On Mon, Sep 20, 2010 at 01:16:51PM -0400, Frank McGeough wrote:
> >
> > On Mon, Sep 20, 2010 at 9:31 AM, Daniel Stenberg <daniel_at_haxx.se> wrote:
> >> On Mon, 20 Sep 2010, Alan Wolfe wrote:
> >>
> >>> To expand on that, i'm thinking if you are making a request with the same
> >>> URL after disconnecting as you made before disconnecting, the response could
> >>> be in the cache and so you are getting the same results back.
> >>
> >> No, they can't.
> >>
> >> libcurl has no HTML/data cache support so if it gets contents, it gets it
> >> over the network.
> >>
> >> For a more detailed analysis exactly what's going on in Johnny's case, we
> >> need detailed info of what actually happens that "once in every 80
> >> requests".
> >>
> >> --
> On Sep 20, 2010, at 1:03 PM, Alan Wolfe wrote:
> >
> > I was thinking of caching happening at the OS level. Is that a possibility?
>
> Anything is possible but there is no o/s that I'm aware of that does
> that.

Solaris does this, but the support isn't enabled by default. See the
nca(1) manpage for more details. I would be surprised if this is the
case that's being hit, though.

-j
-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-library
Etiquette: http://curl.haxx.se/mail/etiquette.html
Received on 2010-09-23