Description of problem:
The cpu_l3_cache measures aren't collected by default, because they aren't in polling.yaml file of ceilometer_agent_compute container. When adding cpu_l3_cache to polling.yaml and restarting the service, the measures start being collected, but if rebooting the machine, the changes are lost.
Version-Release number of selected component (if applicable):
python-ceilometerclient-2.9.0-0.20170811140520.3b4e35a.el7ost.noarch
python-ceilometermiddleware-1.1.0-0.20170824160203.0cea730.el7ost.noarch
openstack-ceilometer-collector-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
openstack-ceilometer-compute-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
openstack-ceilometer-common-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
openstack-ceilometer-notification-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
openstack-ceilometer-ipmi-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
python-ceilometer-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
openstack-ceilometer-central-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
puppet-ceilometer-11.3.1-0.20170923033230.767c543.el7ost.noarch
openstack-ceilometer-api-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
openstack-ceilometer-polling-9.0.2-0.20171013153832.ea962e7.el7ost.noarch
How reproducible:
100%
Steps to Reproduce:
1. start a vm on Intel baremetal with cpu_l3_cache option, libvirt configured to cmt
2.Add cpu_l3_cache to polling.yaml:
docker exec -ti ceilometer_agent_compute cat /etc/ceilometer/polling.yaml | tail -1
- cpu_l3_cache
3. Restart the service:
docker exec -ti ceilometer_agent_compute pkill -HUP -f "ceilometer.*master process"
4. Watch measures being collected after 10 min
5. Reboot compute
Actual results:
Measures are not being collected after reboot
Expected results:
Measures are supposed to remain being collected
Additional info:
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.
https://access.redhat.com/errata/RHBA-2018:0243
Description of problem: The cpu_l3_cache measures aren't collected by default, because they aren't in polling.yaml file of ceilometer_agent_compute container. When adding cpu_l3_cache to polling.yaml and restarting the service, the measures start being collected, but if rebooting the machine, the changes are lost. Version-Release number of selected component (if applicable): python-ceilometerclient-2.9.0-0.20170811140520.3b4e35a.el7ost.noarch python-ceilometermiddleware-1.1.0-0.20170824160203.0cea730.el7ost.noarch openstack-ceilometer-collector-9.0.2-0.20171013153832.ea962e7.el7ost.noarch openstack-ceilometer-compute-9.0.2-0.20171013153832.ea962e7.el7ost.noarch openstack-ceilometer-common-9.0.2-0.20171013153832.ea962e7.el7ost.noarch openstack-ceilometer-notification-9.0.2-0.20171013153832.ea962e7.el7ost.noarch openstack-ceilometer-ipmi-9.0.2-0.20171013153832.ea962e7.el7ost.noarch python-ceilometer-9.0.2-0.20171013153832.ea962e7.el7ost.noarch openstack-ceilometer-central-9.0.2-0.20171013153832.ea962e7.el7ost.noarch puppet-ceilometer-11.3.1-0.20170923033230.767c543.el7ost.noarch openstack-ceilometer-api-9.0.2-0.20171013153832.ea962e7.el7ost.noarch openstack-ceilometer-polling-9.0.2-0.20171013153832.ea962e7.el7ost.noarch How reproducible: 100% Steps to Reproduce: 1. start a vm on Intel baremetal with cpu_l3_cache option, libvirt configured to cmt 2.Add cpu_l3_cache to polling.yaml: docker exec -ti ceilometer_agent_compute cat /etc/ceilometer/polling.yaml | tail -1 - cpu_l3_cache 3. Restart the service: docker exec -ti ceilometer_agent_compute pkill -HUP -f "ceilometer.*master process" 4. Watch measures being collected after 10 min 5. Reboot compute Actual results: Measures are not being collected after reboot Expected results: Measures are supposed to remain being collected Additional info: