Ask Your Question
0

Nova "Conflicting resource provider name" [closed]

asked 2019-05-30 21:14:15 -0500

utdream gravatar image

updated 2019-05-31 13:26:26 -0500

Hi folks,

I've got a lab cluster set up that I'm testing OpenStack on and I'm running into the following error:

2019-05-30 18:53:01.203 23258 ERROR nova.scheduler.client.report [req-82e72f3e-4896-435b-9388-70c22247ca83 - - - - -] [req-69c5c0a6-d8bd-4f8c-bf6d-cfa18e13e487] Failed to create resource provider record in placement API for UUID 124e7c9f-7edb-4574-9675-48f0d686e968. Got 409: {"errors": [{"status": 409, "request_id": "req-69c5c0a6-d8bd-4f8c-bf6d-cfa18e13e487", "detail": "There was a conflict when trying to complete your request.\n\n Conflicting resource provider name: anchovy.lab.local already exists.  ", "title": "Conflict"}]}.

... but I'm not sure how to address this. I've tried removing the server using "openstack compute service delete" but it didn't appear to have any affect.

Found this thread: https://ask.openstack.org/en/question/97941/how-to-remove-compute-host-in-controller-unfortunately-i-removed-that-host-but-name-still-available/ (https://ask.openstack.org/en/question...)

... but after re-adding the instance again I ran into the same issue. I'm at a loss.

Any direction on this would be greatly appreciated.

Thank you in advance!

THE SOLUTION ==============================================================

I had to install the EPEL repo to install pip which would then allow me to install 'osc-placement' which then provided the placement resource provider commands to use with the 'openstack' command.

Fix based on Bernd's excellent advice below:

# yum -y install epel-release
# yum -y install python-pip
# pip install osc-placement
# . admin-openrc
# openstack resource provider list
+--------------------------------------+-----------------------+------------+
| uuid                                 | name                  | generation |
+--------------------------------------+-----------------------+------------+
| 1824de38-b376-4d05-8c42-3c5a65adb6cc | anchovy.lab.local     |         11 |
+--------------------------------------+-----------------------+------------+
# openstack resource provider delete 1824de38-b376-4d05-8c42-3c5a65adb6cc
# ssh 192.168.253.40
# systemctl restart libvirtd.service openstack-nova-compute.service

... and things seem to be working just fine now. Thank you Bernd!

edit retag flag offensive reopen merge delete

Closed for the following reason the question is answered, right answer was accepted by utdream
close date 2019-06-06 13:06:35.630680

1 answer

Sort by ยป oldest newest most voted
1

answered 2019-05-31 00:22:49 -0500

updated 2019-05-31 00:24:33 -0500

It means that compute node anchovy tries to register with the Placement service, but the database already contains a resource provider named anchovy.

Ensure osc-placement is installed, then remove the old anchovy record with openstack resource provider delete. After this, restart nova-compute on anchovy.

I really like that hostname.

edit flag offensive delete link more

Comments

haha. They're little 1U's so the name is appropriate. ;) I have not reviewed the placement stuff yet and admittedly I'm still an openstack newb. Thank you for your help! Will try this and report back.

utdream gravatar imageutdream ( 2019-05-31 12:50:06 -0500 )edit

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

Stats

Asked: 2019-05-30 21:14:15 -0500

Seen: 76 times

Last updated: May 31