Document ClusteredComputeManager

Document the option needed to enable the ClusteredComputeManager in
nova.conf.

The commit a719472d added a custom compute manager to Ironic which
prevents some startups problem (e.g do not cause n-cpu to die at startup
if Ironic is not running yet) and workaround some frictions that we have
with Nova which assumes that each n-cpu is the only one responsible for
an instance (which is not true in the Ironic model).

Change-Id: I26bf5a1d3b634788044991e4524f64acf23bed9a
This commit is contained in:
Lucas Alvares Gomes 2014-04-22 11:50:21 +01:00
parent 19019e1953
commit 6f7c487527

@ -283,6 +283,10 @@ Compute Service's controller nodes and compute nodes.*
#reserved_host_disk_mb=0
reserved_host_memory_mb=0
# Full class name for the Manager for compute (string value)
#compute_manager=nova.compute.manager.ComputeManager
compute_manager=ironic.nova.compute.manager.ClusteredComputeManager
2. Change these configuration options in the ``ironic`` section.
Replace: