NFS and DNS alias

I have an old NetApp filer that exports volumes via NFS for my UNIX hosts.
We want to migrate it to the new filer and keep name of the old one as DNS alias to the new filer name with new IP address.
The question is will UNIX NFS client that still pointing to the old filer name be able to mount the volume from the physically new filer? Cashed MAC addresses discussion aside, please. I understand that.
Or to show it in the example

# nslookup filer2 
Name:    filer2
Address:  1.1.1.1
Aliases:  filer1

# nslookup 1.1.1.1
Name:    filer2
Address:  1.1.1.1

On the UNIX client:

mount filer1:/example /mnt

Will this last command work or it will end up with error message "permission denied", because of the mismatch between used alias and reverse DNS?
Some links to the theory of this process will also be welcomed.

Best regards
Leonid

If it's a cname, I would think it would be no problem at all. Things to look out for would be anything doing any kind of local name caching.... but apart from that, shouldn't be a problem. I haven't actually tried what you are doing though..... so I could be wrong.... (don't you just hate answers like that!)

unmount
repoint name to new IP
remount

arp caching should not be an issue.

It is not what I was looking for, gentlemen.
I know how to handle the situation. I wanted to know why it may or may not work.
If some one could share the ideas or point to the right sources of the information.

Best regards
Leonid

Ouch... you're mean...

It WILL work unless your "NFS" server is doing something weird. And I am actually doing this. We have a server that has a CNAME... and you can mount the server using either name (using NFSv3).

1 Like

OK. Thank you. I have to stage experiment and trace mount command and then read RFS. :wink:
Thank you everyone.
Leonid