[openstack-dev] [TripleO] Candidacy for Deployment(TripleO) PTL
robertc at robertcollins.net
Sat Sep 21 01:07:08 UTC 2013
I would like to run for the OpenStack Deployment (TripleO) PTL position.
I am the current (initial) PTL, and have helped shepard the TripleO
project from an interesting proposition back in San Diego 2012 through
to the current set of lean tooling + dedicated uses of other OpenStack
components & a certain amount of mind bending bootstrap logic that we
have :) I'd love the privilege of continuing to shepard the program
forward as we gain adopters, features and streamline performance.
My view on PTL is that it is an enabling position: working with the
team to identify blockers, constraints issues and weak points in the
(design/code/architecture) and then respectively remove/socialise and
On a technical basis I'm familiar with the entire stack we're dealing
with: I've spent years doing low level system administration, network
design and administration, cluster management, Python development
(across CLI, libraries, web APIs and web apps) : there are few
surprises lurking here :).
However as Russell says PTL is much more than technical chops: it's a
social position, looking at program health - are patches stalling, are
contributors getting reviews, are patches suffering churn, do we have
our priorities straight?
Icehouse is the first opportunity we'll have to have a fully
functional story, and even now thats still a fairly aggressive goal :
my focus over the next 6 months will be getting us out of fire-drill
mode and into the steady-state that is needed for confident
deployments... in addition to fleshing out the features required for
Tuskar, overcloud upgrades and undercloud upgrades. That means, in big
* Full CI gating
* Deployable CD gating for our users
* Persistent storage on nodes
* At least active-passive deployment support out of the box
* And close work with Heat to deliver upgrades of the undercloud Heat.
I like Russell's thoughts on having delegated leads for broad projects
: I'm not sure we're at that point in TripleO, but I think it's
something we need to keep an eye on with the merge of Tuskar and
TripleO - while the overall program vision is entirely relevant for
all the projects, there is quite a bit of technical difference between
e.g. diskimage-builder and the Tuskar horizon plugin, so we need to be
sensitive to both the bandwidth problem and context problem, to ensure
all our projects are well supported.
What have I missed? I don't know, but I'm sure there is more you'd
like to know about me or what I'd do as PTL for the next 6 months :).
Ask and I shall answer :).
( Note that that a) is missing the TripleO repositories - doh, will
investigate when I'm home, and b) still thinks I'm at Canonical, which
Robert Collins <rbtcollins at hp.com>
HP Converged Cloud
More information about the OpenStack-dev