[sheepdog] [PATCH] tests: add a test to test vdi snapshot functionality concurrently

[MORITA Kazutaka]

At Wed, 19 Sep 2012 14:23:37 +0800,
Liu Yuan wrote:
> 
> On 09/19/2012 01:44 PM, Liu Yuan wrote:
> > With current object implementation, collie command will try to read object
> > cache first because we change the semantics of no flag CMD_CACHE into 'writethrough'
> > instead of 'no object cache'.
> > 
> > I think before 0.5.0 release, we need to fix this regression.
> 
> I don't find an easy way to fix this regression on collie command. I think the easiest way
> is to restore old semantics, then current collie code wouldn't need a single change.
> 
> then '-w object:mode={writethrough,writeback},size={size},directio' operate on object cache and
> '-w disk' will enable current disk write back semantic.
> 
> how do you think of it, Kazum?

I'm still testing and trying to understand what is the problem, but
looks working almost correctly, at least for now.  The only problem I
encountered is that sheep pulled deleted vdi objects to the local
cache directory and this is the reason why the master branch doesn't
pass tests/044.  However, it is not a fatal problem, I think, because
sheep just only caches the info that the object is deleted.

When object cache is enabled, collie accesses object with a
writethrough mode - is it a really problem?  What is the scenario
where data inconsistency actually happens?

Thanks,

Kazutaka