Open Stack

Hi,

I'm trying to setup a little nova cluster with 3 nodes :

- 1 controller node running all services but compute
- 2 compute nodes running compute and network only

I have faced a lot of issues, but I can't understand this one. When I 
start nova-scheduler on controller node, the process dies instantly with 
this error :

CRITICAL nova [-] duplicate option: scheduler_host_manager

So I can't start an instance, it remains stuck in building state. I have 
checked my nova.conf file, and I only see this line :

--scheduler_driver=nova.scheduler.simple.SimpleScheduler

What does means this error, how can I solve it ?

Thanks,

Ben