On January 22, [EMAIL PROTECTED] wrote:
> We have a LAN with about 40 Linux systems on it. We use the Berkeley
> "customs" suite to perform parallelized builds of our product. So we
> hammer NFS pretty hard; 30-40 machines can be simultaneously reading
> and writing a single build tree through NFS.
>
> We upgraded one of our developers to 2.2.18. We also upgraded all of
> our systems to the latest nfs-utils (0.2.1) and mount (2.10m) packages
> from <http://nfs.sourceforge.net/>.
>
> As a result, we now have a large number of 2.2.17 NFS clients reading
> and writing to a 2.2.18 NFS server. The server (the developer's
> desktop system) is SMP, and all of our systems have ECC memory.
>
> This developer is now regularly seeing two problems which began with
> the 2.2.18 upgrade. First, remote clients occasionally get "stale NFS
> file handle" errors for no apparent reason. Second, some of the files
> are being corrupted.
Could you retry after applying patches from
http://www.cse.unsw.edu.au/~neilb/patches/knfsd-2.2/
particularly patch-B-sema.
Thanks,
NeilBrown
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
