Index ¦ Archives ¦ Atom

Placement Update 19-14

Placement update 19-14 is here. There will be no 15, 16 or 17 due to various bits of travel. There will be some PTG-related summaries.

Most Important

The Virtual Pre-PTG is in full swing and making some good progress towards making sure that we only hit the hard stuff at the in-person PTG. Apologies if it has been a bit overwhelming. The hope is that by paying the price of a bit more whelm now we will have less whelm at the PTG. If you have questions, please ask.

Links to all the threads are in the PTG etherpads:

There's also a retrospective etherpad.

What's Changed

  • The 0.12.0 release of os-traits is pending. It switches that package to using the independent release policy. os-resource-classes will get the same treatment when we next need to release it.

Specs/Features

There are four specs in flight in the placement repo and one pending to be ported over from nova:

There are also several nova-specs that were visited in the nova spec review day earlier this week. Some are listed below.

Bugs

osc-placement

osc-placement is currently behind by 13^w11 microversions. -1 since last week. Support for 1.19 has just merged. Oh wait, no, -3. 1.21 has just merged. There was nothing to do for 1.20.

Pending changes:

Main Themes

More work remains in the pre-PTG discussions to try to drive towards some themes. The specs above capture some of it, but it appears like a lot of the work will be a) supporting other projects doing things with Placement, b) fixing bugs they discover.

Other Placement

Mostly specs in progress (listed above) for now.

  • https://review.openstack.org/#/c/645255/ This is a start at unit tests for the PlacementFixture. It is proving a bit "fun" to get right, as there are many layers involved. Making sure seemingly unrelated changes in placement don't break the nova gate is important. Besides these unit tests, there's discussion on the PTG etherpad of running the nova functional tests, or a subset thereof, in placement's check run.

    On the one hand this is a pain and messy, but on the other consider what we're enabling: Functional tests that use the real functionality of an external service (real data, real web requests), not stubs or fakes.

    There's a pre-PTG thread for this.

  • https://review.openstack.org/641404 Use code role in api-ref titles

  • https://review.openstack.org/#/q/topic:refactor-classmethod-diaf A sequence of refactorings, based off discussion in yet another pre-PTG thread.

Other Service Users

New discoveries are added to the end. Merged stuff is removed.

End

There's a lot going on.

© Chris Dent. Built using Pelican. Theme by Giulio Fidente on github.