switching NFS back and forth - nodeA, nodeB. Sometimes it works, but not always the NFS client comes across the following:
Oct 4 11:27:00 big-7 nfs: [ID 236337 kern.info] NOTICE: [NFS4][Server: nfs][Mntpt: /mnt]NFS op OP_RENEW got error NFS4ERR_STALE_CLIENTID causing recovery action NR_CLIENTID. Client also suspects that the server rebooted, or experienced a network partition. Oct 4 11:27:00 big-7 nfs: [ID 581112 kern.info] NOTICE: [NFS4][Server: nfs][Mntpt: /mnt]NFS Starting recovery for mount /mnt (mi 0xffffffff8b0a4000 mi_recovflags [0x21]) on server nfs, rnode_pt1 <null string> (0x0), rnode_pt2 <null string> (0x0) Oct 4 11:27:00 big-7 nfs: [ID 236337 kern.info] NOTICE: [NFS4][Server: nfs][Mntpt: /mnt]NFS op OP_WRITE got error NFS4ERR_STALE causing recovery action NR_STALE. Oct 4 11:27:00 big-7 nfs: [ID 286389 kern.info] NOTICE: [NFS4][Server: nfs][Mntpt: /mnt]File ./cp-b7 (rnode_pt: ffffffff8da66b50) was closed due to NFS recovery error on server nfs(failed to recover from NFS4ERR_STALE NFS4ERR_STALE) Oct 4 11:27:00 big-7 nfs: [ID 941083 kern.info] NOTICE: NFS4 FACT SHEET: Oct 4 11:27:00 big-7 Action: NR_STALE Oct 4 11:27:00 big-7 NFS4 error: NFS4ERR_STALE Oct 4 11:27:00 big-7 Suspected server reboot. Oct 4 11:27:00 big-7 nfs: [ID 532616 kern.info] NOTICE: [NFS4][Server: nfs][Mntpt: /mnt]Can't recover clientid on mount point /mnt (mi 0xffffffff8b0a4000) due to NFS error 5, for server nfs. Marking file system as unusable. Naturally since nodeA is down the clientID invalidated. But nodeB was up in the meantime, shouldn't client get a new clientID and continue writing transparently ?? Thanks, Oliver This message posted from opensolaris.org
