Re: client side internal request support

On Mon, 2002-08-12 at 16:30, Adrian Chadd wrote:
> On Mon, Aug 12, 2002, Robert Collins wrote:
>
> > I get identical results with this code, and with HEAD with polygraph.
> > AFAIK I've caused no regressions.
> >
> > So, anyone for a little bit of peer review?
> >
> > I'd really like to get this into HEAD asap...
>
> I'll be running a -HEAD "production" proxy next week, which I can
> run this on. polygraph doesn't really seem to test error/boundary
> conditions very well.. :-)

True. We -really- need a polygraph style (multiple clients, multiple
servers) test bed for squid. So we can build a list of all the corner
cases we know of and have the test bed probe for them.

> I'll be looking at it then. I'll commit a couple of things to -HEAD to
> fix the missing-data-near-EOF issue just so there's a reference to
> look at for the fix before this ESI stuff goes in.

I don't think I've fixed in in this ESI stuff. I've fixed one instance,
but there are a few similar ones IIRC (where the client reply code
assumes that what has been recieved from the store has been forwarded to
the client).

Rob