Re: 2.3 SMP overlapping writes and NFS
Jamie Lokier (lkd@tantalophile.demon.co.uk)
Thu, 5 Aug 1999 17:03:20 +0200
Alan Cox wrote:
> > Regardless of committing to stable storage, write atomicity can be used
> > for database synchronisation between clients. Of course any sensible
> > client would synchronise on a single byte... I don't know if any real
> > applications depend on atomic writes.
>
> NFS clients in NFSv2 do short term data caching so that doesnt work too
> well either. The normal approaches for NFSv2 are link() and mkdir(). (open
> O_EXCL doesnt work for some buggy old servers)
By "applications" I mean boxes which use only NFSv2 and don't do
cacheing. Could be one somewhere, but maybe we don't care anyway.
BTW some clients will do long term data cacheing if the modtime doesn't
change, even to local disk -- and as the modtime's only set in seconds
on Linux that can happen even after the data has changed.
-- Jamie
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/