Cloned from launchpad blueprint https://blueprints.launchpad.net/ceilometer/+spec/predictive-failure-metric.
We would like to have a metric to monitor and predict host failures. This would require changes to the Ceilometer code:
These are the changes required in Ceilometer files (6 lines total).
Add line to ceilometer entry_points.txt file:
host_pfa = ceilometer.compute.notifications.cpu:HostPfa
Add line to ./compute/notifications/cpu.py
"""Handle predictive failure alert message."""
metric = 'host.pfa'
sample_type = sample.TYPE_GAUGE
unit = 'bool'
Specification URL (additional information):