Open Stack

On 05/01/2016 05:03 PM, Steven Dake (stdake) wrote:
> Ryan had rightly pointed out that when we made the original proposal 
> 9am morning we had asked folks if they wanted to participate in a 
> separate repository.

In Keystone, we are going to more and more repositories all the time.  
We started with everything in Keystone server, then split out the 
python-keystoneclient repo, then keystonemiddleware, and now 
keystoneauth.  Kerberos requires a separate auth repo, too, just due to 
package dependencies.  Multiple repos are not a bad thing.  The Policy 
store, and  the drivers behind identity are all candidates for future 
refactoring.

Splitting a repo is not a big deal, but it is easier to do up front than 
to retool.



I think starting with a separate, but supported repository makes things 
much easier.

Kolla is 2 things:

1.  Creation of Containers for deploying the base openstack services.
2.  Actual deployment of the same

I would argue that the architecture for this should be something like 4 
repos:

1.  Container production.  Assuming a single toolchain here.
2.  Kubernetes deploy
3.  Ansible deploy
4.  Mesos deploy
5.  kolla-deploy-common.

Python in general makes it hard to have more than one upstream library 
from a repo, so you really want to think about what it looks like from 
PyPi first and organize based on that.

If anything can be pulled out into its own repo, it should.

Yeah, it makes development a bit more of a pain, but there are ways to 
mitigate that.  git subprojects might be a painful one, but it is not 
the the only approach.

Over time, I would expect both the Ansible and Kubernetes repos 
themselves to be split into finer repos, with Ansible plugins and 
Kubernetes modules being separately managed.



>
> I don't think a separate repository is the correct approach based upon 
> one off private conversations with folks at summit.  Many people from 
> that list approached me and indicated they would like to see the work 
> integrated in one repository as outlined in my vote proposal email. 
>  The reasons I heard were:
>
>   * Better integration of the community
>   * Better integration of the code base
>   * Doesn't present an us vs them mentality that one could argue
>     happened during kolla-mesos
>   * A second repository makes k8s a second class citizen deployment
>     architecture without a voice in the full deployment methodology
>   * Two gating methods versus one
>   * No going back to a unified repository while preserving git history
>
> I favor of the separate repositories I heard
>
>   * It presents a unified workspace for kubernetes alone
>   * Packaging without ansible is simpler as the ansible directory need
>     not be deleted
>
> There were other complaints but not many pros.  Unfortunately I failed 
> to communicate these complaints to the core team prior to the vote, so 
> now is the time for fixing that.
>
> I'll leave it open to the new folks that want to do the work if they 
> want to work on an offshoot repository and open us up to the possible 
> problems above.
>
> If you are on this list:
>
>   * Ryan Hallisey
>   * Britt Houser
>
>   * mark casey
>
>   * Steven Dake (delta-alpha-kilo-echo)
>
>   * Michael Schmidt
>
>   * Marian Schwarz
>
>   * Andrew Battye
>
>   * Kevin Fox (kfox1111)
>
>   * Sidharth Surana (ssurana)
>
>   *  Michal Rostecki (mrostecki)
>
>   *   Swapnil Kulkarni (coolsvap)
>
>   *   MD NADEEM (mail2nadeem92)
>
>   *   Vikram Hosakote (vhosakot)
>
>   *   Jeff Peeler (jpeeler)
>
>   *   Martin Andre (mandre)
>
>   *   Ian Main (Slower)
>
>   * Hui Kang (huikang)
>
>   * Serguei Bezverkhi (sbezverk)
>
>   * Alex Polvi (polvi)
>
>   * Rob Mason
>
>   * Alicja Kwasniewska
>
>   * sean mooney (sean-k-mooney)
>
>   * Keith Byrne (kbyrne)
>
>   * Zdenek Janda (xdeu)
>
>   * Brandon Jozsa (v1k0d3n)
>
>   * Rajath Agasthya (rajathagasthya)
>   * Jinay Vora
>   * Hui Kang
>   * Davanum Srinivas
>
>
>
> Please speak up if you are in favor of a separate repository or a 
> unified repository.
>
> The core reviewers will still take responsibility for determining if 
> we proceed on the action of implementing kubernetes in general.
>
> Thank you
> -steve
>
>
> __________________________________________________________________________
> 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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20160502/503def21/attachment.html>