Index ¦ Archives ¦ Atom

Placement Update 18-26

Posted on: Fri 29 June 2018

Category: openstack – Tags: placement

This is placement update 18-26, a weekly update of ongoing development related to the OpenStack placement service. This is an expand version.

For the next few weeks the "Specs" section will not be present. When we start reviewing specs for Stein, I'll add it back in.

Most Important

Nested allocation candidates are getting very close, but remain a critical piece of functionality. After that is making sure that we are progressing on the /reshapher functionality and bringing the various virt drivers into line with all this nice new functionality (which mostly means ProviderTrees).

All that nice new functionality means bugs. Experiment. Break stuff. Find bugs. Fix them.

Speaking of bugs: a collection of problems—not covered by tests— with consumer generations was discovered this week. Also a problem with the limit functionality on GET /allocation_candidates and how that works when force_hosts is being used. Fixes are in progress but these issues are a strong indicator of our need to make sure that we are experimenting with things: it's where features integrate with each other that problems are showing up.

What's Changed

Quite a lot of bug fixes and bug demonstrations have merged in this week, but features mostly stable.

Bugs

Questions

As far as I can tell there was no discussion on last week's question, so here it is again:

In IRC [last week] we had an extensive discussion about being able to set custom resource classes on the resource provider representing a compute node, outside the virt driver. At the moment the virt driver will clobber it. Is this what we always want?

Main Themes

Documentation

This is a section for reminding us to document all the fun stuff we are enabling. Open areas include:

  • Documenting optional placement database. A bug, 1778227 has been created to track this. This has started, for the install docs, but there are more places that need to be touched.

  • "How to deploy / model shared disk. Seems fairly straight-forward, and we could even maybe create a multi-node ceph job that does this - wouldn't that be awesome?!?!", says an enthusiastic Matt Riedemann.

  • The when's and where's of re-shaping and VGPUs.

Nested providers in allocation candidates

As far as I can tell the main thing left here is to turn it on in a microversion. That code is at:

Consumer Generations

There are new patches in progress on this, related to the bugs that were discovered:

There are a patches left on the consumer generation topic, some tidy ups, and some stuff related to healing allocations:

Is someone already working on code for making use of this in the resource tracker?

Reshape Provider Trees

This allows moving inventory and allocations that were on resource provider A to resource provider B in an atomic fashion. The blueprint topic is:

There are WIPs for the HTTP parts and the resource tracker parts, on that topic.

Mirror Host Aggregates

This needs a command line tool:

Extraction

A while back, Jay made a first pass at an os-resource-classes, which needs some additional eyes on it. I personally thought it might be heavier than required. If you have ideas please share them.

An area we will need to prepare for is dealing with the various infra and co-gating issues that will come up once placement is extracted.

We also need to think about how to manage the fixtures currently made available by nova that we might need or want to use in placement. Some of them might be worth sharing. How should we do that?

Other

18 entries last week. 24 now.

End

A butterfly just used my house as a south to north shortcut. That'll do.

Posted on: Fri 29 June 2018

Category: openstack – Tags: placement

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