Buyer Beware: A Major Change in NFS is about to happen

[Steve Dickson]

On 09/29/2009 07:29 PM, Jesse Keating wrote:
> On Tue, 2009-09-29 at 19:16 -0400, Paul W. Frields wrote:
>>>> I think that what we need, Steve, is some sort of information
>> about
>>>> what testing has happened up to this point that satisfies FESCo
>> that
>>>> this change the equivalent of moving the needle from 99% complete
>> to
>>>> 100% complete, as opposed to moving from 90% complete to 91%
>> complete
>>>> (with testing being the last 9%).
>>>>
>>> Not a problem... I want to and will work closely with FESCo to make
>>> sure we know all the ins and outs of this change... believe me my
>>> slate is cleared for few weeks to be as responsive as possible 
>>> for this new feature... In the end this will be a good thing... 
>> [...snip...]
>>
>> I want to be perfectly clear that I'm not sounding an "all clear" on
>> this by any means.  If your answer here means that this change hasn't
>> been thoroughly tested, you're going to have a hard time convincing
>> anyone that it should be turning over on Beta freeze day. 
> 
> Well Steve, obviously some testing had to be done right?  You wouldn't
> change filesystem things without testing it, or else how would you know
> if it was successful or not?
Of course... testing has to be done... 

> 
> My main concern is with installer, installing from NFS shares from older
> servers, say RHEL5.  How will anaconda handle mounting?  Will there be
> odd errors that are difficult to figure out?  Has this been tested in
> the anaconda environment at all?
This issue is this... when the the F12 does a mount to a linux server 
and that linux server is *not* configured with a  "/ *(ro,fsid=0)" 
export, the mount will fail with ENOENT (or No such file or directory).
If the server does have that export, things will work as expected...
So my advice is to added that one line to your rhel5 server and every
thing should as expected... or may even better... ;-) Another workaround
is to added the '-o v3' mount options... would that be hard?

How will anaconda handle mounting? I don't know since I believe anaconda
has its own mounting code so it does its own thing... and in the end
that will mostly likely work, since that code didn't change...

Will there be odd errors that are difficult to figure out? I do not believe
so... the NFS community has done a lot of work on making sure error messages are
readable and understandable... Try using the '-v' mount flag, it truly
give you a nice insight on what is happening... 

Has this been tested in the anaconda environment at all? No. How do I 
do that and I will run those tests... 

steved.