Ask Your Question
0

failed to launch VM after a clean install

asked 2019-09-02 20:57:12 -0600

Xibuka gravatar image

updated 2019-09-02 23:48:46 -0600

Hi

I'm using an Intel NUC and devstack to create an OpenStack environment.

With a clean Bionic Ubuntu OS, and a very simple local.conf file,

[[local|localrc]]
ADMIN_PASSWORD=ubuntu
DATABASE_PASSWORD=ubuntu
RABBIT_PASSWORD=ubuntu
SERVICE_PASSWORD=$ADMIN_PASSWORD
HOST_IP=20.0.0.19
FIXED_RANGE=10.4.128.0/20
FLOATING_RANGE=20.0.0.128/25
LOGFILE=$DEST/logs/stack.sh.log

I saw the deployment has finished normally, and I can access the dashboard with no issue.

However, when I going to create a test instance, it shows I below error.

Error: Failed to perform requested operation on instance "test", the instance has an error status: Please try again later [Error: Build of instance ce4aed72-821c-4887-9bd9-56cb8db175dc aborted: Invalid input received: Invalid input received: Connector doesn't have required information: initiator (HTTP 400) (Request-ID: req-d6f47d5b-a3c9-43ae-9c7a-46966fafe2cb)].

EDITED I created the instance on web UI, with a tiny flavor and a cirros image, leave other option by default. After set "Create New Volume" to No, the instance can boot up now.

From syslog I can see below errors from cinder-volume. So it looks like a cinder issue

Sep  3 04:11:07 nuc0 cinder-volume[30604]: #033[01;31mERROR cinder.volume.targets.iscsi [#033[01;36mreq-ab85c07b-f0eb-475b-aecf-392160646ef0 req-51584f72-0103-4576-b535-8242a33883a9 #033[00;36mdemo None#033[01;31m] #033[01;35m#033[01;31mThe volume driver requires the iSCSI initiator name in the connector.#033[00m

Sep  3 04:11:07 nuc0 cinder-volume[30604]: #033[01;31mERROR oslo_messaging.rpc.server [#033[01;36mreq-ab85c07b-f0eb-475b-aecf-392160646ef0 req-51584f72-0103-4576-b535-8242a33883a9 #033[00;36mdemo None#033[01;31m] #033[01;35m#033[01;31mException during message handling#033[00m: InvalidInput: Invalid input received: Connector doesn't have required information: initiator

Any suggestions?

edit retag flag offensive close merge delete

Comments

How did you launch the instance?

The reference to "initiator" makes me think that something went wrong with Cinder volumes. While you will want to fix Cinder eventually, you can launch instances without creating volumes.

Check the Nova and Cinder logs for errors related to the request ID.

Bernd Bausch gravatar imageBernd Bausch ( 2019-09-02 22:35:18 -0600 )edit

Hi Bernd, I created the instance on web UI, with a tiny flavor and a cirros image, leave other option by default. As you suggested, After set "Create New Volume" to No, the instance can boot up now. For detail, I updated the description of my question.

Xibuka gravatar imageXibuka ( 2019-09-02 23:41:21 -0600 )edit

1 answer

Sort by » oldest newest most voted
0

answered 2019-09-02 23:57:49 -0600

Xibuka gravatar image

Refer to https://ask.openstack.org/en/question/118921/unable-to-attach-volume-due-to-error-invalidinput-invalid-input-received-connector-doesnt-have-required-information-initiator/ (here), I found a workaround for this, just start the iscsd service.

$ sudo systemctl start iscsid

Not sure why this works, but now my VMs can be created with no issue.

edit flag offensive delete link more

Comments

Volumes are often connected to compute nodes via iSCSI, thus you need the iscsid on your cloud's compute nodes. For some reason, it was not running on your deployment.

Perhaps a race condition, but I am glad you have a solution.

Bernd Bausch gravatar imageBernd Bausch ( 2019-09-03 00:57:41 -0600 )edit

Ye, thanks for the help Bernd :-D

Xibuka gravatar imageXibuka ( 2019-09-03 01:23:49 -0600 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

Stats

Asked: 2019-09-02 20:57:12 -0600

Seen: 24 times

Last updated: Sep 02