Cliff Woolley wrote:
>
> On Wed, 7 Mar 2001, Greg Ames wrote:
>
> > It would be nice to prove that our dir_cleanup seg fault is really gone.
>
> I suppose this isn't "100% proof"...
[...]
> Blammo, segfault.
> Bill fixed this by unregistering the cleanup in apr_dir_close() regardless
> of its success or failure when called directly. Makes sense to me.
>
> <shrug>
>
Yes, my Vulcan persona says this seems very logical. My Missouri
persona says "Show me!". I'd like to see us run 48 hours without this
seg fault, since it was happening about once a day.
Greg
