diff --git a/candidates/pike/OpenStackAnsible/andymccr.txt b/candidates/pike/OpenStackAnsible/andymccr.txt new file mode 100644 index 00000000..df9b577a --- /dev/null +++ b/candidates/pike/OpenStackAnsible/andymccr.txt @@ -0,0 +1,49 @@ +I would like to announce my candidacy for PTL of the OpenStack-Ansible project +for Pike. + +As a community, we achieved a lot during this short cycle and I would like to +continue this momentum into the Pike cycle. I’m motivated to keep working to +improve the stability and usability of the project, and to continue to work with +this great community. + +During the Ocata cycle we achieved on the goals of standardizing testing +further, adding upgrade testing, and starting the work to separate out the +stages of the deployment. This serves as a great point to innovate further and +improve the way in which we do upgrades and deployments. I’d also like to +highlight the great work that happened to add CentOS 7 support, and tested Ceph +integration. + +This cycle I would like to focus on standardization within the roles, as well as +improving the manageability of the project as a whole. The following are some +specific items that I believe will help us to align with these focus points: + +1. Standardize deployments of API services. + By utilizing a web server (Apache or NGINX) we can deploy API services + in a uniform way. This will allow us to be flexible with regards to IPv6 and + SSL integration (at the API server layer), whilst ensuring that all roles can + implement this in the same way. Additionally, we can adopt a mature upstream + role, for the web server, which will reduce our code footprint. + +2. Reduce the templated files we carry. + Several template files that are carried (specifically policy.json files) can + be retrieved from the upstream repositories, and still apply the + "config_overrides" for those files. This will reduce the need to mirror + upstream changes to static files, and ensure we are running the appropriate + version of the file for the specific release. + +3. Closer integration with upstream projects. + Deployment projects carry their own challenges each cycle - we need to ensure + we further the project itself, whilst staying up to date with changes made + in upstream projects. We've done reasonable job of this, but it is often + inconsistent because we have no process or initiative around this. We + discussed briefly at the summit around having a tighter coupling of community + member to specific upstream project. I'd like to trial this during the Pike + cycle, with the aim being to improve the overall quality and consistency of + the roles, whilst keeping up to date with upstream changes. + +I look forward to working with everybody through the Pike cycle! + +Thanks again, +Andy McCrae + +IRC: andymccr