Paul M Dunphy wrote:
> Then I set up a Linux system at home here and I'm trying to get the
> same thing working. Here's what I get as an error on NT, using FTP
> Software's ON-NET32 Interdrive client:
>
> "An error occurred in the network provider Interdrive NT. 1208: The
> PCNFSD is not accessable on the remote server. Contact your system
> administrator (or try logging in as user name NOBODY.)"
>
> The NFS share shows up
> under the correct NFS server (my Linux system) when you try to map a
> network drive on the NT system, so it appears to be a permissions problem.
> I use the correct name and password for /home/pdunphy . . . even tried
> root and the root password. No luck. Always the same error.
Hm. . .
>
>
> Being new to the Unix side, I assume I've got something wrong
> there. Here's what I did:
>
> Edited /etc/exports so that it contains the following line only
>
> /home/pdunphy 10.0.0.1(rw,no_root_squash)
Just to fill in some gaps -- does 10.0.0.1 refer to the NT? (it should) My
/etc/exports file also contains the option 'insecure' after the read/write
part, which short-circuits the login process altogether and just assumes that
the remote machine (in your case 10.0.0.1) has permission to map the drive no
matter which user calls for it. Even if that isn't agreeable to you in the
long run (worried about security infiltrations from your home machine much?),
you might want to try it just to narrow things down. If it fixes your problem
then at least you know that it was a user-to-server problem and not a
machine-to-server problem. Or something like that.
Also in the gap-filling arena, are you having any other network-related
problems? Can you ping, ftp and telnet from the NT to the Linux box? (I
assume so, but ya gotta ask. . . )
> This tells me the mountd and nfs services are running, but I'm curious
> why there are four instances of mountd reported . . . all the documentation
> and HOWTOs I could find indicated two of mountd and two for nfs. Is this
> the problem? If so, how does one fix it?
I may be confused here, but I think that on the "server" side, mountd doesn't
care about what nfs is exporting; it only cares on the client side. Somebody
correct me if I'm wrong about this.
Although that does raise another possibility for debugging: Can you mount your
nfs export on the _Linux_ machine? For example, edit your /etc/exports and
replace 10.0.0.1 with *, or *.your_local_domain.com (whatever will give your
Linux box host-level permissions for the export), restart the nfs daemon, and
then do this:
# mkdir /mnt/tmp
# mount -t nfs <your_linux_box>:/home/pdunphy /mnt/tmp
?
At least if you can do that, then you know the nfs daemon is working.
>
>
> There is one other thing that I found strange. I ran glint and looked
> at the networking packages, specifically the nfs-server daemon. When you
> do a verify, it reports it as nfs-server:2.2beta29:2 and all is ok except
> the /etc/exports file indicates a checksum,size,time problem.
This is normal for any file you've edited from the original. The checksums,
etc. are only recorded in the table from the original distribution; rpm
doesn't know when you've changed a file manually. I don't think you need to
worry about this being an indication of a problem.
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject.
