[PATCH] some improvements to the defer-open test

[Jeremy Allison]

On Wed, Aug 06, 2014 at 04:46:12PM +0200, Michael Adam wrote:
> Hi,
> 
> the defer-open tests is one of the prominent causes
> of flapping tests. I have discussed with Metze some
> possibilites of improving the timing handling in this
> test, unifying it with some other timing-critical tests.
> 
> This is not done yet, but for a start, attached find
> a patchset that improves the defer-open test so that it
> hopefully does not create unexpected errors any more
> (but accurately fails with torture-assert), and gives
> more helpful error messages in that case.
> 
> Review/push appreciated!

LGTM - pushed.