| Christoph Hellwig | f99d4fb | 2016-03-04 20:46:17 +0100 | [diff] [blame] | 1 | |
| Daniel W. S. Almeida | 98600b7 | 2020-01-10 20:24:30 -0300 | [diff] [blame] | 2 | ================================== |
| Christoph Hellwig | f99d4fb | 2016-03-04 20:46:17 +0100 | [diff] [blame] | 3 | pNFS SCSI layout server user guide |
| 4 | ================================== |
| 5 | |
| 6 | This document describes support for pNFS SCSI layouts in the Linux NFS server. |
| 7 | With pNFS SCSI layouts, the NFS server acts as Metadata Server (MDS) for pNFS, |
| 8 | which in addition to handling all the metadata access to the NFS export, |
| 9 | also hands out layouts to the clients so that they can directly access the |
| 10 | underlying SCSI LUNs that are shared with the client. |
| 11 | |
| Randy Dunlap | e70cc71 | 2020-07-03 20:20:14 -0700 | [diff] [blame] | 12 | To use pNFS SCSI layouts with the Linux NFS server, the exported file |
| Christoph Hellwig | f99d4fb | 2016-03-04 20:46:17 +0100 | [diff] [blame] | 13 | system needs to support the pNFS SCSI layouts (currently just XFS), and the |
| 14 | file system must sit on a SCSI LUN that is accessible to the clients in |
| 15 | addition to the MDS. As of now the file system needs to sit directly on the |
| 16 | exported LUN, striping or concatenation of LUNs on the MDS and clients |
| 17 | is not supported yet. |
| 18 | |
| 19 | On a server built with CONFIG_NFSD_SCSI, the pNFS SCSI volume support is |
| 20 | automatically enabled if the file system is exported using the "pnfs" |
| 21 | option and the underlying SCSI device support persistent reservations. |
| 22 | On the client make sure the kernel has the CONFIG_PNFS_BLOCK option |
| 23 | enabled, and the file system is mounted using the NFSv4.1 protocol |
| 24 | version (mount -o vers=4.1). |