On Mon, 23 Mar 2009, Dan Fandrich wrote:

> Looks like a legitimate problem to me. Does this patch work for you?
>
> --- easy.c 9 Mar 2009 12:21:47 -0000 1.134
> +++ easy.c 24 Mar 2009 05:10:01 -0000
> @@ -1095,8 +1095,8 @@
> if(ret1 == -1)
> return CURLE_AGAIN;
>
> - if(n1 == -1)
> - return CURLE_RECV_ERROR;
> + if(ret1 != CURLE_OK)
> + return ret1;
>
> *n = (size_t)n1;

Hey Dan,

Will you commit this or is there anything that prevents it from going in?

-- 
  / daniel.haxx.se