Results of a Robustness Regression tests and Performances campaign on linux-2.6.22-rc5-CITI_NFS4_ALL-1.diff

[J. Bruce Fields]

On Tue, Jul 31, 2007 at 03:40:28PM +0200, Le Rouzic wrote:
> The results of a Robustness Regression tests and Performances campaign 
> on linux-2.6.22-rc5-CITI_NFS4_ALL-1.diff  are available at : 
> http://nfsv4.bullopensource.org/
> Those results include kerberos flavors krb5, krb5i, krb5p on Intel 
> x86_64 machines.

Thanks.  Unfortunately, these bugs that can only be reproduced by
running several tests simultaneously for some time are a lot of work to
analyze!  And there's always some small chance that the problem can turn
out to be elsewhere.

If there was somebody that could take a closer look at them, and find
smaller test cases, or at least find some hints as to why they happen
when they do, that'd be a help.

The mount stuff shouldn't as hard to track down, though.  Is it
reproduceable with the very latest mount code from nfs-utils?

--b.