11:21 a.m.
Hi,
The TL;DR version is: sending SIGHUP to the neutron server in the
neutron_api docker causes it to hang. Sometimes. [1]
I would really appreciate if someone could verify that what I'm doing is
not something pathological: I'm noticing that the neutron api server dies
every now and then on a rdo tripleo stack for stable/queens, and I _think_
it's because logrotate triggers SIGHUP on the process.
I'm not quite sure if this is neutron, or some misconfiguration in the
services or black magic voodoo with events.
My deployment looks something like this:
openstack overcloud deploy --templates \
-r ~/templates/roles_data.yaml \
-n ~/templates/network_data.yaml \
-e
/usr/share/openstack-tripleo-heat-templates/environments/network-isolation.yaml
\
-e /usr/share/openstack-tripleo-heat-templates/environments/docker.yaml \
-e
/usr/share/openstack-tripleo-heat-templates/environments/enable-swap-partition.yaml
\
-e
/usr/share/openstack-tripleo-heat-templates/environments/enable-swap.yaml \
-e
/usr/share/openstack-tripleo-heat-templates/environments/storage-environment.yaml
\
-e
/usr/share/openstack-tripleo-heat-templates/environments/ceph-ansible/ceph-ansible.yaml
\
-e
/usr/share/openstack-tripleo-heat-templates/environments/ceph-ansible/ceph-rgw.yaml
\
-e
/usr/share/openstack-tripleo-heat-templates/environments/ceph-ansible/ceph-mds.yaml
\
-e
/usr/share/openstack-tripleo-heat-templates/environments/services-docker/cinder-backup.yaml
\
-e
/usr/share/openstack-tripleo-heat-templates/environments/services-docker/ironic.yaml
\
-e /usr/share/openstack-tripleo-heat-templates/environments/docker-ha.yaml \
-e ~/templates/environment.yaml \
-e ~/templates/static-ips-for-external.yaml \
-e ~/templates/ceph-config.yaml \
-e ~/templates/ironic-config.yaml \
-e ~/containers/docker_registry.yaml \
-e ~/templates/role_mappings.yaml \
--libvirt-type kvm \
--ntp-server ntp.is.co.za
Absolutely any feedback would be welcomed!
Jan
[1] https://bugs.launchpad.net/neutron/+bug/1780139