NFS Clustering

LifeKeeper is able to provide excellent high availablity for NFS, where failovers can occur without the end user having to reconnect, and without data loss.

LifeKeeper is able to work in an active/active or active/passive configuration for NFS, allowing for either maximum usage of computing resources, or for maximum performance in the event of a failover.

The NFS Server Recovery Kit provides a High Availability NFS service in hierarchical cooperation with the Filesystem Recovery Kit and the IP Recovery Kit that are part of the LifeKeeper core product.

The kit ensures that an IP resource and a file system resource containing the shared mount point are always in-service on the same virtual server in the cluster. Clients who mount the file system using the LifeKeeper-protected IP resource can continue processing files on the volume virtually uninterrupted while the actual export service is switched between servers in the cluster (either manually or in response to a failure).

Because NFS works over UDP, there is no stateful connection to be broken/reinitiated after a failover, therefore the NFS recovery is often used to demonstrate LifeKeeper - our favourite being a space shuttle launch video played from a client which has a LifeKeeper protected NFS mount. When failovers take place, at worst there is a small delay.

Technical details

NFS clients use entities called filehandles to refer to files on the server. The kernel HANFS component of this kit ensures that when a filesystem is switched from one node to another, the filehandle presented by the client is still accepted by the new server.

All files on the file system become temporarily unavailable while a switchover or failover is in progress, but they become available again transparently when the resource transfer is complete.

For a switchover, this can take between 5 and 30 seconds. For a failover, the recovery time depends on how long it takes to repair the file system. It is strongly recommended that you format the underlying disk volume with a Journaling File System (JFS), which is extremely robust to failure and can be repaired in a few seconds.

You may also choose to use a Linux file system (ext2) as the underlying file system, but in that case failover times will be much longer.

NFS Configurations

LifeKeeper can be used in an Active/Active or Active/Standby configuration depending upon requirements. In an Active/Active configuration all cluster nodes are utilised, resulting in no idle servers (maximising return on investment). Alternatively, Active/Standby can be used if there are no other applications which require LifeKeeper's protection.

The image below, shows LifeKeeper in an Active/Active configuration, where there are two LUNS configured on the shared storage device. At any one time, LifeKeeper will ensure that only one machine is capable of accessing a particular LUN through it's use of SCSI reservations, this avoids the requirement for STONITH devices and guarantees data integrity.

Should one server, or service fail, LifeKeeper will initiate a failover on one of the remaining cluster nodes.

The next image shows how LifeKeeper can be used over a WAN environment through use of the LifeKeeper data replication software, which is capable of replicating data in real time over a network (either a local gigabit grade link, or a lower speed WAN). This type of configuration allows for true offsite disaster recovery, where clients are not required to reconfigure after a failover.

This solution can also be used on a local network (LAN) to avoid the requirement for shared storage, which can be an expensive addition to a solution. Through use of the LifeKeeper data replication software, and high speed network cards, it is possible to increase the return on investment, although it should be pointed out that data replication can only occur between two nodes, so it is not possible to have more than one standby node.

Original Source: www.openminds.co.uk