ClusterNFS on sourceforge. So
http://clusternfs.sourceforge.net/
>
> TIA,
> chippo
>
> PS: What I really want to know (which I'll hopefully glean from the
> code/URL when located) is:
> - was this a user-space or kernel-space implementation of NFS
UserSpace
> - how much mission it'll be to get the same (or similar) functionality
> on a 2.4 kern.
Not if I can help it. All the functionality you need already exists.
Each client mounts a common root filesystem.
A very early rc script mounts a per-client filesystem on
e.g. /perclient.
The rc script can use hostname or IP address or MAC address or
whatever it wants to choose the actual filesystem to mount:
mount -t nfs -o nolocks server:/clientfilesys/`hostname` /perclient
Then any file that needs to be per-client becomes a symlink into
/perclient.
About the only thing that this doesn't work for is files the are
unlinked and re-created often. I suspect if there are any of those
there is a workable solution.
NeilBrown
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/