Ask Your Question
0

No valid host was found

asked 2015-06-01 12:57:00 -0500

donaldmize gravatar image

updated 2015-06-01 14:01:38 -0500

dbaxps gravatar image

I am trying to get a simple cirrus VM up on a kilo implementation on CentOS with two nodes. One controller and one compute. I get a 'No valid host was found' in the nova-conductor.log file.

My nova service list output is:

+----+------------------+-----------+----------+---------+-------+----------------------------+-----------------+
| Id | Binary           | Host      | Zone     | Status  | State | Updated_at                 | Disabled Reason |
+----+------------------+-----------+----------+---------+-------+----------------------------+-----------------+
| 1  | nova-consoleauth | anders-38 | internal | enabled | up    | 2015-06-01T12:52:53.000000 | -               |
| 2  | nova-conductor   | anders-38 | internal | enabled | up    | 2015-06-01T12:52:53.000000 | -               |
| 3  | nova-scheduler   | anders-38 | internal | enabled | up    | 2015-06-01T12:52:53.000000 | -               |
| 4  | nova-cert        | anders-38 | internal | enabled | up    | 2015-06-01T12:52:53.000000 | -               |
| 5  | nova-compute     | anders-21 | nova     | enabled | up    | 2015-06-01T12:52:57.000000 | -               |
| 6  | nova-network     | anders-21 | internal | enabled | up    | 2015-06-01T12:52:57.000000 | -               |
+----+------------------+-----------+----------+---------+-------+----------------------------+-----------------+

[root@anders-21 ~]#

my nova hypervisor-stats output is:
[root@anders-21 ~]# nova hypervisor-stats
+----------------------+-------+
| Property             | Value |
+----------------------+-------+
| count                | 1     |
| current_workload     | 0     |
| disk_available_least | 91    |
| free_disk_gb         | 104   |
| free_ram_mb          | 14789 |
| local_gb             | 105   |
| local_gb_used        | 1     |
| memory_mb            | 15813 |
| memory_mb_used       | 1024  |
| running_vms          | 1     |
| vcpus                | 8     |
| vcpus_used           | 1     |
+----------------------+-------+

Any ideas? Thanks! Don

edit retag flag offensive close merge delete

7 answers

Sort by ยป oldest newest most voted
0

answered 2015-06-19 10:42:29 -0500

Kui.Li gravatar image

Did you resolve the issue? Can you post the error part of nova-scheduler.log, nova-conductor.log and nova-compute.log?

edit flag offensive delete link more
0

answered 2015-06-19 02:46:51 -0500

AntonioA gravatar image

Sometimes you can receive that errors when the resources on your environments are not enough. I had this error and I have updated the ram allocation attributes into nova.conf (controller => ram_allocation_ratio).

edit flag offensive delete link more
0

answered 2015-06-15 17:44:43 -0500

donaldmize gravatar image

updated 2015-06-18 16:54:03 -0500

I had the wrong interface for my public interface in the nova.conf file on the compute node. Originally I had that value set to the interface that connected with the private management network. I corrected that and now my instance comes up fine!

edit flag offensive delete link more
0

answered 2015-06-17 15:08:24 -0500

sgordon gravatar image

updated 2015-06-17 15:08:58 -0500

The NoValidHost error is encountered when the scheduler filters eliminate all available hosts as potential targets for the instance. To confirm which filter(s) are rejecting hosts set debug=True in the /etc/nova/nova.conf on the node(s) running the scheduler and restart the scheduler service. Then re-attempt instance creation and watch the /var/log/nova/scheduler.log file for lines like:

[req-... None] Starting with 3 host(s)
[req-... None] Filter RetryFilter returned 3 host(s)
[req-... None] Filter AvailabilityZoneFilter returned 3 host(s) 
[req-... None] Filter RamFilter returned 2 host(s)
...
[req-... None] Filtered [(localhost.localdomain, localhost.localdomain) ram:3208 disk:7168 io_ops:0 instances:1] _schedule ...
[req-... None] Weighed [WeighedHost [host: (localhost.localdomain, localhost.localdomain) ram:3208 disk:7168 io_ops:0 instances:1, weight: 1.0]] ...

Note that these lines were redacted somewhat for brevity. From this output you can determine which filter(s) eliminated hosts.

edit flag offensive delete link more
0

answered 2015-06-04 07:43:27 -0500

no valid hosts error comes for many reasons, kindlly follow the link and go through the troubleshooting steps.

No Valid Hosts

edit flag offensive delete link more

Comments

I'm not running Neutron, just the legacy Nova Network. My clocks on both machines are sync'd up.

donaldmize gravatar imagedonaldmize ( 2015-06-10 17:23:41 -0500 )edit
0

answered 2015-06-02 07:49:17 -0500

Vinoth gravatar image

This should be the possible issues,

All node should have same " TIME & TIME_ZONE"

Check nova.conf in all nodes for the given IP or host name.

edit flag offensive delete link more
0

answered 2015-06-03 04:20:06 -0500

nethawk gravatar image

There are many reasons which can cause this error. You can check the log file "nova-scheduler.log" to find whick filter prompt the error.

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: 2015-06-01 12:57:00 -0500

Seen: 2,038 times

Last updated: Jun 19 '15