Ask Your Question
0

Nova-Comput not connect to Nova Controller !?! [closed]

asked 2013-12-13 02:43:21 -0600

rahuk gravatar image

Hey Guys,

we have issues with the Nova-Compute Nodes in Havana.

The Services is correct installed. with Nova service-list we get :

+------------------+------------------+----------+---------+-------+----------------------------+-----------------+
| Binary           | Host             | Zone     | Status  | State | Updated_at                 | Disabled Reason |
+------------------+------------------+----------+---------+-------+----------------------------+-----------------+
| nova-cert        | openstackc-core1 | internal | enabled | down  | 2013-12-12T15:55:49.000000 | None            |
| nova-conductor   | openstackc-core1 | internal | enabled | up    | 2013-12-13T08:37:51.000000 | None            |
| nova-consoleauth | openstackc-core1 | internal | enabled | down  | 2013-12-12T15:55:50.000000 | None            |
| nova-scheduler   | openstackc-core1 | internal | enabled | up    | 2013-12-13T08:37:51.000000 | None            |
+------------------+------------------+----------+---------+-------+----------------------------+--------

there is no nova-compute service....

The Log of Nova-Compute:

   2013-12-13 09:29:39.109 29437 ERROR nova.servicegroup.drivers.db [-] Recovered model server connection!
2013-12-13 09:29:45.069 29437 ERROR nova.openstack.common.periodic_task [-] Error during ComputeManager.update_available_resource: Compute host 2 could not be found.
Traceback (most recent call last):
  File "/usr/lib/python2.7/dist-packages/nova/openstack/common/rpc/amqp.py", line 461, in _process_data
    **args)

  File "/usr/lib/python2.7/dist-packages/nova/openstack/common/rpc/dispatcher.py", line 172, in dispatch
    result = getattr(proxyobj, method)(ctxt, **kwargs)

  File "/usr/lib/python2.7/dist-packages/nova/conductor/manager.py", line 450, in compute_node_update
    prune_stats)

  File "/usr/lib/python2.7/dist-packages/nova/db/api.py", line 216, in compute_node_update
    return IMPL.compute_node_update(context, compute_id, values, prune_stats)

  File "/usr/lib/python2.7/dist-packages/nova/db/sqlalchemy/api.py", line 108, in wrapper
    return f(*args, **kwargs)

  File "/usr/lib/python2.7/dist-packages/nova/db/sqlalchemy/api.py", line 668, in compute_node_update
    compute_ref = _compute_node_get(context, compute_id, session=session)

  File "/usr/lib/python2.7/dist-packages/nova/db/sqlalchemy/api.py", line 502, in _compute_node_get
    raise exception.ComputeHostNotFound(host=compute_id)

ComputeHostNotFound: Compute host 2 could not be found.
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task Traceback (most recent call last):
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task   File "/usr/lib/python2.7/dist-packages/nova/openstack/common/periodic_task.py", line 180, in run_periodic_tasks
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task     task(self, context)
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task   File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 4863, in update_available_resource
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task     rt.update_available_resource(context)
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task   File "/usr/lib/python2.7/dist-packages/nova/openstack/common/lockutils.py", line 246, in inner
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task     return f(*args, **kwargs)
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task   File "/usr/lib/python2.7/dist-packages/nova/compute/resource_tracker.py", line 325, in update_available_resource
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task     self._sync_compute_node(context, resources)
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task   File "/usr/lib/python2.7/dist-packages/nova/compute/resource_tracker.py", line 356, in _sync_compute_node
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task     self._update(context, resources, prune_stats=True)
2013-12-13 09:29:45.069 29437 TRACE nova.openstack.common.periodic_task   File "/usr/lib/python2.7/dist-packages/nova/compute ...
(more)
edit retag flag offensive reopen merge delete

Closed for the following reason the question is answered, right answer was accepted by dheeru
close date 2013-12-13 10:38:54.192879

add a comment

3 answers

Sort by ยป oldest newest most voted
1

answered 2013-12-13 06:27:53 -0600

rahuk gravatar image

We have found the problem, the old controller communicate with rabbitmq and the new controller do the same. Compute-node 1&2 get the information from rabbit for the old controller. After we Stop all services on the Old controller an restarting the service nova-compute it works !

edit flag offensive delete link more

Comments

Good that you found the answer.

Anand TS gravatar imageAnand TS ( 2013-12-13 07:07:06 -0600 )edit

You had controller running at the same time ?

dheeru gravatar imagedheeru ( 2013-12-13 10:41:00 -0600 )edit

Hey there, yes the old controller and the new are running at the same time, and communicate with rabbit. That was the error. The Compute has search the old controller because he get the wrong information from rabbitmq. Stop all services on the old controller has solved the problem.

rahuk gravatar imagerahuk ( 2013-12-14 05:49:39 -0600 )edit

Hi, how did you correct this because i have the same error ?

fahmi gravatar imagefahmi ( 2014-05-18 15:21:49 -0600 )edit
add a comment
2

answered 2013-12-13 04:26:15 -0600

Anand TS gravatar image

updated 2013-12-13 04:27:56 -0600

Hi,

From the logs it shows compute host 2 could not be found. It means nova-compute service is not reachable or down. You need to restart nova-compute service in each node.

You can also make other service up by debug command.

#cd /usr/bin
#./nova-cert

Regards, Anand

edit flag offensive delete link more
add a comment
0

answered 2013-12-13 05:03:16 -0600

rahuk gravatar image

Hey and thanks for the answer,

nova-compute service is started on both nodes, logs are the same. Only TRACE nova.servicegroup.drivers.db ServiceNotFound: Service 6 could not be found. and the Information that node1 is not found.

We have make a new controller with a different ip-adress. I thing the node is search the old controller. I have purge the nova-compute installation on both nodes and then install it again with new configs with the correct new ip.

Is the information for the compute/controller store in rabbitmq ? Or can I completely remove the old installation/Informations?

edit flag offensive delete link more
add a comment

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

subscribe to rss feed

Stats

Asked: 2013-12-13 02:43:21 -0600

Seen: 7,890 times

Last updated: Dec 13 '13

Related questions

ironic with keystone v3

Connectivity breaks between compute and controller [closed]

Using blade servers with attached storage for compute nodes

During the deployment the libvirt close connation with nova-compute. libvirtError: End of file while reading data: Input/output error\n' [closed]

Found 0 unmapped computes in cell: 88fdcc1f-ddea-4505-a6b1-394c10708dee

nova-network state XXX

How are actions on VMs handled?

can we know where a particular VM is running (on which compute node)?

nova-compute is down in nova service-list

instance resizing not working with xenserver 6.2