Forcibly Unmounting NFS Filesystems

forcibly unmounting nfs filesystems

preventative steps

The Highly Available NFS component of MC/ServiceGuard is a toolkit that enables you

to create NFS packages that run on highly available servers. With MC/ServiceGuard

NFS, an NFS server package containing exported filesystems can move from one node

(the primary node) to a different node (the adoptive node) in the cluster in the event of

failure. After MC/ServiceGuard starts the NFS package on the adoptive node, the NFS

filesystems are re-exported to the clients. Any client-side applications accessing these

filesystems will hang temporarily while the NFS server package is started on the adoptive

node. Once the package is running on the adoptive node, the client-side applications

can continue to seamlessly access their files without having to be restarted.

Looking back at the example on pages 7 through 10, remember that one of the critical

steps in setting up a temporary “surrogate” server involved configuring the IP address of

the original NFS server on the temporary node. MC/ServiceGuard NFS performs a

similar function when migrating NFS packages between nodes in the cluster.

Each highly available NFS package is assigned a dedicated IP address, and it is this IP

address that the NFS clients use when mounting the exported filesystems associated with

the package. When a hardware or software failure occurs on the primary node, and this

failure causes MC/ServiceGuard to decide that a package migration is needed, the

exported filesystems and IP address associated with this NFS package move to an

adoptive node in the cluster. Since both the exported filesystems and the server’s IP

address migrate between systems, NFS clients do not require any special configuration to

work in this environment because they simply retransmit their NFS requests to the same IP

address both before and after the package migration. The client systems have no idea

that a different server node in the MC/ServiceGuard cluster is now running the NFS

package and is responding to their requests.

By employing a highly available cluster of NFS servers that can share responsibility for a

pool of exported filesystems, the likelihood of encountering a situation where NFS clients

are blocked due to an unavailable NFS server is drastically reduced.