[Lustre-discuss] NFS Export Issues
William Olson
lustre_admin at reachone.com
Tue Jul 20 14:07:26 PDT 2010
>> I've tried it explicitly with the no_root_squash option and it still
>> behaves the same way..
>> What I find really frustrating is that if I unmount lustre, I can mount
>> the same nfs export, no problems. As soon as lustre is mounted to that
>> directory I can no longer mount that nfs export.
>> I don't understand where it's failing..
>>
>> Thanks for all your help so far, any more ideas?
>>
> Have you tried to add "fsid=xxx" to your exports line? I think with recently
> Lustre versions (I don't remember the implementation details) it should not be
> required any more and so it should not with recent nfs-utils and until-linux
> (the filesystem uuid is automatically used with those, instead of device
> major/minor as fsid), but maybe both type of workarounds conflict on your
> system?
>
>
Yeah, I've tried it with fsid set and unset, same result either way.
> You also might consider to simply use unfs3, although performance will be
> limited to about 120MB/s, as unfs3 is only single threaded. It also does not
> support NFS locks.
>
>
I would rather not if I don't have to.. While I only need the nfs to
work for a short time I can't see a single threaded solution working
very well, this will be the backend filesystem for a production email
cluster, we're transitioning from another solution to lustre and need
the nfs to play it's part while we move from one to the other.. They
both have kernel version requirements so I can only setup one file
system directly on the cluster machines at a time unless one of them is
mounted with NFS..
> If it still does not work out, you should enabled lustre debugging, nfs
> debugging and you probably should use wireshark to see what it going on.
>
>
This may indeed be my next step in the troubleshooting process.. I was
hoping it was something obvious that somebody had run into before but it
seems this is a "special case".. Oh well, time to dig into the dirty
details I guess.. As always though, if anybody has any other ideas or
can suggest a better way to achieve what I'm trying to do, please let me
know.
> Hope it helps,
> Bernd
>
>
More information about the lustre-discuss
mailing list