[Wien] weird parallel problem

[Laurence Marks]

Even though I have an updated kernel, we still sometimes see problems.
For instance, if I edit case.in1 while lapw1 is running on a child
node the changes do not get propogated to the child (maybe because it
thinks it's version is more recent) unless one does a cluster-fork
copy. In /etc/exports I'm using

/export      20.0.0.0/255.0.0.0(rw,sync,no_wdelay,no_subtree_check,no_root_squash)

It was async by default which was a disaster. Not sure what is best,
never could figure this out and never found anyone who really seemed
to know (experts here dislike nfs, say don't use it).

Caveat: I'm not very good with nfs

On 11/16/06, Martin Kroeker <martin at ruby.chemie.uni-freiburg.de> wrote:
> >Sorry, but I have to disagree.
> >
> >Maybe it was never mentioned, but the "NFS-bug" problems NEVER appeared
> >with the ifc7 executables (on our sites), so it is specific to ifort AND
> >some Linux-Kernel version.
>
> I stand corrected. SuSe 9.2 was using kernel 2.6.8, and postings
> in the archive implicate 2.6.x before November 2005 (i.e. x < 14).
> Assuming it is/was a client bug, some "interesting" fixes for "lseek"
> operations apparently went into 2.6.12 (cf.
> http://client.linux-nfs.org/Linux-2.6.x/2.6.12-rc2/linux-2.6.12-37-file_add_lseek.dif).
> It is not clear to me whether these are also fixed in the latest kernel rpm
> update for 9.2 (2.6.8-24.25), though this at least mentions a "nfs cache
> consistency bug" fixed in october 2005).
> I cannot claim to have noticed such problems in several SuSE 9.2 nfs
> networks, but then none of them was used for distributed wien calculations
> or anything else involving ifort9.
> --
> Dr. Martin Kroeker            martin at ruby.chemie.uni-freiburg.de
> c/o Prof.Dr. Caroline Roehr
> Institut fuer Anorganische und Analytische Chemie der Universitaet Freiburg
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>


-- 
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
http://www.numis.northwestern.edu