Merge "Update Docs"
This commit is contained in:
Zuul
Gerrit Code Review
commit
2a18b5a12e
74
doc/source/charts.rst
Normal file
74
doc/source/charts.rst
Normal file
@ -0,0 +1,74 @@
|
||||
======
|
||||
Charts
|
||||
======
|
||||
|
||||
To include any of the custom charts from Browbeat in a scenario, the following lines will have to be included in the python file of the program.
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
import sys
|
||||
import os
|
||||
sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '../reports')))
|
||||
from generate_scenario_duration_charts import ScenarioDurationChartsGenerator # noqa: E402
|
||||
|
||||
The custom charts will appear in the "Scenario Data" section of the Rally HTML report.
|
||||
|
||||
Chart - add_per_iteration_complete_data
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
This plugin generates a stacked area graph for duration trend for each atomic action in an iteration.
|
||||
To include this chart in any scenario, add the following lines at the end of the run() function of the scenario in the python file.
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
self.duration_charts_generator = ScenarioDurationChartsGenerator()
|
||||
self.duration_charts_generator.add_per_iteration_complete_data(self)
|
||||
|
||||
The graphs will appear under the "Per iteration" section of "Scenario Data" in the Rally HTML report.
|
||||
The resulting graphs will look like the images below.
|
||||
|
||||
.. image:: images/Per_Iteration_Duration_Stacked_Area_Chart/Iteration1.png
|
||||
:alt: Iteration 1 Chart
|
||||
|
||||
.. image:: images/Per_Iteration_Duration_Stacked_Area_Chart/Iteration2.png
|
||||
:alt: Iteration 2 Chart
|
||||
|
||||
.. image:: images/Per_Iteration_Duration_Stacked_Area_Chart/Iteration3.png
|
||||
:alt: Iteration 3 Chart
|
||||
|
||||
.. image:: images/Per_Iteration_Duration_Stacked_Area_Chart/Iteration4.png
|
||||
:alt: Iteration 4 Chart
|
||||
|
||||
Chart - add_duplicate_atomic_actions_iteration_additive_data
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
This plugin generates line graphs for atomic actions that have been executed more than once in the same iteration.
|
||||
To include this chart in any scenario, add the following lines at the end of the run() function of the scenario in the python file.
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
self.duration_charts_generator = ScenarioDurationChartsGenerator()
|
||||
self.duration_charts_generator.add_duplicate_atomic_actions_iteration_additive_data(self)
|
||||
|
||||
The graphs will appear under the "Aggregated" section of "Scenario Data" in the Rally HTML report.
|
||||
The resulting graphs will look like the images below.
|
||||
|
||||
.. image:: images/Duplicate_Atomic_Actions_Duration_Line_Chart.png
|
||||
:alt: Duplicate Atomic Actions Duration Line Chart
|
||||
|
||||
Chart - add_all_resources_additive_data
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
This plugin generates a line graph for duration data from each resource created by Rally.
|
||||
To include this chart in any scenario, add the following lines at the end of the run() function of the scenario in the python file.
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
self.duration_charts_generator = ScenarioDurationChartsGenerator()
|
||||
self.duration_charts_generator.add_all_resources_additive_data(self)
|
||||
|
||||
The graphs will appear under the "Aggregated" section of "Scenario Data" in the Rally HTML report.
|
||||
The resulting graphs will look like the images below.
|
||||
|
||||
.. image:: images/Resource_Atomic_Actions_Duration_Line_Chart.png
|
||||
:alt: Resource Atomic Actions Duration Line Chart
|
||||
@ -15,7 +15,7 @@ Contents:
|
||||
installation
|
||||
usage
|
||||
plugins
|
||||
ci
|
||||
charts
|
||||
developing
|
||||
contributing
|
||||
|
||||
|
||||
@ -42,31 +42,10 @@ On the Undercloud
|
||||
[stack@undercloud ~]$ git clone https://github.com/openstack/browbeat.git
|
||||
[stack@undercloud ~]$ source stackrc
|
||||
[stack@undercloud ~]$ cd browbeat/ansible
|
||||
[stack@undercloud ansible]$ ./generate_tripleo_inventory.sh -l
|
||||
[stack@undercloud ansible]$ sudo easy_install pip
|
||||
[stack@undercloud ansible]$ sudo pip install ansible
|
||||
[stack@undercloud ansible]$ vi install/group_vars/all.yml # Make sure to edit the dns_server to the correct ip address
|
||||
[stack@undercloud ansible]$ ./generate_tripleo_inventory.sh -l
|
||||
[stack@undercloud ansible]$ ansible-playbook -i hosts.yml install/browbeat.yml
|
||||
[stack@undercloud ansible]$ ansible-playbook -i hosts.yml install/shaker_build.yml
|
||||
|
||||
.. note:: Your default network might not work for you depending on your
|
||||
underlay/overlay network setup. In such cases, user needs to create
|
||||
appropriate networks for instances to allow them to reach the
|
||||
internet. Some useful documentation can be found at:
|
||||
https://access.redhat.com/documentation/en/red-hat-openstack-platform/11/single/networking-guide/
|
||||
|
||||
(Optional) Clone e2e-benchmarking repository and deploy benchmark-operator
|
||||
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
e2e-benchmarking is a repository that is used to run workloads to stress an Openshift
|
||||
cluster, and is needed to run shift-on-stack workloads in Browbeat.
|
||||
|
||||
To enable the e2e-benchmarking repository to be cloned and benchmark-operator to be deployed,
|
||||
set install_e2e_benchmarking: true in ansible/install/group_vars/all.yml.
|
||||
After that, add the kubeconfig paths of all your Openshift clusters in the ansible/kubeconfig_paths
|
||||
file. Move the default kubeconfig file(/home/stack/.kube/config) to another location so that it isn't
|
||||
used for all Openshift clusters. After that, run the command mentioned below.
|
||||
|
||||
[stack@undercloud ansible]$ ansible-playbook -i hosts.yml install/browbeat.yml
|
||||
|
||||
(Optional) Install Browbeat instance workloads
|
||||
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
@ -100,7 +79,7 @@ has been installed. To skip directly to this task execute:
|
||||
|
||||
::
|
||||
|
||||
$ ansible-playbook -i hosts install/browbeat.yml --start-at-task "Check browbeat_network"
|
||||
$ ansible-playbook -i hosts.yml install/browbeat.yml --start-at-task "Check browbeat_network"
|
||||
|
||||
|
||||
|
||||
@ -111,28 +90,8 @@ has been installed. To skip directly to this task execute:
|
||||
|
||||
::
|
||||
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts install/collectd.yml
|
||||
[stack@undercloud-0 ansible]$ ansible-playbook -i hosts.yml install/collectd.yml
|
||||
|
||||
(Optional) Install Rsyslogd logging with aggregation (not maintained)
|
||||
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
|
||||
First configure the values rsyslog values and elasticsearch parameters in
|
||||
`ansible/install/group_vars/all.yml`. If you have a large number of hosts
|
||||
deploying an aggregator using `ansible/install/rsyslog-aggregator.yml`
|
||||
is strongly suggested. If you have a small scale, change the value
|
||||
rsyslog_forwarding in `all.yml` to `false`. Once things are configured
|
||||
to your liking deploy logging on the cloud using the `rsyslog-logging.yml`
|
||||
playbook.
|
||||
|
||||
Firewall configuration for the aggregator is left up to the user. The logging
|
||||
install playbook will check that the aggregator is up and the port is open if
|
||||
you deploy with aggregation.
|
||||
|
||||
::
|
||||
|
||||
[stack@ospd ansible]$ vim install/group_vars/all.yml
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts install/rsyslog-aggregator.yml
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts install/rsyslog-logging.yml
|
||||
|
||||
(Optional) Install Browbeat Grafana dashboards
|
||||
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
@ -144,16 +103,16 @@ http://docs.grafana.org/http_api/auth/#create-api-token
|
||||
|
||||
::
|
||||
|
||||
[stack@ospd ansible]$ # update the vars and make sure to update grafana_apikey with value
|
||||
[stack@ospd ansible]$ vi install/group_vars/all.yml
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts install/browbeat.yml # if not run before.
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts install/grafana-dashboards.yml
|
||||
[stack@undercloud ansible]$ # update the vars and make sure to update grafana_apikey with value
|
||||
[stack@undercloud ansible]$ vi install/group_vars/all.yml
|
||||
[stack@undercloud ansible]$ ansible-playbook -i hosts.yml install/browbeat.yml # if not run before.
|
||||
[stack@undercloud ansible]$ ansible-playbook -i hosts.yml install/grafana-dashboards.yml
|
||||
|
||||
(Optional) Install Browbeat Prometheus/Grafana/Collectd
|
||||
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
|
||||
::
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts install/grafana-prometheus-dashboards.yml
|
||||
[stack@undercloud ansible]$ ansible-playbook -i hosts.yml install/grafana-prometheus-dashboards.yml
|
||||
|
||||
Make sure grafana-api-key is added in the `install/group_vars/all.yml`
|
||||
|
||||
@ -166,24 +125,31 @@ client side and Elasticsearch on the server side. Set the `cloud_prefix` and `es
|
||||
|
||||
::
|
||||
|
||||
[stack@ospd ansible]$ # update the vars
|
||||
[stack@ospd ansible]$ vi install/group_vars/all.yml
|
||||
[stack@ospd ansible]$ # install filebeat
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts common_logging/install_logging.yml
|
||||
[stack@ospd ansible]$ # install and start filebeat
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts common_logging/install_logging.yml -e "start_filebeat=True"
|
||||
[stack@undercloud ansible]$ # update the vars
|
||||
[stack@undercloud ansible]$ vi install/group_vars/all.yml
|
||||
[stack@undercloud ansible]$ # install filebeat
|
||||
[stack@undercloud ansible]$ ansible-playbook -i hosts.yml common_logging/install_logging.yml
|
||||
[stack@undercloud ansible]$ # install and start filebeat
|
||||
[stack@undercloud ansible]$ ansible-playbook -i hosts.yml common_logging/install_logging.yml -e "start_filebeat=True"
|
||||
|
||||
Not mantained (Pre-Pike): Run Overcloud checks
|
||||
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
|
||||
|
||||
(Optional) Install Kibana Visualizations
|
||||
----------------------------------------
|
||||
|
||||
1. Update install/group_vars/all.yml (es_ip) to identify your ELK host.
|
||||
2. Install Kibana Visualizations via Ansible playbook
|
||||
|
||||
::
|
||||
|
||||
[stack@ospd ansible]$ ansible-playbook -i hosts check/site.yml
|
||||
[stack@undercloud ansible]# ansible-playbook -i hosts.yml install/kibana-visuals.yml
|
||||
...
|
||||
|
||||
Your Overcloud check output is located in results/bug_report.log
|
||||
Now navigate to http://elk-host-address to verify Kibana is
|
||||
installed and custom visualizations are uploaded.
|
||||
|
||||
Install Browbeat from your local machine
|
||||
----------------------------------------
|
||||
|
||||
Install Browbeat from your local machine (Not Manintained)
|
||||
----------------------------------------------------------
|
||||
|
||||
This installs Browbeat onto your Undercloud but the playbook is run from your
|
||||
local machine rather than directly on the Undercloud machine.
|
||||
@ -651,17 +617,3 @@ entirely on the number of metrics and your environments capacity. There is a
|
||||
Graphite dashboard included and it is recommended to install collectd on your
|
||||
monitoring host such that you can see if you hit resource issues with your
|
||||
monitoring host.
|
||||
|
||||
(Optional) Install Kibana Visualizations
|
||||
----------------------------------------
|
||||
|
||||
1. Update install/group_vars/all.yml (es_ip) to identify your ELK host.
|
||||
2. Install Kibana Visualizations via Ansible playbook
|
||||
|
||||
::
|
||||
|
||||
[root@dhcp23-93 ansible]# ansible-playbook -i hosts install/kibana-visuals.yml
|
||||
...
|
||||
|
||||
Now navigate to http://elk-host-address to verify Kibana is
|
||||
installed and custom visualizations are uploaded.
|
||||
|
||||
@ -5,6 +5,50 @@ Plugins
|
||||
Rally
|
||||
~~~~~
|
||||
|
||||
Scenario - dynamic-workloads
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
Dynamic workloads are workloads that aim to simulate a realistic Openstack customer environment, by introducing elements of randomness into the simulation. A list of the different workloads that are part of this Browbeat Rally Plugin is mentioned below.
|
||||
|
||||
VM:
|
||||
|
||||
- create_delete_servers: Create 'N' VMs(without floating IP), and delete 'M'
|
||||
randomly chosen VMs from this list of VMs.
|
||||
- migrate_servers: Create 'N' VMs(with floating IP), and migrate 'M' randomly
|
||||
chosen VMs from this list of VMs across computes, before resizing them.
|
||||
- swap_floating_ips_between_servers: Swap floating IPs between 2 servers. Ping
|
||||
until failure after dissociating floating IPs, before swapping them. Ping until
|
||||
success after swapping floating IPs between 2 servers.
|
||||
|
||||
Octavia:
|
||||
|
||||
- create_loadbalancers: Create 'N' loadbalancers with specified 'M' pools and 'K'
|
||||
clients per Loadbalancer.
|
||||
- delete_loadbalancers: Deletes 'M' loadbalancers randomly from 'N' loadbalancers
|
||||
- delete_members_random_lb: Deletes 'M' members from a random loadbalancer
|
||||
|
||||
Trunk(pod simulation):
|
||||
|
||||
- pod_fip_simulation: Simulate pods with floating ips using subports on trunks and
|
||||
VMs. Create 'N' trunks and VMs and 'M' subports per trunk/VM. Ping a random subport
|
||||
of each trunk/VM from a jumphost.
|
||||
- add_subports_to_random_trunks: Add 'M' subports to 'N' randomly chosen trunks. This
|
||||
is to simulate pods being added to an existing VM.
|
||||
- delete_subports_from_random_trunks: Delete 'M' subports from 'N' randomly chosen
|
||||
trunks. This is is to simulate pods being destroyed.
|
||||
- swap_floating_ips_between_random_subports: Swap floating IPs between 2 randomly
|
||||
chosen subports from 2 trunks.
|
||||
|
||||
Provider network:
|
||||
|
||||
- provider_netcreate_nova_boot_ping: Creates a provider Network and Boots VM and ping
|
||||
- provider_net_nova_boot_ping: Boots a VM and ping on random existing provider network
|
||||
- provider_net_nova_delete: Delete all VM's and provider network
|
||||
|
||||
Shift on Stack:
|
||||
|
||||
shift_on_stack: Runs specified kube-burner workload through e2e-benchmarking. e2e-benchmarking is a [repository](https://github.com/cloud-bulldozer/e2e-benchmarking.git) that contains scripts to stress Openshift clusters. This workload uses e2e-benchmarking to test Openshift on Openstack.
|
||||
|
||||
Context - browbeat_delay
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
@ -44,77 +88,3 @@ Scenario - nova_boot_persist_with_network_volume_fip
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
This scenario creates instances with a nic, a volume and associates a floating ip that persist upon completion of a rally run. It is used as a workload with Telemetry by spawning many instances that have many metrics for the Telemetry subsystem to collect upon.
|
||||
|
||||
Charts
|
||||
^^^^^^
|
||||
|
||||
To include any of the custom charts from Browbeat in a scenario, the following lines will have to be included in the python file of the program.
|
||||
.. code-block:: python
|
||||
|
||||
import sys
|
||||
import os
|
||||
|
||||
sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '../reports')))
|
||||
from generate_scenario_duration_charts import ScenarioDurationChartsGenerator # noqa: E402
|
||||
|
||||
The customc charts will appear in the "Scenario Data" section of the Rally HTML report.
|
||||
|
||||
Chart - add_per_iteration_complete_data
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
This plugin generates a stacked area graph for duration trend for each atomic action in an iteration.
|
||||
To include this chart in any scenario, add the following lines at the end of the run() function of the scenario in the python file.
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
self.duration_charts_generator = ScenarioDurationChartsGenerator()
|
||||
self.duration_charts_generator.add_per_iteration_complete_data(self)
|
||||
|
||||
The graphs will appear under the "Per iteration" section of "Scenario Data" in the Rally HTML report.
|
||||
The resulting graphs will look like the images below.
|
||||
|
||||
.. image:: images/Per_Iteration_Duration_Stacked_Area_Chart/Iteration1.png
|
||||
:alt: Iteration 1 Chart
|
||||
|
||||
.. image:: images/Per_Iteration_Duration_Stacked_Area_Chart/Iteration2.png
|
||||
:alt: Iteration 2 Chart
|
||||
|
||||
.. image:: images/Per_Iteration_Duration_Stacked_Area_Chart/Iteration3.png
|
||||
:alt: Iteration 3 Chart
|
||||
|
||||
.. image:: images/Per_Iteration_Duration_Stacked_Area_Chart/Iteration4.png
|
||||
:alt: Iteration 4 Chart
|
||||
|
||||
Chart - add_duplicate_atomic_actions_iteration_additive_data
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
This plugin generates line graphs for atomic actions that have been executed more than once in the same iteration.
|
||||
To include this chart in any scenario, add the following lines at the end of the run() function of the scenario in the python file.
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
self.duration_charts_generator = ScenarioDurationChartsGenerator()
|
||||
self.duration_charts_generator.add_duplicate_atomic_actions_iteration_additive_data(self)
|
||||
|
||||
The graphs will appear under the "Aggregated" section of "Scenario Data" in the Rally HTML report.
|
||||
The resulting graphs will look like the images below.
|
||||
|
||||
.. image:: images/Duplicate_Atomic_Actions_Duration_Line_Chart.png
|
||||
:alt: Duplicate Atomic Actions Duration Line Chart
|
||||
|
||||
Chart - add_all_resources_additive_data
|
||||
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|
||||
|
||||
This plugin generates a line graph for duration data from each resource created by Rally.
|
||||
To include this chart in any scenario, add the following lines at the end of the run() function of the scenario in the python file.
|
||||
|
||||
.. code-block:: python
|
||||
|
||||
self.duration_charts_generator = ScenarioDurationChartsGenerator()
|
||||
self.duration_charts_generator.add_all_resources_additive_data(self)
|
||||
|
||||
The graphs will appear under the "Aggregated" section of "Scenario Data" in the Rally HTML report.
|
||||
The resulting graphs will look like the images below.
|
||||
|
||||
.. image:: images/Resource_Atomic_Actions_Duration_Line_Chart.png
|
||||
:alt: Resource Atomic Actions Duration Line Chart
|
||||
|
||||
@ -4,15 +4,16 @@ Usage
|
||||
|
||||
Run Browbeat performance tests from Undercloud
|
||||
----------------------------------------------
|
||||
For Running the workloads from Undercloud
|
||||
|
||||
::
|
||||
|
||||
$ ssh undercloud-root
|
||||
[root@ospd ~]# su - stack
|
||||
[stack@ospd ~]$ cd browbeat/
|
||||
[stack@ospd browbeat]$ . .browbeat-venv/bin/activate
|
||||
(browbeat-venv)[stack@ospd browbeat]$ vi browbeat-config.yaml # Edit browbeat-config.yaml to control how many stress tests are run.
|
||||
(browbeat-venv)[stack@ospd browbeat]$ ./browbeat.py <workload> #rally, shaker or "all"
|
||||
[root@undercloud ~]# su - stack
|
||||
[stack@undercloud ~]$ cd browbeat/
|
||||
[stack@undercloud browbeat]$ . .browbeat-venv/bin/activate
|
||||
(.browbeat-venv)[stack@undercloud browbeat]$ vi browbeat-config.yaml # Edit browbeat-config.yaml to control how many stress tests are run.
|
||||
(.browbeat-venv)[stack@undercloud browbeat]$ ./browbeat.py <workload> #rally, shaker or "all"
|
||||
|
||||
Running Shaker
|
||||
---------------
|
||||
@ -77,7 +78,7 @@ you:
|
||||
|
||||
::
|
||||
|
||||
$ ansible-playbook -i hosts install/kibana-visuals.yml
|
||||
$ ansible-playbook -i hosts.yml install/kibana-visuals.yml
|
||||
|
||||
Alternatively you can create your own visualizations of specific shaker runs
|
||||
using some simple searches such as:
|
||||
@ -94,7 +95,7 @@ If filebeat is enabled in the browbeat configuration file and filebeat was previ
|
||||
|
||||
::
|
||||
|
||||
$ ansible-playbook -i hosts common_logging/install_logging.yml
|
||||
$ ansible-playbook -i hosts.yml common_logging/install_logging.yml
|
||||
|
||||
as explained in the installation documentation, then
|
||||
|
||||
|
||||
Loading…
Reference in New Issue
Block a user