Accessing a Symbolics LMFS from a UN*X machine via NFS

[cjl@ai.mit.edu (Chris Lindblad)]

    Date: Fri, 13 Jul 90 16:10 CDT
    From: lgm@iexist.att.com

    Our system administrator is trying to set up NFS (Genera 8.0) and can't
    find any explicit documentation on how to mount a Symbolics file system
    on a Sun so that it is accessible to UN*X users.  The Sun reportedly
    needs not only a host name but a "file system name."  None of "", ">",
    and "/" seem to work.  Has anyone done this successfully?  Are there any
    other gotchas in making this work?

This should work:
  % mount acme-smbx:/ /acme-smbx

Symbolics computers do list "/" as their exported filesystems, so 
  % showmount -e acme-smbx
should print out:
  export list for acme-smbx:
  / (everyone)

This also means you should be able to access Symbolics computers if you have
automount mount point for the -hosts map:

  % cd /net/acme-smbx
  % ls

Are you sure you have the NFS Server system loaded on your machine?  A common
mistake is to only have the NFS Client system loaded.

Here's some debugging hints:

For debugging, you can try on your Symbolics file server:

  (setq nfs:*debug-on-nfs-error* t)

This will cause your file server to drop into the debugger right before it
returns any error response to any NFS Client, so you can see what errors are
occuring.  Note this goes to the debugger for all errors that might occur
during the servicing of an NFS rewuest, even simple ones like file-not-found,
so you shouldn't run normally with this variable set, but only set it when
debugging.

If you want to get a log of unexpected server errors that occur during normal
operation, the NFS Server is integrated into the automatic-bug-report
mechanism, so if you create a >Server-Bug-Log> directory on your file server
and set these three variables:

(setq dbg:*autobug-recipients* (list "bug-acme-smbx"))
(setq dbg:*autobug-duplicate-action* :ignore)
(setq dbg:*autobug-enable* t)

You can have your file server recording unexpected errors that happen during
normal operation, and send mail messages to a system manager when an out of
the ordinary error occurs.