[Gluster-users] Mounting a volume - isolating directories = Stumped.

[James Cornman]

Thomas,

In this situation, and I do not now if this is the viable answer, but it was
my thought that you could simply make other volumes using the proper
directories as the brick location, and mount via the new mount point name
instead of the root volume share. I think your disk availability percentages
will be skewed as its two volumes sharing the same bricks but other than
that, it should work. Perhaps someone else can shed some light too as to
whether this method will work.

Also, you can try to mount using --bind option to logically mount one sub
directory to another (mount --bind /GFSVOLUME00/dir1 /dir1; gluster brick:
x.x.x.x:/dir1 -- not tested though)


-James


-- Quoted Text --

As so far I have 15 servers feeding data to my four node Gluster-FS
(cluster), at this current time all of my scripts are faceted to mount
the volume and move data into its directory, only issue I see with
this is that half of these clients should not be able to go one level
back from its 'home' directory when transfering data - I wanted to see
if there was a way to isolate these specified clients to mount the
volume to their specified directories without having to cd two
directories deep...

ex:
mount - t glusterfs servnode001.b.c:/GFSVOLUME00 /mnt/backups
$ > ls
dir1/ dir2/ dir3/

$ > cd dir3/
$ > ls
data... data...

This is disconcerting for me since I do not want servers to have full
access...I've attempted to look at permissions, and have yet to find
anything...

Care to shine some light for me?

Thanks,

-- 
Thomas Liakos


-- 
*James Cornman
**Chief Technology Officer*
james at atlanticmetro.net
212.792.9950 - ext 101
*
Atlantic Metro Communications*
*Data Centers • Managed Hosting • Internet Access • Dark Fiber*
Follow @atlanticmetro on Twitter for daily specials.
www.atlanticmetro.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20110701/541a610c/attachment.html>