[sheepdog] [PATCH 0/2] don't remove objects when they might be necessary

[Bastian Scholz]

Second try, these time to the list ;-)

Hi Kazutaka,

Am 2012-08-25 20:09, schrieb MORITA Kazutaka:
> We shouldn't remove objects until object recovery completely 
> finishes.
> With this patch, even if we wrongly stop more sheeps than the
> redundancy level at the same time, sheepdog can recover objects
> automatically after they start up.  Even if sheepdog crashes
> unexpectedly, we can recover objects from the stale directory 
> manually
> at worst.

Thanks for your work :-)

But could you please give me a short Info, how the
cluster behave on the following situation?

5 Sheeps in 5 zones (0-4) and a vdi with 2 copies and
two object.

Obj 0 is on sheep 0 and 4
Obj 1 is on sheep 3 and 4

Now I kill sheep 3 and 4, Obj 0 can be recovered,
but Obj 1 will fail...

What happens to the VM in this situation, Obj 0
can be read and write normally and Obj 1 will fail
and give a kind of I/O Error to the VM, I guess?

And when I restart sheep 3 and/or 4 later, Obj 1
will be recovered (from the stale directory) so the
VM can access it normally?


Thanks

Bastian