On 29 May 2015 at 13:17, Andriy Gapon <[email protected]> wrote:
> On 29/05/2015 20:06, Matthew Ahrens wrote: > > On Fri, May 29, 2015 at 10:00 AM, Richard Yao <[email protected] > > <mailto:[email protected]>> wrote: > > > > It seems that no one is particularly interested in a strict point in > time > > version, > > > > To be clear, I'm not saying that I'm uninterested in that, just that it > seems > > hard/complicated to achieve in combination with other requirements > > (specifically, not needing to have the entire state in memory at once). > (And I > > don't know that anyone else has weighed in.) > > Additionally, what good is a consistent snapshot of the children if by the > time > you get it any number of changes might have happened. It's not like we are > preventing the changes from happening while we look at or iterate the list. > The set of edge cases when the list output is a accurate at a given instant in time should be a subset of what they are now. My view is that subset would be desirable improvement over what we have now. That said, actually achieving it is hard and no one else seems particularly worried about it. I'll do my best to document it and hope people using the API are careful.
_______________________________________________ developer mailing list [email protected] http://lists.open-zfs.org/mailman/listinfo/developer
