no instance data found in start-local

asked 2014-10-07 07:21:34 -0500

PatRock gravatar image

updated 2014-10-09 03:06:25 -0500

My OpenStack worked completely fine for several days, but suddenly I can't create instance completely. I'm watching 3 logs (nova-api.log, keystone.log & neutron.log) at the time i create the instance and nothing appears. I got no clue what's wrong and just the error message from the instance console at the boot.

http://postimg.org/image/wx8wy0hu5/

Something that I noticed is that the token ttl is wrong. when I get a token via keystone token-get it's already expired 1 hour ago. I changed the expire time to 86400 seconds (24hours) so now the token expires in 22hours! so keystone uses a wrong time, but thats not reason for this error isn't it?

After 24min in the keystone.log

WARNING keystone.openstack.common.versionutils [-] Deprecated: keystone.middleware.core.XmlBodyMiddleware is deprecated as of Icehouse in favor of support for "application/json" only and may be removed in K.

New line in nova.log when I start another instance.

 WARNING keystoneclient.middleware.auth_token [-] Authorization failed for token

After one day off i got some messages in the logs when i try to create a instance via nova boot

nova-api.log

WARNING keystoneclient.middleware.auth_token [-] Authorization failed for token

keystone.log

WARNING keystone.common.wsgi [-] Authorization failed. The request you have made requires authentication. from 10.0.9.10

neutron.log

WARNING keystoneclient.middleware.auth_token [-] Authorization failed for token

10.0.9.10 is my controller. So i see that i can't get authenticate, I just don't know why. Could there be something wrong with my openstack.rc file?

think i found something in nova-scheduler.log

WARNING nova.scheduler.host_manager [req-02dd9235-49e0-47d3-96e5-d53f31982d1c a9ed4eb9161244bdb9306f589f301e08 ad948ee0c33749508d76b95f33e7e02d] Host has more disk space than database expected (4gb > -33gb)

my dashboard says under hypervisions that 40GB from 14GB is used. All 40 are on the compute2 node. How can this be possible when the node only has 7GB of storage? Is something wrong with the database?

edit retag flag offensive close merge delete

Comments

Did you change any passwords for any of the service users such as nova,cinder,neutron? Your instance isn't getting networking which means that it isn't getting a NIC attached.

mpetason gravatar imagempetason ( 2014-10-07 13:06:51 -0500 )edit

I have reset my vm-controller out of a snapshot, but i haven't changed any passwords manually. How can i controll that the passwords are still correct? commands like nova keypair-list are working.

PatRock gravatar imagePatRock ( 2014-10-08 02:55:00 -0500 )edit