Message-Id: <199708261447.HAA18759@mailhost1.cac.washington.edu> From: "James W DeRoest" <deroest@cac.washington.edu> To: "AST Webpage" <astweb@u.washington.edu> Subject: *WE* broke NFS on AIX 4.2.1 Date: Tue, 26 Aug 1997 07:46:40 -0700
-----Original Message----- From: Douglas Luft <dbluft@u.washington.edu> To: Eliot Lim <eliot@u.washington.edu> Cc: Ken Lowe <ken@u.washington.edu>; Yonah Karp <yonah@u.washington.edu>; jim deroest <deroest@cac.washington.edu>; Thomas W Remmers <remmers@u.washington.edu> Date: Monday, August 25, 1997 12:14 PM Subject: Re: *WE* broke NFS on AIX 4.2.1
I talked with Eliot a little about the problem and proceeded to review the differences between the Aix base and our modified rc scripts. The problem shows up when we enable nfs port checking ('nfso -o portcheck=1' in rc.nfs).
dbl
On Fri, 22 Aug 1997, Eliot Lim wrote:
> Date: Fri, 22 Aug 1997 18:56:37 -0700 (PDT)
> From: Eliot Lim <eliot@u.washington.edu>
> To: Ken Lowe <ken@u.washington.edu>
> Cc: Yonah Karp <yonah@u.washington.edu>,
> jim deroest <deroest@cac.washington.edu>,
> Douglas Luft <dbluft@u.washington.edu>,
> Thomas W Remmers <remmers@u.washington.edu>
> Subject: *WE* broke NFS on AIX 4.2.1
>
>
> i stumbled onto a vanilla preinstalled AIX 4.2.0 box (dante100) where
> it could export to a 4.2.1 box just fine.
>
> i then proceeded to set it up as a production dante by installing products
> step by step (rather than a whole slew all at once). after each step
> i would test the mount and it would be working just fine.
>
> i ended up installing all IBM patches and products (including DCE)
> that are found on all our 4.2.0 systems.. i.e. i brought dante100 up
> to exactly the same AIX level as our production systems. the mounts
> were still working.
>
> i then proceeded to install our local products. instead of running
> tulsa_weekly i did a tulsa_prodinst of each product individually.
> i finally got to the product called "base" and it was here that
> the mounts started to fail after this was installed. i initially
> suspected it was grabports but it was not..
>
> i'm in the process of tracking it down and i'm pretty close to finding
> the smoking gun.. but i think this is pretty good confirmation that the
> problem is not on the AIX side.
>
>
> eliot
>
>
>
>