This document describes the steps to test zero downtime
upgrade process.
Change-Id: Ieafd65da959f6153524f1f3ab0522025b94cbef4
Co-Authored-by: Anusha Unnam <anusha.unnam@intel.com>
Update the documents, making the following changes:
* Reference Fedora 24 instead of the outdated Fedora 21
* Don't reference the removed 'rejoin-stack' script
* Let devstack do as much work for us as possible (NUMATopologyFilter)
* Don't overcomplicate anything (installing emacs, using non-standard
devstack install dir)
* Use '$' prefix for non-root commands
* Update XML with latest examples
* Use the 'openstack' client, rather than the 'nova' client
* Random formatting fixes
...and, finally, stop hating on poor, innocent vi.
Change-Id: Idc89a5c0617ccf7044e1cfc2b5d7f5ff4004b5bf
Move all scheduler options into their one of two groups. Many of the
options are simply renamed to remove their 'scheduler_' prefix, with
some exceptions:
* scheduler_default_filters -> enabled_filters
* scheduler_baremetal_default_filters -> baremetal_enabled_filters
* scheduler_driver_task_period -> periodic_task_interval
* scheduler_tracks_instance_changes -> track_instance_changes
Change-Id: I3f48e52815e80c99612bcd10cb53331a8c995fc3
Co-Authored-By: Stephen Finucane <sfinucan@redhat.com>
Implements: blueprint centralize-config-options-ocata
The libvirt NUMA testing docs were written before the final code
was merged, so the example JSON shown for the database fields
did not match reality. It was also hard to read the JSON since
there were no line breaks & indentation. To make matters even
worse, the final example was just plain wrong - the data shown
was a mixture of a guest with 1 NUMA node and a guest with 2
NUMA nodes.
Change-Id: I51e4df2d9fe252ba65f6a37d84f125e18b44e239
* Explain these docs are for trunk (copied from ironic)
* All the docs in this repo are meant to be developer docs, so having a
devref inside of the docs is redundant and just makes the docs more
complicated to navigate. Move everything out of the devref folder and
link to everything from main index.
* Move man pages into separate section. The man pages are pretty sparse
* right now, we should either make them useful or just delete them
* Remove dead docs from unused_docs list in doc/source/conf.py
* Shuffle docs landing page, move common referees to the top (API,
hypervisor support matrix), Add a introduction section and more. The
hope is the updated layout makes this document easier to navigate.
* Use maxdepth of 1
* Rename a few sections with what are hopefully better names
The next step is to prune out outdated documents and further cleanup
this page.
Change-Id: Iff453e47ccc902a0e72b1a5f6ce1ee939ff3a1a0
Jenkins