Re: [Rdo-list] [RDO][Instack] heat is not able to create stack with instack
On Wed, Jul 23, 2014 at 03:39:38PM +0530, Peeyush Gupta wrote:
Hi James,
I reinstalled the undercloud and retraced all the steps with debug=True
in /etc/nova/nova/conf.
Here are the logs:
Nova-compute logs:
2014-07-23 05:59:32.239 2915 TRACE nova.compute.manager [instance:
ba65d12e-b664-4ad9-93fa-992a2886ab01] InstanceDeployFailure: PXE deploy
failed for instance ba65d12e-b664-4ad9-93fa-992a2886ab01
2014-07-23 05:59:32.239 2915 TRACE nova.compute.manager [instance:
ba65d12e-b664-4ad9-93fa-992a2886ab01]
2014-07-23 06:05:43.572 2915 WARNING nova.compute.manager [-] Found 4 in
the database and 0 on the hypervisor.
The WARNING here looks suspicous. Did you use instack-delete-overcloud to
delete a previous overcloud? Before trying another deployment, can you run
instack-delete-overcloud? That also deletes baremetal nodes, which will be
re-created the next time you run instack-deploy-overcloud*.
Nova-api logs:
'identity_uri' instead.
2014-07-23 05:45:12.903 2947 ERROR nova.wsgi [-] Could not bind to
0.0.0.0:8773
2014-07-23 05:45:12.903 2947 CRITICAL nova [-] error: [Errno 98] Address
already in use
For this one, I'd suggest stopping the openstack-nova-api service and verifying
that there are indeed no nova-api processes running. If there are, kill those
off. Then start openstack-nova-api.
Nova-scheduler logs:
_claim_test\n "; ".join(reasons))\n',
u'ComputeResourcesUnavailable:
Insufficient compute resources: Free memory 0.00 MB < requested 2048
MB.\n']
Deleting the previous baremetal nodes by using instack-delete-overcloud should
hopefully clear this error up as well.
--
-- James Slagle
--