Open Stack

On Wed, Aug 10, 2016 at 9:21 AM, Jim Rollenhagen <jim at jimrollenhagen.com> wrote:
> Hi Ironicers,
>
> This email serves as a reminder (and a bit of a call to action) about our
> third party CI policy:
> http://specs.openstack.org/openstack/ironic-specs/specs/not-implemented/third-party-ci.html
>
> 1) When I went to find a link to the policy, I realized that it isn't in our
>    developer docs, but only in our specs repo. Can someone volunteer to
>    document this in our dev docs?

Apparently Kurt has this covered: https://review.openstack.org/#/c/353102/

Thanks Kurt!

// jim

> 2) A number of drivers do not have third-party CI up, let alone reporting
>    on patches. Unless someone moves quickly, I strongly suspect the following
>    drivers will be dropped from our tree before the end of Newton. These are
>    the setup.cfg names.
>
>       agent_amt
>       pxe_amt
>       fake_amt
>       agent_iboot
>       pxe_iboot
>       fake_iboot
>       agent_wol
>       pxe_wol
>       fake_wol
>       agent_vbox
>       fake_vbox
>       pxe_vbox
>       pxe_seamicro
>       fake_seamicro
>       pxe_drac
>       fake_drac
>       pxe_snmp
>       fake_snmp
>       pxe_msftocs
>       fake_msftocs
>
>    It's important to note that some ironic folks have taken the burden of
>    maintaining some untested drivers in an out-of-tree repo, however this
>    is not an official OpenStack project, and not part of the ironic governance.
>    https://git.openstack.org/cgit/openstack/ironic-staging-drivers
>
> 3) The SSH drivers (pxe_ssh and agent_ssh) that we use for some testing
>    currently are planned to be dropped.  First, we need to update
>    project-config to make sure all of those jobs are moved to an equivalent
>    *_ipmitool job, and drop the _ssh jobs. Then we can go ahead and remove
>    the drivers. I'd like a volunteer for this as well, but am happy to take
>    it on as needed.
>
> 4) The drivers that use pyghmi (pxe_ipminative and agent_pyghmi) currently are
>    not tested in ironic's CI. We have multiple jobs for each of the ipmitool
>    drivers. Instead of making new jobs, we could just move one of the ipmitool
>    drivers to use the pyghmi drivers (since they both use IPMI, it should be
>    simple to do so). As with above, I'm happy to do this if there are no
>    volunteers.
>
> Thanks for reading (and hopefully volunteering!).
>
> // jim