On Wed, Dec 10, 2008 at 05:23:32PM -0500, Kyle McDonald wrote: > > from what I understand it is something that stops to exist with NFS > > v4. I've seen it disappear in any vendor's NFS v4 implementation, > > where they used to have that behaviour with NFS v3.
Right, NFSv4 does not use the MOUNT protocol. > So what's the benefit of v4 again? It's only been complicating things > for me. and don't list mirror mounts. Yuck - That's what the automounter > is for. I see how your use of NFSv3 was complicated by switching to NFSv4, but NFSv4 solves a great lot of problems that NFSv3 created or left unsolved. Mirror mounts get around some very obnoxious limitations relating to the MOUNT protocol: a) MOUNT is not run over port 2049 (think firewall traversal), c) noone implemented RPCSEC_GSS support for MOUNT, d) v3 clients tend to cache MOUNT protocol results and use only a small subset of them (e.g., the Solaris automounter has a fixed maximum of 1KB or 4KB, I forget what it is exactly; if you have too many shares, then you lose), d) v3 clients tend to cache MOUNT protocol results for very long times, so that when you add shares they don't become universally visible for a long time. And I'm just getting warmed up... > Now sharemgr won't even let you share a filesystem as -o vers=3. So the > only way to fix this is to globally limit the max version to 3 in > /etc/default/nfs. Really? What about the share(1M) command? Anyways, could you shorten these paths in some other way, such as, say, by mounting these filesystems where you currently put symlinks? Nico --
