TrueSight Pulse OpenStack Plugin

This plugin grabs metrics from the openstack node where it is started and parses the data to be able to integrate into boundary. To be able to start, ceilometer should be well configured on the machine and credentials needs to be created. Additional metrics can be added from the ceilometer by editing the plugin.py and adding or replacing different mapping tuple objects.

Prerequisites

Supported OS

OS	Linux	Windows	SmartOS	OS X
Supported	v

• OpenStack version Mitaka
• Ceilometer >= 2.3.0

For Centos/RHEL/Ubuntu

• To install OpenStack on centos v7.x/rhel v7.x see instructions
• To install OpenStack on ubuntu v14.04(trusty) see instructions

TrueSight Pulse Meter versions v4.0 or later

• To install new meter go to Settings->Installation or [see instructons|https://help.boundary.com/hc/en-us/sections/200634331-Installation].
• To upgrade the meter to the latest version - [see instructons|https://help.boundary.com/hc/en-us/articles/201573102-Upgrading-the-Boundary-Meter].

Plugin Setup

For CentOS 7.x & RHEL 7.x and Ubuntu v14.04(trusty)

To get ceilometer configuration such as host, port and more, goto below file.

Default path for ceilometer configuration file is "/etc/ceilometer/ceilometer.conf"

Plugin Configuration Fields

Field Name	Description
service_tenant	The tenant to get into the service panel for OpenStack
service_endpoint	The endpoint to get into the service panel for OpenStack
service_user	The user to get into the service panel for OpenStack
service_timeout	The timeout to get into the service panel for OpenStack
service_password	The password to get into the service panel for OpenStack
pollInterval	The polling interval (in milliseconds) to call the openStack collector. Default value for this is 60000 ms, we recommend not have smaller value.

Metrics Collected

Metric Name	Description
OS_CPUUTIL_AVG	Average openstack CPU utilization on the running node
OS_CPUUTIL_SUM	Summary of total openstack CPU utilization
OS_CPUUTIL_MIN	The minimum openstack CPU utilization by VMs running on the node
OS_CPUUTIL_MAX	The maximum openstack CPU utilization by VMs running on the node
OS_CPU_AVG	Average CPU time used by the openstack VMs
OS_CPU_SUM	Total CPU time used by the openstack VMs
OS_INSTANCE_EVENT_SUM	Total running instances events
OS_INSTANCE_EVENT_MAX	The maximum number of instances events
OS_VOLUME_EVENT_SUM	Total volumes created events
OS_VOLUME_EVENT_AVG	Average volumes created events
OS_IMAGE_SIZE_SUM	The total amount of space used by the created images
OS_IMAGE_SIZE_AVG	Average amount of space used by the created images
OS_DISK_READ_RATE_SUM	The total amount of disk read rate on all VMs running on the node
OS_DISK_READ_RATE_AVG	Average amount of disk read rate on all VMs running on the node
OS_DISK_WRITE_RATE_SUM	The total amount of disk write rate on all VMs running on the node
OS_DISK_WRITE_RATE_AVG	Average amount of disk write rate on all VMs running on the node
OS_NETWORK_IN_BYTES_SUM	The total amount of network incoming bytes to all VMs running on the node
OS_NETWORK_IN_BYTES_AVG	Average amount of network incoming bytes to all VMs running on the node
OS_NETWORK_OUT_BYTES_SUM	The total amount of network outgoing bytes from all VMs running on the node
OS_NETWORK_OUT_BYTES_AVG	Average amount of network outgoing bytes from all VMs running on the node
OS_MEMORY_RESIDENT_SUM	The resident memory sum consumed by the VMs from the Physical Host
OS_MEMORY_RESIDENT_AVG	The resident memory avg consumed by the VMs from the Physical Host