bug#25680: [PATCH] copy: Avoid race when creating hard link over recently-created file

[Mike Crowe]

On Friday 10 February 2017 at 11:55:35 -0800, Paul Eggert wrote:
> On 02/10/2017 11:18 AM, Mike Crowe wrote:
> >-  if (link_failed && replace && errno == EEXIST)
> >+  while (link_failed && replace && errno == EEXIST)
> 
> This could cause 'cp -f' to loop forever, if an attacker keeps creating hard
> links. Is this a new vulnerability? I don't recall any other way that
> copying from a finite source could take forever.
> 
> One possible solution would be to loop for just a few times, and then give
> up with a diagnostic.

Indeed, that's why I wrote:
> There is at least some risk that the while loop in
> create_hard_link might spin forever, but something else would have to be
> recreating the file again rather quickly. It might make sense for me to
> add a maximum number of retries.

Do you think that if I added such a limit and diagnostic then the patch
would be acceptable?

Thanks for the review.

Mike.