Open Stack

-----Original Message-----
From: Matthew Thode <prometheanfire at gentoo.org>
Reply: prometheanfire at gentoo.org <prometheanfire at gentoo.org>,
OpenStack Development Mailing List (not for usage questions)
<openstack-dev at lists.openstack.org>
Date: September 7, 2016 at 08:12:15
To: OpenStack Development Mailing List (not for usage questions)
<openstack-dev at lists.openstack.org>
Subject:  [openstack-dev] [requirements][FFE] global-requirements
update to positional to 1.1.1

> https://review.openstack.org/366631
>
> The combination of oslo.context 2.9.0 + positional 1.0.1 (which is the
> current minimum requirement) results in various unit test failures in
> barbican, related to parsing of request headers in generated contexts
> for unit testing. Updating to 1.1.1 resolves this issue.

So I take it someone has verified that the request headers used in the
faked(?) contexts in these tests will never be seen in the real world
then? I looked at the tests that James linked in the barbican channel
when they were looking for help debugging this and those looked like
*functional* tests, not unit tests. That doesn't give me any
confidence that this is *just* a testing issue.

> This is specifically affecting barbican and RDO testing (from discussion
> and the review).

I believe this is also affecting Ubuntu's backing of Newton-3, but
James can correct me if I'm wrong.

> The reason I think an FFE is needed is because downstream packagers,
> while encouraged to package based on upper-constraints sometimes don't.
> Meaning they'd miss something like this.
>
> Arguments against are that this will have knock on effects down the line
> (will require re-releases and re-re-releases because of updating things
> like keystone (this is deep in the depgraph)), so is bad from a release
> team work point of view. Also, I think this just effects testing, so
> the impact of this is more minor than something more serious (not JUST
> breaking testing).

That's one aspect of the conversation. The other is that we *claim* to
support a minimum version of positional which we don't actually
support (and it seems like we either can't or won't). We should have
the *correct* minimum version specified. While I think this is the
*correct* approach, I also realize that the release team is probably
against this for more *pragmatic* reasons and I respect those and the
release team immensely. I'd like them to weigh in here as well.

--
Ian Cordasco