[rhelv6-beta-list] Does the NFS install with ISO images still work?

[Nico Kadel-Garcia]

On Mon, May 3, 2010 at 4:17 AM, Simon Matter <simon.matter at invoca.ch> wrote:
>> Sort of. It's a little more work than it used to be, but not much.
>>
>> https://bugzilla.redhat.com/show_bug.cgi?id=584472
>>
>> http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6-Beta/html-single/Installation_Guide/#id4005600
>
> Thanks, that explains my issue. BTW, I can not find the diskboot.img
> mentioned here
> http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6-Beta/html-single/Installation_Guide/#sect-New_Users-Alternative_Boot_Methods
> I guess it's because the docs have not been updated.
>
> Simon

I also recommend very strongly against the NFS installation approach.
While it may have some speedups, it often leaves dangling NFS mounts
recorded on the server, which makes "showmount" show all sorts of
debris. This is particularly bad when a DHCP driven kickstart system
does not pre-allocate DHCP reservations for kickstart configured
servers, so the new host bounces from a DHCP address at kickstart time
to a server-style static configuration, and the NFS mount is never
completely forgotten until someone has a harsh word with the NFS
server. And if you don't own the NFS server (for example, if it's a
NetApp owned by another engineering crew), it's very difficult to
clean up.

It's also not *that* much faster, and HTTP is easier to load balance
for the servers.