Ask Your Question

Fail to Introspect/Provision nodes - Ironic cleaning_network_UUID

asked 2017-01-05 03:20:22 -0500

pup_seba gravatar image


I'm trying to deploy RedHat OpenStack 10 (Newton) with a Director (TripleO). I was able to deploy the Director but I am unable to provision the nodes. I can see them registered ("ironic node-list", shows both of them). But when I try to provision or introspect them ( openstack baremetal introspection bulk start ), the process fails.

I can see in both my physical nodes that the are able to boot from PXE agent, that they get an IP address from my OpenStack Director node, but then it shows an error saying that the agent failst to start as there is no boot device.

/var/log/messages in the Director machine shows this error which I don't know how to solve.

Jan 5 08:25:01 os-dir01 ironic-conductor: 2017-01-05 08:25:01.503 1264 ERROR ironic.conductor.manager InvalidParameterValue: You must provide a valid cleaning network UUID in [neutron]cleaning_network_uuid configuration option.

Any clues? I'm kind of lost right now and the guide ( ( I'm following for the installations says absolutly nothing about this parameter :(

Any help is more than welcome.

Regards, Seb

edit retag flag offensive close merge delete



What reports neutron net-list on undercloud node ?

dbaxps gravatar imagedbaxps ( 2017-01-05 08:56:29 -0500 )edit

3 answers

Sort by ยป oldest newest most voted

answered 2017-01-05 16:19:38 -0500

rcernin gravatar image

For anyone looking on this issue, you would need configure cleaning as described in

  • (

And don't forget on step 1 configuring the network first for cleaning

  • (
edit flag offensive delete link more

answered 2017-01-05 06:33:17 -0500

pcaruana gravatar image

Hello Seb, Can you show the relevant part of yourconfiguration? as part of the Configure Bare Metal service

" Warning

The cleaning_network_uuid and provisioning_network_uuid parameters are required for the neutron network interface. If they are not set, ironic-conductor fails to start. "


edit flag offensive delete link more

answered 2017-01-09 08:39:40 -0500

pup_seba gravatar image


You are absolutly right. I was following RedHat install guide which said nothing regarding that configuration. What i did, was to set to "false" the "clean_node" value in the undercloud.conf before deployment (i was working with a virtualized tripleO so that rollback was easy). So now, when performing the introspection/provising, I don't need the cleanning network and it's working. Since cleanning the nodes or not is not a concern to me at this point, I'll stick with this workaround although I already added to my bookmarks your references.

Thanks again for your help! Regards, Seb

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2017-01-05 03:20:22 -0500

Seen: 1,182 times

Last updated: Jan 09 '17