This document describes new section in task config.
This changes needed for reliability and density testing
of OpenStack clusters.
Change-Id: I39fed4b820710b74cf902363dd1c683622f89ca7
This adds SLA plugin that finds minimum and maximum duration of
iterations completed without errors during Rally task execution.
Assuming that minimum duration is 100%, it calculates
performance degradation against maximum duration.
Example config:
sla:
performance_degradation:
max_degradation: 75
Spec: sla_pd_plugin
Change-Id: Ieedba7be72364f5599a3c0cf79f5f494a7391ea0
This is document describes Rally SLA performance
degradation plugin.
This is plugin needed for reliability and density testing
of OpenStack clusters.
More info about testing available in openstack/performance-docs
Change-Id: Ic23d4e21e5bacbc949990364956edf06ac7e99f7
- Since we supports plugin bases now, lets use shorter name and rename
"file-exporter" to just "file".
- this patch updates the help message and docstring for connection string
for "file-exporter"/"file" plugin
- TaskExporter located in the task module, so it is no need to store word
"Task" in the class name
Change-Id: Ic8a9a765e7365ff108c88a67995b9bf26736e54e
Changes:
* rename spec to "class-based-scenarios.rst"
* fix JSON code example format to make it valid
and correctly displayed in HTML
* highlight names of modules
Change-Id: Ice11cff93ee3d64f51afdf2aa33f4d1ed25cb6b3
This spec proposes improvements related to atomic actions:
* rename term "atomic actions" into "actions"
* store actions as list of dicts instead of OrderedDict
* introduce nested actions saved in "children" container
* start and end timestamps
* get rid of duration time (since we have timestamps)
Blueprint: improve-atomic-actions
Change-Id: Ic22c00cf40c36205628dd3bf18b0f09f92fad8c0
These will be used to do out-of-band cleanup in the near future, so
they are no longer specific to the cleanup context.
Change-Id: I5b5c7c89b04a080c487936cc3df0df24929ced31
Implements: blueprint cleanup-refactoring
Spec 'improve_scenario_output_format.rst' is implemented,
so it is moved to doc/specs/implemented/
Also there were some updates to this spec to conform
improvements that are actually done:
* Each output data set has 'description' text passed to
Scenario.add_output() while adding the output.
Description is displayed in HTML report under chart title.
* Method Scenario.add_output() has changed (schema
verification is added)
* Subtab 'Detailed' is renamed to 'Per iteration'
Change-Id: Iddf0b0f5edea8002493d47131994f46202363ef5
The patch introduces Workload model
instead of regular dict that was used to describe
scenario configuration in TaskEngine.
As a side effect it allows to fix subtask representaion
in generated reports and other rally outputs.
Change-Id: I62148707c89f575f83da11e34d0d193e716e8249
This proposes to make resource type converters into proper plugins,
stop depending on OpenStack as much as is possible without modifying
validation, and make some improvements to handle an edge case
converting images.
Change-Id: If078332230df417b3967f086815796c7a96fa882
We have issue with current format there are 2 things called "task",
"task" is file with bunch of "task" that are run.
I would prefer to split these 2 things:
task - is file that is run by "rally task" command
subtasks - is actually single test case described in task file
Change-Id: Ib1a2f39bb91cfb5b0ba0b3d15be53d3a3241bfc7
Rally has billions of plugins and it becomes really hard for newbies
to understand how Rally code is organized and how it works.
So let's split framework and plugins
Example https://review.openstack.org/#/c/176092/
Change-Id: I6980c0dcbeeabf79e774c69cd27e5e4efca0fc50
Most of OpenStack projects have specs.
Specs are documents in .rst format with fixed structure.
They allow developers to use standard gerrit code review approach
to discuss and align future changes in project.
This patch includes:
- doc/specs directory
- doc/specs readme files
- spec template.rst
- unit tests that checks that specs are regarding template
Change-Id: I31e0dbb8bcbc68ca40c67c643d44b27359906b9c
