Archives


- Beowulf
- Beowulf Announce
- Scyld-users
- Beowulf on Debian

NFS question

Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.

Search

Don Holmgren djholm at fnal.gov
Fri Jul 13 08:45:59 PDT 2001


In the (old) user space nfs server, there was the following switch available to
nfsd (from the man page for nfs-server-2.2):

       -r or --re-export
              Allow remotely mounted file-systems to be exported.
              This can be used to turn a machine  into  a  multi-
              plier  for NFS or Novell servers. Caution should be
              used when re-exporting loopback NFS mounts  because
              re-entering the mount point will result in deadlock
              between the NFS client and the NFS server.

              I should be noted that (on Linux) nfsd looks at the
              major  device number of the file system to find out
              whether it is a remote volume; if the major  number
              is not 0, it assumes the file system is local. How-
              ever, not only remote file systems use major number
              0, also procfs does. If you choose to re-export NFS
              file systems, beware that this potentially includes
              /proc  if  you  have the file system root exported.

I've used this successfully when a mirror I operated ran out of space and I
needed to quickly borrow additional space w/out adding a disk (i.e., a 3rd
system, analogous to X, supplied extra space via NFS which the mirror, HEAD,
exported to the world).  But, this was a read-only export.  I'd be leary of
doing this for a less restricted export.

Don Holmgren



On Fri, 13 Jul 2001, Robert Sand wrote:

> Christoph Wasshuber wrote:
> > 
> > I have a fundamental NFS question. Assume
> > that we have three computers, X, HEAD, NODE1.
> > 
> >         X ----- HEAD ----- NODE1
> > 
> >       /vol      /mnt/v1    /mnt/v2
> > 
> > X     is some computer on a larger network
> > HEAD  is also member of this larger network and
> >       knows about X and NODE1
> > NODE1 is one of my beowulf nodes and does not
> >       know anything about the larger network.
> 
> Not a chance.  Your reasoning sounds good but here is were the problem
> starts.  X is sharing the vol via nfs to any number of systems and HEAD
> is one of those systems.  HEAD is mounting the nfs filesystem on /mnt/v1
> therefore /mnt/v1 is an nfs filesystem.  The nfs server on head can not
> share a fiel system that does not belong to it and it shouldn't because
> of the implicit security and permission problems involved in it.  The
> nfs server on HEAD can only share filesystems that are local to it, be
> they HDD's, floppy's, CDROM's, or ZIP drives but never an NFS mounted
> filesystem.
> 
> -- 
> Robert Sand.
> mailto:rsand at d.umn.edu
> University of Minnesota, Duluth
> Information Technology Systems and Services
> 144 MWAH
> 218-726-6122        fax 218-726-7674
> 
> "Walk behind me I may not lead, Walk in front of me I may not follow,
>  Walk beside me and we walk together"  UTE Tribal proverb.
> 
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
> 







More information about the Beowulf mailing list