NFS Cluster Failover via Reverse LIN Map Translation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In NFS environments, failover from a source cluster to a target cluster often results in client instability due to the need for clients to unmount the source cluster and remount the target cluster, causing delays and lack of transparency in the migration process.
Innovation Solution
Establish a connection between the NFS client and the target cluster using a dynamic IP address, mirroring the source cluster's files and directories, and generating a reverse LIN map to translate source cluster identifiers to target cluster identifiers, allowing seamless failover without unmounting and remounting the NFS export.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If clients are failed over from source cluster to target cluster using traditional NFS methods, then failover capability is achieved, but client stability deteriorates due to unmount and remount requirements
Solution Approach 1:
The patent introduces an intermediary mechanism (NFS server on target cluster acting as a gateway) that translates client requests containing source cluster FSIDs and LINS into target cluster identifiers. This mediator enables seamless failover by handling the identification translation transparently, preventing client instability while achieving failover capability.
Solution Approach 2:
The patent changes the parameter mapping between clusters by creating a translation layer that maps source cluster FSIDs and LINS to target cluster equivalents. This parameter transformation allows clients to maintain their original connection parameters while the system internally redirects to the target cluster, resolving the contradiction between failover capability and client stability.
2Adaptability or versatility
If clients unmount and remount NFS export during failover, then cluster migration is achieved, but operation time increases due to remount delays
Solution Approach 1:
The patent performs preliminary actions by pre-establishing the NFS export on the target cluster with identical parameters (FSID, LINS mapping) before failover is needed. The target cluster is prepared in advance to accept client connections with the same identifiers used for the source cluster, enabling instant failover without remount delays.
Solution Approach 2:
The patent maintains continuity of useful action by keeping the NFS export mounted and accessible on the target cluster with identical identification parameters. Clients experience continuous access to their NFS mounts during failover because the target cluster presents the same FSID and LIN structure, eliminating interruptions and remount delays.
3Adaptability or versatility
If target cluster uses different FSID and LIN identifiers, then cluster independence is maintained, but client compatibility deteriorates after failover
Solution Approach 1:
The patent introduces an intermediary translation layer that maps target cluster FSIDs and LINS to source cluster identifiers. The NFS server on the target cluster acts as a gateway, receiving client requests with source cluster identifiers and internally translating them to target cluster equivalents, thus maintaining both cluster independence and client compatibility.
Solution Approach 2:
Instead of requiring clients to adapt to target cluster identifiers, the patent inverts the approach by having the target cluster maintain its independent identifiers while translating outward to match source cluster identifiers. This inversion allows the target cluster to preserve its independence while presenting source-compatible interfaces to clients.
Data Source
AI summary
Implementations are provided herein for establishing a failover cluster in a distributed file system that upon the occurrence of failover event, allows clients to actively migrate during ongoing file system activity from a source cluster to a target cluster without having to unmount an NFS export from the source cluster and remount the NFS export on the target cluster. Upon the occurrence of a failover event, clients can be rerouted to a target cluster that contains mirrored copies of the data the client was expecting to be stored on the original source cluster. However, in attempting to access the data, without unmounting and remounting an export, the client will continue to reference source cluster FSID and LIN identifiers when making NFS calls to the target cluster. Thus, implementations are provided herein for translating the requests of an NFS client on a target cluster after a failover event has occurred. A client can make NFS calls using FSID and LIN information of the original source cluster, and that information can be translated, e.g., through a reverse LIN map, to the unique LIN's of the target cluster, such that any failover migration of the client is transparent to the client.