Ran into some trouble... when installing more compute nodes, I acidentially left the local_ip located in /etc/neutron/plugins/ml2/linuxbridge_agent.ini the same for the additional compute nodes. Now a
openstack hypervisor list
shows:
+----+---------------------+-----------------+--------------+-------+
| ID | Hypervisor Hostname | Hypervisor Type | Host IP | State |
+----+---------------------+-----------------+--------------+-------+
| 1 | ccloud1 | QEMU | 10.131.39.41 | up |
| 2 | ccloud2 | QEMU | 10.131.39.40 | up |
| 3 | ccloud3 | QEMU | 10.131.39.40 | up |
| 5 | ccloud4 | QEMU | 10.131.39.40 | up |
+----+---------------------+-----------------+--------------+-------+
I have taken the following action to try to fix this, but it does not work, using ccloud4 as an example:
On controller:
openstack compute service set --disable ccloud4 nova-compute
On ccloud4:
systemctl stop nova-compute
systemctl stop neutron-linuxbridge-agent
On controller:
openstack compute service list
# above is to get ID, it was 11
openstack compute service delete 11
openstack network agent list
# it still shows up there, lets delete it by UUID
openstack network agent delete 433153bd-896d-45c3-b6ae-1ac5cf9cf6b3
At this point a openstack hypervisor list and openstack network agent list show it is gone, so far so good, lets add it back in:
On ccloud4:
systemctl start nova-compute
systemctl start neutron-linuxbridge-agent
On controller:
su -s /bin/sh -c "nova-manage cell_v2 discover_hosts --verbose" nova
openstack compute service list
openstack network agent list
openstack hypervisor list
Annnnd, it is still there listed at IP address 10.131.39.40. Not sure how to delete and then add back ccloud2,3,4 and have them at a new IP address.
Just to be clear, the actual IP address of ccloud2,3,4 were always correct (10.131.39.42,43,44, respectively), it was only in the /etc/neutron/plugins/ml2/linuxbridge_agent.ini file that it was in error.
I have also performed these steps with no luck.
Any pointers would be appreciated.
