Open Stack

On Mon, Dec 7, 2015 at 11:07 PM, Ken'ichi Ohmichi <ken1ohmichi at gmail.com>
wrote:

> Hi,
>
> Today, we have virtual API document sprint and we are facing a problem
> related to HTTP203 (Non-Authoritative Information).
> The api-site(http://developer.openstack.org/api-ref-compute-v2.1.html)
> contains HTTP203 as one of valid status codes for each API operation,
> but nova implementation code doesn't contain the status code and
> invalid from some nova developers' viewpoints.
>
> In addition, tempest also doesn't consider HTTP203 cases at all now.
> If nova returns HTTP203 response, tempest consider the response as
> invalid on its own tests.
>
> I am imaging this status code comes intentionally for proxying thing,
> so I feel it is nice to keep HTTP203 description in the api-site.
> If so, I would be nice to have a gate job which makes HTTP203 cases
> and tempest needs to be updated for covering these cases.
>

Interesting. I recently went through all the WADL because they had 200 203
in combination, and separated out all the 203 into separate responses. This
change was helpful for ensuring the accuracy of the future migration of the
docs. The 203 codes were from the Compute API specs originally written 5
years ago, and then as people copied from Compute the 203 spread.

I think what you're saying is that providers could use a 203 for some use
case, but a proxy server would have to provide the 203?

Anne




>
> Any comments are welcome.
>
> Thanks
> Ken Ohmichi
>
> __________________________________________________________________________
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>



-- 
Anne Gentle
Rackspace
Principal Engineer
www.justwriteclick.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20151208/287611a0/attachment.html>