Ask Your Question
0

Can only create instances with 0 as disk size

asked 2018-10-02 09:35:35 -0500

renanb02 gravatar image

Hi, I'm facing some issues when creating instances in my Queens installation. I can't create any unless they have 0 as root disk size (no matter which image/RAM/vcpus I use). It started to happen suddenly, without any apparent reason. Every time I start an instance (using the cli or horizon), I get the following error in my nova-conductor.log:

2018-10-01 14:39:34.933 32764 WARNING nova.scheduler.utils [req-0b7021d4-467d-4941-b8a6-fa3a54a27a10 87d4ed2e896a46ae9fd109a0708d3997 de3d53be33b3467d83226a80a012876f - default default] Failed to compute_task_build_instances: No valid host was found. 
Traceback (most recent call last):

  File "/usr/lib/python2.7/site-packages/oslo_messaging/rpc/server.py", line 226, in inner
    return func(*args, **kwargs)

  File "/usr/lib/python2.7/site-packages/nova/scheduler/manager.py", line 139, in select_destinations
    raise exception.NoValidHost(reason="")

NoValidHost: No valid host was found. 
: NoValidHost_Remote: No valid host was found.

I have enough resources available: https://www.dropbox.com/s/mv2ywi40uh45vlj/Screenshot%20from%202018-10-02%2011-26-13.png (https://www.dropbox.com/s/mv2ywi40uh4...) (I cannot upload files yet, so I'm posting this URL).

Everything seems to be okay, all nova and neutron services both in the controller and the compute are up and not logging anything unusual.

Does anyone know what might be the reason for this error?

Thank you.

edit retag flag offensive close merge delete

Comments

1

Check the scheduler log; it should tell you why it can't find a host (you may have to configure debug logging). Also check the Placement service - this is where capacities are recorded. Placement now has a CLI.

Bernd Bausch gravatar imageBernd Bausch ( 2018-10-02 10:23:16 -0500 )edit

1 answer

Sort by ยป oldest newest most voted
0

answered 2019-08-27 13:42:04 -0500

renanb02 gravatar image

Erasing all nova databases and syncing all data again solved the problem. Not sure if this was the best way to solve, but things got back to normal after that.

edit flag offensive delete link more

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: 2018-10-02 09:35:35 -0500

Seen: 64 times

Last updated: Aug 27 '19