NFSv4 Active-Active Cluster Queries

[J. Bruce Fields]

On Mon, Oct 09, 2006 at 04:14:15AM -0600, Rajat Upadhyaya wrote:
> >>> On 7/20/2006 at 7:22 PM, in message
> <20060720135206.GC20263 at fieldses.org>, "J.
> Bruce Fields" <bfields at fieldses.org> wrote:
> > On Thu, Jul 20, 2006 at 12:11:26AM -0600, Rajat Upadhyaya wrote:
> >> I have been working on setting up NFSv4 in cluster scenario. But
> when
> >> it comes to an Active-Active setup, it seems there are a few
> problems. 
> > 
> > Yes.  Even active-passive failover is problematic right now, mainly
> > because the method we use for storing reboot recovery information
> > (analagous to the state recorded by statd) is still in flux.
...
> 
> We had done some testing on NFSv4 in Active-Passive scenario & no
> problems were reported
> (http://linux-nfs.org/pipermail/nfsv4/2006-July/004715.html ). Is
> this still an issue? Can you give some details about the same? We have
> basically tried out Scenario 1 given at
> http://wiki.linux-nfs.org/index.php/Recovery_and_migration . 

Yes, that should work.  The reboot recovery problems alluded to above
are a bit subtle; you're not going to hit them in practice.
Unfortunately, though, solving them is going to require changing the
reboot recovery system in an backwards-incompatible way (so working
setups now will have to be changed at some point.)

> BTW is there any TODO list for NFSv4 which is up-to-date, where all
> such items are reflected?

There are some TODO lists.  They could be better.  See:

http://wiki.linux-nfs.org/index.php/To_do

for links.

--b.