Applies against 2.6.33-rc6 pnfs-all branch. In a DLM cluster, writing to a node other than the node where the open call occurred and where meta data is cached will have performance implications when the write causes meta data changes that need to be propagated to the open call node. Return NFS4ERR_BADIOMODE for LAYOUTGET requests with LAYOUTIOMODE4_RW iomode. Plus a couple of error code fixes. 0001-pnfsd-fix-NFS4ERR_BADIOMODE-in-layoutget.patch 0002-pnfsd-fix-pnfs_export_operations-layoutget-valid-err.patch 0003-pnfsd-DLM-file-layout-only-support-read-iomode-layou.patch 0004-pnfsd-fix-DLM-file-layout-no-device-return.patch Allow the client to fail a RW iomode layout and still grab a RO iomode layout. 0005-pnfs-set-failed-layout-bit-per-iomode.patch Testing: Connectathon tests pass. Basic test shows RW iomode layout for 'bigfile' fails with NFS4ERR_BADIOMODE with writes going through the MDS, and RO layout for 'bigfile' is obtained with reads going to the DS. -->Andy -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
