DescriptionPablo Iranzo Gómez
2016-09-01 10:45:51 UTC
Description of problem:
After a controller cluster restart (testing the reboot of controllers), many openstack services didn't come up, particularily nova-api.
Pacemaker was disabled to try starting the service manually but it reaches timeout and many services appear as zombie
Version-Release number of selected component (if applicable):
openstack-nova-api-2014.2.3-75.1359197.el7ost.noarch
openstack-nova-cert-2014.2.3-75.1359197.el7ost.noarch
openstack-nova-common-2014.2.3-75.1359197.el7ost.noarch
openstack-nova-conductor-2014.2.3-75.1359197.el7ost.noarch
openstack-nova-console-2014.2.3-75.1359197.el7ost.noarch
openstack-nova-novncproxy-2014.2.3-75.1359197.el7ost.noarch
openstack-nova-scheduler-2014.2.3-75.1359197.el7ost.noarch
python-nova-2014.2.3-75.1359197.el7ost.noarch
python-novaclient-2.20.0-1.el7ost.noarch
Actual results:
nova-api times out and leaves zombie proceses, this makes pacemaker not to start other services and OSP platform not to be fully available
Expected results:
nova-api should start
Additional info: