Ask Your Question
1

nova-compute does not start

asked 2014-11-11 16:49:30 -0600

dgalvao gravatar image

Hello there,

I'm installing a new openstack environment on Centos 7, following the instructions from this guide: http://docs.openstack.org/juno/install-guide/install/yum/content/ (http://docs.openstack.org/juno/instal...)

I installed the controller node services (mariadb, rabbitmq, keystone, glance and nova, except nova-compute) on a VM and had no issues. Everything is working as described by the install guide.

But I'm having problems with the nova-compute service starting on the compute node, which runs on a real server also with Centos 7. I followed the instructions in the guide (here: http://docs.openstack.org/juno/install-guide/install/yum/content/ch_nova.html#nova-compute-install (http://docs.openstack.org/juno/instal...) ) but the service fails to start. I enabled debug mode but I don't see any obvious errors in the log either.

I've tried running manually, and this is what I get:

[root@cha-bl465-06:(cha-bl465-06) ~]# /usr/bin/nova-compute --debug --config-file /etc/nova/nova.conf --log-file /var/log/nova/nova-compute.log
2014-11-11 16:17:24.882 4224 DEBUG nova.servicegroup.api [-] ServiceGroup driver defined as an instance of db __new__ /usr/lib/python2.7/site-packages/nova/servicegroup/api.py:65
2014-11-11 16:17:25.056 4224 INFO nova.virt.driver [-] Loading compute driver 'libvirt.LibvirtDriver'
2014-11-11 16:17:25.062 4224 INFO nova.openstack.common.periodic_task [-] Skipping periodic task _periodic_update_dns because its interval is negative
2014-11-11 16:17:25.109 4224 INFO oslo.messaging._drivers.impl_rabbit [req-ae205550-f345-4951-aa05-f3425333336b ] Connecting to AMQP server on cha-ops-xen-v04:5672
2014-11-11 16:17:25.132 4224 INFO oslo.messaging._drivers.impl_rabbit [req-ae205550-f345-4951-aa05-f3425333336b ] Connected to AMQP server on cha-ops-xen-v04:5672
2014-11-11 16:17:25.140 4224 INFO oslo.messaging._drivers.impl_rabbit [req-ae205550-f345-4951-aa05-f3425333336b ] Connecting to AMQP server on cha-ops-xen-v04:5672
2014-11-11 16:17:25.165 4224 INFO oslo.messaging._drivers.impl_rabbit [req-ae205550-f345-4951-aa05-f3425333336b ] Connected to AMQP server on cha-ops-xen-v04:5672

It seems to connect fine to the rabbitmq service in the controller but then it just hangs there...

Any help will be much appreciated.

Daniel

edit retag flag offensive close merge delete

Comments

What do you mean, "it just hangs there"? What do you expect it to do? Until an instance is scheduled on that host, nothing else is going to happen. That looks like a successful start.

larsks gravatar imagelarsks ( 2014-11-11 20:48:29 -0600 )edit

I expected to see it listed when I run nova service-list in the controller, but I only see the 4 services running in the controller (nova-consoleauth, nova-cert, nova-scheduler, nova-conductor)

dgalvao gravatar imagedgalvao ( 2014-11-12 08:40:33 -0600 )edit

4 answers

Sort by ยป oldest newest most voted
2

answered 2014-11-12 14:55:49 -0600

dgalvao gravatar image

Issue solved. This was caused because the /var filesystem in the controller node dropped below 1GB and rabbitmq blocked publishers:

from the rabbitmq logs:

=INFO REPORT==== 12-Nov-2014::15:14:32 ===
Disk free space insufficient. Free bytes:696029184 Limit:1000000000

=WARNING REPORT==== 12-Nov-2014::15:14:32 ===
disk resource limit alarm set on node 'rabbit@cha-ops-xen-v04'.

**********************************************************
*** Publishers will be blocked until this alarm clears ***
**********************************************************

After I increased space in that partition and re-started services I now see the nova-compute service listed in the controller.

edit flag offensive delete link more

Comments

i got the same problem ,the rabbitmq logs is /var/log/rabbitmq/rabbit@localhost.log?

Fei gravatar imageFei ( 2014-11-18 03:00:03 -0600 )edit
0

answered 2016-03-28 05:03:03 -0600

enis454545 gravatar image

I have just solve that problem(devstack) .

In my environment the problem related with pip. If you ask me why i don't have any idea why it's happened. I don't know

While installing devstack it's download pip for us from repositories of CentOS or Ubuntu for python-pip package. After 15 day I realised that pip version couldn't download xmlrpc_client and few packages from pip repository

To solve this problem download
1. delete your pip if you download how ever you download

apt-get remove python-pip #or yum version of that code

2. List item

wget https://bootstrap.pypa.io/get-pip.py 
python get-pip.py

3. uninstall your six package which you download before

pip uninstall six 
easy_install six

than ./unstack.sh and ./stack.sh

and Don't forget!! to watch you RAM by using free command while installing, because there are few problem about RAM it's hard to understand if problem occurs about RAM. I needed 6GB RAM for all install

edit flag offensive delete link more
0

answered 2015-03-30 18:51:19 -0600

Noreen gravatar image

did any body solve it?

Can you give the solution of above problem?

Thanks,

edit flag offensive delete link more

Comments

try this :

 systemctl restart openstack-nova-compute.service
 systemctl enable openstack-nova-compute.service

then look nova's log .

silently gravatar imagesilently ( 2016-11-17 01:17:52 -0600 )edit
0

answered 2014-11-11 21:21:24 -0600

Cool.. If I understood you correctly you are unable to start the service using service openstack-nova-compute start right? But manually you are able to connect to AMQP on RabbitMQ.

That is fine. Run manually and try to boot an instance. See if everything goes well or not? And also let us know what is the error you are getting while executing service openstack-nova-compute restart command.

edit flag offensive delete link more

Comments

With systemctl I get:

[root@cha-bl465-06:(cha-bl465-06) ~]# systemctl start openstack-nova-compute.service Job for openstack-nova-compute.service failed. See 'systemctl status openstack-nova-compute.service' and 'journalctl -xn' for details. [root@cha-bl465-06:(cha-bl465-06) ~]#

dgalvao gravatar imagedgalvao ( 2014-11-12 08:44:39 -0600 )edit

And when I run it manually, it connects to rabbit-mq, but I expected it to show when I do nova service-list, as described in the documentation.

dgalvao gravatar imagedgalvao ( 2014-11-12 08:46:31 -0600 )edit

Even i'm facing the same error. It does not appear on the controller node. But when i check the status of compute-nova status on the compute node, it says it is running. How do we solve that?

Pauravi gravatar imagePauravi ( 2015-03-11 21:07:49 -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: 2014-11-11 15:23:58 -0600

Seen: 8,240 times

Last updated: Mar 30 '15