Well we disabled the aggregation and the error seems to persist. I'm currently running snoop port nfs on the xen clients and on the nfs server. I hope i'm able to catch the error.
Karl > We have 3 XEN Sun Ray servers running sxce b130 > connecting. The home directories are located on an > OpenSolaris 134 box. > > It seems that every day atleast one of the XEN SUN > Ray boxes shows NFS errors on the client such as > > Apr 15 13:10:22 sunray-1 nfs: [ID 581112 kern.info] > NOTICE: > [NFS4][Server: nfsserver-1][Mntpt: /home/lwidish]NFS > Starting recovery for > mount /home/lwidish (mi 0xffffff05e5fda000 > mi_recovflags [0x22]) on > server nfsserver-1, rnode_pt1 <null string> (0x0), > rnode_pt2 <null string> > (0x0) > Apr 15 13:10:22 sunray-1 nfs: [ID 273629 kern.info] > NOTICE: > [NFS4][Server: nfsserver-1][Mntpt: /home/lwidish]NFS > Recovery done for > mount /home/lwidish (mi 0xffffff05e5fda000) on server > nfsserver-1, > rnode_pt1 <null string> (0x0), rnode_pt2 <null > string> (0x0) > Apr 15 13:10:22 sunray-1 nfs: [ID 581112 kern.info] > NOTICE: > [NFS4][Server: nfsserver-1][Mntpt: /home/lwidish]NFS > Starting recovery for > mount /home/lwidish (mi 0xffffff05e5fda000 > mi_recovflags [0x22]) on > server nfsserver-1, rnode_pt1 <null string> (0x0), > rnode_pt2 <null string> > (0x0) > > What ends up happening on the client side is all the > firefox/thunderbird processes crash. > > There are no errors on the NFS server. > > The XEN server has link aggregation turned on. We are > going to turn that off tonight. > > We have also noticed that we do not seem to have a > problem when we set the mount option to -sec=sys when > other XEN SUN Ray servers connect to a older adaptec > snap server. > > I'm wondering if there is an outstanding bug or work > around I can apply? > > Thanks > Karl -- This message posted from opensolaris.org _______________________________________________ nfs-discuss mailing list [email protected]
