One of the storage admins was talking about this issue during lunch....  The 
really big filesystem (bfs) we moved to NetApp, has continued to grow...to 
where it is now bigger than a single tray.  So, now we have bfs1, bfs2, bfs3, 
bfs4, bfs5, bfs6 being served up from the NetApp, and the application 
developers added some storage management feature to the app to support this.

One of the neat things about the zfs appliance, is the nfs migration 
feature....we make the old share on netapp read-only, point all the clients to 
the zfs appliance a creates a cache of the metadata and starts copying the data 
over in the background (or on demand if a client requests something that hasn't 
been copied over yet).  At least that's the theory....storage admin is 
grumbling that NetApp uses /vol/<filesystem> on the exports and ZFS appliance 
is using /export/<filesystem> and wants the ZFS appliance to do /vol before 
trying the NFS migration in production.  Guess I don't understand why the name 
of the export is so important.

Where we mount stuff is neither named /vol nor /export.

Oracle guy says we could consolidate those filesystems in the move, but we'll 
need the application developers to change the app back.  It requires the 
developers to poke around if one of the bfs# stores fills up to move stuff to 
another store.  Not sure why the ASAs didn't get the developers to make it so 
they could manage it themselves.

They were also talking about how NetApp licenses every single radio 
button....manager says its 3 pages long of license keys for the features we 
got.  And, Sun gives everything...and when an update brings new features, it 
just shows up...don't have to buy it.  So far sounds like Oracle isn't going to 
change that, even though they stopped doing a lot of things that Sun was losing 
(or not making) money doing.  Like .edu discounts.

Oh, we ask different groups to estimate their storage needs regularly so we can 
plan upgrades....well, the people that use 'bfs#'...gave us a graph of their 
estimated needs....and it shows that they have exponential growth plans and 
want us to give them infinite storage in 5 years.  What was supposed to happen 
when you fill up a single ZFS pool?

----- Original Message -----
> On Thu, 16 Jun 2011, John Stoffel wrote:
> > So push comes to shove, Netapp does NOT offer an easy way to
> > expand/grow/migrate your data transparently from one set of
> > disks/shelves to another.
> 
> It's not easy, no. Depending on what you're trying to accomplish,
> exactly, it may be possible doing a cluster failover and giveback,
> though.
> It's not really supported, especially for customers, but professional
> services still has a few tricks up their sleeves.
> 
_______________________________________________
Tech mailing list
[email protected]
https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to