Hi, all
I use openstack to manage vcenter, In my compute node, There encounter an error:
2015-01-06 11:09:34.605 23956 CRITICAL nova [-] NotFound: All clusters specified ['clusterA'] were not found in the vCenter
2015-01-06 11:09:34.605 23956 TRACE nova Traceback (most recent call last):
2015-01-06 11:09:34.605 23956 TRACE nova File "/usr/bin/nova-compute", line 10, in <module>
2015-01-06 11:09:34.605 23956 TRACE nova sys.exit(main())
2015-01-06 11:09:34.605 23956 TRACE nova File "/usr/lib/python2.6/site-packages/nova/cmd/compute.py", line 72, in main
2015-01-06 11:09:34.605 23956 TRACE nova db_allowed=CONF.conductor.use_local)
2015-01-06 11:09:34.605 23956 TRACE nova File "/usr/lib/python2.6/site-packages/nova/service.py", line 274, in create
2015-01-06 11:09:34.605 23956 TRACE nova db_allowed=db_allowed)
2015-01-06 11:09:34.605 23956 TRACE nova File "/usr/lib/python2.6/site-packages/nova/service.py", line 147, in __init__
2015-01-06 11:09:34.605 23956 TRACE nova self.manager = manager_class(host=self.host, *args, **kwargs)
2015-01-06 11:09:34.605 23956 TRACE nova File "/usr/lib/python2.6/site-packages/nova/compute/manager.py", line 597, in __init__
2015-01-06 11:09:34.605 23956 TRACE nova self.driver = driver.load_compute_driver(self.virtapi, compute_driver)
2015-01-06 11:09:34.605 23956 TRACE nova File "/usr/lib/python2.6/site-packages/nova/virt/driver.py", line 1300, in load_compute_driver
2015-01-06 11:09:34.605 23956 TRACE nova virtapi)
2015-01-06 11:09:34.605 23956 TRACE nova File "/usr/lib/python2.6/site-packages/nova/openstack/common/importutils.py", line 50, in import_object_ns
2015-01-06 11:09:34.605 23956 TRACE nova return import_class(import_value)(*args, **kwargs)
2015-01-06 11:09:34.605 23956 TRACE nova File "/usr/lib/python2.6/site-packages/nova/virt/vmwareapi/driver.py", line 400, in __init__
2015-01-06 11:09:34.605 23956 TRACE nova % self._cluster_names)
2015-01-06 11:09:34.605 23956 TRACE nova NotFound: All clusters specified ['clusterA'] were not found in the vCenter
But in my vcenter, the cluster clusterA has added correctly.
Any help?
Thanks for your time.
From the screenshot, "awcloud" seems to be the name of the vSphere datacenter, with a standalone host in it.
You need to create a vSphere cluster, and put that single host into the cluster, and use the name of the cluster in your nova.conf .
Hi, can you provide information about what version of OpenStack is installed?
Did you install VOVA, or install via some other mechanism?
Hi danwendlandt
I'm not use VOVA, I had install OpenStack use RDO, and the version is Icehouse.
Here I have three nodes, controller node run nova, glance and keystone.
network node only install nova-network, and compute node install nova-compute.
I have test in other hypervisor,nova-computer process runs well, create a vm it runs OK.
Ok, posting the nova.conf for the nova-compute node (with password blanked out) and a screen shot of the vCenter inventory would be useful. In older version of OpenStack, there were issues with finding clusters if there were multiple vSphere datastores, but that has bene fixed. Any chance this cluster is in a folder or some sort in you vSphere hierarchy?
OK , I have grep some default conf and comments, nova.conf is like this:
[DEFAULT]
amqp_durable_queues=False
rabbit_host=192.168.1.124
rabbit_port=5672
rabbit_hosts=192.168.1.124:5672
rabbit_use_ssl=False
rabbit_userid=guest
rabbit_password=guest
rabbit_virtual_host=/
rabbit_ha_queues=False
notification_driver=
notification_topics=notifications
rpc_backend=nova.openstack.common.rpc.impl_kombu
notify_api_faults=False
state_path=/var/lib/nova
report_interval=10
service_down_time=60
rootwrap_config=/etc/nova/rootwrap.conf
auth_strategy=keystone
glance_api_servers=192.168.1.124:9292
metadata_host=192.168.1.124
flat_interface=eth1
lock_path=/var/lib/nova/tmp
debug=True
verbose=True
log_dir=/var/log/nova
use_syslog=False
compute_driver=vmwareapi.VMwareVCDriver
novncproxy_base_url=http://192.168.1.124:6080/vnc_auto.html
vncserver_listen=0.0.0.0
vncserver_proxyclient_address=192.168.1.126
vnc_enabled=True
volume_api_class=nova.volume.cinder.API
sql_connection=mysql://nova@192.168.1.124/nova
image_service=nova.image.glance.GlanceImageService
[baremetal]
[cells]
[conductor]
[database]
[hyperv]
[image_file_url]
[keymgr]
[keystone_authtoken]
[libvirt]
virt_type=qemu
inject_partition=-1
live_migration_uri=qemu+tcp://nova@%s/system
cpu_mode=none
[matchmaker_ring]
[metrics]
[osapi_v3]
[rdp]
[spice]
[ssl]
[trusted_computing]
[upgrade_levels]
[vmware]
[xenserver]
[zookeeper]
[VMWARE]
task_poll_interval=5.0
use_linked_clone=True
host_ip=192.168.1.83
host_password=vmware
host_username=root
api_retry_count=5
cluster_name=awcloud
maximum_objects=100
And I will show you vcenter view, cluster name awcloud has replace as clusterA in the previous(for some purpose).
From the screenshot, "awcloud" seems to be the name of the vSphere datacenter, with a standalone host in it.
You need to create a vSphere cluster, and put that single host into the cluster, and use the name of the cluster in your nova.conf .
OK , Thank you for your help, This is my mistake.
I had correct my config, nova-comoute run normally.
Hi :
I met the same pro with you
I use openstack mitaka to manage vcenter, In my compute node, There encounter an error:
ERROR nova NotFound: The specified cluster 'test_vcenter_cluster' was not found in vCenter
But in my vcenter, the cluster test_vcenter_cluster has added correctly.
Any help?