We need to provide templates to fully test production architecture of single node deployment (all-in-one), with director being a separate node.
This architecture should replace our current CI scenario of 1 controller + 1 compute.
We may be able to reuse the role we generated for the standalone installer (no undercloud), but need to confirm that the expected service list matches what is desired for this deployment.
We need a complete list of services that we would expect that could possibly be deployed in this architecture (1undercloud+1node). Additionally do we need Ceph support?
The OpenStack services that will be enabled by default are:
* Nova (and related)
* Neutron (and related)
The default configurations are assumed for each of those unless the user provides additional environment files.
I suppose we have to supply hardware requirements for such all-in-one machine in documentation (also need that info for automation infrastructure to define proper VM)
Additionally - what about ceph support? Do we plan to support it in any configuration?
So this configuration is 1 undercloud + 1 all-in-one node. Ceph should work if properly configured though this configuration is not HA in any way so it's up to the user to understand this. The supportability of ceph in this architecture will need to come from the storage folks. The framework should support deploying it. Ideally the minimum hardware for a CI test would be the standard undercloud config + 1 overcloud node with 16GB ram + 4 vcpu. The actual requirements for the node would depend on the users expected load.
Tested and automated
Polarion Test plan:
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.