Compute VMware nodes does not support any controllers recomposition

[Vyacheslav Tabolin]

Customer can add or remove controllers any time but compute VMware nodes still have initial rabbitMQ host information. Those nodes may lost connection to HA cluster.

[Bug Checker Bot (bug-checker)]

(This check performed automatically)
Please, make sure that bug description contains the following sections filled in with the appropriate data related to the bug you are describing:

actual result

version

expected result

steps to reproduce

For more detailed information on the contents of each of the listed sections see https://wiki.openstack.org/wiki/Fuel/How_to_contribute#Here_is_how_you_file_a_bug

[Igor Gajsin (igajsin)]

There are deployment graph for compute-vmware http://paste.openstack.org/show/496780/ and compute http://paste.openstack.org/show/496781/. I mean that problem is located totally not in vmware_dvs plugin. Mark is as invalid for that project.

[Vyacheslav Tabolin (slavchick)]

Initially the problem located in fuel vmware compute. At least nova.conf and nova-compute daemon must contain correct data about rabbit hosts. Since version 2.1 vmware_dvs plugin has agent connected over amqp. It means that neutron.conf must be updated too and agent restarted.
That's why DVS plugin affected.

[Igor Gajsin (igajsin)]

There is no bug in the core. It provides enough flexibility to specify tasks's behaviour.

[Igor Gajsin (igajsin)]

Can't reproduce on 9.0 #495