instance stuck in build, AMQPLAIN login refused: user 'openstack' - invalid credentials

asked 2018-05-31 12:52:53 -0600

mike99201 gravatar image

CentOS 7 with one controller and one compute node. Openstack Queens build. Launching a new instance is stuck in build status.

The nova-conductor error log reports:

2018-05-31 13:21:06.548 6059 ERROR oslo_messaging.rpc.server MessageDeliveryFailure: Unable to connect to AMQP server on ops1:5672 after None tries: (0, 0): (403) ACCESS_REFUSED - Login was refused using authentication mechanism AMQPLAIN. For details see the broker logfile.

Strange it says "after None tries", as I can see it trying in the rabbitmq log:

=ERROR REPORT==== 31-May-2018::13:21:06 === Error on AMQP connection <0.2581.0> ([2620:111:A001:201:10:168:0:141]:49302 -> [2620:111:A001:201:10:168:0:141]:5672, state: starting): AMQPLAIN login refused: user 'openstack' - invalid credentials

I've double, triple, quadruple checked the transport_url rabbitmq password in the nova.conf as well as the neutron.conf on both the controller and the compute node, it is absolutely 100% correct. No special characters or anything. I've made sure the rabbitmq user 'openstack' has set_privileges correctly, and I even tried adding 'openstack' to the admin group in rabbitmq, no dice. All of my compute services start just fine, and report they are up, which just confirms to me that my transport_url in nova.conf is correct:

# openstack compute service list
| ID | Binary           | Host        | Zone     | Status  | State | Updated At                 |
|  1 | nova-consoleauth | ops1.scilan | internal | enabled | up    | 2018-05-31T17:34:03.000000 |
|  2 | nova-conductor   | ops1.scilan | internal | enabled | up    | 2018-05-31T17:34:00.000000 |
|  3 | nova-scheduler   | ops1.scilan | internal | enabled | up    | 2018-05-31T17:34:04.000000 |
|  7 | nova-compute     | ops3.scilan | nova     | enabled | up    | 2018-05-31T17:34:02.000000 |
|  8 | nova-console     | ops1.scilan | internal | enabled | up    | 2018-05-31T17:34:00.000000 |

This access refused / invalid credentials message only occurs right when trying to launch an instance, no other time. I've been at this for days and I'm out of ideas, can anyone point me in the right direction?

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2018-06-05 07:26:57 -0600

mike99201 gravatar image

Either the password or the transport_url are stored in the nova_api sql database. Update or recreate the database. I don't know why the database wouldn't update itself from the config file to deal with password changes.

edit flag offensive delete link more


The method of checking this, and fixing the matter have later been documented in the answer in ( Adding this comment because I found this post before I found the mentioned solution.

LarsON gravatar imageLarsON ( 2019-09-23 14:13:02 -0600 )edit

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2018-05-31 12:52:53 -0600

Seen: 2,069 times

Last updated: Jun 05 '18