server Fault See other posts from Server Fault or by alessandra
Published on 2011-11-20T11:23:21Z Indexed on 2011/11/21 1:54 UTC
Read the original article Hit count: 545

I have a gpfs cluster composed by 10 linux nodes, managed by a primary server A, which also act as nsd server for a first stack of disks. I attached a new jbod to one of the nodes (call it node B), which I would like to become a nsd server for this new stack of disks, but still be included in the cluster so that the disks are available to all the nodes. Node B is connected to the cluster via ethernet. How can I make the new nsd seen by all the nodes of the cluster?

I can create the new nsd but when trying to create the filesystem on node B it the command mmcrfs times out. It looks like the nodes of the cluster cannot understand the filesystem location even if I specify them attached to server B in the description file.

Would it be better to remove node B from the cluster, create a cluster on its own with its attached filesystem and connect it remotely with the previous cluster? Or a clustered NFS solution would apply better? Can you please give me any suggestion?

© Server Fault or respective owner

Related posts about clustering

Related posts about distributed-filesystems