Struggling on NFS problem

[Rick Macklem]

On Wed, 5 May 2010, Cheng-Lin Yang wrote:

> Hi Jeremy,
> I certainly will, it seems to upgrading to RELENG_8 will be a highly possible solution for the issue. Therefore, I already scheduled a upgrade window on this Friday night and will keep updating you guys the result once the upgrade is done.
>
> I would also like to know that is it ok to update the source up to RELENG_8 by applying all NFS patches at the same time? Since it's a production machine, I have to make sure these two steps can be perform at the same time without any concern.
>

The patches at people.freebsd.org/~rmacklem are already in stable/8, so
you shouldn't need to apply them.

> Finally, just curious that why the problem only shows up on FreeBSD clients?
>
If it is the case where the server doesn't return ESTALE when a file
has been deleted on the server, the FreeBSD client will loop retrying
the RPC for EIO, but other clients might not. That is the only case
where I can suggest there is a difference, other than the type of
traffic and load different clients will generate.

rick