Re: [PATCH 1/2] t5516/t5601: avoid using `localhost` for failing HTTPS requests

From: Jeff King <peff@peff.net>
To: Johannes Schindelin via GitGitGadget <gitgitgadget@gmail.com>
Cc: git@vger.kernel.org, Derrick Stolee <derrickstolee@github.com>,
	Johannes Schindelin <johannes.schindelin@gmx.de>
Subject: Re: [PATCH 1/2] t5516/t5601: avoid using `localhost` for failing HTTPS requests
Date: Mon, 31 Oct 2022 22:03:58 -0400	[thread overview]
Message-ID: <Y2B+jn4goN4OIcwL@coredump.intra.peff.net> (raw)
In-Reply-To: <Y2BYKxxkG57XAV/1@coredump.intra.peff.net>

On Mon, Oct 31, 2022 at 07:20:11PM -0400, Jeff King wrote:

> > - The DNS resolver, or the local firewall, might take a substantial
> >   amount of time (or forever, whichever comes first) to fail to connect,
> >   slowing down the test cases unnecessarily.
> 
> Right. I think we assume that DNS resolution of localhost is fast-ish,
> as we use it in other https tests. But I could certainly imagine a local
> firewall causing issues (especially as this is real port 443, whereas
> our other tests are usually high ports).

Actually, I am wrong about DNS here. We use a bare 127.0.0.1 in
lib-httpd.sh, so DNS may indeed be the source of the issue here. That
_might_ mean that using the bare IP would be safe here, but I would not
want to bet on it. Using different port numbers, plus not expecting a
listener on the other end, might cause different outcomes than we see in
the other tests.

I do think moving these tests to use a real http server is a better
solution, though. I'll provide a patch in a moment.

-Peff