Meetings/TechnicalCommittee/Ceilometer Gap Coverage

Concern #1: lack of an explicit project mission statement

• Actions:
  • team wordsmithing of an appropriate mission statement: https://review.openstack.org/87526
• Driver: eglynn
• Milestone targets:
  • Achieved by: juno-1

Concern #2: lack of user- and operator-oriented documentation

• Actions:
  • re-work RDO getting started guide in distro-agnostic form: http://openstack.redhat.com/CeilometerQuickStart
  • discern operator preferences for documentation at summit operator feedback session: http://summit.openstack.org/cfp/details/177
  • joint ceilometer/documentation summit session: http://summit.openstack.org/cfp/details/311
• Driver: ildikov

Concern #3: lack of coverage in integration tests

• Actions:
  • detailed analysis of current coverage https://etherpad.openstack.org/p/ceilometer-tempest-status
  • TL;DR: need to solve "performant only with the storage driver we don't gate on" conumdrum
  • summit design session to address test strategy shortcomings http://summit.openstack.org/cfp/details/152
• Driver: nprivalova

Concern #4: lack of production-readiness of sqlalchemy driver

• Actions:
  • summit design session to address inefficiencies inherent in the current data model: http://summit.openstack.org/cfp/details/163
  • useful to pick Mike Bayer's brains also during this session: http://summit.openstack.org/cfp/details/157
  • goal of allowing Juno-derived distros treat sqlalchemy as a first-class citizen
  • detail on known issues and design considerations: https://etherpad.openstack.org/p/ceilometer-sql
• Blueprints:
  • https://blueprints.launchpad.net/ceilometer/+spec/big-data-sql
• Driver: gordc

Concern #5: lack of participation in Grenade-based upgrade testing

• Actions:
  • grenade integration is fairly clear, no need for an explicit summit design session
  • clarity on extrinsic versus instrinsic goals of ceilometer upgrade testing: http://lists.openstack.org/pipermail/openstack-tc/2014-April/000632.html
• Driver: emilienm/cdent